Skip to main content

Full text of "Proceedings of the Annual Convention"

See other formats


This is a digital copy of a book that was preserved for generations on library shelves before it was carefully scanned by Google as part of a project 
to make the world's books discoverable online. 

It has survived long enough for the copyright to expire and the book to enter the public domain. A public domain book is one that was never subject 
to copyright or whose legal copyright term has expired. Whether a book is in the public domain may vary country to country. Public domain books 
are our gateways to the past, representing a wealth of history, culture and knowledge that's often difficult to discover. 

Marks, notations and other marginalia present in the original volume will appear in this file - a reminder of this book's long journey from the 
publisher to a library and finally to you. 

Usage guidelines 

Google is proud to partner with libraries to digitize public domain materials and make them widely accessible. Public domain books belong to the 
public and we are merely their custodians. Nevertheless, this work is expensive, so in order to keep providing this resource, we have taken steps to 
prevent abuse by commercial parties, including placing technical restrictions on automated querying. 

We also ask that you: 

+ Make non-commercial use of the files We designed Google Book Search for use by individuals, and we request that you use these files for 
personal, non-commercial purposes. 

+ Refrain from automated querying Do not send automated queries of any sort to Google's system: If you are conducting research on machine 
translation, optical character recognition or other areas where access to a large amount of text is helpful, please contact us. We encourage the 
use of public domain materials for these purposes and may be able to help. 

+ Maintain attribution The Google "watermark" you see on each file is essential for informing people about this project and helping them find 
additional materials through Google Book Search. Please do not remove it. 

+ Keep it legal Whatever your use, remember that you are responsible for ensuring that what you are doing is legal. Do not assume that just 
because we believe a book is in the public domain for users in the United States, that the work is also in the public domain for users in other 
countries. Whether a book is still in copyright varies from country to country, and we can't offer guidance on whether any specific use of 
any specific book is allowed. Please do not assume that a book's appearance in Google Book Search means it can be used in any manner 
anywhere in the world. Copyright infringement liability can be quite severe. 

About Google Book Search 

Google's mission is to organize the world's information and to make it universally accessible and useful. Google Book Search helps readers 
discover the world's books while helping authors and publishers reach new audiences. You can search through the full text of this book on the web 



at |http : //books . google . com/ 



IJ 



M 



w 
^P^ 






^, 






f^^. 



THE GIFT OF MR. ALFRED GWYNNE VANDERBILT 



UBRARY 

RAILROAD BRANCH 

YOUNG MEN'S CHRISTIAN ASSOCIATION 

30» PARK AVENUE 
NEW YORK 









L 



y. 



-'TV rt 






- •> ^ / 



^j^ m PROPERTY OP jm 




ARTES SCIENTIA VERITAS 



Digitized by 



Google 



Digitized by 



Google 



r 



; 



NATIONAL ASSQCMriON 



• r •■ b 



OF 



CEMENT UiitFt§:Tr.:: 



PROCEEDINGS 



OF THE 



Eighth Annual Convention 

Held at Kansas City, Mo., 
March 11, 12, 13, 14, 15, 16, 1912 



Volume VIII 



EDITED UNDER DIRECTION OF THE PRESIDENT 

BY THE SECRETARY 



PUBLISHED BY THE ASSOCIATION 
1912 



Digitized by 



Google 



• • ••• • ••••••• • 

•• :•: •• •!•• • • • • : •••: 



• ••;• • ••• •• • 

.••••••• •••••• 

• • • • • • ••« • • • •«« 

! . • •• •' 

• • • ••••••* 



fraf^ station 



s y^ 






Comaxan 1915 bt thb Ambbioak Concbbtb I mwitutb 



The ABBOoiBtioii is not reaponaible, as a body, for the statements and opinions 
advanced in iU publications. 



Digitized by 



Google 




CONTENTS. 



PAOfc 

Personnel of Officers 6 

Personnel of Sectional Committees 7 

Personnel of Past Officers 9 

Charter 10 

By-Laws 11 

Summary of Proceedings, Eighth Annual Convention 16 

Annual Address by the President, European Practice in Concrete Con- 
struction — Richard L. Humphrey 31 

Report of the Committee on Reinforced Concrete and Building Laws — 

A. E. Lindau, Chairman 61 

Disciission 158 

The Testing of Reinforced Concrete Buildings Under Load — W. A. 

Slater 168 

The Design of Concrete Flat Slabs— F. J. Trelease 218 

Discussion 251 

The Practical Design of Reinforced Concrete Flat Slabs — Sanford E. 

Thompson 254 

Discussion 274 

The Design of Concrete Grain Elevators — E. Lee Heidenreich 277 

Discussion 288 

Report of the Committee on Measuring Concrete — ^Robert A. Cummings, 

Chairman 290 

Proposed Standard Methods for the Measurement of Concrete Work 301 

Concrete Retaining Walls — John M. Meade 308 

Discussion 311 

Reinforced Concrete Piles — Robert A. Cummin^ 312 

The Handling of Concrete in the Construction of the Panama Canal — S. B. 

Williamson 326 

Use of Concrete in the Fourth Avenue Subway, Brooklyn, N. Y. — Fred- 
erick C. Noble 361 

The Use of Reinforced Concrete in Hypochlorite Water Purification 

Works— Walter M. Cross 372 

Design and Construction of the Estacada Dam — H. V. Schreiber 376 

Unit Cost of Reinforced Concrete for Industrial Buildings — C. S. Allen . . 400 
Reinforced Concrete Convention Hall at Breslau, Germany — S. J. Trauer. 406 
The Suitability of Concrete for Gas Holder Tanks— Herbert W. Ab-ich . . 412 
Protection of Steel in Catskill Aqueduct Pipe Siphons — Alfred D. Flinn. . 424 

A Fireproof School of Concrete — Theodore H. Skinner 444 

Tlie Plresent Status of Unit Concrete Construction — ^James L. Darnell. . . 455 
Discussion 469 

(3) 



24537 Digitized by Google 



Contents. 



PADS 

Report of Committee on Specifications and Methods of Tests for Concrete 

Materials — Sanford E. Thompson, Chairman 473 

Aggregates for Concrete — William M. Kinney 486 

Discussion 497 

Field Inspection and Testing of Concrete — G. H. Bayles 501 

Comparative Tests of the Strength of Concrete in the Laboratory and in 

the Field— Rudolph J. Wig 522 

Discussion 526 

The Necessity for Field Tests of Concrete — Fritz X'on Emperger 530 

Discussion 537 

Report of the C'ommittee on Treatment of Ccmcrete Surfaces — L. C. 

Wason, Chairman 539 

Discussion 547 

Cement Coatings — F. J. Morse 552 

Discussion 561 

Review of the Present Status of Iron Portland Cement — P. H. Hates. . . . 566 

Marine or Iron Ore (Vments — Herman E. lirovvn 578 

Iron Ore Cement — Arthur E. Williams 597 

Discussion on Iron Ore Cement 613 

Flat Slab Concrete Bridges— William H. Finley 616 

Concrete Highway Bridges — Walter Scott Gearhart 621 

Concrete Bridges — Daniel B. Luten 631 

Discussion on Concrete Bridges 641 

Report of Committee on Roadways, Sidewalks and Floors — C. W. Boyn- 

ton, Chairman 644 

Standard Specifications for Portland Cement Sidewalks 645 

Standard Specifications for Concrete Roads and Street Pavements 651 

Standard Specifications for Concrete Curb and C^oncret(> Curb and (Gutter 658 

Standard Specifications for Plain Concrete Floors 665 

Standard Specifications for Reinforced Concrete Floors 671 

Discussion on Concrete Floors 676 

An Improved Concrete Pavement — E. W. Groves 683 

Cement Paving as Constructed at Mason City, Iowa — F. P. Wilson. . . 689 

Discussion on Concrete Roads 694 

Report of the Committee on Building Blocks and Cement Products — 

P. H. Hudson, Chairman 699 

Recommended Practice for Plain Concrete Drain Tile 700 

Recommended Practice for Concrete Architectural Stone, Building 

Block and Brick 703 

Standard Specifications for Concrete Architectural Stone, Building Block 

and Brick 707 

Standard Building Regulations for the Use of Concrete Architectural 

Stone, Building Block and Brick 710 

Method of- Testing Cement Pipe — Arthur N. Talbot and Duflf A. Abrams 713 
Advantages and Durability of Cement Sewer Pipe — Gustave Kaufman. . 720 
The Manufacture and Use of Cement Drain Tile — C^harles E. Sims 727 

Discussion 732 



Digitized by 



Google 



Contents. 



PAGE 

Modern Methods of Manufacturing Concrete Products — Robert F. 

Havlik 740 

Notes on Reinforced Concrete Telegraph Poles — George Gibbs 757 

Concrete Fence Posts — W. J. Towne 765 

Concrete Fence Posts — L. J. Hotchkiss 766 

Discussion 774 

Report of Executive Board 779 

Minutes of Meetings of the Executive Board 786 

Register of Attendance — ^Eighth Convention 7^2 

Subject Index 796 

Author Index 804 

List of Publications 809 

Table of Contents 809 

Price List 818 

PLATES 

The Handling of Concrotc in the Construction of the Panama Canal- 
Williamson. 

I. Fig. 1. — Plan and Cross-section Gatun Locks 328 

II. Fig. 2.— Concrete Handling Plant, Gatun Locks 328 

Fig. 3.— Wall Forms, Gatun Locks 328 

III. Fig. 6. — (>)ncrcte Handling Plant, Mirafloroa Ix)cks 340 

IV. Fig. 7. — Details, C>)ncretc Handling Plant, Miraflores Locks. . . 342 
V. Fig. 10. — Forms for Wall, Pedro Miguel and Miraflores I>ocks. . . . 344 

VI. Fig. 11.— Handling Plant, Pedro Miguel I^cks 346 

VII. Fig. 13. — Material Handling Cranes, Pedro Miguel Ivocks 346 



Digitized by 



Google 



LIST OF OFFICERS 



OF 



THE NATIONAL ASSOCIATION OF CEMENT USERS 

1912. 



PRESIDENT. 

RICHARD L. HUMPHREY. 



FIRST VICE-PRESIDENT. 

EDWARD D. BOYER. 



SECOND VICE-PRESIDENT. 

ARTHUR N. TALBOT. 



THIRD VICE-PRESIDENT. 

EDWARD S. LARNED. 



FOURTH VICE-PRESIDENT. 

IRA H. WOOLSON. 



SECRETARY. 

EDWARD E. KRAUSS. 



TREASURER. 

HENRY C. TURNER. 



SECTIONAL VICE-PRESIDENTS. 
(See page 7.) 



w 



Digitized by 



Google 



SECTIONAL COMMITTEES. 

(The Chainnen are Vice-Presidents of the Association.) 



BUILDING BLOCKS AND CEMENT PRODUCTS. 
P. S. Hudson, Chairman, 

C. K. ABPy ROBERT V. HAYUK, 

P. H. ATWOOD, CHARLES D. WATSON. 

EXHIBITION. 
H. S. DoTLB, Chairman, 

B. F. AFFI4ECK, W. J. ROSBBERRT, JR., 

W. H. BURKE, P. AUSTIN TOMES. 

FIKEPROOFING. 
Rudolph P. Miller, Chairman, 

EDWIN CLARK, JOHN STEPHEN SEWELL, 

W. C. ROBINSON, IRA H. WOOLSON. 

INSURANCE. 
William H. Ham, Chairman, 

A. L. JOHNSON, F. W. MOSES, 

W. H. MERRILL, J. P. H. PERRY. 

MEASURING CONCRETE. 
Robert A. Cummings, Chairman. 

L. H. ALLEN, CHARLES DERLETH, JR., 

ROBERT ANDERSON, THOMAS M. VINTON. 

NOMENCLATURE. 
Peter Gillespie, Chairman, 

E. p. GOODRICH, F. B. TURNEAURE, 

E. J. MEHREN, FRANK C. WIGHT. 

REINFORCED CONCRETE AND BUILDING LAWS. 
Alfred E. Lindau, Chairman, 

W. p. ANDERSON, ARTHX7R N. TALBOT, 

E. J. MOORE, 8ANF0RD B. THOMPSON. 

ROADWAYS, SIDEWALKS AND FLOORS. 
C. W. BoYNTON, Chairman. 

A. G. BIRNIE, C. R. MILLER, 

i, B. LANDFPXD, A. B. SN0DGRAB8, 

(7) 



Digitized by 



Google 



8 



Sectional Committees. 



SPECIFICATIONS AND METHODS OF TESTS P'OR CONCRETE 

MATERIALS. 

Sanpord E. Thompson", Chairman. 

CLOYD M. CHAPMAN, RUSSELL 8. GRBENMAN, 

WILLIAM B. FULLER, ARTHUR N. TALBOT. 

TREATMENT OF CONCRETE SURFACES. 
Leonard C. Wason, Chairman. 

CLOYD M. CHAPMAN, EMILE O. PERROT, 

ALFRED HOPKINS, H. H. QUIMBY. 

EDUCATION. 

Logan Waller Page, Chairman. 
Percy H. Wilson, Secretary. 
A. C. True. Morris Mbtcalf. 

William G. Hartranft. A. Moyer. 

W. A. Holman. J. P. H. Perry. 



Agricultural Experiment Stations. 



Alabama — J. F. Duogar. 
Arizona — Representative not yet 

appointed. 
Arkansas — Martin Nelson. 
California — LeRoy Anderson. 
Colorodo — Alvin Keyser. 
Connecticut — Charles A. Wheeler. 
Delaware — Harry Hayward. 
Florida — ^J. J. Vernon. 
Georgia — LeRoy C. Hart. 
Idaho — W. L. Carlyle. 
Illinois — E. A. White. 
Indiana — A. T. Wiancke. 
Iowa — J. B. Davidson. 
fiansas — W. W. Jardjnk. 
Kentucky — M. A. Scovell. 
I^uisiana — W. R. Dodson. 
Maine — George E. Summons. 
Maryland — W. T. L. Taliaferro. 
Massachusetts — Wiluam D. Hurd. 
Michigan — R. S. Shaw. 
Minnesota — John T. Stewart. 
Mississippi — W. L. Hutchinson. 
Missouri — F. B. Mumford. 
Montana — H. B. Bonebright. 



Nebraska — L. W. Chase. 
Nevada — Representative not yet 

appointed. 
New Hampshire — F. W. Taylor. 
New Jersey — ^J. G. Lipman. 
Neiv Mexico — F. L. Bixby. 
New York—n. W. Riley. 
North Carolina — C. L. Newman. 
North Dakota — ^J. H. Shepperd. 
Ohio — H. C. Hamsower. 
Oklahoma — O. O. Churchill. 
Oregon — H. D. Scudder. 
Pennsylmnia — Frank D. Gardener. 
Rhode Island — G. E, Adams, 
South Carolina — W. R, Perkjns. 
South Dakota — A, N. Hume. 
Tenntissee — C, A. Moores, 
Texas — S. S. McMjllant. 
IJtah—F, S. Harris. 
Vermont — J. W. Elliot. 
Virginia — H. L. Price. 
Washington — O. L. Waller. 
West Virginia — I. S. Cook. 
Wisconsin — Charles A. Ocock. 
Wyoming — J. C. F|tterer. 



Digitized by 



Google 



PAST OFFICERS. 



President. 



1905 John P. Given. 

(Presiding Officer First Convention.) 

1905-11 Richard L. Humphrey. 



First Vice-President. 



1905 a. l. goetzmann. 
1906-9 Merrill Watson. 
1909-11 Edward D. Boyer. 



Second Vice-President. 



1905-6 John H. Fellows. 
1907-10 M. S. Daniels. 
1911 Arthur N. Talbot. 



Third Vice-President. 



1905 H. C. QuiNN. 
1906-7 0. U. Miracle. 
1908 S. B. Newberry. 
1909-11 E. S. Larned. 



Fourth Vice-President. 



1905-7 A. Monsted. 
1908-9 George C. Walters. 
1909-10 F. A. NoRRis. 
1911 Ira H. Woolsen. 



Treasurer. 



1905 A. S. J. Gammon. 
1906^11 H. C. Turner. 



Secretary. 



1905-6 Charles C. Brown. 

1907 W. W. Curtis. 

1908-9 George C. Wright. 

1910 Edward E. Krauss (Acting) 

1911 Edward E. Krauss, 



(9) 



Digitized by 



Google 



CHARTER 

OF 

THE NATIONAL ASSOCIATION OF CEMENT USERS. 



KNOW ALL MBN BY THBSB PRBSBNTS, That we, the under- 
signed, all of whom are citizens of the United States, and a majority of 
whom are residents of the District of Columbia, have associated our- 
selves together for the purpose hereinafter set forth and desiring that 
we may be incorporated as an Association under sub-chapter three (3) of 
the Incorporation Laws of the District of Columbia, as provided in the 
Code of Law of the District of Columbia, enacted by Congress and ap- 
proved by the President of the United States, do hereby certify: 

/• Nmm€. The name of the proposed corporation is "The National 
Association of Cement Users." 

2. Term oi Bxiateace. The existence of the said corporation shall 
be perpetual. 

3. OtiectMm The particular business and objects of the said corpora- 
tion shall be to disseminate information and experience upon and to 
promote the best methods to be employed in the various uses of cement by 
means of convention, the reading and discussion of papers upon materials 
of a cement nature and their uses, by social and friendly intercourse at 
such conventions, the exhibition and study of materials, machinery and 
methods and to circulate among its members by means of publications the 
information thus obtained. 

4. iacorpormton. The number of its managers for the first year 
shall be fifteen. 

la WItaeMB Whereof, we have hereunto set our hands and seals this 
fourteenth day of December, A. D. 1906. 

RICHARD L. HUMPHREY, (Seal) 

JOHN STEPHEN SEWELL, (Seal) 

S. S. VOORHEES. (Seal) 

Office of Recorder of Deeds, 
District of Columbia. 
This is to certify that the foregoing is a true and verified copy of a 
Certificate of Incorporation, and of the whole of such Certificate as re- 
ceived for record in this office at 9 : 49 A. m., the 19th day of December, 
A. D. 1906. 

In testimony whereof I have hereunto set my hand and affixed the 
teal of this office, this 20th day of December, A. D. 1906. 
(Signed) R. W. DUTTON, 

Deputy Recorder of Deeds, 

LHstrict of Columbia 

(10) 



Digitized by 



Google 



BY-LAWS. 
Article I. 

MEMBERS. 

Section 1. Any person engaged in the construction or 
maintenance of work in which cement is used, or qualified by 
business relations or practical experience to co-operate in the pur- 
poses of this Association, or engaged in the manufacture or sale 
of machinery or supplies for cement users, or a man who has 
attained eminence in the field of engineering, architecture or 
applied science, is eligible for membership. 

Sec. 2. A firm or company shall be treated as a single mem- 
ber. 

Sec. 3. Any member contributing annually twenty or more 
dollars in addition to the regular dues shall be designated and 
listed as a Contributing Member. 

Sec. 4. Application for menibership shall be made to the 
Secretary on a form prescribed by the Board of Direction. The 
Secretary shall submit monthly or oftener if necessary to each 
member of the Board of Direction for letter ballot a list of all 
applicants for membership on hand at that time with a statement 
of the qualifications, and a two-thirds majority of the members of 
the Board shall be necessary to an election. 

Applicants for membership shall be qualified upon notification 
of election by the Secretary by the payment of the annual dues, 
and unless these dues are paid within 60 days thereafter the elec- 
tion shall become void. An extract of the By-Laws relating to 
dues shall accompany the notice of election. 

Sec. 5. Resignations from membership must be presented 
in writing to the Secretary on or before the close of the fiscal year 
^pd shall be acceptable provided the dues are paid for that year, 



Digitized by 



Google 



12 By-Laws. 

Article II. 

OFFICERS. 

Section 1. The officers shall be the President, two Vice- 
Presidents, six Directors (one from each geographical district), 
the Secretary and the Treasurer, who, with the five latest living 
Past-Presidents, who continue to be members, shall constitute 
the Board of Direction. 

Sec. 2. The Board of Direction shall, from time to time, 
divide the territory occupied by the membership into six geographi- 
cal districts, to be designated by numbers. 

Sec. 3. The terms of office of the President, Secretary and 
Treasurer shall be one year; of the Vice-Presidents and the Direct- 
ors, two years. Provided, however, that at the first election after 
the adoption of this By-Law, a President, one Vice-President, 
three Directors and a Treasurer shall be elected to serve for one 
year only, and one Vice-President and three Directors for two 
years; provided, also, that after the first election a President, one 
Vice-President, three Directors and a Treasurer shall be elected 
annually. 

The term of each officer shall begin at the close of the Annual 
Convention at which such officer is elected, and shall continue for 
the period above named or until a successor is duly elected. 

A vacancy in the office of President shall be filled by the senior 
Vice-President. A vacancy in the office of Vice-President shall 
be filled by the senior Director. 

Seniority between persoas holding similar offices shall be 
determined by priority of election to the office, and when these 
dates are the same, by priority of admission to membership; 
and when the latter dates are identical, the selection shall be made 
by lot. In case of the disability or neglect in the performance of 
his duty, of any officer of this Association, the Board of Direction 
shall have power to declare the office vacant. Vacancies in any 
office for the unexpired term shall be filled by the Board of Direc- 
tion, except as provided above. 

Sec. 4. The Board of Direction shall appoint the Secretary; 
it shall create such special committees as may be deemed desirable 
for the purpose of preparing recommended practice and standards 
concerning the proper use of cement for consideration by the 



Digitized by 



Google 



By-Laws. 13 

Association, and shall appoint a chairman for each committee. 
Four or more additional members on each special committee 
shall be appointed by the President, in consultation with the 
Chainnan. 

Sec. 5. It shall be the duty of the Board of Direction to 
audit the accounts of the Secretary and the Treasurer before each 
annual convention. 

Sec. 6. The Board of Direction shall appoint a Committee 
on Nomination of Officers and a Committee on Resolutions, to be 
announced by the President at the first regular session of the 
annual convention. 

Sec. 7. There shall be an Executive Committee of the 
Board of Direction, consisting of the President, the Secretary, the 
Treasurer and two of its members, appointed by the Board of 
Direction. 

Sec. 8. The Executive Committee shall manage the affairs 
of the Association during the interim between the meetings of 
the Board of Direction. 

Sec. 9. The President shall have general supervision of the 
affairs of the Association. He shall preside at the Annual Con- 
vention, at the meetings of the Board of Direction and the Execu- 
tive Committee, and shall be ex-officio member of all conmiittees. 

The Vice-Presidents in order of seniority shall discharge the 
duties of the President in his absence. 

Sec. 10. The Secretary shall perform such duties and fur- 
nish such bond as may be determined by the Board of Direction. 

Sec. 11. The Treasurer shall be the custodian of the funds 
of the Association, shall disburse the same in the manner prescribed 
and shall furnish bond in such sum as the Board of Direction may 
determine. 

Sec. 12. The Secretary shall receive such salary as may be 
fixed by the Board of Direction. 

Article III. 

MEETINGS. 

Section 1. The Association shall meet annually. The time 
and place shall be fixed by the Board of Direction and notice of 
this action shall be mailed to all members at least thirty days 
previous to the date of the Convention. 



Digitized by 



Google 



14 bT-tiAW&. 

Sec. 2. The Board of Direction shall meet immediately 
after the Convention at which it was elected, effect organization 
and transact such business as may be necessary. 

Sec. 3. The Board of Direction shall meet at least twice 
each year. The time and place to be fixed by the Executive Com- 
mittee. 

Sec. 4. A majority of the members shall constitute a 
quorum for meetings of the Board of Direction and of the Executive 
Committee. 

Article IV. 

DUES. 

Section 1 The fiscal year shall commence on the first 
of July and all dues shall be payable in advance. 

Sec. 2. The annual dues of each member shall be five dollars 
($5.00). 

Sec. 3. Any person elected after six months of any fiscal 
year shall have expired, need pay only one-half of the amount 
of dues for that fiscal year; but he shall not be entitled to a copy 
of the Proceedings of that year. 

Sec. 4. A member whose dues remain unpaid for a period of 
three months shall forfeit the privilege of membership and shall 
be officially notified to this effect by the Secretary, and if these 
dues are not paid within thirty days thereafter his name shall be 
stricken from the list of members. Members may be reinstated 
upon the payment of all indebtedness against them upon the books 
of the Association. 

Article V. 

RECOBfBfENDED PRACTICE AND SPECIFICATIONS. 

Section 1. Proposed Recommended Practice and Specifica- 
tions to be submitted to the Association must be mailed to the 
members at least thirty days prior to the Annual Convention, and 
as there amended and approved, passed to letter ballot, which shall 
be canvassed within sixty days thereafter, such Recommended 
Practice and Specifications shall be considered adopted unless at 
least ten per cent, of the total membership shall vote in the 
negative. 



Digitized by 



Google 



Akticle VI. 

AMENDMENT 

Section 1. Amendments to these By-Laws, signed by at 
least fifteen members, must be presented in writing to the Board of 
Direction ninety days before the Annual Convention and shall be 
printed in the notice of the Annual Convention. These amend- 
ments may be discussed and amended at the Annual Convention 
and passed to letter ballot by a two-thirds vote of those present. 
Two-thirds of the votes cast by letter ballot shall be necessary 
for their adoption. 



Digitized by 



Google 



SUMMARY OF THE PROCEEDINGS OF THE EIGHTH 
ANNUAL CONVENTION. 



First Session, Monday, March U, 1912, 8 p. m. 

The Convention was called to order by the President, Richard 
L. Humphrey. 

John Lyle Harrington, Past President, Engineers' Club of 
Kansas City, delivered an address of welcome on behalf of the 
Engineering Interests as follows: 

In the dark ages of industry which extended well into the last century, 
it was the custom for every member of a craft, trade or profession to guard 
jealously and to keep closely secret every item of knowledge he or his associates 
had, in order to secure to himself the whole advantage of it. In addition to 
patenting an invention, it was common to keep secret every possible detail 
of the processes employed, and many businesses were based wholly upon 
secret formulae which were closely held by members of a firm or family, often 
handed down from father to son, and the utmost precautions were taken to 
ensure that employees and even associates, as well as competitors, actual and 
potential, should be kept in ignorance of the methods employed or discoveries 
made. Even up to the present day it is dangerous in some e.stablishments 
for an employee to ask too many questions regarding the methods of manu- 
facture or materials he employs in his work, and in certain lines of manufacture 
a considerable remnant of this old secrecy remains. Here and there the 
possessor of a formula has, like the old alchemist, at the right moment dropped 
his fluid into a molten metal or added his mite to the production of important 
materials and kept the secret and profits thereof to himself. 

But early in the last century the civil engineers of Great Britain met and 
formed an institute for the purpose of disseminating the knowledge acquired 
by its individual members; and in the latter half of the century the engineers 
of this country came to appreciate the advantages of co-operation. The 
first such organizations were few in number and comprehensive in scope, 
but gradually important groups working in special lines came to feel that the 
interests of the broad general organization were too varied to permit adequate 
consideration of the matters which specially occupied their attention and so 
they split off and organized societies of more limited .scope. With the enor- 
mous development of industries based on the appliinl sciences special interests 
so increased in value and importance that the benefits to be derived from close 
association and active discussions of men engaged in them came to be generally 
understood, and group organizations grew apace. 

(10) 



Digitized by 



Google 



Summary of Proceedings, Eighth Convention. 17 

In the course of time, manufacturers began to understand that by keep- 
ing to themselves knowledge of their s|x;cialities they encouraged like action 
on the part of their competitors — and each member of that branch of industrj' 
developed internally only. This limited the development of all and circum- 
scribed the field of the industry. Gradually it became evident that the 
advancement of the industry resulting from the dissemination of knowledge 
of it among those engaged in it more than compensated for the advantage 
secured even by the most successful operation under the secret methods. As 
soon as this fact came to be fully realized the organization of special societies 
for special purposes multiplied rapidly. They reacted upon the industries 
and the accomplishments of one man so pointed out possibilities and so 
stimulated others that development has increased in geometric ratio. 

Among these later organizations developed to special interests, the 
National Association of Cement Users has come to occupy an important 
place. 

Cement in some form has been in use for many centuries, so many that 
its origin is lost in antiquity; but the development of Portland cement and 
the industries and types of construction dependent upon its use are of com- 
paratively recent date. It is hard to believe that less than twenty years ago 
Congress gravely questioned the existence in this country of materials essential 
to the manufacture of first-class Portland cement, and debated whether, in 
view of that condition, it would be justifiable to impose so much duty on 
imported cement as would induce its manufacture in this country. Yet hist 
year we produced nearly 78,(XX),(XX) barrels, which, at the existing low price 
prevailing, brought nearly $(>8,(XX),(XX) at the mills, and the value of con- 
structions in which it wjis used probably reached nearly $500,000,000. 

It is well, therefore, that the users of cement, the men who are respon- 
sible for construction of cement and concrete work worth one-half billion 
dollars per annum, should organize themselves into an association and meet 
to advance their special interests by the discussion of the work they are doing 
and the experiences they have gained. And while it is natural and right that 
interest and enthusiasm for their work should lead chiefly to the exposition 
of the successes achieved, it is quite as important that the defects found in the 
materials, in the constructions and in the methods employed should be 
exposed and discussed, for we learn as much — often more, if we are wise — 
from our failures as from our successes. And it is important that the limita- 
tions of materials and their uses be fully understood and bad results thus 
guarded against, and the character of the practices on th<i whole thus improved. 
The effort to meet the ever-pressing demand for cheap construction leads to 
many failures and to much consequent damage to the industry. To increase 
the stresses or reduce the quality of concrete in order to enable it to compare 
favorably with wood in first cost is an unwarranted, but far too common 
practice. If in any instance its many superior qualities do not justify the 
greater expenditure for good concrete safely stressed, then the cheaper mate- 
rials should be employed. Wilful, foolhardy rLsks are responsible for some of 
the failures, but ignorance of all conditions governing concrete construction 



Digitized by 



Google 



18 Summary op t*ROCEia)iNGS, Eighth ConvenHok. 

is by far the larger factor, and it is the duty and purpose of this organization 
to expose bad construction as it is to urge good methods and good workmanship. 
Concrete construction is peculiarly liable to failure through both ignorance 
and carelessness. Through ignorance because its use in the simpler tjrpes of 
construction leads men to believe its employment universally simple; through 
carelessness because it is so amenable to the employment of unskilled labor 
which needs, but does not always receive, thorough and careful supervision. 
I well remember the county engineer who was confident he had made all the 
plans necessary for the construction of a reinforced concrete bridge when he 
had made a picture on which he had given general dimensions and shown the 
location of some reinforcement. 

Failure to recognize the need of careful inspection of materials and super- 
vision of workmanship is perhaps responsible for the larger portion of failures 
of concrete structures. Steel is manufactured by skilled men working under 
able superintendence and so continuously employed that they come to under- 
stand the processes thoroughly; whereas concrete is conmionly manufactured 
by an itinerant, common laborer and often directed by a foreman whose 
only qualification is his ability to handle men. Constructions of wood, 
brick or stone may be fairly well inspected at any stage or after completion, 
but the character of concrete construction may not be determined until 
after the forms are removed, and remedying defects then disclosed is always 
difficult, often impossible. Too early removal of the forms, thus placing upon 
the concrete stresses which it is yet unfitted to bear; carelessness and imwise 
placing of reinforcement; inadequate provision for expansion and contrac- 
tion due to changes of temperature; weak and leaky forms; inadequate 
tamping; loose methods of depositing concrete, both in air and under water; 
excessive dependence in designing upon empirical methods and fallacious 
load tests; these and many other difficulties must be guarded against and 
overcome. Much disappointment is certainly in store for the owners of 
staff and concrete buildings who have not fully appreciated the difficulties 
in securing construction which will resist the weather. There are many other 
difficulties which must be guarded against and overcome. 

As the use of concrete becomes more general, more attention is given to 
the sightliness of structures built of it, but too often the efforts in this line 
are misdirected. It should be clearly recognized that this material is capable 
of excellent treatment peculiar to itself, and that it is an error to try to 
make it resemble other materials. The old Spanish structures of southwestern 
states and Mexico are beautiful because they are true. They were designed 
frankly to be built of concrete and have forms and finish suitable for that 
material. It may be that with the wider use and fuller development of con- 
crete construction we may improve upon the work of these Spaniards, but 
we have not by any means equalled it as yet. The endeavor to meet the needs 
of the material has too frequently led to the adoption of extreme and grotesque 
forms and finishes which soon weary or offend. 

We are gradually improving the finish of our concrete structures and 
recent work in this line is especially promising, but much remains to be done 
both for the appearance and for the weathering quality of concrete and staff 



Digitized by 



Google 



SUMMART OP PROCEEBtNGft, ElGHTH CoNVfcNTtOI^. 19 

^surfaces. Good finish adds to the cost, but the extra expense is surely jus- 
tified. 

The gfeat advantages of concrete, durability, loW cost as compared with 
other equally durable materials, ease of handling, ease of molding to the forms 
desired, resistance to fire, strength and homogeneity) adaptability to many 
uses, both with and without reinforcement, resistance to acids and decay^ 
attractiveness of appearance and the wide distribution of its constituent 
materials, improvements in methods of construction, and the steadily 
increased scarcity and cost of good timber ensure continual increase in the 
use of concrete. But perhaps the greatest argument in favor of fireproof 
concrete buildings is the enormous fire loss this country sustains through 
the 'excessive use of wood and the equally great cost of fire insurance. Our 
losses by fire have become a national disgrace. 

The influence of this Association should be very great in combating the 
tendency of some constructors to secure business and profit by reducing the 
quality of work. This course surely militates against the interest of cement 
users as a whole and it brings their work as a whole into disrepute. This was 
clearly exemplified recently at Los Angeles where the city engineer recom- 
mended creosoted instead of reinforced concrete piles for dock construction 
and was able to justify his position by citing about a dozen failures of docks 
which were supported on reinforced concrete piles. And generally speaking, 
any improper use of the material of construction affects adversely the inter- 
ests of all who engage in the use of that material. There is, therefore, a large 
opportunity for this Association to benefit greatly the interests of its members 
and to benefit the country at large by compelling a high standard of ability 
and integrity among cement users. 

It is thus apparent that this Association has met to deal with subjects 
of the largest consequence and the effects of its discussions will be to improve 
throughout this country and other countries the practices of concrete design 
and construction. The work of this Association is, in no inconsiderable 
measure, the work of the Engineer. He shares largely in the work and the 
responsibilities and the benefits of this convention; hence it is with pleasure 
and cordial good will that on behalf of the Engineers of Kansas City, I extend 
a hearty welcome to the National Association of Cement Users. 

An address of welcome on behalf of the Concrete Interests 
was made by F. W. Pratt, President of the Union Bridge and 
Terminal Railroad Company. 

The President responded: 

The Convention has this year been fortunate in haying two eminent 
engineers express thoughts that will be of good service to the Association 
in its future work. I am sure that you all join me in reciprocating the 
good wishes which Mr. Harrington and Mr. Pratt have extended to us, and 
in the hope that our deliberations may prove of interest and value. 



Digitized by 



Google 



20 Summary of t^ROckEDiNGs, Eighth Convention. 

The following committees of the Convention, appointed by 
the Executive Board, were announced by the President: 

Committee on Nomination of Officers: 

E. J. Moore, Chairman^ New York, N. Y. 
D. A. Abrams, Ames, Iowa. 

John L. Conzelman, St. Ix)uis, Mo. 

B. F. Lippold, Chicago, III. 

Ij. T. Sunderland, KanHa.s CMty, Mo. 

Committee on Resolutions: 

Rudolph P. Miller, Chairman, New York, N. Y. 
P. n. Bates, PittsburKh, Pa. 
Allen Brett, New York, N. Y. 

F. L. Williamson, Kansjus C'ity, Mo. 
Perry H. Wilson, Philadelphia, Pa. 

A paper on *'Th(» Use of Reinforced Concrete in Hypochlorite 
Water Purification Works" was read by Walter M. Ooss. 

(leorge E. Tebbetts presented a paper on *'The Use of Con- 
crete in the New Union Station at Kansas City, Mo.," which 
wjis followed l)y a discussion. 

The meeting adjourncMl until Tuesday at 10.30 A. m. 



Tuesday, March 12, 1912, 10.00 a. m. 

Meeting of the Sections on Mcnisuring ('oncrete. Nomen- 
clature, and Specifications and Methods of Tests for Concrete 
Materials. 

President Richard L. Humphrey in the chair. 

The meeting took the form of a discussion on the deposition 
of mortar with compressed air, its effectiveness and various 
methods of application. 



Second Session, Tuesday, March 12, 1912, 10.30 a. m. 

President Richard L. Humphrey in the chair. 

The report of the Committee on Specifications and Methods 
of Tests for ('oncrete Materials was, in the absence of the Chair- 
man, Sanford E. Thompson, presented by Cloyd M. Chapman. 

Wm. M. Kinney read a paper on "Aggregates for Concrete," 



Digitized by 



Google 



Summary of Proceedings, Eighth Convention. 21 

which was followed by a Topical Discussion on Concrete Aggre- 
gates. 

In the absence in Europe of the Chairman, Robert A. 
Cummings, Edw. D. Boyer presented the report of the Com- 
mittee on Measuring Concrete, which was followed by a dis- 
cussion. The Proposed Standard Method of Measuring Concrete 
was referred back to the Committee, with instructions for a 
revision in the light of the discussion and report at the next 
annual Convention. 



Third Session— Tuesday, March 12, 1912, 8.00 p. m. 

President Richard L. Humphrey in the chair. 
The annual address of the President, entitled **The Use of 
Concrete in Eurojx*,*' was delivered by Richard L. Humphrey. 
The following papers were then read and discussed: 

"The Design and Construction of the Hollow Reinforced 
Concrete Dam of the Portland Railway Light and 
Power Company," by Herman V. Schreiber. 

"Cement Coatings in Color,'' by F. J. Morse. 

The report of the Committee on Concrete Surfaces was 
presented by the C'hairman, L. C. Wason. On motion the pro- 
posed Standard Method for Tests of Waterproofing was referred 
to the Committee on Specifications and Methods of Tests for 
Concrete Materials, for report. Consideration of the proposed 
Standard Specification for Portland Cement Stucco was deferred 
imtil a later session. The changes in the last general report of 
the Committee were approved. 

The report of the Committee on Insurance was, in the 
absence of the Chairman, Wm. H. Ham, read by title. 

The meeting then adjourned until Wednesday at 9.30 a. m. 



Fourth Session — Wednesday, March 13, 1912, 9.30 a. m. 

President Richard L. Humphrey in the chair. 

In the absence of the author, Sanford E. Thompson, the 
President read the paper entitled "The Practical Design of 
Reinforced Concrete Flat Slabs." 



Digitized by 



Google 



22 SUMliART OP PROCEEDINGS; ElGHTH CONVENTION. 

The report of the Committee on Reinforced Concrete and 
Building Laws was presented by the Chairman, Alfred E. Lindau, 
and Arthur N. Talbot, and was followed by a discussion. 

The President then presented the annual report of the 
Executive Board, submitting a proposed revision of the By-Laws, 
which were amended and ordered to letter ballot. The report 
was approved. 

The change of name of the Association was discussed and 
on motion the matter was referred to the Executive Board, with 
authority to effect a change in name if deemed expedient for the 
best interests of the Association. 

The Committee on Nomination of Officers presented the 
following nominations, which were unanimously approved and 
the Secretary instructed to cast the ballot for their election: 

Presidenty Richard L. Humphrey, Philadelphia, Pa. 
First Vice-President, Edward D. Boyer, Catasauqua, Pa. 
Second Vice-President, Arthur N. Talbot, Urbana, 111. 
Third Vice-President, Edward S. Lamed, Boston, Ma«s. 
Fourth Vice-President, Ira H. Woolson, New York, N. Y. 
Treasurer, Henry C. Turner, New York, N. Y. 

The time and place of the Ninth Annual Convention was 
referred to the Executive Board with power to act. 
The meeting then adjourned imtil 3 p. m. 



Fifth Session — ^Wednesday, March 13, 1912, 3.00 p. m. 

President Richard L. Humphrey in the chair. 
The following papers were read and discussed: 

"The Testing of Reinforced Concrete Buildings under 

Load," by W. A. Slater. 
"The Design of Concrete Flat Slabs," by Frank J. Tre- 
lease; in the absence of the author read by Alfred E. 
Lindau. 
"The Present Status of Unit Construction," by James L. 
Darnell. 
In the absence of the author, Theodore H. Skinner, the 
paper on "A Fireproof School of Concrete," was read by title. 
The meeting then adjourned until 8.00 p. m. 



Digitized by 



Google 



SUMMABT OF PROCEEDINGS, ElGHTH CONVENTION. 23 

Sixth Session — ^Wednesday, March 13, 1912, 8.00 p. m. 

President Richard L. Humphrey in the chair. 

The President introduced the Honorable Darius A. Brown, 
Mayor, who extended an address of welcome to Kansas City as 
follows: 

Mayor Brown. — I am glad of the opportunity to appear before you and 
express my appreciation of the fact that men should gather together in this 
city for the purpose of discussing the matters which have brought you here. 
We have heard about the Stone Age, the Wooden Age and the Iron Age, and 
I think it is the consensus of opinion that we are now in the Concrete Age 
and that all of the improved structures are coming to be made of this material. 
We are also in the age of progress and advancement; we are in an age when 
people have adopted the idea that if a thing is worth doing at all it is worth 
doing well and that is the reason why in every branch of human industry, 
in business, in commerce, in the sciences and in the professions, they gather 
together periodically for the purpose of discussing the ways and means of better 
doing the business in which they are engaged. 

I am satisfied that the result of your deliberations will not only be of 
benefit to you in your particular business but will be of benefit to the com- 
munity in which you live and benefit to the American people as a whole. 
I hope one of the results of your deliberations will be that the use of concrete 
will become so perfected that it will not only give us more comfortable build- 
ings and better and more ornamental structures but will decrease the cost of 
those materials to the people; because that is one of the great problems to be 
solved. 

On behalf of the people of Kansas City I want to extend to you a very 
cordial welcome and trust that you will not only be benefited by your delibera- 
tons but that you will have some pleasure while staying in our city. I thank 
you. 

The President responded: 

I am sure we all appreciate the welcome that has been extended by His 
Honor. A city of this size in whose immediate vicinity there is a production 
of Portland cement of about one-tenth of that of the entire country, is a 
good place in which to hold deliberations of this character. In accepting 
the hospitality of this city we do so with a feeling that we will be benefited. 
Our conventions have been held heretofore east of the Mississippi River, 
and I believe that the mingling of the eastern and western ends of this 
great country cannot help but be beneficial to all. I know you will join me 
in extending to Mayor Brown hearty thanks for his welcome. 

A paper on "The Use of Reinforced Concrete on the Wabash 
Railroad," was presented by A, 0. Cunningham, 



Digitized by 



Google 



24 Summary of Proceedings, Eighth Convention. 

The report of the Committee on Treatment of Concrete 
Surfaces was taken up and on motion the proposed Standard 
Specification for Portland Cement Stucco was received as in- 
formation and the Committee instructed to confer with other 
Associations, in order to reach an agreement as to a specifica- 
tion to be presented at the next Convention. 

The following papers were then read and discussed: 

"The Design and Construction of a Reinforced Concrete 
Dome, 220 Foot Span," by S. J. Trauer; in the 
absence of the author read by the President. 
'*Tho Design of Concrete (irain Elevators," by E. Ja'v 

Heidenreich. 
'*The Suitability of Concrete for (liis Holder Tanks/' by 
Herbert W. Alrich; in the absence of the author 
presented by the President. 
"The Necessity of Field Tests for Concrete," by Fritz E. 
Von Emperger; in the absence of the author pre- 
sented by the Secretary. 
The meeting adjourned until Thursday at 10.30 a. m. 



Thursday, March 14, 1912, 10.00 a. m. 

Meeting of the Section on Treatment of Concrete Surfaces. 
President Richard L. Humphrey in the chair. 
The meeting took the form of a topical discussion on the 
coloring of concrete surfaces, contraction, dusting of floors, etc. 



Seventh Session— Thursday, March 14, 1912, 10.30 a. m. 

President Richard L. Humphrey in the chair. 
The following papers were read and discussed: 

"Concrete Highway Bridges," by William Scott Gearhart. 

"Concrete Bridges," by Daniel B. Luten. 

"Flat Slab Bridges," by William H. Finley; in the absence 
of the author read by the President, 
The meeting adjourned until 8.00 p. m, 



Digitized by 



Google 



Summary of Proceedings, Eighth Convention. 25 

Eighth Session — ^Thursday, March 14, 1912, 8.00 p. m. 

President Richard L. Humphrey in the chair. 
The paper on "The Necessity for Good Roads/' by Logan 
Waller Page, was in the absence of the author presented by E. L. 
Eldredge. 

The following papers were then presented : 

"The Necessity of National Aid in Good Roads," by H. C. 

Gilbert. 
"Cement Paving as Constructed at Mason City, Iowa," by 

F. P. Wilson. 
"An Improved Concrete Pavement," by E. W. Groves. 
The report of the Committee on Roadways, Sidewalks and 
Floors, was presented by the Chairman, C. W. Boynton, and 
the following action taken: 

Proposed revisions of the Standard Specifications for Con- 
crete Road and Street Pavements ordered to letter ballot. 

The proposed revision of the Specifications on Sidewalks, 
Curb and Gutter, and the proposed new Specifications for Plaiin 
and Reinforced Concrete Floors, were considered and referred 
to the Committee for report at a later session. 

The meeting then adjourned until Friday at 10.30 a. m. 



Friday, March 15, 1912, 9.00 a. m. 

Meeting of the Section on Roadways, Sidewalks and Floors; 
C. W. Boynton, Chairman of the Section, in the chair. 

The meeting took the form of a discussion on concrete roads, 
concrete floors, the prevention of dusting of floors, concreting in 
freezing weather, etc. 



Ninth Session— Friday, March 15, 1912, 10.30 a. m. 

President Richard L. Humphrey in the chair. 
The following papers were read and discussed: 

"Concrete Fence Posts," by L. J. Hotchkiss. 

"The Design of Reinforced Concrete Retaining Walls," by 
John M. Meade. 



Digitized by 



Google 



26 Summary of Proceedings, Eighth Convention. 

"Advantages and Durability of Cement Sewer Pipe/' by 
Gustave Kaufman; in the absence of the author 
presented by the President. 
"Methods of Testing Cement Pipe," by Duff A. Abrams; 
in the absence of the author presented by W. A. Slater. 
"The Manufacture and Use of Cement Drain Tile," by 
Charles E. Sims. 
The report of the Committee on Cement Products and Build- 
ing Blocks was in the absence of the Chairman, Percy S. Hudson, 
presented by Clarence K. Arp. On motion the Proposed Standard 
Recommended Practice for Cement Tile was referred to letter 
ballot. 

The Committee on Roadways, Sidewalks and Floors, C. W. 
Boynton, Chainnan, reported back the matters referred to it 
and on motion the following were referred to letter ballot: 

Proposed Revisions of the Standard Specification for 

Portland Cement Sidewalks. 
Proposed Revisions of the Standard Specifications for 

Portland Cement Curb and Curb and Gutter. 
Proposed Standard Specifications for Plain Concrete 

Floors. 
Proposed Standard Specifications for Reinforced Concrete 
Floors. 
The report of the Committee on Nomenclature was in the 
absence of the Chairman, Peter Gillepsie, presented by Frank C. 
Wight and was on motion accepted as information. 
The Committee on Education reported progress. 
The Convention then adjourned until Saturday at 10.30 a. m. 



Saturday, March 16, 1912, 9.30 a. m. 

Meeting of the Section on Building Blocks and Cement 
Products, President Richard L. Humphrey in the chair. 

The meeting discussed the materials, methods of manufacture 
and tests of cement drain tile. 



Digitized by 



Google 



SuMMABY OP Proceedings^ Eighth Convention. 27 

Tenth Session — Saturday, March 16, 1912, 10.30 a. m. 

President Bichard L. Humphrey in the chair. 
The paper by George Gibbs on "Some Notes on the Value 
and Comparative Cost of Reinforced Concrete Telegraph Poles," 
was in the absence of the author presented by the President. 

The President read in the absence of the author, Robert A. 
Cummings, the paper on "The Making and Driving of Rein- 
forced Concrete Piles Within Six Days." 

The paper by W. J. Towne on "Concrete Fence Posts," 
was in the absence of the author read by title. 

A paper on "Comparative Tests of the Strength of Concrete 
in the Laboratory and in the Field," was presented by R. J. Wig. 
The following papers were, in the absence of the authors, 
read by title: 

"Field Inspection and Tests of Materials for Reinforced 

Concrete," by G. H. Bayles. 
"Unit Cost of Reinforced Concrete for Industrial Build- 
ings," by C. S. AUen. 
"Notes on "the Deformation in the Webs of Rectangular 
Concrete Beams," by H. C. Berry. 
The President then presented a paper by Alfred D. Flinn 
on "The Use of Cement for Protecting Steel Pipes Along the 
New York Aqueduct." 

Robert F. HavUk presented a paper on "Modem Methods of 
Manufacturing Concrete Products," which was followed by a 
discussion. 

The meeting then adjourned until 8.00 p. m. 



Eleventh Session — Saturday, March 16, 1912, 8.00 p. m. 

President Richard L. Humphrey in the chair. 

The paper on "The Use of Concrete in the Fourth Avenue 
Subway, Brooklyn, N. Y.," was in the absence of the author, 
Frederick C. Noble, presented by the President. 

W. A. CoUings presented a paper on "Reinforced Concrete 
in Agriculture." 



Digitized by 



Google 



28 Summary of Proceedings, Eighth Convention. 

The paper by S. B. Williamson on "The Handling of Concrete 
in the Construction of the Panama Canal," was in the absence of 
the author, presented by the President. 

The following papers were then read and discussed: 

"The Present Status of Iron Portland Cements," by P. H. 

Bates. 
"Iron Portland Cement," by Herman E. Brown; in the 
absence of the author, presented by the President. 
The Committee on Resolutions, Rudolph P. Miller, Chair- 
man, presented the following resolutions, which were unanimously 
adopted: 

(a) Resolved J That a eommittoo bo appoint od by the Executive Board 
to consider the form of all standard specifications or recommende<i practice 
issued by this Association with a view to securing uniformity so far as prac- 
ticable. 

(6) Resolved, That a committee of five members of this Association, of 
which one member shall be Chairman of the Committee on Specifications 
and Methods of Tests for Concrete Materials, be appointed by the President 
to plan a comprehensive and systematic investigation of the aggregates used 
for concrete and to interest State Universities, Experiment Stations and 
other laboratories in carrying out the same. 

(c) Resolved, That the Executive Board be instructed to consider the 
advisability of appointing a Committee to report on Standard Specifications 
for Concrete Highway Bridges and (Culverts. 

(d) Resolved, That the Committee on Nomenclature be instructed and 
empowered to extend its work to include the standardization of the size of 
drawings, the symbols used on same and the graphical representation of 
details. 

Resolved, That the Committee on Cement Products be instructed to 
consider the suggestions and criticisms on building block specifications offered 
at this Convention, to confer with the Committee on Reinforced Concrete 
and Building Laws with a view to reconciling there commendations of 
the two committees, and to report revised specifications to the next 
convention. 

(c) Resolved, That a report be submitted to the next Convention on 
Standard Specifications for Concrete Fence Posts and that the Executive 
Board consider the advisability of having this done by a sub-committee of 
the Committee on Cement Products or by a separate committee. 

(/) Resolved, That the thanks of this Association are hereby tendered 
the officials and the representatives of the local engineering and concrete 
interests for their hearty welcome, to the citizens of Kansas City for their 
co-operation in making this, the Annual Convention, a notable success, and 
to the guests of the Association for their assistance in this success by the 
contribution of their interesting and valuable papers. 



Digitized by 



Google 



Summary of Proceedings, Eighth CoNVENTioN. 2§ 

(g) Reaolvedf That the thanks of this Association are hereby tendered 
to the members who have aided by the presentation of papers, to the several 
committees whose efforts have added this meeting to the long series of suc- 
cessful conventions, to the local and technical press whose recognition of 
the work of this organization is gratefully acknowledged, and to its officers 
but particularly to its President, Mr. Richard L. Humphrey, for his untiring 
devotion to the interest and welfare of this Association. 

The President thereupon declared the meeting adjourned, 
sine die. 



Digitized by 



Google 



Digitized by 



Google 



National Association of Cement Users. 



PROCEEDINGS 

OF THE 

EIGHTH CONVENTION 



This ABSOciation is not responsible, as a body, for the statements and 
opinions advanced in its publications. 



EUROPEAN PRACTICE IN CONCRETE CON- 
STRUCTION. 

Annual Address by the President, 
Richard L. Humphrey.* 

Two years ago, in speaking on the use of concrete in Europe, 
I discussed in a general way the progress that was being made 
and contrasted the conditions prevailing in various parts of 
Europe. It was also pointed out that in the artistic treatment 
of concrete the foreign engineer and architect undoubtedly showed 
greater skill than was shown in this country, and therefore obtained 
much more pleasing results. I further commented on the fact 
that the development of the use of concrete in certain countries 
was very much handicapped by restrictive building laws. It was 
my good fortune to again visit Europe last year and to inspect 
extensively various structures of concrete, covering the most 
important work west of the Russian boundary. I shall use this 
opportunity for enlarging upon my former address, pointing out the 
development and essential points of difference at the present time 
in reinforced concrete construction in this country and in Europe. 

This visit to Europe and the inspection of concrete construc- 
tion during the closing months of last year, was under more favor- 
able conditions than on the occasion of my previous trip, in that 
I was a guest generally of the concrete associations, whose officers 
did all in their power to show me everything of interest. This 
was particularly true of my visit to Austria, where, as the Presi- 



* Consulting Engineer, Philadelphia, Pa. 

(31) 



Digitized by 



Google 



S^ Annual Address by the President. 

dent of this and Honorary Member of the Austrian Concrete 
Association, I was the guest of the latter during my entire stay 
in that country, and was under the devoted personal guidance 
of its Director, Karl Bitner. This gentleman, as you know, was 
a delegate to our New York Convention and one of the speakers 
at the banquet, and I wish again to express my heartiest thanks 
and appreciation for the courtesies ext^^nded to me and the 
unusual efforts which he made to render my visit a pleasant and 
profitable one. It is certainly true that this opportunity of 
inspecting the concrete buildings in Austria was a valuable one 
for the reason that some of the best examples of concn^te con- 
struction are to be found in that country. 

The Austrian Concrete Association has always manifested 
a great interest in the work of our Institute and showed a willing- 
ness to co-operate with us in every possible way. The unusual 
character of the program laid out and the rapidity with which 
various pieces of work were inspected was, I think, intended as 
a tribute to our American characteristics. I submit this inter- 
esting program. (See opposite page.) 

A feature that impressed me most favorably was that my 
visits to the various buildings had been arranged for in advance 
and upon scheduled time. On our arrival at the building we 
were met by the architect or his representative, the builder, and 
the engineer in charge; in many instances the plans of the struc- 
ture were tacked up at some convenient point, and before the 
building was inspected a representative who spoke English ex- 
plained the particular points of interest in the structure. On 
Thursday night I was the guest of the Austrian ('oncrete Institute, 
the Austrian Association of Cement Manufacturers, Austrian 
Society of Engineers and Architects and the Austrian Clay Products 
Association. My here recorded acknowledgment of the signal 
honor conferred upon me but inadequately expresses th(» (l(»pth 
of my gratitude and the extent of my appreciation. The precis- 
ion with which the progran was carried out, the completeness of 
the details and the warm hospitality extended to me by all those 
I met, has left its permanent record — one that I shall never forget. 

Dm-ing my visit to p]ngland it wius my privilege to address 
the Concrete Institute in London, on October 26, 1911, on the 
subject of " Fireproofing** for which I was honored by the award 
of the Institute medal. 



Digitized by 



Google 



OSTERREICHISCHER BETONVEREIN 
WIEN, IV/2, MOLLWALDPLATZ 4 

TELEPHONE NT. 10.597 TELEPHONE Nr. 10.597 



ZEITEINTEILUNG 

anlasslich 

Des Besuches des Prasidenten der National Association of Cement 

Users, Herrn RICHARD L. HUMPHREY, Consulting Engineer, 

Philadelphia. 



Donnerstag, 21. September, 1911. 

9 Uhr 30 Min. Abfahrt vom Hotel Bristol: 

9 " 35 " Kamtnerhofbasar, I. Kamtnerstrasse 

(Ed. Ast ACic); 

10 " 20 " Wohn- iind Geschaftshaus, I. Weihburggasse 7 
(Plachy & Co.); 

10 " 30 " Wohn -und Geschaftahaus, I. Weihburggasse 9 
(G. A. Wayss& Cie.); 

10 " 40 " Wohn- und Geschaftshaus, I. Weihburggasse 10 

(N. Rella & Xcffe) : 

10 " 50 * Lazzenhof, I. Rotenturmstrasse, verlangeter Fleischmarkt 

(Kontrollbalkenversuch) 
(k. k. Oberbaurat Dr. Ing. Friti von Empergcr — Chefingenieur 
Richard Wucakowski) 

11 " 30 *' Wiener Urania, I. Aspernplatz 

(G. A. Wayss&Cie.): 

11 " 55 *' Dreilaufferhaus, I. Kohlmarkt, Ecke Herrengasse 

(Pittel & Brausewettrr) ; 

12 " 00 " Wiener Bankverein, I. Schottenring 

(Ed. Ast A Cie. und N. Rella & Neflfe); 

12 " 30 " Lunch im Rathauskeller: 



Digitized by 



Google 



2 Uhr 00 Min. Autogarage, II. Rembrandtstrasse 29 

(Jancsch und Schnell) ; 

40 " Basaltoidpflaster (2 Jahre alt), XVI. Hasnerstrasee— 
Richard Wagner-Platz 

(Baaaltwprke Radebeulo) ; 
55 " Kirchenbau, XVI. Herbststrasse 66 

(MazEtnerA Cie.); 

25 " Zentralpalast, VI. Mariahilferstrafise — Ecke Kaiser- 
strafTe 

(G. A. Waysa & Cie.) ; 

00 '' Gewerbliche Fortbildungsschule, VI. Mollardgasse 

(N. Rella ANeffe): 

20 " Schokoladefabrik Stollwerck, V. Gaudenzdorfer GOrtel 
43,45 

(Pittol A Braiwewett^r) ; 

30 " Schule, V. Margaretenstrasse 103 

(N. RellaANeffe); 

15 " Kronenbrotwerke, X. Siccardsburggasse 83 

(Ed. Ast A Cie.) ; 

Wahrend der Fahrt zu besichtigen: 
Kabelblocklegung ftir die Telephonleitungen, Unterleitung 
der Strafisenbahn, Stadtbahn als Untergnind- und 
Hochbahn etc. ; 

00 ** Besuch des k. k. Hof-Opemtheaters; 
Souper. 

Freitag, 22. September. 

Fahrt nach Berndorf, Besichtigung der Kirche, Weiterfahrt 
nach Weissenbach, Besichtigung der Kunststeinfabrik; 
Adolf Baron Pittel 

Lunch; 

Weiterfahrt; 

Besichtigung der Zementfabrik Achau; 

Besichtigung der Betriebsanlagen der Wienerberger 
Ziegelfabriks^ und Baugesellschaft (Keramitfabrik und 
Seidlbalkenerzeugung), X. Triesterstrasse 100; 

00 " Souper am Cobenzl, angeboten vom Osterreichischen 
Betonverein, Verein Osteneichischer Zementfabrik an- 
ten, Osterreichischen Ingenieur- und Architektenverein 
und Osterreichischen Tonindustrievcrein (bei schlechtem 
Wetter Souper im Ktlnstlerzimmer des Restauj^ants 
Hopfner). 



8 Uhr 00 Min 


11 


u 


30 


t 


1 


It 


45 


ft 


3 


tl 


00 


tt 


4 


tt 


30 


tt 



Digitized by 



Google 



«9 


" 16 


9 


" 30 


9 


" 45 


10 


" 05 



Samstaflp 23. September. 

8 Uhr 00 Min. Abfahrt vom Hotel Bristol: 

8 '' 30 " Versuchsplatz des Eisenbetonausschusses des OsteireichiB- 
chen Ingenieur- und ArchitektenvereineSi Wien, XIX. 
Muthgasse; 

SchleuBenanlagen in Nussdorf ; 

Waaserturm am Bahnhof HeiUgenstadt 
(N. Rella A Neffe): 

Zigarettenpapierfabrik Schnabl, XIX. HeiUgenstadt 

(N. Rella & Neffe); 

Brtlcke im Zuge der Rampengasse, XIX. 
fH. RelUftCo.); 

10 " 20 ** Wagenhalle der stjidtbchen Strassenbahnen, XVIII. 
Wahringer Gtirtel 
(N. Rella ft Neffe); 

10 " 45 " Physikalisches Institut, IX. Wahringerstrasse — Waisen- 

hausgasse 
(Ed. .\pt<fcCie.); 

11 " 30 " Sargfabrik, XIII. Matznergasse 8 

(A. Porr): 

11 " 50 " Technisches Museum ftir Industrie und Gewerbe, XIV. 

Ecke Winckelmannstrasse und Linzerstrasse 

(A. Porr); 

12 '' 30 '' Basaltoidflaster, XIII. Sch5nbrunner Schloesstrasse 

(Basaltwerke Radebeule); 

Lunch im Parkhotel (Hietzing); 

Landes-Heil- und Pflegeanstalt Steinhof ; 

Kohlenturm und Koksseperationsanlage im Gaswerke 
XXI. Leopoldau 
(H. Rella A Co.): 

4 " 30 " Siloanlage nebst Weichraum und Lagerraum in der 
Malzfabrik Hauser & Sobotka, XXI. Stadlau 

(H. Rella A Co.); 

Sehenswtkrdigkeiten von Wien; 
8 " 00 " Souper Venedig in Wien. 

Eventuell: Sonntag, 24. September. 

Unter Ftlhrung des Herm Ingenieur E. A. Westermann, 
Chef der Firma Wayss, Westermann & Cie., Graz: 

Fahrt tiber den Semmering nach Graz, Besichtigung des 
Landeskrankenhauses Graz und der Briickenobjekte an 
der Bahnstrecke Weiz-Birkfeld. 



12 


" 46 


2 


" 15 


3 


" 30 



Digitized by 



Google 



Montag, 28. Sentember. 

Fahrt nach Retznei, Besichtigung der Zexnentfabrik, 
Weiterfahrt nach Adelsberg, Besichtigung der Adels- 
berger Grotte; 

Fahrt nach Triest, Besichtigung der Hangarbauten, 
Fundierungen etc. 



Dienstag, 26. September. 



8 Uhr 00 Min. Excelsior Palace Hotel 

(Wayas' Wefltcrmann & Cie.) ; 

8 " 30 " Greinitz 

(Maxorana A Cornel) ; 

8 " 45 " Riunione Adriatica 

(Aat. ACie.): 

9 '* 00 " Hanger -Bauten 

(WayM* Freitag & Mciaong); 

10-12 '' Rundfahrt im Hafen; 

Eventuell: Souper in Opcina; 

Fahrt nach Miramare. 



Digitized by 



Google 



Annual Address by the President. 



33 



The various countries of Europe are progressing in the use 
of concrete, but many of the large cities are still handicapped by- 
restrictive building laws; particularly is this true of London, 
where only recently have the London County Council Regula- 
tions permitted the erection of structures that might be termed 
reinforced concrete. Most continental countries show far greater 
skill in the application of concrete than is shown in England, 
where the British conservatism has resulted in heavy structures 
of very simple application. This is also true for the most part 
of the various cities in Germany. In France, in Belgiimi, and 
particularly in Austria a wider and less conservative grasp in the 
use of this material has resulted in the erection of structures 



'AHDi JIMMf^H'^C 



i4*'/l'*"l# KMVffil'J 



FrKFf'dQOFING, 




FIG. 1. — THE CONCRETE INSTITUTE MEDAL. 

which are not equaled anywhere in the world. Certainly at the 
present time in Vienna I believe one may find the most extensive 
use of concrete that is to be found either in this country or Europe. 
The city of Hamburg is perhaps second only to Vienna in the 
number of its reinforced concrete buildings, and these two cities 
are the most progressive spots in Europe. I observed on this 
trip that more concrete buildings were in evidence in the out- 
skirts of London and other English and continental cities than on 
my previous visit; in this country the same is true, probably 
for the same reason, viz., that the laws governing the erection of 
buildings are more liberal outside of than inside of the larger cities. 
The development of the use of concrete is certainly much 
greater in Europe than it was two years ago. I do not think, 
however, that Europe as a whole shows as great a development 



Digitized by 



Google 



34 Annual Address by the t^REsiDEKf. 

as is to be found in this count^>^ In certain parts of Europe, 
Austria, for example, is shown a greater knowledge and skill in 
the use of this material. In England, especially in London, 
where there were practically no concrete buildings to be foimd on 
my visit two years ago, there are now to be found many struc- 
tures of reinforced concrete; although it is true that these are 
not within the limits of the authority of the London County 
Council, which permits only reinforced concrete for floors with 
masonry-bearing walls, and to a limited extent for columns. It is 
probable, since a recent revision of the London County Building 
regulations permits the use of reinforced concrete, that from 
now on many structiu'es wall be erected of reinforced concrete. 

The materials in Europe available for use in concrete are 
still relatively more expensive than labor. As a result, the 
designer, for purposes of economy, finds it desirable to so shape 
his forms as to eliminate as much as possible all material which 
is not required for structural or protective purposes. The effect 
of this is to render the structure less massive and more pleasing 
in appearance. The abundance of extremely cheap unskilled 
labor and the presence of low-priced technically-trained labor is 
one of the great advantages in the erection of concrete struc- 
tures in Europe. This is particularly true as to the foremen 
and labor bosses who, in many cases, especially in Germany, are 
technically-trained men, which is of course unusual in this country. 

Another feature which tends to increased efficiency in the 
erection of concrete structures is the fact that there are govern- 
mental regulations which apply with sufficient rigidity to terri- 
tory outside of the large cities. In large cities the regulations are 
necessarily much more rigid. There is a wholesome respect for 
the law throughout Europe — ^which is lacking in this country — 
with the result that each person concerned conscientiously en- 
deavors to erect the structure in full accordance with the building 
regulations. In some countries, especially Germany, a contractor 
who is found guilty of dishonest practices loses caste and becomes 
discredited, which is, after all, the most effective way of pre- 
venting the construction of dishonest structures. I recall par- 
ticularly a case in Stuttgart where what might properly be called 
a "Quantity" engineer, having assumed responsibility for the 
erection of the building, was arrested and sentenced to several 
years imprisonment by reason of the collapse of the structure of 



Digitized by 



Google 



Annual Address by the President. 35 

which he had assumed the responsibility. Another effective way 
of dealing with this subject is practiced in France, especially in 
Paris, where the building department acts merely as a custodian 
of plans; the law placing the responsibility for the structure on 
the architect, contractor and owner. Under the law, these three 
parties are held to be responsible for the collapse of the structure 
until their innocence is established. The use of cheap labor, 
especially women, to carry mortar and concrete in small tubs on 
the head, seems to be the usual method but hardly an effective 
one for handling concrete. In the erection of but two buildings 
did I see used the elevating machine, so commonly used in this 
country, for handling concrete. 

The presence of a large number of women laborers on con- 
crete structures was always a source of interest to me as was 
also the pittance that these women were paid for a day's work. 
A recent lecturer in New York stated that in African hunting 
expeditions the camp followers received about sixty cents a month. 
While these women do not receive so little, yet when you con- 
sider that in the one case the men were furnished their food, 
while in the other, the women had to supply it themselves, the 
wage paid the women (in some countries equal to 16 cents and as 
low as 12J cents a day) is extraordinarily low and you can appre- 
ciate why elevating and conveying machinery is doubtless more 
expensive than labor. The almost imiversal limit of about 5 
stories or 22 metres in the height of all structures renders ele- 
vating and conveying machinery relatively unimportant. When, 
however, speed in the erection of concrete structures becomes 
important, Europe will be obliged to resort to mechanical means 
for elevating and conveying concrete, as the method of carrying 
concrete in tubs on the heads of the laborers is too slow for proper 
continuous placing. 

Fig. 2 is a view of the memorial church at Berndorf, the 
industrial village of the Krupp works, just outside of Vienna, 
in which will be seen a nimiber of the women laborers who are 
engaged in carrying mortar and concrete in the manner above 
described. These women, in spite of their skirts, are able to climb 
ladders with almost the same speed as men. 

The use of round timbers instead of sawed, as studs for forms 
and scaffolding, is quite general. Where a splice is necessar>'^, 
the two parts are tied together with rope or chain. They do 



Digitized by 



Google 



38 



AN^UAL^ApbRESS Bt I'&E t^RESlfiENT^. 



not use nails, therefore, these timbers can be used over and over 
again and last for such a length of time as to make the cost rela- 
tively very low. Another interesting construction detail is the 
method of splicing uprights through the medium of two iron 
rectangular iron bands siurounding the uprights, which are clamped 
tightly together by wedges between the inside of the band and the 
face of one of the uprights, which is a very simple and readily 
adjustable device. This particular device was used on the props 
in the construction of the Commercial Museum in Vienna. 

The development in the use of reinforced concrete telephone 




na. 2. — DOME MEMORIAL CHURCH IN BERNDORF, AUSTRIA. 

and telegraph poles is even greater than it was on my previous 
visit and the experience gained has brought forth many orna- 
mental and efficient poles; the tendency for purposes of economy 
is towards a hollow pole, this being of greater necessity in Europe 
than in America; and in my judgment the cost of the concrete 
pole must be materially reduced, to effectually compete with the 
wooden pole in this country. The view shown (Fig. 3) of the 
centrifically molded circular poles in Bad Kosen indicates the 
uniformity and symmetrical shape of this pole. This circular 
form, however, is not necessary, and there are many hollow poles 
of ornamental character of square or octagonal design, notably 



Digitized by 



Google 



Annual Address by the President. 



37 



four in Dresden, in the Exposition of Hygiene. The ornamental 
poles on the Augusta Bridge in Dresden further exemplify the 
beauty of this type of pole, the hollow interior affording an ideal 
place for running the electrical wires. 

The method on making the hollow reinforced concrete pole 




FIG. 3. CENTRIFUGALLY MOLDED CIRCULAR TOWN ELECTRIC LIGHT POLES, 

IN BAD KOSEN, GERMANY. 

consists in placing the mold filled with concrete in a machine 
rotated at the rate of 600 revolutions per minute. The effect of* 
this high speed is to force the concrete against the walls of the 
mold by centrificial action, gradually compacting the concrete 
and forming a hollow space in the center of the pole in which 



Digitized by 



Google 



38 Annual Address by the President. 

the excess water and laitance gathers producing an exterior sur- 
face of a hard and uniform texture which greatly adds to the 
appearance of the pole. The molds are generally kept on the 
pole until properly hardened. 

The poles erected in connection with the Danish railways in 
Copenhagen, Denmark, are hollow, 36 ft. high, and were molded 
by the hand process and the tests showed them to be very stiff and 
capable of resisting high loads. 

It has been found in Europe, especially where lumber is 
becoming scarce, that reinforced concrete poles are much more 
efficient and less costly than wooden poles; that the maintenance 
of the line is less expensive and the permanent life of the pole 
much greater. It appears to me to be evident that in order to 
effect the desired economy in the cost of manufacture of this 
class of concrete products, it is necessary to turn out a great 
many each day; inasmuch as they are coming into general use 
in Europe, there is a constant decrease in the cost of manufacture. 

The poles illustrated in Fig. 4 are those at the plant of R. 
Wolle in Leipsic, Germany. There seems to be a general use for 
concrete poles for carrying high-tension lines, especially where 
the pole must be of considerable height. It is claimed that the 
cost of maintenance of such lines is very much less than for 
wooden poles. This probably accounts for their popularity. 

The continued use of reinforced concrete poles in this country 
leads to the belief that as the number of poles and skill in making 
them increases they will become more serious competitors of the 
wooden pole and will in time replace the other forms of telephone, 
telegraph and electric transmission poles. They can be molded 
to suit the particular conditions of almost any height and can be 
so anchored in the ground as to enable them to maintain a rigid 
position in almost all soils. The high tension transmission line 
poles of reinforced concrete, used in connection with the Penn- 
sylvania Railroad tunnels,* which were erected on a mattress in 
the marshy land of the approaches on the New Jersey side, are an 
illustration of the superior excellence of this type of pole. 

Another matter which was of considerable interest to me 
was the form of chimney developed by Captain F. Mohl in Copen- 
hagen. This consists of a four-leaf-clover section at the base 

♦See rroc, N.A.C.U., Vol. VIII, p. 769, 



Digitized by 



Google 



Annual Address by the President. 



39 



which gradually merges mto a circular at the top. A num- 
ber of these chimneys have been erected, some of considerable 
height, and it is said that they are more economical and much 
more suitable than the ordinary circular stack. 

The reinforced concrete barges shown in the illustration were 
photographed (Fig. 5) at the place of manufacture in Livomo, 
Italy. These particular barges were used for handling coal and 
seemed to be in every way thoroughly satisfactory. When the 
Italian government first used reinforced concrete for armor plate, 




FIG. 4. — TYPES OF REINFORCED CONCRETE POLES AT PLANT OF R. WOLLE, 

LEIPSIC, GERMANY. 

there was much amusement manifested in this country and the 
average person believed that the weight of this material would 
sink the boat. Its use, therefore, to form the entire hull of a 
boat would seem even more quixotic. Barges and other vessels, 
especially battleships, are made entirely of steel, which is heavier 
than concrete; and when you consider that the floating of the 
vessel is a question of buoyancy depending on the lightness of 
the material and character of the air-tight compartments, it is 
evident, I think, that any material properly designed will b^ 



Digitized by 



Google 



40 



Annual Address by the President. 



suflSciently buoyant for practical purposes. In the case of the 
armament of concrete, which is much lighter than steel, the thick- 
ness can be much greater and the toughness of the former material 
renders it a better protection. I believe with the development 
of the art of reinforced concrete boat construction, these vessels 
will in the future come into general use and will prove most ser- 
viceable and economical, both as to first cost, maintenance and 
durability. 

It is not so very many years ago that the concrete barge was 
a novelty and regarded by many as a freak application of cement. 
However, a few years' trial of these boats has resulted in striking 
economy and in many places I found concrete barges being used. 






FIG. 5. — REINFORCED CONCRETE BARGE FACTORY AT LIVORNO, ITALY. 

especially of the canal boat type, for handling coal and other 
materials. It has been found that the durability and serviceability 
of these boats render their ultimate cost very much less than 
boats made of any other material. There have been a number 
of such boats used in this country, notably in connection with 
the construction of the Panama Canal, and in my judgment 
there will be an even greater use of them in the future. 

The constant study of the reinforced concrete railroad tie 
(with somewhat disappointing results at the present time) shows 
a desire for a tie of this type. In parts of Europe where steel or 
wooden ties are readily obtained at reasonable cost, the concrete 
tie does not make much headway. In other parts where ties of 



Digitized by 



Google 



Annual Address by the President. 



41 



timber and steel are expensive, for instance in Italy, where the 
steel and wood tie is at least as expensive as the concrete, there 
has been considerable development and the officials informed me 
that more than 300,000 were in use in the Italian railways and 
that there were contracts for upwards of a million. The tie, 
however, has not a very great life when used in main line ser- 
vice, where it is subjected to the frequent passage of heavy loco- 
motives. In what are termed secondary lines and sidings, ties 




FIG. 6. — ^REINFORCED CONCRETE TIES IN MAIN LINE ITALIAN RAILWAY AT 
PORTO NACCIO, ITALY. 

of reinforced concrete are reasonably effective and in Italy have 
as much as six or seven years of life. 

I inspected some railroad ties just outside of Rome and 
foimd that these ties (see Fig. 6), which had been in service for 
about two years, were not wearing very well. A number of ties 
having crushed just inside the rail. 

The method of fastening the tie consists of the use of a 
wooden block, cylindrical in shape, which is driven into the hole 
molded in the tie and au ordinary wood screw which fastens 



Digitized by 



Google 



42 



Annual Address by the President. 



the rail to the tie. When the threads in the block are worn by 
use, the block is replaced. 

Fig. 7 is an illustration of the section of track of the London 
and Southeastern Railway at Knockholt station in which rein- 
forced concrete ties have been used. These ties, I think, are not 
of as good design as those used in the Italian railways. They 
have been in service about two years. A number of them have 
failed in the manner shown in the illustration (Fig. 8) by the 
concrete crushing just inside the rail. It is my opinion that the 
difficulty in reinforced concrete ties is in a lack of proper analysis 




FIG. 7. — REINFORCED CONCRETE TIES IN SECTION OF LONDON AND SOUTHEASTERN 
RAILWAY AT KNOCKHOLT STATION. 

of the stresses, and that a tie could be so designed as to properly 
care for these stresses and thus prevent the breaking down of the 
tie in the manner just referred to. With this point cared for the 
life of the tie would be greatly prolonged. 

It was the universal opinion of track men that through the 
use of reinforced concrete ties the cost of maintenance could be 
materially reduced and the alignment of the track much more 
readily maintained. It is, however, on curves that the ties are 
least effective and their Ufe very brief. Another objection seems 
to be that the use of the concrete tie usuallj^' results in a rigiditjr 



Digitized by 



Google 



Annual Address by the President. 



43 



of roadbed which is extremely undesirable from an operating 
point of view, with the result that many devices have been tried 
with a view to introducing some elastic medium which will absorb 
the shock of passing locomotives. 

At the Exposition in Turin were shown a numl^er of rein- 
forced concrete ties with wood blocks, fiber cushions, and other 
similar elastic shock al)sorbers, placed in the tie with a view to 
increasing its life. In most cases the design of the tie seemed 




FIG. 8. — MANNER OF FAILURE OF REINFORCED CONCRETE TIES IN LONDON AND 
SOUTHEASTERN RAILWAY. 

to be at fault; many of them had been developed by mechanics 
not skilled in structural designing, with the result that the rein- 
forcement was not properly placed with regard to the stresses, 
especially those of impact; and it appears to me that the con- 
crete tie problem can only be solved with a due consideration 
for these stresses. 

Every one of the railroad officials who has had to do with 
concrete ties in Europe feels sure that a tie will be developed 
which will overcome the objections above indicated and redyc^ 



Digitized by 



Google 



44 Annual Address by the President. 

the initial cost to a point where its life and reduced cost of main- 
tenance will make it the cheapest railroad tie. 

The objection in this country to the reinforced concrete tie, 
namely, its rigidity under moving loads and the consequent objec- 
tionable hammering to the locomotive, can be cared for by the 
introduction of an elastic cushion in the shape of fiber or wood. 
I believe the concrete tie will be the tie of the future and its first 
cost will prove of minor consideration when the reduced cost of 
maintenance and durability is considered. 

Another interesting development in the use of concrete is in 
connection with sewers or conduits of circular or elliptical form 
in which the walls are made up of segments of concrete, which 
after being placed in position are grouted or cemented together, 
forming a solid ring. There were a number of cases where unusual 
economy had been effected through the use of these sewers. In 
some of them the segments were grooved along the axis and 
around the circumference in which the reinforcement was placed. 

The use of reinforced concrete for the lining of sewers and 
tunnels, in my opinion, is a very important application. The 
possibility of molding these blocks and placing them in position, 
and filling the space between the roof and the ring with con- 
crete forms a very simple and effective method of tunnel lining, 
the application of which is cheaper than brick and also cheaper 
than concrete construction where tight forms must be constructed 
and maintained in position until the concrete has properly hard- 
ened. In the segmental method, with the completion of the ring, 
the concrete backing may be readily placed in position and but 
little shoring will be necessary until the concrete has set. 

Another extremely interesting matter was the use of rein- 
forced concrete pipe of varying lengths which was laid as shown 
in Fig. 9 to conform to the general contour of the ground, the 
connections and adjustment being effected by means of loose 
sleeves slipped aroimd the joints. After the pipes and sleeves 
are in position the spaces between the sleeves are grouted solidly 
to the pipe This appUcation is in advance of anything we are 
doing in this country. Sections of reinforced concrete pipe laid 
in this manner, in lengths of 12 feet or more, would have great 
possibilities for use in pressure lines. The pipe in the illustration 
were made in a machine not unlike a lathe. The reinforcement 



Digitized by 



Google 



Annual Address by the President. 



45 



Was placed around a core rotating horizontally and a very stiff 
mortar thrown against it. After the mortar had been built out 
to the requisite thickness, the pipe was wrapped with cotton 
bands four or five inches in width. The pipe was then removed 
and the wrapping kept wet until the mortar had set, when the 







1 % 




>.2 "^i 


A.4. .' 



FIG. 9.- 



-REINFORCED CONCRETE PIPE OF VARYING LENGTHS IN USE IN 
SWITZERLAND. 



bands were pulled off. This method of construction naturally 
requires a great deal of labor and even if considered desirable 
would be entirely too expensive for use in this country. 

Fig. 10 shows a cement products yard in Vienna and par- 
ticularly illustrates the concrete ducts which are being used in 



Digitized by 



Google 



46 



Annual Address by the President. 



that city. These ducts have longitudinal recesses in which rein* 
forcement is placed and the remaining space filled with cement 
mortar; this stiffens the ducts and holds them in .position. It 
seemed to me there was an excellent opportunity for this type 
of construction in this country. When properly cemented in 
position, the ducts formed were in every way desirable and I am 
told that the cost is much less than that of terra cotta or other 
material. 

During my last visit in Vienna I had occasion to refer to the 
use of concrete pavements and. I find that this type is coming 




FIG. 10. — CEMENT PRODUCTS PI/ANT OF PITTEL AND BRAUSEWETTER AND 
E. GAERTNER IN VIENNA, AUSTRIA. 

into gradual use. The pavement inspected on my previous visit 
and reported to you two years ago had been laid four or five 
years and during my last visit I found these concrete roadways 
still in excellent condition and that they had not been repaired. 
The city of Vienna was engaged in laying considerable yardage 
of these pavements, especially around the Royal Palace. This 
pavement was of great width and was entirely of concrete. I am 
absolutely convinced of the durability of these pavements and 
repeat what I have stated in my previous address, that I believe 
they will come into general use. 



Digitized by 



Google 



Annual Address by the President. 



47 



The tank of reinforced concrete in use at the coal mines, 
in Rotherham, England, to remove the coal dust from the water 
coming from the coal washery is of considerable interest. This 
water is pumped into the tank, the coal dust is removed and 
the clean water returned to the washery. This coal dust is used 
in briquetting coal. Steel tanks are not available for this purpose 
because the sulphur in the coal would so corrode the steel as to 
make its life extremely short. The concrete tank has proved 
highly satisfactory not only in its resistance to the action of 
sulphur but in its water-tightness. 

I commented on the artistic use of reinforced concrete in 
bridge construction in my previous address, but the matter is 
so striking that I cannot help referring to it again. The innum- 




FIG. 11. — BIEMORIAL REINFOKCED CONCRETE BRIDGE CONNECTING HISTORICAL 
AND ART EXPOSITIONS OVER TIBER, IN ROME, ITALY. 

erable bridges throughout Europe, carefully designed, carrying 
railroad as well as ordinary highway traffic, are monuments to 
the ability of the European engineer. The designs are for the 
most part graceful and show a wide diversity in artistic treatment. 

A structure of great beauty is the Memorial Bridge (Fig. 11) 
in Rome connecting the Historical and Art Expositions, located on 
opposite sides of the Tiber. This bridge was built by the Henne- 
bique Construction Company and is an example of European 
engineering skill. By reason of floods, it was necessary to con- 
struct the centering on which this bridge was built, of reinforced 
concrete — ^which is unusual and the first application of the use 
of reinforced concrete for centering that has come to my at- 
tention. 

The bridge has a span of 100 meters and is built in imitation 



Digitized by 



Google 



48 Annual Address by the tRBSil)ENt. 

of the native Travertine stone; the beauty of this is fully as 
great as the artificial Travertine stone to be found in the New 
York station of the Pennsylvania Railroad. This structure was 
built by the Italian government, and many severe tests were 
applied with but extremely slight deflections. 

The manner in which this bridge was tested is also of interest. 
A commission was appointed to conduct the tests for the govern- 
ment. These tests consisted in moving heavy steam rollers and 
marching soldiers in a solid mass over the bridge; particularly 
interesting was the loading of the bridge solidly with soldiers 
and studying the effect of their cadenced step on the structure. 




FIG. 12. — REINFORCED CONCRETE TRUSS 113.8 FT. SPAN USED IN MAIN RAILr 

ROAD STATION IN LEIPSIC, ERECTED FOR TESTING IN COSSEBAUDE, 

NEAR DRESDEN, GERMANY. 

I think it may be stated without fear of contradiction that 
reinforced concrete structures properly designed are less affected 
by vibratory and cadenced loads than any other structure. It 
is particularly noticeable that in structures of identical design 
as to carrying capacity, those of steel show more movement than 
those of reinforced concrete. 

A most interesting structure is the truss (Fig. 12) erected 
initially by the firm of Dyckerhoflf and Widman, at their plant in 
Dresden. It was erected for testing the actual strength and justify- 



Digitized by 



Google 



Annual ADUREgd by fHE President. 4§ 

ing the use of a truss of this tjrpe in the main railroad station in 
Leipsic. The arch was loaded in various ways and the deflections 
observed, and the tests were so satisfactory that it was successfully 
used in the structure. It is unusual in this coimtry to go to the 
expense of erecting a structure of this kind and applying tests in 
order to satisfy the officials that it is amply safe; It would, how- 
ever, be an effective means of preventing failures. 

A matter that impressed me was the development of struc- 
tures, many of them articulated, composed of separately molded 
parts, and the views herein presented exemplify some of the more 
remarkable of these structures. The practice of casting mem- 
bers and then pinning them together and encasing the connection 
in concrete is a development that seems to meet with favor, and 
while this may not appeal to many of the conservative engineers 
of this country, it seems to me that when this pin connection is 
of the same design as pin connected steel members, there is no 
reason why this type of structure, when properly designed, should 
not be more serviceable by reason of the concrete covering which 
makes it more durable than a structure of steel. The concrete 
trestle used in connection with the mine at Floreiffe, Belgium, 
the lower portion of this is used for the county roadway and the 
upper portion for the handling of cars to the "tipple" from the 
coal mine, is an illustration of a remarkable development of 
this principle. I was informed that the parts of this structure 
were separately molded and afterward erected in place in a very 
short space of time and that the cost was considerably less than 
a similar structure of steel or of timber. There is a tendency 
towards systems requiring separately molded members, and it 
seems to me that such systems afford an opportunity for economy. 
The weak points in such structures are the joints, and with the 
same attention to these connections that is given to steel struc- 
tures there is no reason why the joint in a concrete structure 
should not be stronger than the weakest part of the separate 
members. 

A development of the Visintini system was to me somewhat 
surprising, since this system is very little used in this country; 
the early attempts to introduce it were unsuccessful, chiefly 
because of its cost as compared with that of other systems of 
reinforced concrete construction. In Europe it has a wide appli- 



Digitized by 



Google 



50 Annual Address by the President. 

cation for buildings and bridges. Fig. 13 shows the bridge 
being erected in Copenhagen, Denmark, with girders of system 
Visintini. This bridge, erected in connection with the new sta- 
tion that the State Railways were building, carries a street with 
two Hues of electric trams over the railroad. The ornamental 
portion is entirely of molded concrete which was not subsequently 
treated. Where labor is inexpensive the cost of forms is not so 
important as the cost of material. In this country, where condi- 
tions are just the reverse, this system proves uneconomical. The 
bridge consists of deep girders spanned by beams all of the Visin- 
tini system. The latter spaced solidly so as to form a slab, under 
the railroad tracks and 20 inches apart between the tracks. I 




FIG. 13. — HUmWAY BRIDCIK OF REINFORCED CONC^RETE OVER STATE RAILWAY 
TRACKS, COPENHAGEN, DENMARK. 

think this bridge might be said to be a type of unit construction. 
Certainly the bridge, which was nearly completed, presented an 
extremely beautiful, ornamental appearance and far more desir- 
able than a similar structure of steel. I believe that this method 
of construction, which is in a measure illustrated in the flat slab 
construction in use in railroad bridges in this country, possesses 
great possibilities in the matter of appearance and speed of erec- 
tion. All the beams of the system could be molded and the 
structure then erected continuously, replacing perhaps an old 
structure, without interfering with the traffic. 

An interesting type of girder construction is shown in Fig. 14, 
a highway bridge at Desna, Austria, which consists of a series of 
Visintini girders supported on piers of rather unusual construction. 



Digitized by 



Google 



Annual Address by the President. 



51 



The application of this system for arch ribs of a building is illus- 
trated in (Fig. 15) the ceiling of a church erected in Aussig, 
Austria, and which I think is also an application of unit system. 
The lower flange of the rib of the arch is extended so as to support 
the Visintini beams which form the ceiling. Another type of 
structure which shows a trend in Europe that conservative struc- 
tural engineers in this country might call dangerous is illustrated 
Fig. 16. This is a trussed bridge practically a development of 
the Visintini system, which forms a central span, with Visintini 
girders and beams used in the approaches. 




FIG. 14. — REINFORCED CONCRETE HIGHWAY BRIDGE, DESNA, AUSTRIA. 

The photograph, Fig. 17, shows the coal bunkers of the 
extensive plant of the city gas works in Vienna. The one on 
the left side has most of the forms removed and illustrates the 
generally pleasing character of the structure, which is notable in 
view of the fact that it must be massive, in order to carry a large 
quantity of coal. The attempt to render this structure pleasing 
and ornamental could well be emulated by our American designers. 
The exterior surface is dressed with pneumatic tools and a color 
effect has been obtained which is not unlike that attained in the 
construction of the Connecticut Avenue Bridge in Washington. 
D. C. 



Digitized by 



Google 



52 



Annual Address bv thE President* 




no. 15. — PORTION REINFORCED CONCRETE ROOF OF CHURCH, AUSSIG, AUSTRIA. 




FIG. 16. — REINFORCED CONCRETE TRUSS HIGHWAY BRIDGE NEAR VIENNA, 

AUSTRIA. 



Digitized by 



Google 



Annual Address by the PREsroENT. 



53 



We have, of course, many water towers in this country, but 
the thing that most impresses me in connection with those in 
European countries, is the thinness of the tank walls and the 
fact that they are constructed without the aid of waterproofing; 
the density of the concrete and the manner of reinforcing is suf- 
ficient to so distribute the cracks as to render them water-tight. 

In connection with buildings abroad it would seem that more 
attention is given to the exterior and interior finish, even in 
factory buildings, than is given in this country; the skill used in 




FIG. 17. — REINFORCED CONCRETE COAL BUNKERS, CITY GAS WORKS, VIENNA. 

secimng a pleasing finish, even considering the low-priced labor 
of Europe, is very rarely more expensive than the rough finish 
commonly used here. The tendency toward flat slabs and the 
elimination of as many beams as possible shows, I believe, an 
unmistakable turn, which is reflected in this country in the recent 
development of the flat slab type of construction. In many 
places the elimination of beams results in paneled effects through 
the use of girders between the columns and mortising and mold- 
ing the connection between the slab and girder in such a way 
as to produce pleasing ornamental effects. There is also a ten^ 



Digitized by 



Google 



54 



Annual Address by the President. 



dency to panel the slab itself, which relieves its flatness and adds 
no little to the beauty of the interior finish. The use of much 
higher ceilings than are commonly used in this country renders 
flat slabs unnecessary for the distribution of light. The effect 
of the paneling is somewhat that of the Roman barrel arch, and 
this latter type of construction may be seen in the ceiling decora- 
tion of many European structures. 

An excellent exterior finish having the appearance of gray 
stone was to be seen in the C'igarette Paper Factory in Vienna. 
This structure had been erected many years and the weathering 
had been so uniform as not to in any way mar its beauty. This 
building was illustrated in my address describing my trip of two 




iiillri 



111. 
liiUjiiiliii fliirrn 





111 

Mi 

iii 
nil 



FIG. 18. — TRADE SCHOOL OF REINFORCED CONCRETE, VIENNA, AUSTRIA. 

years ago and is again referred to because after an interval of two 
years the exterior finish of the building remains unchanged. The 
entire ornamentation is of concrete and serves as an excellent 
illustration of the artistic possibilities of this material. 

The Trade School in Vienna (Fig. 18) is also an excellent 
example of the artistic treatment of concrete. The paneling and 
ornamental work are most excellent in character, and for this 
reason do not call forth the criticism which once was so rife in 
this country, where the crude structures which we erected left 
much to be desired from the aesthetic point of view. It is fre- 
quently the practice in Europe, especially in Vienna, to cast 
monolithic walls and tool them afterwards, in a manner to pro- 



Digitized by 



Google 



Annual Address by the President. 



55 



duce the effect of stone. This, however, does not appeal to me 
so strongly as the structure in which there is no attempt to 
imitate stone, but where the material is used to stand for what 
it is, producing pleasing ornamentation and a surface uniform 
in color and texture and free from those stains and cracks which 
are frequently seen in structures of concrete in this country. 




FIG. 19. — CITY HALL OF REINFORCED CONCRETE, NEAR BADEN, AUSTRIA. 

Fig. 19 is a view of the City Hall at Weikersdorf near Baden, 
Austria. The entire wall, with ornamentation, is of concrete 
without any attempt to imitate other material, which I think 
illustrates that this is the proper way in which to use concrete 
and that when so used the results are much more effective than 
where it is used in imitation of other materials. 

A favorite form of exterior decoration is to apply phvster 



Digitized by 



Google 



56 



Annual Address by the PREsroENT. 



to the rough concrete and mold it in a manner to produce artistic 
effects. 

The photograph of the Villa Figari in Genoa, Italy, shown 
in Fig. 20, is an illustration of the splendid possibilities in the 
artistic use of reinforced concrete. This graceful, beautiful struc- 
ture needs no comment. The erection of this Villa involved a 
number of interesting problems iVi design which I think were met 




FIG. 20. — VILLA FIGARI, GENOA, ITALY. 

more successfully through the use of this material than would be 
possible through the use of any other. There are a number of 
plants in this country engaged in making ornamental concrete 
products of a very high order, and fully equal to some of the 
best work done in Europe, but the American designer does not at 
the present time seem to appreciate the adaptability of concrete 
for use in the ornamental structural portions of buildings; it 
seems to me that development along the lines illustrated in the 
Villa Figari opens a wonderfully promising field. 



Digitized by 



Google 



Annual Address by the President. 



57 



As the architect and designing engineer more fully realize 
the architectural possibilities of concrete and apply them in build- 
ing construction with a due appreciation for the aesthetic, there 
will result such graceful structures as will entirely meet the 
criticisms and objections to the many cumbersome, ungraceful 
and unattractive structures which are being erected at the 
present time. 

The desire to render even ordinary structures beautiful is 




FIG. 21. — REINFORCED CONCRETE CAR BARNS OF CITY OF VIENNA, AUSTRIA. 

particularly observable in Europe and it is, therefore, unusual 
to see even a mill building in which no attempt has been made 
to give it a proper finish. The eye for the beautiful is a matter 
of education and growth and I presume that in time we will 
develop similar tastes so that our mill buildings will be given 
artistic finishes. It should be borne in mind, however, that the 
cost of this work in Europe is considerably less than in this 
country, by reason of the cheap labor and because of the greater 
quantity that is done; they have acquired more skill and there 



Digitized by 



Google 



58 Annual Address by the President. 

are a great many more skilled men capable of doing this class 
of work. 

The car barns of the city of Vienna, shown in the illustration 
(Fig. 21), is another development in the use of concrete which 
I think will become general in the future. This structure is built 
entirely of concrete, including the roof which has not been water- 
proofed. I understand that the cost of this structure was mate- 
rially less than the cost of other types of construction. The 
span and lines of the girder was a matter of interest and is char- 
acteristic of the skill of the Austrian engineer. 

A building of reinforced concrete which was brought to my 
attention during my visit two years ago, and which was then 
under construction, is the Urania Building in Vienna, which is 
devoted to the development of popular musical education. It 
has a rather difficult problem in acoustics which is of moment 
to those interested in concrete. The main auditorium has a floor 
and ceiling of reinforced concrete of considerable thickness; 
above and below this auditorium are smaller halls for musical 
entertainment. The large auditorium contains an organ, and 
there is no connection between it and the halls above and below, 
yet the Director states that when there is a concert in progress 
in the large auditorium it is impossible to hold a concert in either 
of the other auditoriums, because the sound of the music in the 
large hall is so pronoimced as to seriously interfere with the per- 
formance in the other halls. This subject has received a great 
deal of consideration in Vienna and methods are now in progress 
to eliminate, if possible, by some insulating medium, the trans- 
mission of sound. 

Another matter of interest was the work which the Austrian 
Association is doing in investigations and tests, not unlike those 
of our own Committee on Reinforced Concrete, except that the 
value of the work accomplished is greater because the work is 
much more extensive. They have conducted a comprehensive 
series of tests of columns and beams, the cost of the work being 
defrayed by the Austrian cement manufacturers, and has been 
under the auspices of representatives of the government, the 
Society of Engineers and Architects, the Austrian Concrete Asso- 
ciation and the Austrian Association of Cement Manufacturers. 
The principal feature of the tests which were being conducted at 



Digitized by 



Google 



Annual Address by the President. 



59 



the time of my visit was a study of the effect of the weight of 
wall in restraining the ends of beams; a number of different 
methods were being tried and it had been found that the stiff- 




PIG. 22. — A RESTRAINED REINFORCED CONCRETE BEAM AFTER TESTING. 




FIG. 23. — METHOD OF APPLYING LOAD IN TESTING REINFORCED CONCRETE 

BEAMS. 

noss of the beam was materially increased through the weight 
of the wall. 

Fig. 22 illustrates one of these beams after test. In some 



Digitized by 



Google 



60 Annual Address by the President. 

cases the beam rested on a pier, and in others it extended over; 
in still others it was imbedded in concrete, or laid up in a brick 
wall so as to approximate practical conditions. 

Fig. 23 shows the manner of loading and observing the deflec- 
tions. The method of applying the load is perhaps of interest. 
The load to be apphed was carried by hydraulic jacks, the lower- 
ing of which brought the weight of this super-imposed load on 
the beam. 

One thing that was apparent on my recent trip, and which 
must be a matter of great gratification to every member, is the 
high regard in which this Association and its work are held in 
Europe. It is taken generally as a model, and in many cases 
the character of the work done here and the value of its pro- 
ceedings and discussions are so appreciated as to result in a 
number of Europeans becoming members in order to secure its 
publications. 

There is also a cordial feeling of cooperative good-will between 
the various concrete organizations of Europe and our Association, 
which I hope may be fostered and that it may be possible through 
the interchange of delegates, papers, and in other ways, to extend 
this cordiality so as to obtain the advantages of the development 
in the art of concrete construction in various parts of the world. 

This Association stands for education in the development 
of the proper use of concrete, and certainly it should be a part 
of its policy to encourage international co-operation, to the end 
that the development in this country may proceed with a full 
knowledge of what is being done by our foreign competitors. 



Digitized by 



Google 



REPORT OF THE COMMITTEE ON REINFORCED 
CONCRETE AND BUILDING LAWS. 

In accordance with the program of tests on completed 
stnictures presented to the Association at the last convention, 
four tests have been made, two of them on standard forms of 
reinforced concrete floors in which the concrete slab is supported 
by one or more intermediate beams which in turn are framed 
into or supported by girders carrying the load to the columns. 
Of the other two tests one is on the girderless or flat slab floor, 
and the other on a combination of tile and concrete floor, in which 
the reinforcement is placed in two directions. 

Making arrangements for the tests* and computing the 
results have taken up all the available time of the Conmiittee; 
the test data is therefore presented to the Association without 
any attempt at generalization. 

The tests on the Wenalden and the Turner-Carter buildings 
were made under the direct supervision of A. N. Talbot; the 
other tests were made by F. J. Trelease of the Research Depart- 
ment of the Corrugated Bar Company, with the assistance of 
W. A. Slater of the Illinois Engineering Experiment Station. 

Part I. Tests op Two Reinforced Concrete Buildings 
OF the Beam and Girder Type. 

Preliminary, — These tests were undertaken for the purpose 
of obtaining information on the action of the composite structure 
of concrete and steel under load in a reinforced concrete building 
constructed under the usual conditions of work. Many tests 
have been made of separate reinforced concrete members, but 
little attention has been given to the measurement of stresses and 
deformations in the completed building and to the determination 
of their actual amount and distribution and of the effect of one 
part of the structure upon another. Load-deflection tests are 
common and are of value in judging of the quality of workman- 
ship and in giving confidence in the structure, but they throw 
littie light on the stresses developed in the different parts or 

(61) 



Digitized by 



Google 



62 Report of Committee on Reinforced Concrete. 

upon their distribution. A variety of views have been advanced 
on the relation between the bending moment at a section at the 
support and that at the middle of the beam, on the amount of 
arch action which may be developed in the structure, on the dis- 
tribution of the stresses across a floor slab acting as the flange 
of a T-beam,»on the restraint of girders and beams, etc. These 
tests were undertaken in an effort to obtain some information 




FIG. 1. — INSTRUMENTS AND TOOLS. 

on such matters as well as to find something of the general action 
of reinforced concrete structures as a whole. 

The general method of test followed the plan outlined by 
Arthur R. Lord in the paper, A Test of a Flat Slab Floor in a 
Reinforced Concrete Bxdlding* presented at the New York Con- 
vention. Holes were cut in the concrete until the reinforcing 
bars were bared. Gauge holes were then drilled in these bars, 
at distances apart to give the proper gauge lengths. Where 
measurements of deformation of the concrete were desired, holes 
were cut in the concrete and a metal plug inserted in which the 

♦ See Proceeding, Vol. VII, p. 156. 



Digitized by 



Google 



Report op Committee on Reinforced Concrete. 63 

gauge holes were later drilled. These gauge lines were selected 
in places where it was thought that critical stresses would be 
determined. In some places for one reason or another the rein- 
forcing bars were inaccessible and it was impracticable to obtain 
measurements to give information which would have been of 
interest. In some cases a series of gauge lines were used to de- 



/>-o/T» C rt> //." 




Central ho/^ 



Eccentric hole 

Finishing 
tooL 

FIG. 2. — SHOWING EXTEN80METER AND GAUGE HOLES. 



termine the change of stress or distribution from one point to 
another as at the end of a restrained beam and across the floor 
slab between beams. 

The measurements were made by means of Berry exten- 
someters of the form developed at the University of Illinois. The 
extensometer is shown at the bottom of Fig. 1. The instrument 



Digitized by 



Google 



64 Repor* of Committers on Reinforc*sd CoNCRETti. 

reads to Wajf in. and is estimated to TJi^nis in. Its make-up is 
shown in Fig. 2, as are the gauge holes. In making a measurement 
the legs of the instrument are inserted in the gauge holes, a read- 
ing taken, the instrument taken out and again inserted and read, 
and this proceeding repeated until a number of readings without 
serious discrepancies are found. The operation of making a 




FIG. 3. — TAKING AN OBSERVATION. 

measurement is shown in Figs. 3 and 4. The instrument is of 
a simple character, but its use requires unusual care and skill on 
the part of the manipulator. The method of using the instrument 
as well as the necessary general conditions attending such tests 
are comprehensively discussed in the paper presented at this 
convention by W. A. Slater on Tests of Reinforced Concrete Build- 
ings under Load.* 

Acknowledgment. — ^These tests were undertaken through the 

* See page 168.— Ed. 



Digitized by 



Google 



Beport of Committee on Reinforced Concrete. 



65 



eflforts of the Committee on Reinforced Concrete and BuUding 
Laws of the Association, in co-operation with the Engineering 
Experiment Station of the University of Illinois. The money 
to defray the expense of the test was arranged for by the Presi- 
dent and Treasurer of the Association. The contractors for the 
two buildings co-operated in the tests. The technical part of 
making the tests was done by members of the staff of the Engineer- 
ing Experiment Station of the University of Illinois. 




PIG. 4. — TAIONG AN OBSBBVATION. 

Comments, — ^A few words on the basis and limitations of 
such tests may not be out of place. The measurements and ob- 
servations are subject to some uncertainty; they are not exact 
or precise — some erratic readings must be expected. The measur- 
ing instrument is used under unfavorable conditions. The gauge 
holes are deep in the concrete and the measurements may be 
interfered with by dust or other obstructing matter. Great care 
and much skill is necessary in making observations. Each test 
of this kind made has shown advances in accuracy and certainty^ 



Digitized by 



Google 



66 Report of Committee on Reinforced Concrete. 

and further experience ought to show further progress. It must 
be understood that the structure itself is not entirely homogeneous 
and that all parts of it do not act alike. Further, the structure 
itself is tied together so closely that stress in one portion may be 
modified or assisted by another portion which may not be thought 
to afifect it, and this in an unknown amount. The modulus of 
elasticity of the concrete in the structure may not be known. 
The load-deformation lines may be irregular and imperfect. This 



r 



3S £2 05 OQ SS EH ^ fin ir 
r^l2ie!I62I!SS!10SB!!Ti3 

^ 12 EB SE SI fSi SS BH !S! SQ EH! IS It 
■*2?i £2 URUBtSi EESfSSiS! EH !!!!& ^ 
fTT 177? TTTj mi oil iin im fiii fin iiii if ji 

^n pti« Iff if fffp pfF IB" fvn im nn M I ^ 




FIG. 5. — THE WENALDEN BUILDING. 

all means that care must be taken in the interpretation of results. 
Important information will be brought out by such tests, as these 
tests show, and tests of special features of construction and an 
accumulation of data on the action of the structure as a whole 
will be worth many times the cost of the work. 

Wenalden Building Test. 

Building, — The Wenalden Building, Fig. 5, is a ten-story 
reinforced concrete structure at 18th and Lumber Streets, Chicago. 



Digitized by 



Google 



Report of Committee on Reinforced Concrete. 67 

It was built by the Ferro-Concrete Construction Company, 
Cincinnati, Ohio, in accordance with the plans and specifications 
of Howard Chapman, architect. It is now occupied by Carson, 
Pirie, Scott and Company, dry goods merchants, as a warehouse. 




Elevation of Intermediate Beam. 




Elevation of Column Beam. 





ri^iri^^:i^4;!a 



View of Girder. 

Fia. 6. — GENERAL POSITION OP REINFORCEMENT. 

The building is of the beam and girder type. The floor panels 
are 15 ft. by 20 ft. The girders are placed between columns in 
the short direction. Floor beams extend the long way of the panel, 
there being two intermediate beams built into and supported by 
the girders and a column beam built into and supported by the 



Digitized by 



Google 



84 Report of Committee on. Reinforced Concrete. 







if^.,4,ij^fer-^',^^-,4^^ 



II* 



// 



k 



I! 



I 



I 



'I I 
r 



^.1 



8L-— L-.^ ^-— IJ 



□ 



in 



i 



^ij u u — ' — u u u ir^ 

FIG. 17.— PLAN SHOWING LOCATION OF GAUGE UNES ON UNDER SIDE OF FLOOR. 



Digitized by 



Google 



Report op Committee on Reinforced Concrete. 85 



1 



i 11 



i^ 1^ ' *A a44a ^^a: ■» j*»^ 



^ a 



I 



I 



rn 



S"?! 



rt 






^ 



jAV 




— u u — 

FIO 18. — ^PLAN 8H0WIN0 



— u u u u \r^ 

LOCATION OF GAUGE UNES ON X7PPBB BIDE OF FLOOB. 



Digitized by 



Google 



86 Report of Committee on Reinforced Concrete. 

Preparation for the Test — ^A week was used in preparing for 
the test. Platforms supported by scafifolding for the use of 
observers were built on the second floor. Independent of this 
was a framework, which was supported by the second floor, for 
use in making measurement of deflections. The boxes for hold- 
ing the sand were constructed, this being facilitated by a power 
saw located on the second floor. Considerable time was con- 
sumed in drilling holes in the concrete to bare the reinforcement. 







p^J^^^^J^^u, 



I L 






-!fr-J - 






FIG. 19. — LOCATION OF SAND BOXES AND FLOOR CRACKS. 

In some cases this was found to be at a considerable depth from 
the surface. In all nearly two hundred holes were cut in the con- 
crete. Holes were drilled in the reinforcing bars, as heretofore 
described, for use as gauge points. The gauge length was made 
8 in. The position of the gauge lines for the reinforcing bare 
is shown on Figs. 17 and 18 by the even numbers. For use in the 
measurement of deformations of the concrete, holes about \ in. 
in diameter and 1 in. deep were drilled in the concrete and steel 



Digitized by 



Google 



Report of Committee on Reinforced Concrete. 87 




FIG. 20.— VIEW OP SAND BOXES. 




ip--»'»f 



FIG. 21. — ^VIEW OF TEST LOAD IN TURNER-CARTER BUILDING. 



Digitized by 



Google 



88 Rbpokt op Committee on Reinfobced Concrete. 




^00 
None ^ 



I 

O ^ Ok ^ ^ ^ ^ _. . _ ^ _, 



%%%%%_%%%%% 



no. 22. — LOAD DEFORMATION DIAGRAMS FOR UNDER BIDE OF BEABC8 AT END. 



Digitized by 



Google 



Report of Couuhtee on Reinforced Concrete. 89 



Ahne 




Do/or-ma'f'lon p^rUnti' o, 



^ => § 5 ^ S S 
i ^ S § § § 




no. 23.— LOAD DEFORMATION DIAGRAMS FOR UNDER SIDE OF BEAMS AT END. 



Digitized by 



Google 



90 Report op Committee on Reinforced Concrete. 

plugs were inserted and set in plaster of Paris. Gauge holes 
for receiving the points of the extensometers were drilled in these 
plugs with a No. 54 drill. The position of the gauge lines is shown 
in Figs. 17 and 18 by the odd numbers. The gauge length was 
Sin. 

The deflections were measured between a steel ball set in 
the under surface of the beam and a ball attached to the frame- 
work previously described. The measurements were made by 
means of the micrometer shown in Fig. 1. 

Method of Loading. — The test area was on the third floor. 
The loading material was damp sand which was placed in bot- 
tomless boxes. These boxes were of various sizes and were placed 
in such a way as to give a well distributed load. The general 
size of the box was 4 ft. 6 in. wide, 8 ft. long and 4 ft. 6 in. deep. 
Fig. 19 shows the position of the boxes and the test area. Fig. 
20 is a view with the sand boxes ready for loading. The boxes 
were made small enough to permit a good distribution of load 
even though part of the weight of the sand might be carried by 
arching and friction down the sides. The test area covered three 
full panels and parts of four others, in all equivalent to five panels. 
A loading space was chosen which it was thought would give the 
fullest stresses over the girders and beams on which the principal 
measurements were made. In removing the load the outer panels 
were unloaded first in an attempt to determine the relation be- 
tween single panel loading and group loading. The load applied 
was the equivalent of 300 lb. per sq. ft., double the design live 
load. 

Before beginning the test, a calibration of the heaviness of 
the sand was made by weighing the sand which had been shoveled 
into a box of 16 cu. ft. capacity placed on the scales. It was 
found that there was a difference of about 10 per cent, in the weight 
of sand which had been thrown in loosely and sand which was 
packed somewhat. During unloading, the entire contents of 
three of the sand boxes (about 500 cu. ft.) were weighed. This 
gave an average of 88.6 lb. per cu. ft., agreeing closely with the 
weights of the unpacked sand previously weighed, and this value 
was used in the calculation of loads. 

On a part of the area where the boxes were not carried to 
a suflicient height and where the space was not covered adequately 
by them, cement in sacks was used as loading material. 



Digitized by 



Google 



Report of Committee on Reinforced Concrete. 



91 



The supply of sand for the loading had previously been de- 
livered on the same floor, the piles being kept at least one panel 
away from the location of the test area, and this was distributed 
over sufficient floor space that the stresses in the beams of the 
test area could not be affected. In applying the load the sand 
was wheeled in barrows and dumped into the boxes. As the 
sand was placed, the sides of the boxes were rapped to break the 
adhesion of the sand. Some leveling of the sand in the boxes 



Abne 

None ^ 

^aoo 






I 



^200 







125 



Qi 15 JVi 



I 



£i}L 



\ 



\ 



ZM 



^| 



2J3 




FIG. 24. — ^LOAD DEFORMATION DIAGRAMS FOR UNDER SIDE OF BEAMS AT END. 

was done, but there was little compacting by tramping or other- 
wise. 

Making the Test, — A very important element of a test of 
this kind is the initial observation for fixing the zero point of the 
test readings. Three sets of observations for a number of gauge 
lines were made before the beginning of the test, on the after- 
noon of September 10 and the forenoon of September 11. 
Where discrepancies were found new observations were made. 
Even with this number of observations there are uncertainties 



Digitized by 



Google 



92 Report of Committee on Reinforced Concrete. 



Mono 



zSOO 







^00 



£ 
\AII ^200 

\aii \/oo 






0^ 



^^00 

^ )EFHt \dO0 

\Af/ ^2O0 

v%w'^ 

£ ^300 

\£diH^300 

\KrHi ^00 

^All ^300 

%An ^200 
5 






3*4 



31 Z 



\ 




Y 



z: 



^ 



3ve 



1 



390 




3U f'C 



^ S § 



^ ^ ^ 



n 




§ 



JC€ 




§ ^ « > - « « ^ 




3f€ 



t-. 



5 



&« b 



X 
B 



FIG. 25. — ^LOAD DEFORMATION DIAGRAMS FOR UPPER SIDE OF BEAMS AT END. 



Digitized by 



Google 



Repokt op Committee on Reinfokced Concrete. 93 

in some initial readings. Experience confirms the view that 
before any load is placed the initial readings which have been 
taken should be worked up and observations repeated untiT all 
discrepancies and uncertainties have been removed. 

Readings were taken immediately after the completion of each 



if ^00 

^>^// %200 
£ ^^OO 

£ ^300 
\£6iH^^300 

\defhi ^O 



'AH 



^AH ^2tt? 




nQ.26. — ^LOAD DEFORMATION DIAGRAMS FOR UNDER SIDE OF BEAMS AT MIDDLE. 

increment of load and again immediately before the beginning 
of placing another increment of load. This usually corresponded 
with evening readings and morning readings. A series of readings 
was also taken with the full test load on. These extended over a 



Digitized by 



Google 



94 Report of Committee on Reinforced Concrete. 



^A// <300 

M>n9^ 

£ ^300 

^£0LH^300 

^>f// ^300 
^An ^20O 



JO J 



X 5 « ;^ 



S/S 




307 



§ ^ 



3?S 



305\ 



30.' 







^ 15 JVi > S> 
^ d ^ d ^ 



§§ 



FIO. 27. — LOAD DEFORMATION DIAGRAMS FOR UPPER SIDE OF BEAMS ATBODDLI. 



£ ^00 
^£AH^O 

%An '^200 




§ S <S 



a S g 5 g S § 

o C? C5 ^ ^ c[ c> 

FIG. 28. — LOAD DEFORMATION DIAGRAMS FOR UNDER SIDE OF GIRDERS AT END. 



Digitized by 



Google 



Repobt of Committee on Reinforced Concrete. 



96 




Abne^ 0, 






FIG. 29. — ^LOAD DEFORMATION DIAGRAMS FOR UPPER SIDE AND UNDER SIDE OF 

GIRDERS AT MIDDLE. 






aoo 



^A// 



A// voo 
Uorm 



£ ^300 
\DEFm 



^t 



%A!I [^20? 

^A/l ^00 

Nfme^ Or 



§ § ^ 



/£»/' 



3: 



T 



I 



/ZO^ 



/Z7S 



;s^ <\| K^ ^ >^ CVl 
§ S ^ § ^ ^ 



t 



I 



/a 03 



I 



/k*/S 






I S * 



FIG.30. — LOAD DEFORMATION DIAGRAMS FOR CONCRETE ON UNDER SIDE OF SLAB. 



Digitized by 



Google 



96 Report of Committee on Reinforced Concrete. 



TaBLB IV. — SCHBDULE OF LOADING OPERATIONS IN TuRNBR-CaRTBB 

Building Test 
Loading Schedule, 



Day. 



Sunday. . . . 
Monday... 
TuMday... 
Wednesday 

ThUTBday.., 



Date. 



9-10-11 



0-11-11 



9-12-11 



9-13-11 



9-14-11 



Obeervationa. 



Load, 
lb. per 
sq.ft. 



100 



200 



300 



Hours. 



lb. per 
sq.ft. 



12 m 

to 

2 p. M. 

7.20 

to 

12 m. 

e.30 

to 

8.16 A. M. 

6.20 

to 

8.20 A. M. 



100 



200 



300 



8.00 

to 

8.30 A. M. 



I 



ling. 


ObservaUons. 




Load, 




Hours. 


lb. per 
sq. ft. 


Hours. 


1.30 


100 


6.10 


to 




to 


6.00 p. M. 




8.00 p. M. 


10.30 A. M. 


200 


3.10 


to 




to 


3.00 p. M. 




6.30 p. M. 


9.00 A. M. 


300 


3.60 


to 


800 


to 


3.30 p. M. 




6.80 p. M. 

1030 

to 

11.30 p.m. 




300 


3.00 

to 

3.30 p. M. 



Unloading Schedule, 



Friday 

Saturday. . 
Monday... 



Tuesday. 



9-15-11 



9-16-11 



9-18-11 



9-19-11 



Wednesday. 



300 



7.30 300 on D. 3.30 

to E, F, H, to 

I 9.30 A. M. and /. 7.30 p. m. 



300 on D. 7.20 
if, F. H, to 

and J. 9.15 a. m. 



300 on £ 
andff. 



9-20-11 300 on ^ 
I only. 



6.15 a.m. 

to 
9.20 A. M. 



8.30 A. M. 


Zero. 


to 




12.30 p. M. 





300 on £ 9.30 
and H. to 

I 11.45 a.m. 

300 on ^ 9.30 a. m. 
only. to 

* 12.00 m. 



1.00 

to 

3.40 p. M. 



300 on Z), 

E, P, H, 

and I. 

300 on £ 
andff. 



3O0onE 
only. 



300 on £ 
only. I 



Zero on 
all bays. 



8.00 

to 

8.30 p. M. 

6.30 

to 

8.00 p. M. 

12.16 

to 

1.60 p. M. 

4.16 

to 

8.00 p. M. 

4.60 

to 

6.60 p. M. 

4.00 

to 

6.40 p. M. 



Digitized by 



Google 



Report of Committee on Reinforced Concrete. 97 

period of 48 hoxirs. A similar method was used in the process of 
removing the load. 

Table IV shows the loading schedule. The load was applied 
in increments of 100 lb. per sq. ft. based upon the whole test area. 
The application of the load consumed three days. The full load 
was left on 48 hours. The unloading schedule is also shown in 
Table IV. In the unloading, the load on panels B and C were 
first removed, then the load on panels Z), F, and 7, followed by the 






\300 



\200 

voo 



1-4// 



Nbne O 




FIG. 31 



-LOAD DEFORMATION DIAGRAMS FOR CONCRETE ON UPPER SIDE OF SLAB. 



removal of the load on panel H, Fig. 21 is a view at a load of 
300 lb. per sq. ft. over the test area. The total load was over 
500,000 lb. 

Personnel of Testing Staff. — ^All instrument readings were 
made by W. A. Slater and H. F. Moore, of the stafif of the Engi- 
neering Experiment Station of the University of Illinois. Mr. 
Slater had immediate charge of the test as a whole. A. N. Talbot 
was present during the work of preparing for the test. Three 
others assisted in the work of recording and reducing data. 



Digitized by 



Google 



98 Repokt op Committee on Reinforced Concrete. 



Table V.— Strbss Indications in Turner-Cartbh Building Test. 



Member. 


Gauge Line. 

220 
244 
304 
318 
310 

202 
206 
230 
234 
236 
238 
240 
222 
224 
214 


Reinforcement. 


Gauge Line. 

269 
311 

265 
267 
281 
293 
301 
305 
313 
315 


Concrete. 


End of girder 

Middle of girder 

Rnd of beam 


'siooo 

9,000 
8,000 
8,000 
4,000 

"7]666 

11,000 
9,000 
8,000 
8,000 

11,000 

5,000 

5,000 

5,000 

—3,000 


900 
Little 

' 1,166 

1,100 

1,000 

800 


u 


It 


It 


Middle of beam 


350 


it 


350 


ft 


200 


it 


300 


li 




tt 




It 




Rent up bar in girder 








Bent UD bar in beani 









Table VI. — Maximum Stresses and Moment Coefficients 
in Turner-Carter Building Test. 



Member. 



Girder, End — 
" End.... 
" Middle . 
" Middle. 



Intermediate Beam, End — 
End.... 
Middle. 
Middle. 



Column Beam, End. . . . 
End.... 
Middle. 
Middle. 



Reinforcement. 



Btreea. 



31,000 

ik'ySdo 

8,000 

Sl,500 

8,000 

18,600 

11,000 

lOfiOO 



17,000 
10,000 



Coefficient- 



1/lB 

1/12 
0.05 

1/12 
0.03 
1/12 
0.05 

1/12 

1/12 
0.05 



Concrete. 



Stress. Coefficient. 



IfiOO 
900 

too 

Little 

1^00 

1,100 

S80 

350 

1,200 
950 
S50 
225 



1/12 
0.06 
1/12 



1/12 
0.07 
1/12 
0.077 

1/12 
0.064 
1/12 
0.054 



Digitized by 



Google 



Report of Committee on Reinforced Concrete. 



99 



Deformations and Stresses. — The results of observations on 
various gauge lines for the beams and girders are plotted in Figs. 
22 to 29. Fig. 30 gives the deformations in the concrete on the 
under side of the floor slab and Fig. 31 those on the upper side 
Fig. 32 records measurements made on the bent-up bars and 
stirrups. 

As already stated, the location of the gauge lines is shown on 
Figs. 17 and 18, the odd numbers referring to measurement on the 



S^300 



A. 



thne 

£ 



Ahne 



200 

too 


E ^300 
\CdLH^O 

\defhi \zoo 

^AN ^00 

^All ^2tt7 

^AH %00 

None^ 0, 



^(£. 



7 



/i^^^ 



1^ 



^ 




^'S 



t 



/ 



^/4- 



Zi*C 




^^a 



2/'^l 



S 



De "fonma -f-ion per Un // or L en^-f-h. 



no. 32. — ^LOAO DEFORMATION DIAQRAMS FOR BENT-UP BARS AND BTIRRUPe. 

concrete, the even numbers to measurement on the reinforcement. 
The numbers in the two hundreds are gauge lines on the under 
side or second story side, and the numbers in the three hundreds 
are on the upper side or third story side. 

Stresses and bending moment coefficients are tabulated in 
Tables V and VI. 

The suggestions given for caution and care in interpreting 
measurements should be applied to this test. 



Digitized by 



Google 



100 Report of Committee on Reinforced Concrete. 

Beams. — For the tensile stresses in the reinforcement at the 
middle of the intermediate beams at the full load of 300 lb. per 
sq. ft., the highest stress observed ^as 11,000 lb. per sq. in. and 
the average stress recorded may be said to be 8,500 lb. per sq. in. 
At the ends of the intermediate beams, the highest stress observed 
in the reinforcement was 8,000 lb. per sq. in., and the general 
value may be said to be 7,500 lb. per sq. in. Using the assump- 
tions for resisting moment ordinarily taken in design calculations, 
these stresses may be considered to correspond to a bending 




^aioffce from C0k/m/7-j 



•/nch^iS 



^ § 35 5 > 5k 



no. 33. — ^DIAGRAM SHOWING DISTRIBUTION OF COMPRESSIVE DEFORMATION IN 
BOTTOM OF COLUMN BEAM. 

moment coefficient of .05 Wl for the maximum stress at the middle 
of the beam and .03 Wl for the maximum stress at the end of the 
beam, if the tensile strength of the concrete be not considered. 
Assuming a modulus of elasticity for the concrete of 2,500,000 
lb. per sq. in., the concrete on the compression side of the beams 
at the middle showed a compressive stress of 350 lb. per sq. in. 
and at the end of the beam 1,100 lb. per sq. in. It is apparent 
that the total compressive stress in the concrete is greater than 
the total tensile stress in the reinforcement of the beams. A 
possible explanation is that end thrust exists, involving so-called 



Digitized by 



Google 



Report op Committee on Reinforced Concrete. 101 

arch action in the beams and floor structure, and that the tensile 
stress is relieved by the presence of this thrust. The tensile 
strength of the concrete must have a large effect on the resisting 
moment. The coefficient of Wl in the bending moment, necessary 
to give a compressive stress equal to the maximum measiu^ in 
the concrete, on the assumptions made, is .077 for the middle of 
the beam and .07 for the end of the b^am. These coeflScienta 
are lower than the value 1/12 usually assumed in design of such 
beams. 

Girders, — For the tensile stresses at the middle of the girders 
the observations showed about 8,000 lb. per sq. in. in the reinforce- 



' O O ' - ' 

o o o 




whncofrvm y/rder' inches ^ 

FIG 34. — DIAGRAM SHOWING DISTRIBUTIGN OF COMPRESSIVE DEFORMATION IN 
INTERMEDIATE BEAM. 

ment at the middle. This corresponds to a bending moment 
coefficient of .05, again neglecting the tensile strength of the con- 
crete. The reinforcement at the end of the girder was inaccessible. 
Assuming a modulus of elasticity of 2,500,000 lb. per sq. in., 
the concrete on the compressive side of the beam at the support 
showed a compressive stress of 900 lb. per sq. in. The reading 
at the middle of the beam showed very little compression. Assum- 
ing that the loads on the girder are concentrated at the points 
where the intermediate beams are connected, and making the same 
assumption of distribution of stress as before, the coefficient of 
bending moment was .06. It seems probable that the compres- 
sion at the middle of the span must be distributed over a consid- 



Digitized by 



Google 



102 Report op Committee on Reinforced Concrete. 

erable width of floor, or larger readings of compression would have 
been obtained. 

Decrease in Compression with Distance from Support. — ^In four 
beams measurements of compressive deformations were taken at 
a series of gauge lines from the support to a location near the point 
of inflection. The position of these points is shown in Pig. 17. 
The gauge lines No. 223, 225, 227, 229, 231, and 233 are on one 
side of column No. 6, and 281, 283, 285, 287, 289, and 291 are on 
the other side of colxmin No. 6. It may be expected that there 
will be full restraint for the end of the beams. Gauge lines 243, 
241, 239, 237, and 235 are on one side of a girder and 293, 295, 
297, 299, and 1201 are on the other side. The unit-deformations 
for these gauge lines at loads of 200 lb. per sq. ft. and 300 lb. per 
sq. ft. are plotted in Figs. 33 and 34. 

The measurements recorded for the column beams show con- 
siderably more compressive stress than do those for the intermediate 
beams, perhaps one-third more. This difference in stress may be 
due partly to the deflection of the girder, and to the deflection of 
the intermediate beam between its support and a point opposite 
the end of the column beam, which would permit a larger part 
of the load to be carried by the column beam. It may be due 
somewhat to the fact that reinforcing bars are bent down from a 
point at the end of the column beam, while in the intermediate 
beams the bars run horizontally for a foot from the face of the 
girder. 

The direction of the lines in Fig. 33 and Fig. 34 indicates a 
zero stress at about 45 in. from the face of column in the column 
beams and at about 50 in. from the face of the girder in the inter- 
mediate beams. In both cases the results locate the point of 
inflection at about 0.22 of the clear span. 

T'Beam Action. — The distribution of compressive stresses 
in the T-beam formed by a beam and the floor slab (which involves 
the distances away from the beam for which compressive stresses 
are developed) has been a fruitful source of discussion. Measure- 
ments parallel to the axis of the beam were taken on the upper 
surface of the floor slab immediately above beams and at intervals 
between them. These gauge lines are No. 315, 317, 319, 321, 323, 
325, and 327 (see Fig. 17). The deformations are shown in Figs. 
27 and 31. The amount of these deformations at points across the 



Digitized by 



Google 



Report of Committee on Reinforced Concrete. 103 

slab for loads of 200 lb. and 300 lb. per sq. ft. is shown in Fig. 35. 
It is apparent that a somewhat higher stress existed in one beam 
than in the other. Taking this into consideration, the compres- 
sive stress varies quite uniformly from one beam to the other, and 
the full width of the floor slab may be said to be effective in taking 
compression. The overhang (counting to the midpoint between 
beams) is 6} times the thickness of slab. It will be noticed that 
the conclusions are the same as given for the Wenalden building 
test. 

Readings were also taken on the under side of the floor slabs 
parallel to the beams at three places (No. 1205, 1211, and 1213), 
but the conditions attending the location of these points do not 
permit conclusions to be drawn. 

Floor Slab. — Measurements were taken on the floor slab in 



.0002 






■_ 


^^— 


o 


"^o"" 







.ooot 
o 


— 




— , 


--— 


— 


""">« 


- — 


' 



-^^i2/^ 



:; ^ « B! s 



SM 



no. 35. — DIAQBAM SHOWING DISTRIBUTION OF COMPRESSIVE DEFORMATION 
ACROSS FLANGE OF TEE BEAMS. 

the direction of its span at three places on the under side and at 
one place on the upper side immediately above one of the lower 
measurements. These gauge lines were Na 277 on the under side 
of the slab close to a girder (Fig. 17), No. 279 on the under side of 
the slab 5 ft. from the edge of the girder, No. 309 (Fig. 18) on the 
upper surface immediately above No. 279, and No. 1203 (Fig. 17) 
on the under side half way between girders. The measurements 
are plotted in Figs. 30 and 31. As might be expected from being 
close to the girder and near the level of its neutral axis, No. 277 
showed little deformation. The pair of gauge lines (No. 279 and 
309) shows less deformation than would be calculated by the ordi- 
nary beam formula, but perhaps not less than would be the case 
if the tensile strength of the concrete is considered to be quite 
effective. The reading of No. 1203 was even smaller than 279. 



Digitized by 



Google 



104 Report of Committee on Reinforced Concrete. 

All the stresses found in the floor slab were low. The deformations 
parallel to the beams were discussed under T-beams. 

Bond Stresses, — ^At the ends of the beams the reinforcing bars 
lapped over the center line of the girder a distance of 15 in. An 
effort was made to determine whether there was a movement of 
one of these bars with reference to the adjoining concrete and with 
reference to the adjoining bar; also whether the deformation in 
the stub end of the reinforcing bar was the same as in the adjoin- 
ing bar. Fig. 36 shows the location of the reinforcing bars with 
reference to each other, and the position of the gauge lines. No. 
312-14 in comparison with No. 312 and 314 will indicate any 
relative movement of one bar with respect to the other, and No. 
312c and 314c in comparison with No. 312 and 314, respectively 
will indicate any movement of the bars with respect to the con- 
crete. 

It appears possible that the initial reading of No. 314 is 



Jrife/ plug m concrt*f^ 




no. 36. — ^ARRANQEMENT OF GAUGE POINTS TO TEST FOB MOVEMENT OF BAR 
RELATFVE TO CONCRETE. 

slightly in error, and the remarks already made about quantitative 
interpretation of results and the chances for variations in stresses 
in adjacent bars or in adjoining concrete should be borne in mind 
in studying the results. It seems evident that No. 314 (on the 
lapped bar) records considerable less stress than No. 312. The 
measurements indicate a possibility that the right-hand point 
of gauge line No. 314 has moved to the right relatively to the right- 
hand point of No. 312, though this amount may not be more than 
the amount of initial slip necessary to develop the requisite bond 
stress. The measurements taken have no bearing on whether 
the left-hand point of No. 314 has moved. The measurements 
also indicate that there was no motion of the left-hand point on 
the reinforcing bar (No. 312 gauge line) relatively to the concrete 
at its side, though it must be borne in mind that the point taken 
was so close to the bar that only slip and not distortion of concrete 
could be measured. 



Digitized by 



Google 



Report of Committee on Reinforced Concrete. 105 

Web Deformations, — No diagonal tension cracks were visible 
on any of the beams or girders. 

In girder 4 measurements were taken on the diagonal portion 
of a reinforcing bar, one of the bars which is provided to take 
negative bending moment. This is shown in Fig. 17, Section K-K. 
The gauge lines are No. 222, 224, and 226 The position of the 
gauge lines is also shown in Fig. 16. The measurements are plotted 
in Fig. 32. It was impracticable to measure the deformation at 
a point closer to the support. The measurements show about 
the same stress at No. 222 and 224, perhaps 5,000 lb. per sq. in. 
The stress at No. 226 is materially less. It is not improbable 
that there was tension in this rod throughout its length. As there 
was considerable compression measured in the gauge lines on the 
bottom of the girder below No. 222, it seems probable that a 
crack was formed in the top of the floor slab somewhere above No. 
222, but as this space was filled in with bags of cement no observa- 
tion was made during the test, and inspection of this space after 
the load was removed seems to have been overlooked. At the 
other end of the girder, near column 6, a fine test crack was found 
on the upper surface of the floor 2 in. from the face of the column 
extending across the width of the girder and beyond. This 
extended through the floor. A similar crack was observed on 
girder 3 near column 15. 

Gauge line No. 228 is on a stirrup (see Fig. 16). This stirrup 
is in an inclined position. It is not known what bar it is intended 
to be connected with, nor whether there. is connection with a 
tension bar. The gauge line is in a region of the beam where hori- 
zontal compressive stresses may be expected. The measurement 
in the stirrup at the first increment of load shows tension (see Fig. 
32) and subsequent increments give compression. It should be 
noted that readings could not be taken on the upper end of the 
stirrup. If the upper ends are merely bent out into the floor slab 
it is hard to see that the stirrup may be expected to be useful in 
transmitting web stresses. 

In beam 9 (see Fig. 17, Section L-L, gauge line No. 218) 
measurement was taken on the diagonal portion of a reinforcing 
bar which is carried through the girder at its top and a few inches 
beyond. See also Fig. 16. This shows a tension of 3,000 to 5,000 
lb. per sq. in. (See Fig. 32.) This bar was inaccessible from the 



Digitized by 



Google 



106 Report of Committeb on Reinforced Concrete 



Non9 



\DEnff \$00 
'^A/t <5O0 

\a// \lOO 







^ 8 § S % ^ 

Def/ecr/on^ /n inches. 

FIG.— 37. LOAD DEFLECTION DIAGRAMS. 



Digitized by 



Google 



Report op Committee on Reinforced Concrete. 107 

top of the floor, but the gauge lines on the companion bar (No. 
324 and 318) show about 5,000 and 9,000 lb. per sq. in. Measure- 
ments in the diagonal portion of a single-bend bar (gauge lines 
No. 216 and 214, Fig. 17) which extends only to the center of the 
supporting girder indicate a small compression in the bar (see 
Fig. 32). A stirrup, which like the one in the girder was close to 
the end of the beam and was inclined so that its lower end was 
nearer the support than its upper, showed shortening of the stir- 
rup (see gauge line No. 212, Figs. 16, 17, and 32). In both cases, 




FIQ, 38. — CABINET PROJECTION SHOWING BEAMS AND GIRDERS AND POSITION OF 

TEST CRACKS. 

the arrangement was such that the stirrup could hardly be effec- 
tive. 

The amount of the vertical shear in the beams and girders 
was such that diagonal tension cracks might be expected except 
for the small tensile stresses in the top of the girder and the end 
constraint which seems to have been developed in both beams and 
girders. 

Deflections. — ^The deflections of the beams (including that due 
to deflection of girder) and the deflections of girders are given in 
Fig. 37. The location of the deflection points is shown in Fig. 19. 



Digitized by 



Google 



108 Report of Committee on Reinforced Concrete. 

The effect of time upon the deflection is shown by the increase 
in deflection under constant load. The change when portions 
of the load had been removed may be due in part to the time 
element and in part to the effect of location of the load on the 
panels. The deflections seem relatively small, especially when 
compared with deflections obtained in laboratory tests of beams 
carrying the same loads. The effect of the time element is indi- 
cated on these diagrams. The conditions were such that the sup- 
ports were subject to possible displacement by workmen. 

Columns. — Readings were taken on the four faces of Column 
No. 6 just below the girders, but the results are not consistent 
enough to warrant attempting drawing conclusions. 

Test Cracks. — Fine tension cracks were observed in the lower 
part of the beams and girders. The location of the observed 
cracks is shown on Fig. 38. The appearance of these fine cracks 
is similar to those observed in laboratory tests. They would 
not be noticed without specially careful examination. 

The floor cracks already mentioned indicate the development 
of the tensile stresses in the beams and girders at the support. 

The limitation of space and time have prevented the pre- 
sentation of other matters which were observed in the tests. 
For example, the observations on deformations during the 48 
hours time with the full loading showed in general a slight increase 
in the deformations in the reinforcement and in the concrete. 
It is hoped to take up some of these matters at another time. 
It was not possible to give full attention to every feature upon 
which information was sought, and in some cases isolated points 
were used with a view of determining tendencies, and in these 
naturally there is less certainty in the indications. 

Part II. Test of a New Type of Flat Slab Floor 
Construction. 

Building. — The Powers Building, Fig. 39, is a three story 
and basement warehouse located at Minneapolis, Minn. The 
exterior walls of the building are bearing walls and there is one 
row of columns through the middle of the building. The floors 
are flat slabs, reinforced in two directions with high elastic limit 
deformed bars. Fig. 40 shows a basement and first floor plan 



Digitized by 



Google 



Report op Committee on Reinforced Concrete. 109 

of the building, Fig. 41 the reinforcement in the floor tested and 
Fig. 42 the reinforcement in place. 

General Outline of Test — ^The first floor of the building was 
selected for the test by the Building Inspector of the City of 
Minneapolis, as this represented an acceptance test for the new 




FIG. 39. — POWERS BUILDING, MINNEAPOLIS, MINN. 

design. At the time of the test the floor was about 3 months old. 
Four panels were loaded to 200 lb. per sq. ft. (the design load) 
and then two panels were loaded to twice this amount. Cement 
in bags was used for the load, piled in piers to prevent arching. 
The load was arranged as shown in Fig. 44 until the design load 
was reached, while for the maximum load Panels 2 and 3 only 



Digitized by 



Google 



no Report of Cobimittee on Reinforced Concrete 









50 



.J^ 



«Si 



open^ 



i Bl 



ytasement Columns • Zo'square % 



first ^ory Columns- id'square. 



io\o' 



7%^ Rough 3lab except 
Rectangles over Columns 
which are iwo inches 
1 thicker 

_ jji^Cemenf Finish 



-4 Tile Partition in 
first Story only. 



— f — 



l_.j:— J 



•H 



^.. 



/9'-6' 






'I 

I 

p 

i 

I 

I 

I 

i 



no. 40. — PI.AN or powers building, baskmxnt and rbbt floob. 

/Google 



Digitized by ^ 



Report of Committee on Reinforced Concrete. Ill 




dors extend ^ same dirsction as lettering indicating them ■ straight A bent 
bars alernate. - StmigHt bars extend 6' past margin of panel ana bent bars 
extend to quarier-jpoinr ofadjdcenf panel 

There one onfy ho layers of bars m the top of the siaband two ^ 
vss\ lowers in the bottom. The bent bars extending across the building fbrm% 
^ the bottom tiyer in the middle oftfjepanel and the top layer over the 



the pat 
Mm 






i^&QBlMBi 



TIQ. 41. — DIAGRAM 8H0WINQ ARRANGEMENT OF REINFORCEMENT. 



Digitized by 



Google 



112 Report of Committee on Reinforced Concrete. 

were loaded. The wide aisles shown were necessary for the 
accommodation of the mstruments and observers. All tabulated 
loads per square foot and the loads used in plotting are in every 
case the total load on the panel divided by the area of the panel; 
the intensity of the load under the piers, of course, is greater. 
Deflections of the various panels were measured under different 
stages of loading and also the deformation of the reinforcement 
and concrete due to these loads. 




FIG. 42. — REINFORCEMENT IN PLACE, POWERS BUILDING. 

Instruments. — The deflections were measured with a deflec- 
tometer, see Fig. 1. 

Deformations in the reinforcement and concrete were meas- 
ured with an 8-in. Berry strain gauge. Fig. 43, which reads direct 
to 1/2000 of an inch, or by estimation to one-fourth of this 
amount. As slight variation is possible, 5 readings were taken 
at each point and the average of these assumed to be correct. 
Readings were taken at intervals throughout the test on stand- 
ard bars and on standard points placed in imstressed portions of 
the concrete; the temperature corrections so observed have been 



Digitized by 



Google 



Report of Committee on Reinforced Concrete. 113 

made to the readings. There are many diflSculties mvolved in 
measuring deformations under conditions such as exist in a test 
of this kind. In view of these difficulties extreme accuracy can- 




Frame 





'Ame6 Gauge 



1 1 

■Plunger I 

ii 



Lever 



Hxed Leg 



^35 Hole 




^55 Hole 



no. 43. — BEBBT STRAIN QAUOE. 

not be hoped for, but serviceable results can be obtained by a 
trained observer. Observations were made by W. A. Slater, 
University of Illinois, and F. J. Trelease of the Research Depart- 
ment, Corrugated Bar Company. 



Digitized by 



Google 



114 Report of Committee on Reinforced Concrete. 




FIG. 44. — ARRANGEMENT OF LOADING. 



Digitized by 



Google 



Report of Committee on Reinforced Concrete. 115 

Application of Load, — ^The test continued for 9 days, July 27 
to August 4, inclusive, 1911. The loading was in 10 stages as 

Table VII. — Schedule of Loading Operations. 



Stage. 


Load ic 


Place 


Loadio 


Poiinds 


per Panel and per Square 
Foot. 




From 


To 


Panel 1. 


Panel 2. 


Panel 3. 


Panel 4. 


1 
2 
3 

4 
5 
6 


1.45 P.M., 7-27 
11.10 A.M., 7-27 
4.45 P.M., 7-28 
10.45 A.M.. 7-29 
3.40 P.M.. 7-29 
7.00 P.M.. 7-29 
3.15 P.M.. 8-1 
8.20 a.m., 8-2 
4.30 P.M.. 8-2 
10.35 a.m.. 8-3 


8.40 A.M., 7-28 
2.00 P.M.. 7-28 
8.45 A.M., 7-29 
1.25 P.M.. 7-29 
5.40 P.M.. 7-29 
11.00 a.m.. 7-31 
4.25 P.M.. 8-1 
9.15 a.m.. 8-2 
7.15 A.M.. 8-3 
3.45 P.M.. 8-3 


13,300 
33.440 
53,200 
34.200 
15,200 


52 
132 
210 
135 

60 


13.300 51 
33,440 128 
53.200 204 
72,200 276 
91.200 1 350 
106,400i 408 


L^.iiLtO 
3^S.].|.0 
f.:j,JiiO 

r-'.-iiO 

ItKLiOO 
lOf-MK) 

Sa.Ui'O 


51 
129 
206 
280 
354 
414 
198 


13,300 
33.440 
53,200 
34,200 
15.200 


51 
128 
204 
131 

58 


7 






96,900 1 370 
106.780' 408 






8 










9 






53,010 
21.660 


202 
83 




10 




























Note.— Started placing Stage 1 at 10 A. M.. 7-27-1911. All load removed at 4.20 P. M.. 
8-3-1911. 

will be noted from Table VII, which shows the arrangement of the 
load at various stages and the length of time each load remained on 




FIG. 46. — DESIGN LOAD IN PLACE ON FOUR PANELS. 

the floor. Readings were first taken on all points with the floor 
unloaded and then a load approximately equivalent to 50 lb. 



Digitized by 



Google 



116 Report of Committee on Reinforced Concrete. 



■X 



•7 



• 6 



•5 



•4 



• J 



• 2 



'^ r 

I 



i& 



! L 



I I 



•/J 



lj2is^ 



I 



•iz 



•10 



I 

f 

I 

— -rf/7 






i/4 



I I 



•<S 



;-— ^ 



•27 

• 26 

• 25 

•24 •3? 

• 22 •SO 
•21 ^29 
•20 ^28 

• 19 



L 



:l . J 



FIG. 46. — ^LOCATZON OF DEFLECTION POINTS. 



Digitized by 



Google 



Report op CoMBniTEE on Reinforced Concrete. 117 

per sq. ft. was applied over 4 panels, Fig. 44. Another set of 
readings was taken and the load increased to 125 lb. per sq. ft. 
This method of alternate loading and reading obtained through- 
out the test. When the desigp load of 200 lb. was reached, how- 
ever, the load on Panels 1 and 4 was moved by stages to Panels 
2 and 3, so that finally Panels 1 and 4 were completely unloaded 
and Panels 2 and 3 were loaded to about 400 lb. per sq. ft. The 











Table VIII. — Deflections in 


Inches. 










Set 


1 


2 


3 


4 


6 


6 7 1 8 ' 9 


10 1 11 


12 


13 


14 


15 


16 


Load 


DL 


1 


1 


2 


3 


3 4 5 1 6 


6 1 6 


7 


9 


9 


10 


DL 


^ 




































-.005 
.005 
.014 
.004 
.000 
-.002 
-.001 
-.016 
.004 
.038 
.000 
.001 
-.001 
.008 
.016 
.017 
.012 
.010 
.013 
.022 
.021 
.020 
.028 
.019 
.016 
.008 
-.001 
.014 
.013 
.015 
.017 
.018 


.000 
.015 
.026 
.034 
.005 
.010 
.007 
-.003 
.012 
.043 
.000 
.002 
.012 
.022 
.019 
.028 
.023 
.019 
.028 
.035 
.033 
.037 
.038 
.041 
.027 
.021 
.006 
.023 
.022 
.018 
.029 
.031 


.004 
.032 
.057 
.055 
.003 
.012 
.008 
.000 
.017 
.039 
.006 
.005 
.015 
.051 
.054 
.067 
.060 
.049 
.068 
.085 
.081 
.079 
.090 
.106 
.069 

!oio 

.061 
.169 
.065 
.079 
.078 


.Oil 
.052 
.079 
.066 
.008 
.010 
.003 
.009 
.019 
.046 
.014 
.009 
.013 
.111 
.126 
.142 
.133 
.118 
.137 
.205 
.213 
.224 
.236 
.249 
.154 
.321 
.021 
.158 
.176 
.179 
.196 
.196 


' ; ' 
; , 1 

.016' .026' .043 .065 
.060 .091 .143, .197 
.090 .139, .217 .294 
.073, .102 .144 .195 
.013, .0171 .035' .050 
.008 .010 .015 .020 
.003 .005, .0061 .009 
.0121 .007 .002:-. 002 
.021: .028 .037, .046 
.056, .067' .076 .101 
.014, .020' .033' .035 
.009, .011] .009' .008 
.013' .013, .0151 .015 
.126 .134, .1631 .152 
.146' .163' .182. .209 
.155 .175' .220; .242 
.149; .163' .190' .198 
.138 .141 .144' .137 
.1681 .1631 .1451 .136 
.238 .251, .2731 .283 
.255, .278 .320| .348 
.264' .294' .346, .374 
.278 .303' .326 .363 
.2861 .289 .318' .316 
.156 .179 .1661 .148 

.341! .326, .3061 

.0291 .034 .027, .023 
.1841 .196| .2061 .229 
.211, .2301 .2661 .291 
.214 .234 .273 .305 
.228' .248 .285 .309 
.229, .236 .256 .268 


.082 
.237 
.351 
.243 
.063 
.028 
.014 
.000 
055 


.07. 

•:«4 

.070 



.059 
.174 
.271 
.196 
.070 
.026 


.040 
.112 
.159 
.102 
.047 
.012 
.010 
-.003 
.031 
.070 
.023 
.006 
.011 
.120 
.194 
.210 
.177 
.136 
.130 
.272 
.327 
.350 
.350 
.313 
.149 


.040 
.110 
.159 
.108 
.050 
.016 

-!6d6 
.030 
.066 
.030 
.007 

".lis 

.187 
.207 
.169 
.130 
.131 
.260 
.308 
.333 
.312 
.304 
.151 


.042 
.114 
.161 
.113 
.048 
.016 
.014 
.003 
.026 
.059 
.026 
.007 
.015 
.095 
.156 
.171 
.141 
.110 
.120 
.226 
.265 
.284 
.286 
.269 
.130 


'!i55 




.OOi 


-.002 




10 
11 


.112 
.049 
.009 
.022 
.135 
.216 
.240 
.205 
.147 
.132 
.307 
.378 
.413 
.401 
.349 
.148 
.137 
.027 
.243 
.319 
.341 
.342 
,287 


.120 


.104 


.056 


12 
13 


.010 


.008 


.009 


14 








15 








16 






.131 


17 








18 








19 
20 
21 


.148 


.161 
.320 


.282 


22 
23 


.421 


.424 


.214 


24 
26 
26 


'!i49 


.363 
.156 


"iw 


27 






.022 
.214 
.287 
.299 
.302 
.270 


";266 
.268 
.284 
.290 
.261 


.021 
.179 
.242 
.254 
.255 
.230 




28 
29 








30 








81 








82 

















load was next removed from Panel 3 and finally from Panel 2, 
readings being taken at intervals during the unloading. Fig. 45 
shows the design load in place on the four panels. 

Data. — ^Deflection readings were taken at 32 points. Fig. 46. 
The deflections are given in Table VIII and are plotted in Figs. 
47 and 48. The load plotted in these figures and in the stress 
curves is the load per square foot on the central panels. The 
stage of loading can be obtained by referring to Table VIII. 



Digitized by 



Google 



118 Report of Committee on Reinforced Concrete. 




6 aJ ai o 37 oB aS o ST a? 53" 
Deflection In Inches 



FIG. 47. — DEFLECTION CUBVK8. 



Digitized by 



Google 



Report of Committee on Reinforced Concrete. 119 

Deformation of the reinforcement was read at 30 points and 
of the concrete at 31 points. The location of these points is 
shown in Figs. 49 and 50. Table IX gives the embedment to 
center of bar; thickness of the slab and finish coat; net thickness 



^ 



Loaded Portion 



^; ^Me■»> q J>vlM>« LWA^a^J;Jft^;.w;^■l!<^aw■i^^^^^^^^^ 



% 




FIG. 48. — SECTIONS OF FLOOR SHOWING DEFLECTIONS. 

of the rough concrete slab, deducting the thickness of the finish 
coat from the total thickness of slab, as the finish coat was found 
to be loose in many places and did not have much load-carrying 
capacity; and also the effective depth of the reinforcement on 
which the strain was measured. 



Digitized by 



Google 



120 Rbfobt of Comhtttee on Reinfobcisd CoNCBinv. 



o4« 



my 



"H r 



1 



I 



\^- 



I I 



III- 



ItO 



116 




I 



?05 



.„:r-.j 



/a5 



All Oauge Length6 - (S"' 



%I07 



k. 



m 



w 



?oer 



£ 






•-• Indicates Points on Top of 5lab 

o-o Indicates Points on bottom of 5lab 

no. 40. — LOCATION OF DEFORMATION GAUGS POINTS OR REINFORCEMENT. 



Digitized by 



Google 



Report of Committee on Reinforced Concrete. 121 










^ 




1 — " 








' 




f" 


IT 






• 












— 




_L_. 


! 
1 

i 








1 


I — 
1 


m "^ ''^ 






1 


1 
1 






H 


L. 












4<»> 


r^ 












I 1 






[" 


1 












1 
1 
I 

1 


J 














r- 


415-^ 405^ 
J3/5 .,p 


305 

* 304 
1 






All Gauge Lengths 


Q' 




1 

1 
I 
1 


^r 


^ 


^311 307 


1 

r-' 



•-• Indicates Points on Top of J lab 

o-o Indicates Points on dottom of 5lab. 

nOt. 60.— LOCATION OF DBFORMATION GAUGE POINTS ON CONCRETE. 



Digitized by 



Google 



122 Report of CoMBniTEE on Reinforced Concrete. 

The observed elongations and corresponding unit stresses, 
based on an assumed modulus of 30,000,000, in the reinforcement 
are given in Table X. Unit reinforcement stresses are plotted 
against the load per square foot on Panel 2 in Figs. 51 and 62. A 
series of readings taken by Mr. Slater on one of the bars over the 
colunm cap shows that a bond stress of 135 lb. per sq. in. under 
maximum loading existed at that point. 

Table IX. — Dimensions at Deformation Gauge Points. 





Embed'- 


Thickness of 


Net Thickness of 
Slab, in in. 


Effective Depth 
of Bars, in. 


Gauge - 
Point. 


ment of 
Bar to 












Slab. 


Finish 




From 




From 




center, in. 


in. 


Coat. in. 


Actual. 


Plans. 


Actual. 

7V. 


Plans. 


101 


IVi 


9Vs 


1 V4 


7V« 


7«/4 


7»i,* 


102 


1 


OVa 




7Vi 




8V. 


8V4* 


103 


l"/if 


9Vt 


" 


7Vi 


" 


7Vi« 


7 v.* 


104 


IVm 


91. 


• « 


7Vt 


*• 


7»»/i« 


8V4* 


105 


IVt 


8Vs 


•• 


7 V. 


" 


7V. 


7Vi» 


106 


1V4 
1 "/!• 














107 


"difii" 


;. 


"sv;."' 


;• 


'■7Vi"" 


"Viu*" 


108 


1 Vt 














109 


2Vi« 


"sifV" 


•* 


"7V." 


*• 


"6«A.' ■ 


"7ij\i" 


111 


l»/w 


8»»/w 


•' 


7 "/!• 


•* 


7V. 


7V.« 


112 


IVw 


9 


** 


7V4 


" 


7Vw 


8V4* 


118 


1 »/!• 

0»'w 














114 


"svV" 


:. 


"7V«"* 


" * * * " 


"7»/i;" 


"svii" 


115 


lu/is 














116 


IVs 














118 


1V4 


"s'^'lu" 


•• 


**7»Vii" 


" 


••7U/1'." 


"SVii" 


120 


OVw 

l>/» 














122 


"si/V" 


*"*"'"' 


■'7Vi " 




"Vvi" 


■'s 1/4* ' 


201 


3Vw 


lOV. 


IV4 


9V. 


"ovi'" 


7Vi« 


8». 


202 


3Vi 


lOVi 


l»/4 


9V. 


9V4 


7V4 


9 


203 


2V8 


9V. 


n/4 


7V. 


7V4 


7 


6Vi 


204 


2Vw 


11 Vi« 


l»/4 


9>*/i« 


9V4 


8V« 


9 


205 


3»/» 


10 v» 


1V4 


9 v. 


9 "A 


7V. 


8Vi 


206 


3Vi« 


10 "/w 


1 Vi 


9Vi« 


9«/« 


7Vs 


9 


207 


2Vi 


11 1/4 


1 


10 V4 


9V4 


8Vi/ 


8V« 


208 


2>/i« 


8"/m 


IVii 


7Vb 


7V4 


6Vt 


7 


200 


8 1/4 


9 


l»/4 


7V4 


7V4 


5V4 


6V« 


210 


2 Vie 


91/1. 


l»/l« 


8 


7V4 


6Vt 


7 



* Assuming 1 \ in. finish coat. 

Table XI gives the observed elongations and corresponding 
miit concrete stresses. These concrete stresses are based on an 
assumed modulus of 2,000,000 and are plotted against loads per 
square foot in Figs. 53 and 54. 

A summary of the unit stresses is given in Table XII, the 
stresses giveii being the maximum probable values. 

During the test, cracks were carefully searched for with an 
electric light and recorded as found. Their location is shown 
in Fig. 55. Most of these cracks were very minute and difficult 



Digitized by 



Google 



Report op Committee on Reinforced Concrete. 123 



Tablb X. — Deformations in Inches and Reinforcement Stresses in 

Lb. per Sq. In. 



Set 


1 


2 


1 
3 


4 


5 


6 


7 


8 


9 


10 


11 


12 


13 


14 


15 


16 


LoMl 


DL 


DL 


1 


1 


2 


3 


3 


4 


5 


6 


6 


7 


8 


9 


10 


DL 


Gauge 
Point 
101 

102 

103 

104 

105 








































































0.0 
1685 

0.3 
560 

-0.3 
-560 

-0.5 
-040 

0.4 
750 

•0.7 
1310 

0.8 
1600 

-0.5 
-040 

0.3 
560 

1.2 
2250 

0.8 
1500 

0.8 
1500 

0.5 
940 

0.4 
750 


0.3 
560 

-0.3 
-560 

-0.6 
-1120 

-0.7 
-1310 

-0.8 
-1600 

-0.5 
-940 

-0.2 
-380 

0.3 
560 

-0.1 
-190 

-0.3 
-560 

-0.3 
-*60 

0.1 
190 

-0.6 
-1120 

0.2 
380 

-0.5 
-940 

-0.5 
-040 

-0.9 
-1690 

-0.3 
-«60 

-0.2 
-380 

-0.3 
^60 

1.0 
1880 

1.1 
2060 

05 
940 


-0.1 
-190 

0.6 
1120 

-0.9 
-1690 

-0.4 
.750 

0.5 
040 


-0.1 
-190 

0.6 
1120 

-0.4 
-750 

0.3 
560 

-0.4 
-750 

0.1 
190 

-0.5 
-940 

0.1 
190 

-0.5 
-940 

-1.6 
-^000 

-0.4 
-760 

0.9 
1690 

-1.0 
-1880 

0.3 
560 

-0.1 
-190 

-0.2 
-380 

-0.9 
-1690 

0.5 
940 


0.2 
380 

3.3 
6200 

-0.1 
-190 

8.6 
6740 

-0.5 
-940 

0.2 
380 

0.0 


••••• 


0.7 
1310 

4.2 
7880 

0.9 
1600 

4.7 
8810 


1.1 
2060 

4.3 
8060 

2.7 
5060 

5.0 
9380 


1.6 
3000 

4.3 
8060 

2.7 
5060 

5.7 
10690 


2.0 
3750 

4.2 
7880 

3.5 
6560 

6.0 
11250 


2.2 
4120 

4.4 

8250 

4.2 
7880 

7.0 
13120 


2.4 
4600 

6.6 
10500 

4.8 
9000 

7.7 
14440 


1.9 
3560 

4.0 
7500 

3.9 
7310 

5.7 
10690 

-1.0 
-1880 


1.6 
3000 

3.0 
5620 

3.4 

6380 

4.9 
9190 

-0.3 
-560 


2.8 
5250 

2.7 
5060 

2.0 
3750 

3.7 
6040 

-1.2 




::::: ;:::: 












-2250 


106 




1.5 
2810 


8.0 
15000 


0.6 
1120 


0.8 
1500 




















107 






0.0 


0.0 


-0.5 


















-940 


106 


0.3 
190 

0.2 
380 

-1.8 
-3380 

0.2 
380 

5.3 
9940 

-0.6 
-1120 

1.2 
2250 

-0.8 
-1500 

0.2 
380 

-0.6 
-1120 

2.8 
5250 

1.7 
3190 

1.5 
2810 

1.6 
3000 

7.0 
13120 

0.6 
04G 




0.3 
560 

0.9 
1690 

-2.1 
-3940 


-0.8 
-1500 

1.8 
3380 

-2.4 
-4500 


-3.1 
^10 

2.4 

4500 

-3.0 
-5600 


0.3 
560 

.20 
3750 

-2.5 
-4600 
























109 
110 


2.2 
9120 




0.8 
1500 


0.8 
1500 


1.2 
2250 














111 






-0.3 
^60 

5.9 
11060 


5.6 
10500 


-1.6 


















-3000 


112 
113 


••••• 


6.3 
11810 

0.4 
750 

2.1 
3940 

0.3 
560 

0.7 
1310 

-0.7 
-1310 

2.9 
5440 


5.4 
10120 

1.2 
2250 

2.7 
5060 

0.9 
1690 

1.3 
2440 

-1.4 
-2620 

1.9 
3560 


6.0 
11250 

0.3 
560 

1.5 
2810 

0.1 
190 

1.5 
2810 

-1.5 
-2810 

2.4 
4500 


6.1 
11440 

1.4 
2620 




7.6 
14250 


4.0 
7500 














114 






2.0 
3750 


1.2 
2250 


1 7 










3190 


115 


-0.1 
190 

1.7 
3190 

-0 R 




















116 






1.2 
2250 


1.2 
2250 


0.8 
1500 


117 










-040 












118 


2.1 
3940 






2.7 
5060 




1.3 

2440 


120 


1 0.7 

1310 

1 

0.2 

1 380 

0.8 1.2 
1500 2250 

0.4 2.7 
750, 5060 

-0.4i -0.1 
-750 -190 






0.5 






















940 


122 
201 
202 
203 


1.5 
2810 

7.6 
14250 

-0.1 
-190 


1.4 
2620 

2.0 
3750 

8.1 
J5190 

0.1 
190 


3.0 
5620 

1.8 
3380 

9.2 
17250 

-0.8 
-1500 


3.9 
7310 

3.1 
5810 

9.1 
17060 

-0.5 
-940 


4.0 
7500 

3.7 
6940 

10.1 
18040 

-0.6 
-1120 


3.4 
6380 

3.1 
5810 

9.1 
17060 


9.5 
17810 


1.8 
3380 

2.8 
5250 

7.8 
14620 


2.1 
3940 

2.4 
4500 

5.8 
10880 


2.4 
4500 

2.5 
4690 

5.2 
9750 

-0.2 



Digitized by 



Google 



124 Report op Committee on Reinforced Concrete. 



Table X. — Continued. 



Set 

Load 

Gauge 

Point 

204 

205 

200 

207 

208 

208 



209 



210 



1 


2 
DL 

0.4 


DL 








750 





0.1 





190 





0.4 





750 





-0.8 





-1500 



-0.2 
-380 



-0.1 
-190 



0.3 
560 



0.6 
1120 



0.6 
1120, 



0.7 
1310 



8 I 9 I 10 
5 ' 6 



2.9 9.4 9.7 10.4 11.5 11.7 12.31 12.0 
5440 17620 18190 19500 21560 21940 23060 22500 



0.9 
1690 



1.6 
3000 



-0.5 
-WO 



0.7 
1310 



0.8 
1500 



-0.1, -^.3 -0.3 
-190 -560 -580 



-0.3 
-560 



0.9 
1690 



0.1 
1880 



3.11 7.7 
5810 14440 



0.1 
190 



3.2 

5620 



-0.3 
-560 



5 
2810 



0.6 
1120 



12.2 
22880 



1.7 . 
3190|. 



3.3 
6190 



4.7 5.2 4.0 
8810' 9750 7500 



8.8! 10.4 12 2 12.4 14.4 
16500 19500 22880 23250,27000 



4.6 



-0.5 
-940 



1.6 
3000 



4.9 
9190 



0.7 
1310 



0.5 
940 



0.7| 
1310 



6.4 5.6 6.7 
12000 10500 12560 



0.3 1.0 0.6 1.1 
560 1880 1120 2060 



-0.3 
-560 



2.9; 3.1 
5440; 5810 



4.2 4.4 4 5 
7780 8250 8440 



7.0 6.3i 
13120 11810 



1.9 . 
3560 . 



5.0 . 



15 
10 



16 
DL 



9.7; 7.1 5.80 

18190 13310, 9380 

2.3 2.3I 1.3 

4310 4310 2440 



8.0 
15000 



10.61 3.2 
19880 17440 



-0.8 
-1500 



4.8 3.2 

9000! 6000 

0.6| 0.9 

1120. 1690 



3.2 4.2 -2.9 
6000. 2810-5440 



1.4 
3190 



0.4 
750 



-0.8 
-1500 



Table XI. — Deformations in Inches and Concrete Stresses in 
Lb. per Sq. In. 



Set 



Loadj DL 



Gauge 
Point 
301 



302 
803 
304 
305 
306 
307 
JC8 
309 
810 



2 


3 


DL 


1 




-0.8 
-100 



1 I 2 



-0.3 
-40 



-0.7: 



0.8 -0.1 
100 -10 



-0.5 
-60 



0.1 
10, 



-0.2' 
-25 



-0.2 
-25 



0.5 -0.7 . 
60 -90 . 



-1501. 



0.1 . 
10 . 



-0.2 
-25 

-0.2; 
-25 

-0.2 
-25 

-0.7 
-90 

-0.6 
-76 



0.6. 
75,. 



-0.9'. 

-no . 



-1.61 
-200 



-1.0 
-125 



0.1 
10 


-0.8 
-100 


-0.7 
-90 


-2.8 
-350 


-0.3 
-40 


-0.9 
-110 


-0.7 
-90 


-08 

-:oo 


1.5 
190 


-03 
-40 

i 


-0.7 
-90 


-2.3 



-6.8 
-850 



-1.4 
-175 



-3.9 
-490 



0.3 
40 



-1.1 
-140 



-0 1 
-10 



11 12 



-1.2 . 
-160 . 



-1.5 . 
-190 . 



10 











-2.0 
-250 








-1.8 
-225 

-1.2 





0.3; 
40 



,^.,1. 



.1 -0.3 -0 2 
i -40, -25, 



-1.7 -2.6 

-210 -325 -360: -425 



-0.1 -Oil 
-10 -10. 



-2.9 -3.4 



-3.1 
-390 



0.3 

40 



0.2 
25 



0.4 . 
50 . 



16 
DL 



-0.2 
-26 



0.2 
25 



0.5 
60 



-2.9 -0.7i -1.6 
-360 -90 -190 



0.1 
10 



-3.2-3.9! -.7 -2.3 

-400 -360 -340 -290 



-6.8' -8 3 -8.6 -9.4 -8.5; -8.7 
-850 -1040 -1075 -1175-1060-1090 



-1.8 
-225 



-5.6 
-700 



-1.7 
-210 



-5.61 
-700 



-1.4 -1.7. 
-175 -2IO1. 



-6.3 -7 9 
-790 -990 



-7.1 . 



. -0 1 -0.4 

.1 -10 1 -50 



-7.1 
-890 



-6.2 -4.7 
-775 -590 



-0.1 
-10 



-5.1 -4 5 -2.7 
-640 -560 -340 



Digitized by 



Google 



Report of Committee on Reinforced Concrete. 125 



Table XI. — Continued. 



Set 


1 
DL 


2 


3 

1 

-0.7 


4 
1 


5 


6 


7 


8 


9 


10 1 11 

1 


12 

7 


13 


14 


15 


16 


Load 


DL 

-0.4 
-50 

0.6 
75 

-0.6 
-75 

-0.4 
-50 

-0.4 

-0.1 
-10 

0.4 
50 

-«.5 
-60 

-0.1 
-10 

-0.2 
-25 

-0.2 
-25 

0.5 
60 

-0.6 
-75 

0.5 
60 

-0.6 
-75 

0.1 
10 


2 1 3 


3 


4 5 

1 

1 
i 


8 


9 


10 


DL 


Point 


































































-0.6 
-60 

-0.8 








-0.3 
-40 

-2.1 
-260 

-1.4 
-175 


-0.5 
-60 

-1.5 
-190 

-1.0 
-125 


-0.6 




-90 

-0.4 

-50 

-0.8 

-100 

-1.5 





1 


1 






-75 


312 


1 
-1ft 


-2.9' -3.5 
-3601 -440 

1 


-4.1' -5.5 
-510 -690 

1 






-1.0 




-lOO' -225 






-135 


313 


1 
-1.3, -1.6 
-160 -200 

-0.41 -0.8 
-60| -100 

-1.21 -1.9 
-160 -240 

-0.9 -10 









-1.2 






1 








-150 


314 














0.1 




-190 

-1.1 
-140 

-0.6 
-75 


-0.1 
-10 








1 










10 


315 
401 





-1.3 
-160 

-10 
-125 


-3.1 
-^90 

-0.5 
-60 


-2.9 -3.6 
-360 - 460 

^).4' 


-3.9 
-490 


-3 2 
-400 


-2.4 
-300 


-3.0 
-375 


-2.1 
-260 

-1.1 




-110 -125 - - 


-60 




'..'.'.'.\..y.'. 




-140 


402 


-0 6i -1 6 




1 










-2.4 








-75| -200 

-0.51 -1.0 
-60| -125 

-I.9I -3 2 
-240 -400 








i 










-300 


403 


-0.1 
-10 

-1.0 
-125 

0.1 
10 

-0.5 
-60 

-0.2 
-25 

-0.8 
-100 

0.2 
25 

-0.2 
-25 

-0.1 
-10 

-0.2 
-25 

-0.5 
-60 


-0.5 
-60 

-1.0 
-125 

-0.9 
-110 


-1.1 
-140 


-1.4 
-175 

-4.5 
-560 


-2.3 
-290 

-3.3 
-410 


1 
-2.4 -3.4 
-300 -425 

1 
-2.6 






-2.1 
-260 


-1.2 
-150 


-0.8 








-100 


404 






-3 2 




-325 ... 




::::: ::.:. 




-400 


405 


0.2 
25 

-1.1 
-140 

-0.1 
-10 

-1.4 
-175 


-0.2 
-25 

-2.6 
-325 

-0.2 
-25 

-3.0 
-375 

-0.9 
-110 

-2.4 
-300 

-0.3 
-40 

-3.1 
-i390 

-1 n 


1 










-1 8 




















-225 


406 


-3 3 

-410 

-0.5 
-60 

-3.4 
-425 


-2.9 
-360 

-0.7 
-90 

-3.5 
-440 


-3.1 
-390 

-0.9 
-110 

-3.8 
-475 


1 
-2.9, -2.9 
-360' -360 

1 
-0.8, -1.6 

-no: -200 

1 
-3.7, -4.7 

-460, -590 

1 






-3.2 
-400 


-2.7 
-340 


-2.4 








-300 


407 






-1.6 













-200 


408 
400 


-4.7 
^90 





-3.6 
-450 


-2.6 
-325 


-1.9 
-240 

-1 5 














j 










-190 


410 


-0.5 
-60 

-0.7 
-W 

-0.6 
-75 


-0.3 
-40 

-0.2 
-0.6 


-2 5 


-9. 7 


-3.1 

-390 


-3.4 -3.6 
-425 1 -450 

1 


-4 5 








-1.0 




-310; -340 


-560 








125 


411 









-1.1 




1 




' 1 


::::: ::::: 






-140 


412 
413 


-3.2' -3.3 
-4001 -410 

-1.5' -2.0 
-1901 -250 

'h).8 

1 -100 

-1.4' -1.4 
-175| -175 

1 


-3.8' -3.4 -4.0 
-4751 -I25I -.500 

-2.6 -2.7, -3.3 
-325' -340l -410 

-1.3 -0.9, -0.8 
-160' -110| -100 

-1.9' -2.4| -3.1 
-240| -3001 -390 

i 


-1.4I 

-650| 

1 


-3.8 
-475 


-2.4 
-300 


-1.6 
-200 

-1.6 




-75[ -125 
-0.5' -0 8 










-200 

0.2 
25 

-1.6 
-200 

0.1 
10 


414 













! 


-60 

-0.4 
-50 

-0.1 


-100 

-0.7 
-90 

-0 ? 










415 
417 


-0.2 
-25 

-0.1 


-0.7 


-3.2 
-400 







-2.6 
-325 


-1.7 
-210 




-"1 


-10 -25 


i""[ 


.........j. ... .......... 

























Digitized by 



Google 



126 Report of Committee on Reinforced Concrete. 




— — d MM y 

Un/t Tension in Jteei 



I 



400* 


) 




j 


=*, 1 




{ 








,/ 


n 








/ 




/ 








1 






r 
















/ 


/ 




/ 








0* 


ic 


00 5000 


[»7 , 

5000 


ifd 1 

5000 




\or 


1 5000 







Unit Tension in sreel 



^ 400 
^JOO 
5 ?00 
I /OO 
S 



p I J -i t 


~- t X t 


r^ ^ 4 - t 





ST 
^500 

5 ?«> 



/OO 



•^ 



107 



loe I09 



ikW" 



/// 



unit 5fre55 in 5tee\ 



3^ 



115 



114 



\II5 



-TSSr 



^400 



^00^ 
500 



1 /a? 

S 



//7 



_L 



(' 3000 



M 



z. 



116. 



J- 



I 



IZO 



-L 



6000 MOO 



Unit 3tre3S in 3teel 
FIQ. 51. — LOAD-DEFORMATION DIAGRAMS FOR REINFORCEMENT. 



Digitized by 



Google 



Report of Committee on Reinforced Concrete. 127 




o 3600 6 foooo 

Unit Tension in 3feel 



^400 



I /oo/- 



I 



f 



ZIO 

-1- 



ZIZ 




6^ tOtOOO b MXXX) O 5000 

Unit Tension in 3teel 



?I6 




O to 20 30 '40 

Distonce from Capital in Inches 



^400 



2 Joo 
I o 



< 










v\ 


1 — ^\ 












r 




/ 








J 


f 












J 






/ 




J 




^ 










^ 










M 


y 


/ 


Z 


Z 




i^zta 


( 


"^ 








6 A 

Unii 


^ 3tr 


ess 


AM 

in Jj 


^eel 


m 


IM I 


) t 




}6t 


WO 


tdi 


\dd 






Unit Stress in Steel 

PIG. 62. — LOAD-DEFORMATION DIAGRAMS FOR REINFORCEMENT. 
READINGS AND CURVES BY W. A. SLATER. 

Digitized by 



Google 



128 Report op Committee on Reinforced Concrete. 

to trace, and were only of such magnitude as would be expected 
to accompany the reinforcement stresses observed. 




^500 
S zoo 



Unit Compression //? Concrete 



too 












y 












\T 
















/ 












I 








L 




1 




/ 




1 




I \ 














9 .,/ 


^1 


> 


3/s 


? 


u 


11 






dA 


» c 


) 5 


00 i 


■> A 


00 c 


3t 


30 C 


) 300 C 


> 600 







Unit Compression m Concrete 




Unit Compression in Concrete 

Fia. 53. — ^LOAD-DEFORMATION DIAGRAMS FOR CONCRETE. 



Examination of the plotted deflections and stresses shows a 
change of inclination of the curves at a load of 200 lb. per sq. ft. 
Take, for instance, point 202, which is typical of this condition. 



Digitized by 



Google 



Repobt of Committeb on Reinfobced Concrete, 129 

At 200 lb. per sq. ft. the stress in this bar was 14,000 lb. per sq. in.; 
the load producing this stress was that of stage 3. 




Unit Compression in Concrete 




4aL 



fam loos 



JUX) 
; 300 



Ebb 6 Ad 6 ido 5 
Unit Compression In Concrete 




407 4408 



So- 



J40i \ 

o soo 



I 



I ^^f—l — T 

:^ \40§ \4IC 

« % lo h A 



\4II 



300 



Wt 



6 SCO o 355 soo o soo' 
unit Compression in Concrete 



1 4/3 \4K \4/5\ 
> soo 500 o soo' 



FIQ. 54. — LOAO-DBFOBMATION DIAGRAMS FOR CONCBBTB. RBAAINQ8 
AND CURVES BT W. A. SLATER. 

When the load from the adjoining panels was shifted and cor- 
responded to 400 lb. per sq. ft., the stress was only 19,000 lb. per 
sq. in., instead of some 32,000 obtained by producing the curve to 



Digitized by 



Google 



130 Report of Committee on Reinforced Concrete. 

the 400 lb. line at that slope which obtained below. This indicates 
that a part of the load is carried by the panels adjoining the loaded 
one. 

The same result is indicated in the case of point 104, which 
is at the bottom of the slab near the middle of one of the panels. 
The fact that the adjoining panels assist in carrying a loiad placed 

Table XII. — Summary op Stresses. 



Location. 




Gauge 
Point. 


Design Load, 
200 lb. per sq. ft. 


400 
lb. per«q. ft. 




D.L.* 


L. L. 


Total. 


L. L. 


Total. 


Reinforcement Streui 
Over Column Head 


Long Span 


202 
204 
Av. 


2.000 
2.500 
2.250 


14.000 
17.500 
15.750 


16.000 
20.000 
18.000 


19.000 
23.000 
21.000 


21,000 
25.600 
23,260 


Short Span 


201 
206 
Av. 


1.500 
1,000 
1.250 


3.000 
3.000 
3.000 


4.500 
4.000 
4.250 


7,000 
1,000 
8,600 


8,600 
11.000 
9,750 


Bottom of Slab at Side of 
Panel 


Long Span 


102 
112 
Av. 


1.000 
1.600 
1.250 


7.000 
11.500 
0.250 


8.000 
13.000 
10.600 


9.000 
13.600 
11.260 


10,000 
15.000 
12.500 




Short Span 


101 


500 


1.000 


1.500 


4.000 


4.600 


Bottom of Slab at Middle of 


Long Span 


104 
122 
Av. 


500 

1.000 

750 


7.500 
4.000 
6.750 


8.000 
6,000 
6.500 


12.000 
9.600 
10.760 


12.600 
10.600 
11.600 


Panel 

Concrete Streaees, 
Bottom of Slab at Colxmm 


Short Span 
Long Span 


103 
109 
Av. 

306 


500 

1,000 

750 

100 


1.500 
2,500 
2.000 

750 


2.000 
3,500 
2,750 

850 


8.000 
6.600 
7.260 

1.060 


8.500 
7.500 
8.000 

1.160 


Short Span 


315 


100 


250 


350 


450 


660 


Top of Slab at Edge of Panel 


Long Span 


406 
412 
Av. 


100 
100 
100 


300 
400 
350 


400 
500 
460 


360 
600 
426 


460 
600 
625 


Top at Middle 




408 


150 


400 


660 


600 


750 


Bottom of SUb at Wall 


Long Span 


302 
304 
Av. 


50 
100 
75 


150 
250 
200 


200 
350 
275 


260 
400 
326 


300 
600 
400 



* Dead-load Btreases obtained from atreas-deformation curves by projecting curve down to 
no-load line. 

on one panel is also shown by the fact that deflections were ob- 
served in panels adjacent to those loaded^ as shown in Fig. 48 
and by the fact that cracks on the under side of the slab were 
traceable well past the middle of panels adjoining these loads. 

The point of contraflection was foimd to be at 0.21 of the 
clear span from the edge of the column head. 



Digitized by 



Google 



Report op Committee on Reinforced Concrete. 131 




FIG. 55. — PLAN SHOWING LOCATION OF CRACKS. FULL LINES INDICATE CRACKS 
ON TOP AND DOTTED LINES ON BOTTOM OF SLAB. 



Digitized by 



Google 



132 Repobt op Committeb on Reinfobced Concrete. 




. Mm 



FIQ. 66. — BARR BUILDINQ TEST PANEL, ST. LOUIS, MO. 



Digitized by 



Google 



Repobt op Committeb on Reinforced Concbete. 133 

The high values of bond stress observed indicate that con- 
sideration be given to this feature in the design of flat slabs, 
especially when it is realized that no matter what scheme of rein- 
forcement be adopted there will exist stresses at right angles to 
any given bar over the column which tend to destroy the adhesion 
of the concrete to the reinforcement. 

Part III. Test op a Concrete Floor Reinforced in Two 

Directions. 

Test Panel — ^The test was made on a panel representing a wall 
panel of a continuous floor and to approximate existing conditions, 
cantilevers were built on 3 sides. These cantilevers were loaded 




FIG. 57. — BARB BUILDINQ TEST PANEL WITH LOAD OF 380 LB. PER SQ. FT. 

during the test until the tangents to the slab over the supporting 
beams became approximately horizontal, as would be the case in a 
continuous floor under multiple panel loading. 

The panel proper is 25 ft. long by 26 ft. 9 in. wide, Fig. 56; 
and is carried by steel I beams, fireproofed with concrete. These 
beams rest on steel bearing plates on concrete posts at each comer 
of the panel. Seven-inch tile was used with a cover of 2J in. of 
1:1^:3 concrete, making the total thickness of the slab 9 J in. 

The arrangement of tile is such that when alternate rows of 
block-tile and channel-tile are laid, the concrete is formed into 
two series of intersecting T beams, spaced 15 in. centers each way 
and having 3-in. steins. Small furring-tile, laid at the intersection 



Digitized by 



Google 



134 Repobt op Committbb on Rbinporcbd Concrete. 

of the ribs thus formed, complete the tile ceiUng surface. The 
panel was remforced as shown m Fig. 56, designed for a live load 
of 150 lb. per sq. ft. 

The panel was poured on September 1, 1911, imder unfavorable 
conditions, it being necessary to carry on a considerable portion 




PIQ. 58. — PLAN SHOWING METHOD OF LOADING BARR BUILDING TEST PANEL. 

of the work after dark. The weather was very hot during the 
next few days and several of them were holidays, so that the slab 
was not even kept wet. On the whole, conditions under which 
this slab was erected were probably not more favorable than would 
exist in an actual building. 



Digitized by 



Google 



Report of Committee on Reinforced Concrete. 135 

Method of Testing. — Sand in sacks was used for the load, 
Fig. 57, each sack being filled on the scales to weigh 100 lb. The 
loading was started on November 23, 1911, and progressed by 
stages, alternating with readings, until 650 lb. per sq. ft. was 
reached on December 6, 1911. The test was stopped at this 
point because of the expense involved in placing the load upon 
the high piles of sacks. The total load remained in place on the 
panel for 48 hours, after which unloading was started and completed 
5 days later. 

Table XIII. — Stages op Loading. 





Time of Loading. 


Load in lb 


periQ. 


ft. 


Stefe. 




Gantileve 














tm 






Start. 




Finiflh. 


Panel. 










Left. 


Rear. 


Rilht. 






1911 




1911 










1 


Nov. 


23. 2.30 p.m. 


Nov. 


24, 10.00 A. M. 


75.5 


38.3 


38.6 


39.4 


2 


Nov. 


25, 11.00 A. M. 


Nov. 


25. 3.30 p.m. 


76.5 


74,3 


70.3 


74.3 


3 


Nov. 


27, 2.00 p. M. 


Nov. 


27. 4.00 p.m. 


75.5 


93 


98.2 


93 


4 


Nov. 


28. 12.30 p. M. 


Nov. 


28. 6.30 p.m. 


150 


183.5 


194.2 


191.5 


5 


Nov. 


29. 12.30 p. M. 


Nov. 


29. 5.20 p.m. 


247 


183.5 


194.2 


191.5 


6 


Nov. 


30. 1.00 p.m. 


Nov. 


30. 4.50 p.m. 


300 


183.6 


247 


191.5 


7 


Dec. 


1. 10.00 A. M. 


Dec. 


1. 2.10 p.m. 


300 


300 


298 


300 


8 


Dec. 


1. 4.30 p.m. 


Dec. 


1, 6.15 p.m. 


300 


346 


298 


346 


9 


Deo. 


2. 9.30 a.m. 


Dec. 


2. 9.50 a.m. 


300 


372 


298 


372 


10 


Dec. 


2, 10.00 A. M. 


Dec. 


2. 10.15 A. M. 


300 


385 


298 


385 


11 


Dec. 


2. 


Dec. 


2. 12.05 p. M. 


300 


385 


298 


385 


12 


Dec. 


3, 9.00 a.m. 


Dec. 


3. 9.45 a.m. 


300 


385 


298 


385 


13 


Dec. 


3. 9.00 a.m. 


Dec. 


3. 2.00 p.m. 


380 


465 


376 


465 


14 


Dec. 


5, 10.30 A. M. 


Dec. 


5. 4.00 p.m. 


500 


465 


376 


465 


15 


Dec. 


6. 9.40 a.m. 


Dec. 


6. 10.40 A. M. 


527 


465 


376 


465 


16 


Dec. 


6. 11.20 A. M. 


Dec. 


6. 12.15 p. M. 


550 


465 


376 


465 


17 


Dec. 


6. 12.50 p. M. 


Dec. 


6. 1.35 p.m. 


567 


465 


376 


465 


18 


Dec 


6, 2.10 p.m. 


Dec. 


6. 3.00 p.m. 


591 


465 


376 


465 


19 


Dec. 


7, 8.00 a.m. 


Dec. 


7. 9.20 a.m. 


615 


465 


376 


465 


20 


Dec. 


7. 10.60 A. M. 


Deo. 


7. 1.10 p.m. 


650 


465 


376 


465 



Both reinforcement and concrete stresses, as well as deflec- 
tions, were measured throughout the test, exceptionally complete 
sets of readings being taken imder the design load of 150 lb. per 
sq. ft. and under the test load required by the City of St. Louis, 
t. e,, a superimposed load equal to once the dead load plus twice 
the live load, or 380 lb. per sq. ft. 

Deflections were read by measuring the distance between a 
steel plate fastened to the ceiling and a steel rod held in a scaffold 
below. An inside micrometer reading to 0.001 in. was used, both 
plate and rod having countersunk holes to locate exactly the posi- 
tion of the instrument. 



Digitized by 



Google 



136 Report of Committee on Reinforced Concrete. 

The stresses in the remforcement were read by an exten- 
someter which is a modification of the Berry Strain Gauge used to 
read stresses on the concrete, while W. A. Slater read reinforcement 
stresses with a modification of the Berry Strain-Gauge, made at 




FIG. 59. — PLAN SHOWING LOCATION OP DEFLECTION POINTS. 

the University of Illinois, Figs. 1 and 2. Both of these instruments 
have points which are inserted into small holes drilled in the steel 
or in metal plugs set in the concrete and read directly to 1/5000 
of an inch, or by estimation to one quarter of this amount. An 



Digitized by 



Google 



Report op Committee on Reinforced Concrete. 137 





Table XIV.- 


-Deflections, Barb Building Test Panel. 






Unit 
Load. 


76.5; 150 


300 


300 


380 


380 500 


500 


627 


550 


567 


591 


615 


660 


Deflec 

tiOD 

Pomt 


Stage. 


3 


4 


6 


11 


13 


13 


14 


14 


15 


16 


17 


18 


19 


20 


Date 
and 
Hoar. 


^i 

ii 


1^ 




•^1 




ii 


11 


1= 




4 




• si 

^3 






1 






.354 
.266 
.198 
.153 
.130 
.117 
.110 
.127 
.142 
.134 
.296 
.213 
.160 
.104 
.094 
.089 
.078 
.086 
.090 
.091 
.071 
.054 
.043 
.034 
.024 
.019 
.022 
.025 
.030 
.116 
.098 
.082 
.075 
.068 
.073 
.068 
.072 
.082 
071 
.063 
.054 
049 


".Hi 


.903 
.724 
.559 
.428 
.420 
.421 


1.340 1.417!l.456 
1.043{1.099ii 2lfl 












1.456 
1.216 
1.017 
.870 
.952 
1.167 
1.332 
1.631 
1.921 
2.155 




2 


















3 






.794 831 
















4 






.591 
.567 


.611 

.581 
.568 
.658 
.778 
ROT 
















5 




















6 




















7 






.480; .645 
.568' 764 

















8 




















9 






.645 
.693 
.860 
.598 
.438 
.302 
.317 
.310 
.352 
.408 
.460 
.067 
.402 
.331 
.265 
.205 
.172 
.148 
.154 
.165 
.179 
.530 
.448 
.362 
.293 
.262 
.244 
.238 
.204 
.248 
.386 
.324 
.260 
.202 
.177 
.159 
.167 
.186 
.202 


.878 

















20 






.9601 .982 
1 153i 1.222 
.890' .940 
.6361 .668 
.419, .438 
.424) .434 
.4101 .417 
.468, .478 
.5461 .558 
.622| .633 
.073 .078 
.5441 .553 
.4441 .453 
.354 .358 
.268 .274 
.234 .247 
.228' .239 
.2711 .290 
.330, .355 
.3941 .431 
.7251 .742 
603 616 
















10 




















11 






















12 






















13 






















14 






















15 








1 












16 








. 












17 








1 












18 








1 












19 
21 


Col. 


.02 




.100 .103 


.107 


.109 


.114 




.126 


.137 


22 




















23 






















24 


















25 























26 






















27 






















'28 






















29 






















30 






















31 






















32 






.477 
.376 
.351 
.351 
.373 
.415 
.467 
.527 
.432 
.338 
.251 
.228 
.232 
.263 
.324 
.880 
.066 
.155 
.900 


.488 
.384 
.361 
.366, 
.3921 
.442 
.502 
.545 
.446 
.347 
.258 
.238 
.2451 
.285, 
.3501 
.414 
.068 
.1581 

02fi 


















33 






















34 




















35 








I"": 












36 
























87 























38 






















39 






















40 






















41 






















42 























43 






046 


















44 






.053 
.064 
.079 
.095 
.016 
.042 
.104 
.274 
.124 
.229 
.013 
.071 


















45 






















46 






















47 








!!.!. :.... 












48 
49 
60 


Col. 


.003 
.027 


.033' .047 
.118, .119 

' .643 

.7741 .925 

.72.'5 


.085 
.264 


.087 .094 .095 
.2741 .2941 .324 

■ 1- 


.099 
.348 


.103 
.379 


.107 
.414 


.115 
.409 


51 
52 


Center 


.129 


1.29911.331,1.947 
.993 1 0171 


2.6192. 163'2. 327 

1... 1. . 


2.444 


2.658 


2.897 


3.313 


53 
54 
55 


Col. 

"cor* 


.009 
.040 
.039 


.152 
.211 
.114 


.127 
.335 
.078 


.148 
.434 
.105 


.154 
.449' 

.113| 


.162 
.535 
.130 


.175 .176, .171 
.5451 .56ll .586 
.136| .139 .143 


.170 
.601 
.147 


.175 
.627 
.150 


.190 
.659 
.158 


.196 
.702 
.165 














:::::i::::: 


1 
























1 














51 


Corr 

1 


0.080 


.167 


.594 


.641 


.920 


.9441.459 


l.sie 


1.632 


1.777 


1.872 


2.052 


2.252 


2.608 



Laat line of readings are deflections at center of panel assuming supporting beams to be rigid. All 
other readingB eorrected for pier settlement only. 



Digitized by 



Google 



138 Report op Combiittee on Reinforced Concrete. 

8 in. gauge length was used throughout the test. The length 
of such an instrument varies with changes of the temperature of 
the observer's hands, to correct for which readings were taken at 
intervals on standard bars and all temperature corrections neces- 
sary have been applied to the stresses given in the tables and 
plotted in the curves. 

Method of Loading. — To prevent the possibility of the load 
arching, the sacks of sand were arranged in separate piles, as 



&50 
























































^ 


^^ 


^^^ 


^ 


,^' 


--" 






















^ 


^ 




_^ 


^ 


J>^' 








A/)/) 
















^ 


^ 


<: 


^ 

y 


^ 












0\iftlO 












(S 


p\ 


W 


A 


/^ 
























A 


P 


A 


K 


/ 


















^JMD 










/mi 


J 


^U 


L_& 


p.SL±U 


^1 


rstLoad 










a«^^c 








J 


T 


^y 


.^ 










|5<» 








^ 


6 


/ 




























/ 


/ 
































/ 


/ 










Curve / 5ha¥5 defk 


ortlon of h 


Hp 








V 


r - 


siqn 


Lope 


1 




center of the panel wilt) reelect to 
the columns. 

Curve t shows deflection of 
Jhe center of the panel with respect 






S/oo 




f 
















>«4 
















to the supporting beams 

. 1 . 1 1 1 






^Xl. 




































c 




ilea 


4 d. 


6 d. 
in 


6 / 
/nch 


ea. 








/ 











3. 








FIG. 60. — DEFLECTIONS AT CENTER OF PANEL. 

shown in Fig. 57. The wide aisles, Fig. 58, were necessary for 
the accommodation of the instruments and observers. Table 
XIII gives the loading on the various portions of the panel at 
different times, as well as the loads per square foot of slab and 
cantilevers and the date and hour at which each stage was started 
and completed. The loads per square foot are in each case 
obtained by dividing the total load on the panel or cantilever by 
the area, the intensity of load under the piers being much greater. 
Up to the test load required by the City of St. Louis the 



Digitized by 



Google 



Repobt of Committee on Reinforced Concrete. 139 



Deflection Points 

QJS^XJfM^XJl JO 




Deflection Curve along Center- line of Panel in Direction of long Span 
Under Design load {I50*per x^n) 
Deflection Points 



fl J 


«J 


7 S 


6 S 


5 34 33 S 


r s 


JO 


A 


3t 34 




















































































^ 
























^ 


' 








> 


% 


























^ 








y' 




















\ 






y 




did 


















K ^ 




^ \ 




















X # 




^ 1 












•^ 










V w, 




/ !^ 




1-0 








'^ 










>v J? 


y « 




I.I 


















^s. 


" 


^ 


>^k 1 


/« 




















^^ 




.1 



Deflection Cunn along Center- line cf Panel In DIrectloh of long Joan 
under Test Load required by the 3t Louis DIdg Code (5acrperjif.fr.) 

WIQ. 61. — ^DEFLECTION CXTRYE IN DIRECTION OF LONG SPAN. 





1 i 


ilea 




Pomtj 

3 i 




r f 




10 a/ s 


4« 










L 










\ 




^ — 1 








r^ 


-r 








< 


^ 1 


^^ 


i 


.50 


J 


/ 




% 














I 


V 




Dm 


flection 


Cm 


rve 


a/a 
una 


u 


\iisign lSoS (lOL 


^;jr"4^'^' 


Tf Short span 





1 J 


)efk 


vtk) 


nPi 


vnt. 


f 

5 7 i 


> s 


} I 


D 


£ 


V 


s 





^ 


9 






























.to 
























_^ 


























y 




























X 




























/ 


— 


M^l 












•k 








. 




/ 


70 






/ 






■^ 


s 






> 




/ 






r 








> 






B 




^ \ 






/ 












2i 




•4g 




/ \ 


E 






f 














^ 




y 


**» 


uo 

LtO 




















^ 


^ 






















v^ 


^ 






























/JO 



































Defkchon Curve along Center- line of f^jnet in Dined fop of3hort Span 
under Test load required fy the St. Louis 3/dg Code(3d0^prs(f.ft) 

FIQ. 62. — ^DEFLECTION CURVE IN DIRECTION OF SHORT SPAN. 



Digitized by 



Google 



140 ReFOBT of ComUTTEB ON ReINFORCEO Ck)NCRIiTB. 

cantilevers were carefully loaded to horizontality, but from this 
stage on no more load was applied to the cantUevers and they 
rose slightly, thus increasmg the deflection at the center of the 
panel. 



I--!- Kr-««"B55— B» ^ 



gHDnnnnnniiBannnnDD 
TinDDDnnnnnnnnDnnnn 
DnDnnDDnnDnnnDDDnd 
DnDnnDnDDDnnnnnDn 
nnnnnDDnDDnnnnDDDD 
nnDnnnnDDDnnnnDnnn 
DDDDnDnnDDnnDnnDD 
nnnnnnnDDDnnDDnnaD 
DnnDnDDDDDnDnDnDD 
nnannnDDDDDnnDnDnL 

•-• Indicates Points onfop of Slab. 
o-o Indicates faints onBoltomafSab 



aDDDDDDaDDDDDnnDa 

nnnnnnnnnnDnnnnnn 
DaDnanDDnnnnnnnDD ., 
DDanDDnnnnDnnDDnnngj 
pnnnnnDnnDnDannnnn^ 
DDnannnnnnnnnDnnna 

DDDDDDDDQDDDDDDDD 

^nDnannnnaDDnnnnnn 



no. 63. — PLAN SHOWINQ LOCATION OP GAUGE POINTS ON CONCRETE. 

The horizontality of the tangents over the beams was deter- 
mined by measuring deflections of both the slab and cantilevers 
at a series of points along a line perpendicular to the beams. 
The deflections were then plotted and a smooth curve drawn 



Digitized by 



Google 



Report of Committee on Reinforced Concrete. 141 



which showed at a glance whether or not horizontality existed. 
It was often found necessary to shift the load on the cantilevers 
several times before the correct amount and position were 
reached. 



.40) 




4O0 



b 500 Kxx> o mTIoSS ^ , 
Unit Compression in Concrete 



SCO 1000 O JoO /OOO 1500 




tddd d 3^ md o S^ }6do 



Point5ii 



O 300 1000 O 500 

Unit Compression In Cortcnefe 

FIG. 64. — ^DIAGRAM OF STRESSES IN CONCRETE. 



:305- 



Deflections. — Deflections were read at the points shown in 
Rg. 69. A summary of the observations is given in Table XIV. 
There was a settlement of the footings which has been corrected 
for in the summary and in the curves. 



Digitized by 



Google 



142 Report of Committee on Reinforced Concrete. 

The center point deflections are plotted in Fig. 60. The 
deflections shown by the curves represent the deflections of the 
panel with respect to the supporting beams. In order to arrive 
at the deflection of the panel proper, the gross deflections were 
decreased by an amount equal to the average deflection of 



3600 

2400 

2200 

ZOOO 

%I600 

I f600 

^ I400 

^l2O0 

% 

%I000 

. 600 

§ 600 

% 400 

ZOO 
O 























^•*" 


















y^ 


^ 
















^ 


/ 


















/ 


/ 


















^ 


/ 




















/ 




a-*" 


,^^ 


.^ 


hroke 










/ 




/ 


^ 














/ 




y 


















/ 


r 


















/ 


/ 




















y 




















J 













































,000£ J00O4 .0006 .OOOS 

Unit Deformations '(6 gauge length) 



.ooto 



PIG. 65. — STRESS-STRAIN DIAGRAM OF TWO CONCRETE TEST CYLINDERS. 

the supporting beams. Curve 1 shows the gross deflections, 
the dotted curve including settlements occurring during the 
intervals between loads and the solid line giving the true elastic 
deflections (including beam deflections) caused by a uniform rate 
of loading. The dotted portion of Curve 2 ehows the same thing 



Digitized by 



Google 



Report of Committee on Reinforced Concrete. 143 

after correcting to a uniform rate of loading. The solid line, 
Curve 2, should be used in comparing this test with others or as 
a basis for deflection coefficients. 



T^-^-fi 



-f 






' DnnnDnnb" 



otpt09 

-"IBS- 



\^ 



DDDnnnn 

,nnnnDnna 

DDDDnnan 

.nnanannn 

nnnnnaD 



«r »* M 



X 




V 

DDDDDnnnis 



DDDnaaDa 
nnnnnnnnjs 

DDnDDnaDj 

nnnDDDDnT 

DDDnnnnnraannnnDDni^ 

nnnnnnanDaDnnnnnnDT 
aDnnpnaDDDDDn '" 




]0[ 




nnnnnDnnDnnnnDnDDD 

DnnnDnnnanDnnDnDDD 

DDnnnnDDDDDnnDnnDn 

nnnnnnnDnDDDDnnDaa 

nDDDnnnnDDnnnDDnnn 

DnDnDnnnnDDDDDnnDn 

DDDnDnnDDDDDDDaann 

nnnnnnnDDDDDnnaDnn 
^ ^j 



•-• Indicates Points on Top of Slab 
oo Indicates Points on bdttom of5lab . 



PIQ. 66. — PLAN SHOWING LOCATION OP GAUGE POINTS ON REINPORCEMSNT. 

Fig. 61 shows exaggerated scale sections of the panel along 
the long axis for the design load and for the load specified by the 
St. Louis Building Laws. The deflections, from which these are 
plotted, are corrected for settlement of the footings only. Fig. 62 



Digitized by 



Google 



144 Report op Coboiittee on Reinforced Concrete. 

shows similar sections along the short axis of the panel. Inspec- 
tion of these figures will show that the cantilevers were properly 
loaded to obtain horizontal tangents over the supporting beams. 
Deformations and Stresses. — Concrete stresses were read at- 
the points shown in Fig. 63. A summary of the observed stresses 
is given in Table XV; the loadnstress curves at a few of these 
points are plotted in Fig. 64. In these curves and in the sum- 
mary, the stresses are those which were induced in the concrete 
by the loads greater than 75 lb. per sq. ft. At this stage the 
points, at which these stresses were read, worked loose and had 
to be reset. 



Table XV.— Concrete Stresses, Barr Building Test Panel. 




Unit 
Load. 


150 


150 


160 


800 


380 


880 


Gaufce 
Point 


Stage. 


4 


4 


4 


11 


13 


18 




Date 


Nor. 28 


Nov. 29 


Nov. 29 


Dec. 2 


Dec. 3 


Dec. 4 




and Hour. 


7 p.m. 


10 A.M. 


12 m. 


12 m. 


8.30 p. M. 


9.30 p.m. 


401 


R.O. 


-40 


-285 


-81 


-625 


-796 


-765 


402 


R.O. 


+40 


-162 


+ 122 


-430 


-422 


-422 


801 


R-0. 


—195 


—195 


—203 


—943 


—1225 


—1322 


803 


R.O. 


-195 


-504 


-211 


-1005 


-1322 


-1370 


805 


R.O. 


-105 


-350 


-57 


-740 


-935 


-1022 


802 


R.O. 


-26 


-858 


-57 


-723 


-870 


-967 


804 


R.O. 


-122 


-382 


-162 


-796 


-1022 


-1095 


808 


R.O. 


-58 


-252 


-73 


-1005 


-862 


-926 


810 


R.O. 


+65 


-219 


+130 


-740 


-796 


-943 


807 


R.O. 


+146 


-471 


+90 


+130 


+365 


+211 


809 


R.O. 


-49 


-398 


-81 


-292 


-220 


-374 


811 


R.O. 


-80 


-325 


+0 


-560 


-480 


-600 



+ as Tension. 
— »Compreimon. 

The reduction in stress with increasing load appearing in 
some of the curves may be the result of the gradual giving away 
of the concrete in tension and resulting shifting of the point of 
inflection. 

Two test cylinders 8 x 15 in. were taken from the mixer at 
the time the slab was poured and after being cured under the 
same conditions as the panel itself, they were tested just after 
completion of the readings on the panel. The ultimate strength 
and the modulus of elasticity of these were widely different as 
can be seen from the stress-strain diagrams in Fig. 65. In deriv- 
ing the concrete stresses the average modulus was used, being 
taken at 3,250,000. 



Digitized by 



Google 



Report of Committee on Reinforced Concrete. 145 






s 









a 

5 









1^ 



*<»9i 



"si 



1^ 






t4 



ecrt 
4 + 






+ 



II 
++ 

|i" 

++ 



+ 

I 
I 

II II ll II 

e5« SS 5w e«e« 

++ +4- +4- 



++ 



§1 i§ §1 

<DO C4C4 OOOO 

^H«H <^<^ ^4<-l 

++ +4- 4-+ 



4-4- 



4- 

ii 



++ 



II II 

4-4- .4-+ 

11 ii 
-f4- -f4- 



§ 

+ 
8 

r^ - 

? 

§1 11 ii 11 i| 

Soi eoeo «c8 «w «eo 

4-+ -f+ -f+ -f+ -f-h 



§i §1 §§ i§ §§ 

sa S2 sa 8"s a:s 

4-+ 4-4- 4-+ 4-4- 4-+ 



§;i;i§ii;Siil§ 



I I 

4-1 



Digitized by 



Google 



146 Report of CoMHiiTTEB on Reinforced Concrete. 



f 

< 
fit 



B 



PQ 

a 



1 



3 



i 


s 




















+35W 






§ 


s 


^3 






: :' 


i i 








i Hi? 






§ 


2 








! :1 


f : 






: 4 


i +? ?$ f ? j! 




3 


2 

^5 


li 














: ++ i+ :+ :■ 


12 


























1 


2 


9i 










































jl 


1? 


: + 






§ 


^ 


^5 






i § 


: + 








jl 


: 1 
: o» 

: + 


: + 


il 




i 


CO 


ii 






: + 


\^ 








i t 

i i 


: + 


it 


i' 




i 


s 


n 


III 


i ^ II II 1! I| i! 

: ++ ++ ++ ++ ++ :■ 


1 p ti ^ 

i ^s sS ?$ 
^ ++ ++ +^ 


§ 


11 


^ J 

«^^ 


11 i 

^S 3 
++ + 


i |§ ig 11 IS IS il m II il 

i ^i i+ i+ i+ T+ n s2 == %% 

- 1 


g 


•« 


Si 

|i 


+ 


: S 




\ i 
: 1 


\l 


; § 

: + 


: + 


11 


■ + 


: + 

il 

; + 


: g 

: + 

i g 




§ 


•* 


1 

+ 


; + 




: i 








: S 

i t 


•O 


e* 




1 


: 1 


ii 


: o 


\ ? 


: 1 


1 1 

1 7 


: + 


j + 


: + 


1 • 




11 


1 


HI 




























U 


! 


s 


8 


00 




§ 


§ 


55 


§ 


§ 


§ 


i 





IJ 

n n 
+ 1 



Digitized by 



Google 



Rbpobt of CoMMiTTBfi ON Reinfobcbd Concbets. 147 





§ 


S 


li 






3 ii 








r f 












§ 


S 

o 


eo 


:1 


f :■ 


if 












1 


18! 


i if 

l It 


5 i 

r : 










1. 


3 


00 

5: 






u 


f :■ 
















< 
Oh 




; 8 

: !J 

: + 
















i 


i 


s 










1 
















1 

P8 


§ 


j5 


i 

^5 


-: 


i — :- 


:— : 


1 
















< 
H 


i 1 


1 

+ 


11 










i 


2 

CO 








i i 

: 4- 


+ 


!^ 












g 
S 
g 


ii 11 II II 1 
?^ 1? ?? $^ + 


1^1 


i 11 II ll If 

? 77 77 ++ +' 


t 


§ 


*<• 






nil 

4- ++ + 


i 11 i| 11 m \ 


III 
T 77 


H 


I 


i S 


ii 


: 1 

; + 


. § 

1 


i § 


if 


i^ 


i^ 


: ::- 

: + 


: + : 


5 


S 


' 


I: 


1 


i 1 

: + 


\i 


; 1 


















J2 


M 


1 ^ 


': + 




1 

T 
















11 


1 


HI 


















II 


§ § s 


2 


i 


3 


s 


§ 


e 


e 


3 



II 

I I 

+ 1 

I 



Digitized by 



Google 



148 Report op Committee on Reinforced Concrete. 



1 






PQ 



PQ 



QQ 



1 






in 

CO 



IS 






Mi" 






11^ 



i 



ii" 



?l 






11 II II 

tz ++ %% 



i| §1 

++ ++ 



i| §1 11 II II II i 



+f 



Vh «4- ooT ■^co ■"♦"o 
I -I. ' ri-i- -^r" -^-^ 



+ 1 + 



■*« '♦O c« 



++ ++ 



+ 



I I 



I I 



04 04 
i I 



i§ i§ §§ 



I I 



I I 



I I 



T' 



11 

I I 



1 


T§' 


: i 


1 


" : 1 


; 1 


s 


: o 










M 


t>- 




4 


• : + 


-i— ^ 


— 




-:- -: 



s s s ss s; 



e4 *i4 1-4 v^ 



HO 

II n 

+ 1 



Digitized by 



Google 



Report op Committee on Reinforced Concrete. 149 





§ 


s 




: 






















-•— 




§ 


s 
























1 


s 


s 


l1 




1 \ 


i i 

s ; 

+ : 


1 i 

4- : 


1 : 

4- : 
















1 


1 


ss 


i^ 




? 1 


1 ^ 


1 \ 

? : 


? : 


-:-4- 


-; -[ 


-:— f 


i--i- 








o 

s 


§ 


»« 


i.' 










i~F 








§ 




ii 


— : 


















PS 




\ — : 










4—^ 


-:- 


i 


§ 


<<l 








! 


§ 


s 




























I 


§ 


s 




? 


§ ii 11 1 


§ §1 


1 s§ §! 


f +4- + 




? 


s 

1 


"t" t* k * 


1 §§ §? §< 
+ :f+ + + 


7+ : 


1 




% 


^ 


CI . 


1 


11 


i + 


: a 


4- 'r 


11 


if 


i g 


i 1 

it 


: + 


: J2 


-i— 


S 


s 


'i* 






1 




-f 


-i-f 






i; 


u 


g 


S5 


§ 


s 


§ 


§ 


s 


s 


s 


Z 


2 





II 

HO 
I 
+ 1 



Digitized by 



Google 



150 Report op Committee on Reinforced Concrete. 



I 



B 



CQ 



^ 



a 

S3 



§ 


S 


H 






i 1 
; S3 

: -t 




§ 


8 


-* 

^i 






i 1 

1 ^ 


e 


2 








i § 

: or 
: + 


3 


00 












§ 


J; 


12 










§ 


<o 


i^ 










§ 


3 


i 






: + 


§ 


««• 


1^ 


1 

1 


: 1 


ii 

: + 


i 


S 




7 


ji 


jl 


§ 


S 




SI 


1 


§ 


- 


ii 


ill 

eo<4< CO 
1 1 1 


i § 
i ^ 


§ 


•* 


i^ 


+ 




il 


1 


-* 












5 


M 


SB-" 


1 




i U 

: + 


11 


1 


111 












!• 


1 

4 


§ 


s 


9 





HO 
H I 
+ 1 



Digitized by 



Google 



Report op Committee on Reinforced Concrete. 151 

Fig. 66 shows the location of gauge points at which reinforce- 
ment stresses were read. A summary of the observed stresses is 
given in Table XVI, together with the load and time at which 
each set was taken. The two figures given for each point are 
by the two different observers. Figs. 67 and 68 show loadnstress 



700 



r 

S 









J 


















--• 










/ 














^ 


^ 


r 










f 




1 


Xldr^ 


/ 










^ 


:? 






.^^Siid 


• Simmi 




^ 


JSW 












/ 


A 


oint 


201 






/ 


A 


7/nf 


?«? 








/ 




De^ 


^nu 


w/ 




y 




























I 












f 










• 


1 



















200 


w 


4a 


MO 


( 


> 


toe 


w 


^00 


00 









Unit Tension in Steel 



700 
























i 






^600 




















^ 


/^ 








I 


— 


-/"/fw 


, j-^ «i 


tfutim 


Mi 


n^ - 






/ 


/ 










ZySlJiLi^SMaW. 


TSL. 


'"' 


> 


7^ 






■" 




"' 




Point 219 








/ 


P 


■>lnt 


113 










— 


.^S&ML^ 


U)ad 




I 







— 



















/ 










z 












{ 
















( 


u 


nit Ti 


100 

vrsiOi 


40C 

1 In 


100 

Jfee 


, ' 


) 


tXK 


XX) 


40(y 


00 









no. 67. — ^DIAGRAM OF STRESSES IN REINFORCEMENT. 

curves for a few of the bars at critical points and for the struc- 
tural steel supporting beams. 

Fig. 69 shows the distribution of stress in the bars which are 
on the short span. The section shown is taken along the long 
axis of the panel and the stresses plotted are those in the bars 
cut by the section. Similar curves are shown in Fig. 70 for the 



Digitized by 



Google 



152 Repobt of Committee on Reinforced Concrete. 
TOO 
^600 

\600 

T 



s- 



JOO 



^ 



















/ 


r 
















/ 










/ 
















1 


Ux)di 




1 














z 




n 


-r-— - 












/ 


^ 1 1 
Point IZZ 






/ 


Point I2Q 

1 1 








t 


— 





,_D^lgn_Loa<L- _ 


/ 




— . 


— 



















/ 


1 














' 















O tOOOO 40000 

Unit Vuision in Steel 



gOOOO 40000 



^600 



poo 



1»9 

sax? 



100 



^ 



Point ioi-\ 
Outer Beam jr 


r 
















/: 


/ 


« 






/ /" Point 103 
( /Inner Beam 






Point 124 
Outer Beam 


f 










/ 


f 


Iru^A 






V 


^ Point 126 
Inner Bean 


' 




~""~ 


1/ 


bLJdnLfL: 


^t.UJU 


±!S^ 


ipga_ 


■^~"~" 


J 


¥'- 


"" 


■ 






1 


/ Points 101 A 


105 








/Points lZ4d 


l?6 






/ 





— 


_j? 


?^£'7 


Load 


.„ 


-/ 


— 
















r 














{ 














5-^ 


■ tOi 


odd 


4o^ 


000 




c 


) 


zoo 


)00 


40i 


000 







Unit Tension in Steel 



FIQ. 68. — DIAGRAM OF STRESSES IN REINFORCEMENT AND IN STEEL 
SUPPORTINQ BEAMS. 




\40000- 
% 30000 

\toooo 

\ 10000 
O 

no. 69. — ^DISTRIBUTION OF STRESS IN THE REINFORCEMENT CABBTING THE 
LOAD ACROSS THE SHORT SPAN. 



Digitized by 



Google 



Report of CoMinTTEE on Reinforced Concrete. 153 

top bars over the supporting beams. Some of these curves may 
be distorted because the supporting beams tipped up on the 
bearing plates and spread apart at the comers, thus inducing 
tension across the diagonals of the slab. 



V. ' . *".4. ' . ' . ; .".'- ' ^ » .v.m.nij r »...^.ni. i i.-;.-.i ' .i.^\-.n-'. . ■ .,*■•■■..>.• ... I ... ■.. ' •'••• .'..S-'V ^» '.*.■. ' ■• i-. ■ ■ ' • ' •1 *1 




Dbtri button cf Jtress In Top dars o/er the Z5'cf Supporting deam 

f ^ ■•■■ • :■' v-:^|. ' •::;:• ^^•■^^:v;^:^^^T:^.^^•^^•■:•^'^^^*••V••?|• ^' •"•^ 




P0intZ09 Z07 Z05 Z05 ZOI 

Distribution of Stress in Top Dars arer d6'9* 
Supporting Beam- Dars extend into Cant Her er 



10000 




ointzn z 



\SQO 



}\//in^ 



hi30 



Point Zn ZI5 Z15 ZI7 ZI9 
Distribution of Stress in Top 5arj 
over Z6''9'' Spandrel Ream 

FIQ. 70. — DIAGRAMS SHOWING DISTRIBUTION OF REINFORCEMENT STRESS. 

At the 500-lb. load, cracks could be found extending a short 
distance from the comers of the panel along the diagonals. 
Pig. 71 shows the location of cracks at a load of 660 lb. per sq. ft. 



Digitized by 



Google 



154 Report op Committeb on Reinforced Concrete. 

In order to locate the points of inflection and to determine 
the bond stresses; readings were taken along portions of three 
bottom bars. The results of these readings are plotted in Figs. 
72 and 73. 



c 


^'1 1 1 1 


c 




L 




c 




D 




C 




c 




c 




c 




c 




c 




c 




c 




c 




c 




c 




c 




c 




c 


u 1 1 1 1 





I I 

I 
I 
I 



annnnDannDDaai 

gaDDaaaaDannnDQa 

nnnnannnDaanDaaa 

I cgoannsaaDDDDaDnaaa 
I nannnnanDDDDnDDaan 
I nnDDDDannDBoonannn 



f-SDnnnnnDnDanaannna 



D 
D 
D 



n 



I 



aaanDaDnannaDnDDDD 



nnnaDDDaQDnnanna 



DDDn 



DnnaQBaDE3QaDDnDD 
nnnnnDDnnaDaaaDD 
DnDnDDnnannnnnDa 



□DC 



DDD[ 



DDDDD 



! DaDDanDannnannaD! 



aDDDDDDDnaDDDDnn 
DDDDDDDaDDDDa 

DDDDDDannnbg 



3 QEPDDaDDDC 



DD 



JDDDC 




FIG. 71. — PLAN SHOWING LOCATION OP CRACKS. 

A summary of the stresses and deflections at the critical 
points of the slab is given in Table XVII. 

No definite conclusions as to the distribution of stress among 



Digitized by 



Google 



Report of Cobimittee on Reinforced Concrete. 155 




10000 §*^ 

zoooo ^^ 



a^^imyffmfM 



6 



^ o 



^^zoood 



6' 



6' 



8' 




8' 



8' 



8' 



8' 



8' 



Point of inflection 30Cf ' 



Load-/30*. 
tjoacl-300' 

Load-36(f 



no. 72. — VARIATION OF STRESS AND LOCATION OF POINT OF INFLECTION ALONO 
MIDDLE BAR IN DIRECTION OF SHORT SPAN. 



Digitized by 



Google 



156 Report of Committee on Reinforced Concrete. 



Table XVII. — Summary of Principal Stresses and Deflections, Barr 
Building Test Panel. 

Deflections at Cbnteb. 



Design Load 160 lb. per sq. ft. 



St. Louis Building Code 380 lb. per sq. ft. 



0.167 in., or j^„ of Spian 



0.860 in., or gf, of Span 









CoNCBBTB Stresses. 








[>de 
l.ft. 




Gauge 
Point. 


Design T.oad 
160 lb. per sq. ft. 


St. Louis C 
380 lb. per sc 




L. L. 


D.L. 


Total. 


L.L. 


D. L. 


Total. 


Top of Slab at 
Center 


Long Span 
Short Span 


402 
401 


100 
100 


100 
100 


200 
200 


675 
800 


100 
100 


775 
900 


Bottom of Slab at 

Flange of 
Supporting Beam 


Lon 

Short 
Span 


*1Ci 
End 
Free 
End 


304 
801 

300 


325 
400 

160 


150 
200 

75 


475 
600 

225 


1,250 
1,525 

425 


150 
200 

75 


1.400 
1.725 

600 


Bottom of Slab on 
Rib just outside 
of Beam Flange 


LOD 

Short 
Span 


End 
Free 
End 


308 
303 

311 


175 
400 

175 


75 
200 

100 


260 
600 

276 


926 
1.676 

700 


75 
200 

100 


1.000 
1,776 

800 







RsiNrORCElfENT 


AKD Steel Stbbsses. 








Location. 


Gauge 
Point. 


Design T^d 
160 lb. per sq. ft. 


St. Louis Code 
380 lb. per sq.ft. 




L. L. 


D.L. 


Total. 


L. L. 


D.L. 


Total. 


Bottom of Slab at 
Center 


I>ong Span 
Short Span 


122 
113 


3,600 
7,500 


2,000 
3.600 


5,500 ' 19,500 
11.000 , 26,600 


2.000 
3.500 


21.500 
30,000 




Long Span^ 


210 


3,600 


1.600 


6.000 


37,600 


1.600 


39.000 


Top of SUb at 
Support 


Short 


l-ixea 
End 


201 


6.600 


3.600 


10,000 


28.600 


3.600 


32.000 


Span 


Free 
End 


219 


7.600 


4.000 


11.500 


8.600 


4.000 


12.600 




Short Span 


101 & 
103 


7.000 


3.600 


10.600 


17,600 


3,600 


21.000 


Bottom Flange of 


Long 


Spandrel 
Beam 


128 


3.600 


2,000 


5,600 


9.500 


2.000 


11,600 


I Beam 


Span 


Interior 
Beam 


124 & 
126 


7,600 


4,000 


11,600 


22.000 


4.000 


26.000 



load. 



Note. — Stresses due to dead load are obtained by projecting the stress curve to the soro 



Digitized by 



Google 



Report of Committee on Reinforced Concrete. 157 

the two sets of ribs can be reached from the data obtained in this 
one testy although the stresses obtained indicate that about 






nMBSSR^^^^^Sfl^^tSimrS^HIi^ 




FIQ. 73. — VARIATION OF STRESS AND LOCATION OF POINT OF INFLECTION ALONG 
MIDDLE BAR IN DIRECTION OF LONG SPAN. 

65 per cent of the load was carried across the short span of the 
panel, instead of the 55 per cent assumed. 



Digitized by 



Googk 



DISCUSSION. 



Mr. Ash. Mr. L. R. Ash. — I would like to know if experiments have 

been made to determine whether the modulus of elasticity of high 
carbon steel is really more than that of medium steel. Some claim 
that high carbon steel should be used because of the increased 
modulus of elasticity. 

One other question is whether experiments have been made 
to determine the distribution of stresses in a continuous slab with 
a concentrated load. A problem that frequently comes to the 
bridge designer is how to take care of a continuous slab which has 
large dimensions one way and small dimensions the other, for 
example a slab supported between two continuous stringers. 

Mr. Talbot. Mr. Arthur N. Talbot. — ^The modulus of elasticity of 

steel is a fairly definite property. It varies from 29,000,000 lb. 
per sq. in. to 31,000,000, perhaps; the average of tests would 
bring it a little imder 30,000,000 lb. per sq. in. I do not know 
that anything has been foimd which explains or gives the causes 
of this slight variation. I am quite sure, however, from tests 
which we have made, that there is no appreciable difference 
in the modulus of elasticity between the high carbon and the mild 
steel. 

It is an important question to determine to what extent it 
is necessary to reinforce a slab laterally when a concentrated 
load is carried. We have been making tests along this line at 
the laboratory of the University of Illinois for three years and we 
hope to have results to give out soon. I may say now that I 
have been surprised to find how little reinforcement is necessary 
laterally in distributing the stresses sidewise. In these tests we 
have used beams 30 and 48 in. and longer, with depths of 4 to 7 
in. and with widths of 30 to 48 in. These slabs were loaded in 
several ways, by applying the load entirely across the beam, by 
appljring it over half the width of the beam and also over 1/5 of 
the width and over 1/10 the width. In a general way it may be 
said that with as much as ^ of one per cent of lateral reinforcement 
the stresses are distributed laterally over a width of say 8 or 10 
times the depth of the beam. 

(168) 



Digitized by 



Google 



Report of Committee on Reinforced Concrete. 159 

Mr. W. p. ANDEBfiON. — In making the comparison with the Mr. Andenon. 
bending moments, what distance was used for the span. 

Mr. Talbot. — ^A definite length had to be assumed. It was Mr. xaibot. 
thought best to use something more than the clear span and 
3 in. more than the clear span for both girders and beams was 
used. This is somewhat shorter than that given by the usual 
methods of design. 

Mr. W. K. Hatt* (By Letter).— The writer has read this Mr.Hmtt. 
important contribution with a great deal of interest. . 

It is realized that our methods of computing the strength of 
a continuous construction of slabs, Boor beams, and columns of a 
reinforced concrete building is conventional and very largely 
empirical. We need many such tests as are described in this 
report to determine the allowable limits of deflection and the real 
factors of safety obtaining in the usual designs. The common 
practice of loading one panel is evidently misleading since sur- 
rounding panels under load assist in carrying the load. 

The report confirms the opinion many have had that under 
ordinary working loads the tensional stresses in the concrete 
assist the reinforcement in carrying the bending moment and that 
the conventional computal stress of 16,000 lb. will not be found in 
the test floor. Of course imder this distribution the compressional 
stresses will be the critical stresses. 

The ratios between the compressional stresses and the com- 
puted reinforcement stresses in concrete beams reinforced with 
various percentages of metal is an important factor in the work 
of the designer, who is controlled by building laws. This report 
does not comment on the proper values of this ratio. The indica- 
tion from the tests, however, is that in spite of high compressional 
stresses failiu-es in compression are not evident. The recent pub- 
lication of the Bureau of Standards by Messrs. Richard L. Hum- 
phrey and Louis Losse, fixes an extreme fiber stress in compression 
of 1000 lb. per sq. in. at the measured unit stress of 16,000 lb. per 
sq. in. in the reinforcement for 1-2-4 concrete. 

The writer made a careful test of the Franks building in 
Chicago in the siunmer of 1911 and submits report of this test 
as a contribution to the discussion of this subject. 



*PYofeflior of Civil Engineeriiig, Pardue University. Lafayette. Ind. 



Digitized by 



Google 



160 Report of Committee on Reinforced Concrete. 

REPORT ON TEST FOR ACTUAL STRESSES OF THE 
A. J. FRANKS BUILDING, CHICAGO, ILLINOIS. 

Bt W. K. Hatt, Consulting Engineer, Professor of Civil Engineering, 
Purdue University. 

During the latter part of August, 1911, the undersigned loaded panels 
of a reinforced concrete building. Fig. 1, constructed by the Leonard Con- 
struction Company, for Mr. A. J. Franks, and measured the actual deforma- 
tions of the concrete and steel under load. The test was performed upon the 
basis of specifications prepared by the undersigned. 

In brief, the panels were loaded with pig iron in increments, and the 
accompanying deformations were measured in the steel and in the concrete 
at all critical points, with a view to fixing a safe limit of loading and to under- 
standing the mechanical action of the structure. 

TEST STRUCTURE. 

The building tested is a ten-story and basement warehouse intended 
for the printing and paper trades, Dwight Bros. Paper Company, tenants. 
The type of construction used was the Cantilever Flat Slab System, the rein- 
forced concrete setting drawings and shop details being made by the Concrete 
Steel Products Company, Engineers, of Chicago. The architects of the 
building were Richard E. Schmidt, Garden and Martin. The design load 
on the floors was taken at 250 lb. per sq. ft. The panel dimensions were 
19 ft. 4 in. by 20 ft. 3 in. The four panels under observation were in the 
interior of the building and on the tenth floor where the columns were of 
minimum size. As the effective clear span between capital and the eccentric 
action on the columns were the greatest, the location was such as to insure 
the most severe test possible. 

OBSERVERS AND METHODS OF OBSERVATION. 

The observations were made by experienced observers as follows: Pro- 
fessor H. H. Scofield, of Purdue University; Professor W. A. Slater, of the 
University of Illinois Engineering Experiment Station; Mr. W. E. Ensign, 
of the University of Illinois Engineering Experiment Station; with the assist- 
ance of Professor L. W. Weeks, of Purdue University. 

The deformations were determined by the use of extensometers of the 
type devised by Professor H. C. Berry of the University of Pennsylvania. 
On the steel a gauge length of 10 in. was used, which, with the multiplying 
lever in the instrument, gave direct readings of unit deformations of .00002 
in. per in., corresponding to stresses of 600 lb. per sq. in. in the steel, and 
it was possible to estimate clearly fractions of this amount. On the concrete 
readings the gauge length was 6 in. and the direct reading of unit deforma- 
tions was .000033, corresponding to a stress of 133 lb. per sq. in., and it was 
possible to estimate fractions of this amount with accuracy. 

Errors in operating the instrument were reduced to a minimum by tak- 
ing every reading at least five times and by calibrating on a standard bar at 



Digitized by 



Google 



Report of Committee on Reinforced Concrete. 161 

frequent intervals. Four standard bars were used throughout the test, all 
of these being embedded in the concrete and subject to the same temperature 
changes as the rods in the test floor but free from any stress due to applied 
load. By reading on these bars between sets of five or six test readings the 




FIG. 1. — A. J. FRANKS BUILDINQ IN COURSE OF CONSTRUCTION, 
CHICAGO, ILL. 

observations on the materials under test were freed from temperature differ- 
ences and systematic errors. 

The deflections were measured to .0001 in. by use of the deflectometer 
described in a previous test fQr actual stresses. 



Digitized by 



Google 



162 Repobt of Committee on Reinforced Concrete. 

Tensile defonnations were measured throughout the test over 42 gauge 
lengths on the steel reinforcing rods; compressive deformations in the con- 
crete were measured over 26 gauge lengths; deflections were observed at 24 
points, and 27 other readings of deformations were taken throughout the test 
to study such phenomena as the arch and slab action, the distribution of 




FIQ. 2. — ^LOCATION OF GAUGE LBNGTHS IN CONCRBTB BLAB. 

stress, and the eccentric loading on the columns at the edges of the loaded 
area. The location of the gauge lengths is given in Figs. 2 and 3. 

The observations were arranged in groups, each designed to cover 
adequately some particular feature, and synmietrically located observations 
were obtained as a check in every case where possible, in order to cover varia- 
tions in the quality of the concrete at different points. 



Digitized by 



Google 



Report of CoMBflTTEE on Reinforced Concrete. 163 

The conditions were exceptionally favorable to a satisfactory test. 
The building was practically completed at the time of test and the variation 
in temperature was very slight, being about 74** at the start of the test, 
dropping gradually and uniformly to 70°, and rising again at the end to 74°. 




FIG. 3. — LOCATION OF GAUGE LENGTHS IN REINFORCING RODS. 

The slab was poured on June 23, 1911, and was 64 days old when the maximum 
load of 624 lb. per sq. ft. of panel area was placed upon it. 

LOADING. 

The amount of loading was determined by the weight of the pig iron as 
recorded on the weigh bills delivered by the teamsters, and was checked by 
weighing a number of piles of pig iron from the test load on a platform scales. 

The pig iron was piled on the floor in separate piers, Fig. 4, each placed 



Digitized by 



Google 




Digitized by 



Google 



Report of Committee on Reinforced Concrete. 165 

within a two-foot square, so that no arch action existed in the load itself to 
relieve the panel from moment. It was necessary to leave several of these 
squares vacant about columns and to leave an aisle between the columns to 
allow space for observations. The pig iron belonging on these vacant squares 
was distributed over the remaining squares of the same panel. It will be 
apparent that under this procedure the real load intensity, as affecting bend- 
ing stresses and deflections, is the intensity of loading over the loaded area 
(about 90 per cent of panel area), rather than the nominal or average load 
over the entire panel area. It is evident that had the 10 per cent of panel 
area close to the columns been loaded to the same intensity the increase in 
stresses and deflection would have been practically nil. 

All gauge lengths were measured and checked throughout before loading 
was started. When the loading reached certain amounts, distributed evenly 
over four panels, it was discontinued and allowed to stand for 6 hours before 
the observations were made and the loading resumed. The increments of 
load at which measurements were made were as follows: 75, 150, 256 (the 
design load), 312 (intensity, 359), and 624 (intensity, 717) lb. per sq. ft., 
the latter load being applied to two panels only. Readings were also taken 
with 256 on two diagonal panels and 312 on the other two panels, and with 
468 on two diagonal panels and 156 on the others, but without waiting for 
the six-hour interval to elapse before taking readings. The total number 
of complete observations over single gauge lengths was over 2,000, and the 
total individual readings over 10,000. 

STRESSES. 

The stresses were determined from the observed deformations by using 
a modulus of elasticity of 30,000,000 lb. per sq. in. for the steel, and 4,000,000 
lb. per sq. in. for the concrete. The latter value was determined from tests 
of three concrete prisms poured from the concrete in the test slab and tested 
at Purdue University at an age of 77 days. 

Table I gives a summary of the corrected values of the total dead and 
live load stresses observed in the various groups of observations. The detailed 
summary of individual stresses at the various observation points are omitted 
in this report. After the observations had been corrected for temperature 
and observational errors, by use of the standard bar calibrations, load deforma- 
tion curves were plotted for each observation point, the known nature of the 
load deformation curve under flexure being used as a basis. The dead load 
stress has been taken from these curves as equal to the stress caused by an 
equal live load. 

COMMENTS ON THE RESULTS. 

The undersigned is not prepared at the present time to state the signifi- 
cance of the results obtained with respect to the mechanics of this form of 
construction. Such statements would have somewhat of a speculative ele- 
ment and should be separated from the report of test which is one of measured 
facts. 



Digitized by 



Google 



166 Report of Committeb on Rbinfobced Concrete. 

As compared with the design requirements of the Chicago Building 
Code, it is interesting to note that at design load the highest average stress 
in the steel is 5078 lb. per sq. in., while the highest average compressive 



.ft. 



Table I. — Condensed Summary of Actual Stresses. 

All stresses are the final values for total dead and live load, and are given 
in lb. per sq. in. 

Stresses in Slab Rods. 

oJt^JSS.tc., DeBcription. ^3^ 

5, 6, 7, 14, 15, 16 Center of span, diagonal band 1071 

8, 9, 10, 11, 12, 13 Center of span, cross band . . 4539 

31,32,33,34,35,37 Over capital at center col- 
umn, diagonal band 3440 

27, 28, 29, 30 Over capital at center col- 
umn, cross band 4575 

21,22,23 Over capital at comer col- 
umn, diagonal band 1920 

24, 25, 26, 41, 42, 43 Over capital at side column, 

cross band 2690 



Live Load-lb. per 0q. f 
812 624 

1920 

6140 



10095 



4350 9280 

4840 8140 

2280 7540 

3138 5315 



Compressive Stresses in Concrete. 
66, 67, 69, 60, 68, 69, 70. .On slab at center column. . . 

61, 62, 65, 66 On drop at center column. . . 

61, 74 On drop at comer column . . . 

52, 53, 73 On slab at corner column .... 

54, 55, 71, 72 On slab at side columns 



560 650 1206 

677 778 1685 

318 370 1515 

329 378 650 

189 217 420 



Mcudmum Stresses in Columns Due to Eccentric Live Load. 

104 . . Compression in concrete, corner column 680 840 1660 

109. .Compression in concrete, side column 416 512 1000 

84. .Tension in steel, comer column 4980 6000 11620 

99. .Tension in steel, side column 2220 2640 5880 



Deflections in Inches. 

121, 124. .Center of panel— at 6 hours 123 

Center of panel — at 24 hours 

After standing unloaded 6 hours 142 

After standing unloaded 2) days 090 



.156 



.475 
.500 



stress in the concrete is 677 lb. per sq. in. On the basis of safe working 
stress in the steel of 16,000 lb. per sq. in., and in the concrete of 35 per 
cent of the ultimate strength (which averaged over 3250 by tests of prisms) 
or 1100 lb. per sq. in.; it appears that the steel is stressed to 31 per cent 



Digitized by 



Google 



Report op Committee on Reinforced Concrete. 167 

of its safe load while the concrete is stressed to 62 per cent of its safe load. 
It appears, therefore, that the design is overbalanced with an excess of steel. 
At the highest applied load of 717 lb. per sq. ft., the ratio of the steel and 
concrete stresses remains practically unchanged. 

The eccentric action of the test load was most marked on the comer 
columns of the loaded area and was suj£cient to produce a tension in the 
steel of 5000 lb. per sq. in. 

With respect to the strength of the structure, it may be said that a nominal 
load of 624 lb. per sq. ft. of panel, actually 717 lb. per sq. ft. of loaded surface, 
was applied without producing any permanent damage to the building. At 
this load the highest observed average total dead and live load stresses were 
less than 12,000 lb. per sq. in. in the steel and less than 1700 lb. per sq. in. 
on the concrete. 

From a consideration of the data from the above test the writer con- 
cludes that the A. J. Franks Building is amply strong to carry the designed 
load and that the lower floors at least may safely and continuously be loaded 
with considerably more than the designed load. 

(Signed) W. K. Hatt. 



Digitized by 



Google 



THE TESTING OF REINFORCED CONCRETE 
BUILDINGS UNDER LOAD. 

By W. A. Slater.* 

I. Introduction. 

Development of Building Tests. — ^For several years there has 
been a growing demand for tests of full-size structural members. 
A more recent development is the test of structures themselves 
and the measurement of actual stresses in the component parts. 

Load tests have been required by city building departments 
as a condition of acceptance of reinforced concrete buildings and 
have been used by construction companies and engineers to 
demonstrate the adequacy of various designs. Such load tests 
are never continued to destruction, the applied load being gen- 
erally twice the design live load, and emphasis is placed upon 
measurement of deflection and recovery. No measurements of 
stresses are made in such tests and under these conditions the 
safe load can not be fixed upon as a definite ratio of the ultimate 
load. The deflections observed in such tests constitute a very 
inadequate and actually misleading measure of the stresses. 
Slight deflections have been taken to indicate low stresses in 
reinforcement and in concrete, but recent tests in which deforma- 
tions were measured have shown that even with slight deflections 
large stresses are developed in concrete even when the reinforce- 
ment stresses are low. The tendency of building codes was to 
disregard continuity of action in beams in reinforced concrete 
buildings and to specify the design as of simple beams, but even 
in such cases a small amount of reinforcement was placed across 
the support to prevent the opening of large cracks. This rein- 
forcement and the tensile strength of the concrete have been 
sufficient to develop a large stress in the concrete at the support 
which may not have been specifically provided for. Thus the 
so-called conservative attitude of not allowing anything for 
continuity of beams at the support may prove a source of weak- 

* Engineering Experiment Station, University of Illinois. Urbana. 111. 

(168) 



Digitized by 



Google 



Slater on Testing Reinforced Concrete Buildings. 169 



The measurements of deformations in building structures 
eonfirms the truth of this statement. 

Purpose and Scope of this Paper. — ^The reports of all building 
tests in which deformations have been measured deal in the main 
with the behavior of the structure and record the results, and are 
not primarily concerned with the working of the instruments or 
with the methods of making the tests. To conduct a successful 
building test is diflScult, however, and this paper is written in 
order to present information as to methods of testing gained by 
experience and to point out certain respects in which such tests 




1 



Building 





euf'Mhff 





CorMfon Bolklfng 



•RELATIVE SIZE OF FLOOR AREAS TESTED. 



may be conducted more satisfactorily than those which have 
already been made. The following general order of presenting 
the material in hand will be observed: (1) enumeration of tests, 
(2) the planning and preparation for a test, (3) the instruments; 
their construction and use, and the methods of making observa- 
tions, (4) the methods of making calculations and (5) the cost of 
a test. 

The following is a list of tests of building floors in which the 
methods described herein of measuring deformation were used. 
Fig. 1 shows the range in size of these test areas. 



Digitized by 



Google 



170 Slater on Testing Reinforced Concrete Buildings. 

Test No. 1. — Deere and Webber Building, Minneapolis, 
Minnesota, October and November, 1910; flat slab floor with 
four-way reinforcement; built by Leonard Construction Com- 
pany of Chicago, and tested by them with the co-operation of 
the Erigineering Experiment Station of the University of Illinois. 

Test No. 2. — ^Wenalden Building, Chicago, Illinois, June and 
July, 1911. Beam and girder building constructed by Ferro- 
concrete Construction Company of Cincinnati, and tests made 
by co-operation between the National Association of Cement 
Users, the Ferro-Concrete Construction Company and the Engi- 
neering Experiment Station of the University of Illinois. 

Test No. 3. — ^The Powers Building, Minneapolis, Minnesota, 
July and August, 1911; flat slab floor with two-way reinforcement; 
built and tested by Corrugated Bar Company of St. Louis. 

Test No. 4. — ^Franks Building, Chicago, Illinois, August, 
1911; flat slab floor with four-way reinforcement; buUt and 
tested by Leonard Construction Company of Chicago. 

Test No. 5. — ^Turner-Carter Building, Brooklyn, New York, 
September, 1911; beam and girder floor; built by Turner Con- 
struction Company of New York; test made by co-operation 
between National Association of Cement Users, the Turner 
Construction Company and the Engineering Experiment Station 
of the University of Illinois. 

Test No. 6. — Carleton Building, St. Louis, Missouri, October, 
1911; flat slab floor with two-way reinforcement; built and 
tested by Corrugated Bar Company. 

Test No. 7. — ^Barr Building, St. Louis, Missouri, December, 
1911; full size test panel (25 ft. x 26 ft. 9 in.). Terra-cotta tile 
used to lighten construction; gives two-way T-beams with web 
between tile on tension side and concrete flange above the tile; 
two-way reinforcement. Panel built by Corrugated Bar Com- 
pany to demonstrate eflSciency of design proposed for Barr Build- 
ing in St. Louis; test made by Corrugated Bar Company. 

Test No. 8. — Ford Motor Building, Detroit, Michigan, 
February and March, 1912; flat slab floor; built and tested by 
the Corrugated Bar Company. 

These seem to be the only full-size reinforced concrete floor 
tests on record in which deformations in reinforcement and con- 
crete have been measured. The writer was in immediate charge 



Digitized by 



Google 



Slater on Testing Reinforced Concrhtb Buildings. 171 

of Tests Nos. 2 and 5 and had a part in tlie conduct of 'all "tjie 
others except Tests Nos. 6 and 8. I'jie 'motfioas' of t&3tiag 
presented in this paper were developed by the writer as a , result 
of his connection with the tests. These meth'ods*v>^re ji^dft^^'V^ 
increase the accuracy of results, to avoid accidental^ errors^ ai».d t^ 
correct for systematic errors. 

Much credit for the initiative in this type of test ife due Mr. 
A. R. Lord, formerly research fellow at the University of Illinois, 
who was largely instrumental in bringing about the test of the 
Deere and Webber Building, the first in the series named. After 
the presentation of Mr. Lord's paper on the test of the Deere and 
Webber Building, The National Association of Cement Users 
decided to continue the investigation. All of the tests given in the 
above list were conducted on the same general lines as that of the 
Deere and Webber Building. Only the tests of the Wenalden 
Building and the Turner-Carter Building were in the series author- 
ized by the National Association of Cement Users, but the results 
of the tests made by the Corrugated Bar Company on the Powers 
Building and on the Barr Building test panel have been placed 
at the disposal of the Association. The Franks Building test, 
made by the Leonard Construction Company, was an investigation 
planned to give data for an intelligent modification of the Chicago 
Building Code. The other two tests, those of the Carleton Build- 
ing and the Ford Motor Building, were in the nature of investiga- 
tion of special features of design. The methods used in all of 
these tests are essentially the same and have been developed 
at the University of Illinois Engineering Experiment Station. 

Available Literature, — Reports of results of some of these 
tests are available as follows: 

1. Deere and Webber Building. 

Paper by A. R. Lord, "A Test of a Flat Slab Floor in a 
Reinforced Concrete Building." ProceedingSf Vol. VII, 
1911. 

Abstracts: Engineering News, December 22, 1910; Engi- 
neering-Contracting, December 22, 1910. 

2. Wenalden Building and Turner-Carter Building. 
Report of Committee on Reinforced Concrete and Build- 
ing Laws, Part I. Proceedings, Vol. VIII, 1912.* 



♦See pp. 61-167.— Ed. 



Digitized by 



Google 



172 Slater on Testing Reinforced Concrete Buildings. 



• • • • • 



• •• •' 



Abilrkctfi f \Fngin6efi7iy. i?6cord, March 23 and April 13, 
• le/^y i^inf 4r«W Vf ^w's, April 18, 1912; and CetnerU 

••\i: : .. . :-4ffV^ei;w *•* 

•/ ; S?*! JVveiiHuiiaihg and Barr Building Test Panel. 
. . .'. *. *.llep6y; g/ tlgmi^ttee on Reinforced Concrete and Building 
;: V: ( } V Law)a;-i^arteil and III. Proceedings, Vol. VIII, 1912.* 
Abstracts: Engineering News, April 18, 1912; and Engi- 
neering Record, April 20, 1912. 
4. Franks Building. 

(a) Abstracts of paper by W. K. Hattf before Indiana 
Engineering Society, Engineering-Contracting, March 
13, 1912; and Engineering News, April 8, 1912. 
(6) Discussion by W. K. Hatt on Report of Committee on 
Reinforced Concrete and Building Laws, Proceed- 
ings, Vol. VIII, 1912.t 
(c) Trade publication on Cantilever Slabs published by 
Concrete Steel Products Company, Chicago, 111. 

II. Conduct op Tests. 

DEFINITIONS. 

In the following descriptions of tests, many terms will be 
used for which somewhat arbitrary definitions will need to be made. 
These definitions are: 

Gauge Hole: A small hole (.055 in. is here recommended) 
drilled into the steel bar or into the plug inserted in the concrete 
has been termed a gauge hole. It is for the admission of the point 
of a leg of the extensometer. 

Gauge Line: The gauged length connecting a pair of gauge 
holes is termed a gauge line. 

Reading: A reading is a single observation on any gauge line. 

Observation: An observation as here used is the average of a 
number of readings. 

Zero Length of Instrument: The length of the instrument at 
the time of taking the first observation on the standard bar will 
be known as the zero length of tlie instrument. This first observa- 



• See pp. 61-157.— Ed. 
t See pp. 159-167.— Ed. 



Digitized by 



Google 



Slater on Testing Reinforced Concrete Buildings. 173 

tion on the standard bar is not the zero length, but a comparison 
of a subsequent observation with it shows any change from the 
zero length. 

Correction: A correction is the amount which if added alge- 
braically to the observation will give the observation which would 
have been obtained if the instrument had not changed from its 
zero length. 

Series of Observaiiona: The observations taken consecutively 
at a given load without repetitions on any gauge line ar^ defined as 
a series of observations. 

Interval: An interval as used here is the time elapsing between 
consecutive observations, and all intervals in *any series are (for 
lack of more exact information) assumed to be equal. For this 
piupose the average of two consecutive observations on standard 
gauge lines is considered a single observation. 

Standard Gauge Line: This is a gauge line used usually to 
determine changes of length of instrument, of reinforcement or of 
concrete due to other causes than the applied load. Its purpose 
usually is to determine the temperature effect on the instrument, 
but it may be used to detect accidental changes of instrument or 
temperature stresses in the reinforcement of the concrete. Origi- 
nally this gauge line was placed on a steel bar separate from the 
structure, and this gave rise to the term standard bar. In several 
of the later tests, however, the standards have consisted of gauge 
lines placed in the reinforcement and concrete of the structure 
remote from the area affected by the load. Standard gauge line 
is adopted, therefore, as the more general term and any reference 
to the standard bar may be understood to signify the standard 
gauge line on a bar separate from the structure. 

general outline of method of testing. 

After determining what measurements will best give the 
information desired from the test, the gauge lines are laid off on' 
the surface of the concrete and small holes are cut or drilled in the 
concrete at a predetermined distance apart in order to expose the 
reinforcement or allow a metal plug to be inserted, according as 
the measurement is of reinforcement or concrete deformation. 
The metal plugs used are securely held in place by imbedment' 



Digitized by 



Google 



174 Slater on Testing Reinforced Concrete Buildings. 

in plaster of Paris. The gauge holes having been carefully pre- 
pared; a set of zero readings is taken on all gauge lines, an increment 
of the loading material is then added and a second series of observa- 
tions on the gauge lines taken. The difference between the two 
readings on the same gauge line represents the deformation in 
that gauge line. It is possible that this apparent deformation 
may be due partly to temperature changes in the instrument 
instead of stress changes of the material by reason of applied load. 
For this reason reference measurements are made on standard 
unstressed bars made of Invar steel which has a very low coefficient 
of expansion and whose change in length due to temperature 
would therefore bfe very sUght. From these readings on the 
standard bar, temperature corrections are computed as shown in a 
later paragraph and applied to the observations in order to deter- 
mine the actual change in length of the gauge line. Another 
increment of load is then applied and another series of observations 
taken. 

Floor deflections also have been measured in all of these tests, 
but they have been considered as of secondary importance. They 
have been used to throw light on the correctness or incorrectness 
of the deformation readings and to gain some idea of the general 
distribution of stresses throughout a floor. They can apparently 
be depended upon to show with considerable accuracy the pro- 
portional rate of increase of stress, but deflection formulas are so 
imperfect that measurement of deflections can not be depended 
upon to give the actual values of stresses. 

Measurements of dimensions such as span, depth of beams, 
location of observation points, weight of loading material, location 
of cracks, and any other measurements which were considered of 
value in working up results have been carefully taken. The 
measurements taken are usually distributed over and under the 
surface of the floor tested in order to gain an idea of the changes 
occiuring in different parts of the structure. 

The above statement gives in general terms the features of 
any one of the tests dealt with in this paper. There are many 
difficulties to overcome and many chances for error. What 
follows is concerned mainly with the method of overcoming these 
difficulties and avoiding these errors. Most of the statements 
made represent the results of experience on previous building 



Digitized by 



Google 



Slater on Testing Reinforced Concrete Buildings. 175 

tests. Some merely give ideas which it is believed if put into 
operation would be advantageous. 

the planning op a test. 

Each test made will involve individual consideration of the 
choice of area to be loaded, the number and location of gauge 
lines and deflection points, the number of laborers required, the 
loading material to be used and its distribution, and provisions 
for storage of the loading material near the test area without 
appreciably affecting the stresses which are to be measured. 
Other matters will come up for consideration, but in the main 
they will not require different solutions for each test. 

Choice of Test Area. — The area to be loaded should be chosen 
so as to fulfil the following conditions as completely as possible: 

(a) It should be so located as to give conditions in the beams, 
slabs, columns, etc., as severe as will be found anywhere in the 
building when in use. 

(6) It should be free from irregularities of construction. 

(c) It should be as free as possible from disturbances of 
workmen. 

(d) It should be as easily accessible to the loading material 
as possible. 

In most cases some limitation is found on part or all of the 
conditions named. For example, in the test of the Wenalden 
Building it was impossible to find an area entirely free from irregu- 
larities of construction. An industrial track crossed one of the 
panels chosen, and the floor was thicker immediately under this 
track than at other places. On the edge of one or two of the panels 
tested, beams about an inch deeper than the regular beams were 
located. However, none of the measurements assumed to give 
typical results were taken in these panels, and it is believed that 
the stresses in the other panels were not affected appreciably by 
these irregularities. Again, in the test of the Franks Building 
it was not possible to choose a lower floor convenient to the loading 
material. An upper floor was used in order, during the course of 
construction, to make preparation for the test, thus avoiding dig- 
ging in the concrete. However, this choice of a floor fulfilled one 
of the conditions mentioned, in that it gave a much more severe 
test of the columns than a test on a lower floor would have done. 



Digitized by 



Google 



176 Slater on Testing Reinforced Concrete Buildinqs. 

Also, in the test of the Carleton Building at St. Louis the area to 
be tested was specified by the city building department, and there 
was no choice as to location on the part of those making the test. 
Number oj Measurements. — ^The number of measurements to 
be taken will depend upon the nature of the test, the number of 
observers, and the number of laborers. If the test is a part of a 
series by which it is expected to gain scientific information which 
will afford a basis for design, it is likely that it will be deliberate 
enough that a large number of measurements may be taken. 
Such tests were those of the Wenalden Building, the Franks 
Building, the Turner-Carter Building, and the Barr »test panel. 




Pl/^n5how/a4 5 Location ^ 1 > f>L/9AfSMO¥Y*MgLoCATtON 

o^ fio/Afrs OAf Top OF Slab "^y^ or Po/Afrs cr^Sorromor Sl^B 

•• M[>inf on sfevi ' 

0aQg0l9nff/i 8 tnc/»99 I 

FIG. 2. — ^LOCATION OF GAUGE POINTS, CARLETON BUILDING. 

If, on the other hand, the test has more of a commercial 
nature or is a utilization of the opportunity offered by the accept- 
ance test to take some measurements which will show actual 
stresses, or if for any other reason the test is hurried, the number 
of measurements will necessarily be rather small. Of this class, 
the tests of the Carleton Building in St. Louis and of the Ford 
Motor Building in Detroit, Michigan, are good examples. Notice 
was given the engineers only about one day in advance that a 
test would be made on the Carleton Building. Permission was 
obtained from the contractor to expose bars for measurement in 
various points and to erect the necessary scaffolding. The meas- 



Digitized by 



Google 



Slater on Testing Reinforced Concrete BxnLDiNGs. 177 

urements were made more for the purpose of checking the anal- 
ysis upon which the design was based than to form in itself a 
basis of design. Therefore comparatively few observation points 
were used. It is believed that this test is representative of the 
type of test which is practicable on a commercial basis, hence (by 
courtesy of the Corrugated Bar Company) a plan is given in 
Fig. 2 showing the points where measurements were taken. 

Disirihulwn of Measurements. — The arrangement of observa- 
tion points will depend on what are the principal subjects for 
investigation in the test. Whatever the subject of study may be, 
the observation points should be arranged in such a way that a 
curve of deformations may be plotted against distance, showing 




I ^ 5 S § 
§ § § § § 



§ ^ S ^ § 



^ S S 



Deformat/on per Unit of Length. 

PIG. 3. — LOAD-DEFORMATION DIAGRAM FOR SERIES OP GAUGE LINES ON 
REINFORCEMENT, POWERS* BUILDING. 

a gradual progression from the condition at one part of the struc- 
ture to the condition at another, for it is found that there are even 
imder the most careful work, inconsistencies which will make the 
results look doubtful if standing by themselves. The points so 
arranged should be numerous near the place where the measure- 
ments of greatest importance are to be taken, so that the results 
will not depend upon measurements at a single point, or upon the 
average at portions of the structure supposed to be similarly 
situated but in different parts of the building where unknown 
conditions may actually cause a large variation in the phenomena 
of the test. It will not be possible to carry out this plan for all 
subjects of investigation, as the number of observations required 



Digitized by 



Google 



178 Slater on Testing Reinforced Concrete Buildings. 

would usually be impracticably large. Such provisions may be 
made to cover the main lines of investigation, and isolated obser- 
vation points may be used to gain information as to tendencies 
of other portions of the structure, but of course, less reliance must 
be placed on the results of the latter measurements than where 
the larger number of observations is made. It would be advan- 
tageous, as was done in the Powers Building test and also in the 
Barr test panel, for two observers to check measurements on the 




FIG. 4. — LOCATION OF GAUGE LINES, POWERS BUILDING. 

same points. One or both of these checks is very valuable in 
establishing the correctness of observations. 

Figs. 3, 4, 5, and 6 illustrate the former method. Fig. 3 gives 
the load deformation diagrams for several gauge lines in the test 
of the Powers Building. Fig. 4 shows the location of these points 
with reference to the wall and a column. Fig. 5 shows the same 
data plotted as deformation against distance from the column 
instead of against load. It may be seen that the correctness of 
the load deformation curve for one of these points, if standing by 



Digitized by 



Google 



Slateb on Testing Reinforced Concrete Buildings. 179 

itself, might be doubted because of the complete change in the 
character of the curve at a load of 200 lb. per sq. ft. But when 
these deformations are plotted against distance, the results look 
so consistent that it is scarcely conceivable that they are seriously 
incorrect. In the test of the Wenalden Building very high 
deformations were observed in the concrete of the beams near the 
supports; so high that the results were doubted, and as the points 
on the load deformation curves were few and scattering, there 
was often room for doubt. For this reason it was considered 
especially important that evidence which would confirm or dis- 
prove this high compression in the concrete be obtained in the 

V .oooa 
g ^.oood 

^.0003 

^.oooz 

.0001 



. D /stances from Capital in Inches. 

PIG. 5. — DATA OF FIG. 3 PLOTTED AS A DISTANCE-DEFORMATION DIAGRAM, 

POWERS BUILDING. 

test of the Turner-Carter Building; accordingly the method of 
placing observation points at frequent and regular intervals along 
the ends of the beams was used. The deformations measured 
are plotted in Fig. 6 against the distance from the supporting 
column, and the results not only tend to show the correctness of 
these measurements but also to indicate that the high stresses 
observed in the beams of the Wenalden Building were actually 
present. 

Subjects of Investigation, — In the tests discussed in this paper 
deformations have been measured with a view to obtaining 
information on each of the following subjects: 




Digitized by 



Google 



180 Slater on Testing Reinforced Concrete Buildings. 

(a) The values of the moment coefficients at the center and 
support of the beam or slab under mvestigation. 

(6) Relative moments at support for various conditions of 
fixity. 

(c) The extent to which the floor slab acts as a compression 
flange of the floor beam to produce T-beam action. 

(d) Bond stresses. 
{e) Diagonal tension. 
(/) Stresses in columns. 

(gf) Time effect under constant load. 




FIG. 6. — DISTRIBUTION OF COMPRESSIVE DEFORMATION IN BOTTOM OF COLUMN 
BEAM, TURNER-CARTER BUILDING. 

(A) The lateral distribution of stress to parts of the structure 
entirely outside of the loaded area. 

(i) The extent to which reinforcement stresses are modified 
by errors in the assumption that no tension is carried by concrete. 

(j) Stresses in slabs of beam and girder construction. 

Other subjects of investigation have received attention, but 
these are the most important ones. Some phenomena have been 
observed, offering additional problems, of which the determina- 
tion of the amount of arch action present is probably the most 
important. It is important both in itself and because it is 



Digitized by 



Google 



Slater on Testing Reinforced Concrete Buildings. 181 

intimately involved in the determination of moment coefficients. 
A further discussion of this problem is here given. 

The attempts to determine moment coefficients have not 
been entirely successful due to errors of measurement and unex- 
pected variations in similar parts of the structure remote from each 
other. The method has been to measure deformations on both 
reinforcement and concrete at the center and support, and from 
these measurements to determine the total resisting moment 
developed. Equating this resisting moment to a constant K 
times Wl a solution is made for the value of K, The indications 
that arch action has been present have so complicated this that 
even where measurements have appeared quite satisfactory, the 
uncertain amoimt of arch action entering has rendered the value 
of K uncertain. 

A proposed method of determining the amount of arch action 
in any case is to make a special study of the deformations in a 
cross-section at the center of each beam across an entire panel. 
In this study, deformations should be observed on the reinforce- 
ment and at various elevations on the concrete so that the position 
of the neutral axis and of the center of gravity of tensile and com- 
pressive stresses respectively can be accurately located without 
dependence upon the law of conservation of plane sections. By 
this means it should 'be possible to determine if the sum of the 
compressive stresses is in excess of that of the tensile stresses. 
If so, the difiference apparently must be the direct thrust due to 
arch action. The same study can be made, though not so satis- 
factorily, at the ends of the beams. This measurement of thrust 
will require observations on an extremely large number of gauge 
lines, and it would appear important to concentrate the greater 
part of the attention of the test on one panel. 

If the floor be considered to be made up of strip-beams of 
differential width capable of transmitting shear from strip to strip, 
it is not necessary, for perfect beam action, that the sum of the 
tensile and compressive stresses on a cross-section of any one strip 
be zero. However, beam action does require that the sum of the 
tensile and compressive stresses on the total cross-section of the 
beam should be zero, and for this reason it is important to extend 
the investigation sufficiently to determine if appreciable deforma- 
tions are continued out into the panel adjacent to the loaded area. 



Digitized by 



Google 



182 Slater on Testing Reinforced Concrete Buildings. 

Laborers, — The number of laborers which can be used advan- 
tageously will depend on the distance from which the loading 
material is to be transferred, on the size and accessibility of the 
tested area, on the amount of work which can be done by them 
during the intervals between increments of loading while observa- 
tions are being taken, and on the length of time required to take 
a series of observations. The handling of the labor should, if 
possible, not be left to the one in charge of the test, as proper 




FIG. 7. — BRICK AND CEMENT AS A LOAD, WENALDEN BUILDING. 

attention to the conduct of the test demands all of his time. In 
the tests included in this paper the number of laborers has varied 
between wide limits, from 5 or 6 in the Powers test to 30 or 35 
in the Deere and Webber test. 

Loading. — ^In the tests which have already been made, the 
following loading materials have been used: brick, cement in 
bags, loose sand in small boxes, sand in sacks and pig iron. The 
material used will almost always be that which is most easily 
available, because the moving of loading material from any 



Digitized by 



Google 



Slater on Testino Reinforced Concrete Buildinqs. 183 

distance adds very greatly to the cost of the test. Leaving con- 
sideration of cost out of the question, sand in sacks seems to be the 
most satisfactory of the materials above mentioned for loading 
purposes. Some of the qualities of the materials mentioned are 
as follows: 

(a) Brick: Brick spalls and chips in handling, covering the 
floor with dust and jagged particles which cause discomfort to the 
observer in kneeling to take observations. It is important to 




PIG. 8. — SAND IN BOXES AS A LOAD, TUBNER-CABTER BUILDING. 

avoid this because discomfort necessarily decreases the accuracy 
of his observations. This might be avoided by sweeping, but in 
sweeping it is diflScult to avoid getting dirt into holes where obser- 
vations are to be taken, and this is just as troublesome as having 
the dirt on the floor. Fig. 7 shows the use of both brick and cement 
in the same test. Attention is called to the proximity of the cement 
sacks to the beams and girders of the floor above. In some cases 
the intensity of the load would be limited by the height of the 
ceiling if cement and brick are used. 



Digitized by 



Google 



184 Slateb on Testing Reinforced Concrete Buildings. 

(6) Cement: Cement sifts through the sacks and the sacks 
become imtied, scattering cement on the floor, filling observation 
holes and causing much dust in sweeping or cleaning up. The 
dust is injurious to delicate instruments and annoying to observers 
and recorders. 

(c) Loose Sand in Small Boxes: As sand is usually damp, it 
does not have the fault of causing dust and consequently is more 
easily cleaned up than the other materials mentioned. There are, 
however, other objections to it. In filling boxes it is difficult to 
avoid spilling the sand around and between the boxes, and con- 




FIG. 9. — BAND IN SACKS AS A LOAD, BARR BUILDINQ TEST^PANEL. 

sequently filling the observation holes. On account of the great 
difficulty in removing loose sand without spilling a great deal of 
it, it is impracticable to take observations as the load is being 
removed, therefore it is necessary to remove in one increment the 
whole load from a given panel. Fig. 8 shows this method of loading, 
(d) Sand in Sacks: Sand in sacks constitutes a very satis- 
factory loading material. Fig. 9. It was piled up to a height of 
about 9 or 10 ft. and very little inconvenience was caused by the 
sacks becoming untied or by spillmg the sand. The worst difficulty 
encountered, and this exists with all materials handled in sacks, 
is that of the slidmg of sacks on themselves when the load is piled 



Digitized by 



Google 



Slater on Testing Reinforced Concrete Buildings. 185 

high. It can be seen in Fig. 9, above referred to, that bracing 
was necessary to prevent the sand from sHding together and 
filling up the aisles. It is a source of danger to those taking 
observations as, if a slide should occur, it would probably give 
very little warning and might catch the observer while in such a 
position that he could not escape. However, this objection 
would be likely to occur with any material which is piled as high 
as was that in this test. Under any circumstances it is necessary 
that care be taken and undue risks avoided. 

(c) Pig Iron: Pig iron was used as loading material in the 
test of the Franks Building, Fig. 10. From the standpoint of 




i FIG. 10. — pig iron as a LOAD FRANKS BUILDING, 

the making of the test, the worst objection to it is that, as with 
the brick, small particles break off and cause annoyance to observ- 
ers. This is less noticeable than with brick and in other ways 
pig iron is clean. It possesses the great advantage that with its 
use a very heavy load can be applied without piling the load 
extremely high. 

Tin plate in boxes 2 ft. square, each weighing 200 lb., was to 
have been used in a building test. A more nearly ideal loading 
material would probably be hard to find, but unfortunately this 
test could not be carried out. 

The intensity of the loading will depend mainly on the load 



Digitized by 



Google 



186 Slater on Testing Reinforced Concrete Buildings. , 

for which the building was designed. It will not be possible to 
make the load absolutely uniform, as aisles will be necessary for 
the purposes of (a) convenience in placing the load, (6) access to 
gauge lines for the taking of observations and (c) prevention of 







I 



mmk 



li 




FIG. 11. — MOMENT AND SHEAR DIAGRAMS FOR 
THREE ARRANGEMENTS OF LOAD. 

arching in the loading materials. It has been found that it is 
difficult to cover more than about 75 per cent of the actual area 
of the floor, and in many cases less than this will actually be 
covered. Hence in computing the probable height of the load, 
this fact must be taken into consideration. 



Digitized by 



Google 



Slater on Testing Reinforced Concrete Buildings. 187 

Aisles should be so placed that the load, even though partly 
carried by archmg of the material, will cause stresses in the floor 
which are approximately equal to and always as severe as those 
caused by an actual uniform load. Fig. 11 shows the moment 
and shear diagrams which would be obtained by loading a simple 
beam with a total load W distributed over the span in three dif- 
ferent ways, as follows: 

(a) Solid Line: Uniform load W, over full span. 

(6) Broken Line: Same load W distributed over one-half 
of span, giving aisles of equal width at center and support. 

(c) Heavy Dotted Line: Same load W distributed over 
one-half span, half of load being carried by arch action to ends 
of boxes (shown here as concentrated loads W/8), and the other 
half being uniformly distributed over the half span. 

It will be possible in almost any test to arrange the loading 
material in such a way as to come within the limits outlined by the 
three arrangements of load assumed in Fig. 11, and it is seen that 
if this is done, the presence of the aisles or of arching to the sides 
of the boxes or piers, while not affecting the amoimt of the maxi- 
mmn moment and the maximum shear, would tend to cause them 
to exist over greater portions of the span than would be the case 
with an equal imiform load. In this figure aisles equal to one- 
quarter of the span have been assumed. In no case would they 
be as large as this, and, therefore, the moment and shear diagrams 
should actually conform even more nearly to those for uniform 
load than is shown in the figure. 

Arrangement should be made, if possible, to store the loading 
material near the test area to hasten the work of applying the 
load after the test begins. The general rule has been to allow 
loadmg material to be stored as close as one full panel length from 
the test area, but the intensity of the storage load has been kept 
down as much as possible. 

preparation for the test. 

Cutting Holes in Concrete. — ^In all of these tests it is necessary 
to cut holes in the concrete in order to expose the reinforcement. 
Fig. 4 shows a hole cut in the concrete of the Powers Building 
where a series of measurements was taken on a rod passing through 
a column. This cutting has been best accomplished by the use 



Digitized by 



Google 



188 Slater on Testing Reinforced Concrete Buildings. 

of a cold chisel with a very gradually tapering point. This is a task 
for common laborers and a long one for inexperienced men, but it 
has been found that a great deal of speed can be developed by 
practice, hence the importance of completing this part of the work 
with a single set of workmen. 

A saving in mutilation of floors can often be effected by 
planning the test ahead of time and inserting plugs in the concrete 
during construction in the proper position for the gauge lines. 
Removal of the plugs after the concrete has set exposes the rein- 
forcement without the use of a cold chisel. Likewise metal plugs 
may be set in the concrete at the proper positions for the measure- 
ment of concrete stresses and thus save cutting into the concrete 
to place compression plugs. The point has been raised that by 
preparation of this kind a chance is given to the contractor to 
know what panels are to be tested and thus to make the construc- 
tion of that panel better than others. For this reason there is 
room for question as to the advisability of using this method. In 
most of the tests under consideration this point has been taken 
care of by the fact that it was not known until shortly before the 
test what area was to be loaded. It is believed that the saving 
thus effected is not generally sufficient to justify prejudicing the 
test by the use of this method. 

Drilling of the gauge holes will be discussed under the subject 
''Instruments and Observations. '' 

Scaffolding, — A platform supported on some kind of scaffold 
is necessary which will enable the observer to get close enough to 
the floor above to take observations of deflection and deformation. 
This should be at such a height that when the observer stands 
upon it the points where measurements of deformation are to be 
taken will be about one inch above his head. For flat slab con- 
struction this condition is easily obtained (see Fig. 12), but with 
beam and girder construction where there are measurements 
on beams, girders, and the floor slab, the heights df different gauge 
lines are so different that arrangement will need to be made for 
building certain parts of the platform higher than others (see Fig. 
13). It is important that the elevation of the platform should 
be such that the observer can stand erect while taking the readings, 
and yet such that the instrument will not be too high for convenient 
and accurate observation. 



Digitized by 



Google 



Slater on Testing Reinforced Concrete Buildings. 189 

Another framework for the purpose of supporting deflection 
apparatus under the points where measurements of deflection 
are to be taken is also necessary. In order that the movements 
of the observers upon the observation platform may not jar the 
deflection apparatus, the two frameworks must be built indepen- 
dently of each other. In all the tests which have been made, up 
to the present date, these deflection frames have stood on the floor 
and have been braced from one to the other in order to make a 
comparatively rigid framework. Fig. 12 shows scaffolding and 
deflection frames for the Franks test. An objection to this method 
of measuring deflections is that changes of humidity are likely 




FIG. 12. — SCAFFOLDING AND DEFLECTION FRAMES, FRANKS BUILDING. 

to change the length of the wooden posts used, and it is quite 
probable that an improvement could be made in the form of this 
frame. An arrangement which has been suggested consists of 
steel I-beams supported directly by the columns and carrying 
other steel framework on which can be placed the deflection 
apparatus. This would give more nearly a self-contained con- 
struction, and the changes of humidity and temperature would 
not change the deflection readings, except as the length of column 
between the platform thus built up and the floor above is changed. 
Equipment. — The equipment will necessarily consist of the 
following: cutting and drilling tools, portable lights for throwing 



Digitized by 



Google 



190 Slater on Testing Reinforced Concrete Buildings. 

light into observation holes, note books and facilities for doing 
drafting and for reducing data. 

The cutting and drilling tools are sufficiently described in 
other paragraphs. 

Some kind of a portable light is a necessity, as gauge lines are 
often located in dark comers and as observations may be taken at 
any hour of the day or night. The light shown in Fig. 20 is a 




PIG. 13. — PHOTOGRAPH SHOWING VARIATION IN HEIGHT OP GAUGE UNES, 
TURNER-CARTER BUILDING. 

hunter's acetylene light and is quite satisfactory. The light is 
attached to the forehead and may be thrown in various directions 
according to the setting of the clamp attachment. The acetylene 
tank may be attached to the belt or carried in the pocket. 

Loose leaf note books should be provided in which the sheets 
are as large as can be conveniently handled and filed. The forms for 
record shown in Fig. 29 are very conveniently ruled in hectograph 



Digitized by 



Google 



Slater on Testing Reinforced Concrete Buildings. 191 

ink and copied by means of a hectograph. Printed forms might be 
used, but so many differences in detail are made to correspond 
with the particular test in question that this would not be advisable 
as too few sheets of a single form would be required to justify the 
expense of having them printed. 

For the most efficient work in computing results and making 
sketches for records, it is important that an adequate place be 
provided where some privacy may be had, where benches and 
drafting tables may be used and where instruments and other 
equipment may be kept. Fig. 14 shows the temporary office 
which was provided in the Turner-Carter Building test. This 




PIG. 14. — TEMPORARY OFFICE, TURNER-CARTER BUILDING. 

is one of the portable office shanties which the Company moves 
to places where work is being done. The photograph shows the 
interior of the office with the observers and recorders at work 
reducing the data of the test. This added equipment will add 
only slightly to the cost of the test and very greatly to the efficiency 
of the work. Special attention is called to it because there is a 
tendency to neglect this part and to think of it as only a secondary 
matter, whereas it should be considered as one of the most impor- 
tant pieces of equipment. 

Summary of Test Data. — ^A summary of the main features 
of the building tests discussed in this paper is presented in Table I, 
as it is believed that the information given there will be of assist- 



Digitized by 



Google 



192 Slater on Tbbtino Reinforced Concrete 

t. I I S • I * 



BUILDXNGB. 



o 
Q 



! 



<3I 



t 



II 

£1 



go 2 






ja 






iril 



>* 00 I I »« 



0909ClCOe« 09^09 



S S S S 3 S S 



§ § i § ill 



§ i § g § § g § 



1 


11? 


04 lO O 40 ^ 
<«' CO Q «0 <«' 

iH ^ .H 1-1 


100.0 

1.5 

10.0 


1 




2850 
1800 
1070 
1959 
1690 


1270 

440 

2450 



II i 



1 1 



pqpQCf^cS mPQra 



I 

1 



£ pq 



II 
II 



.Si 
& 



•8 

S3 



II 



I 

s -s 



II 






Digitized by 



Google 



Slater on Testing Reinforced Concrete Buildings. 193 

ance in the efficient planning of and preparation for such a test. 
The foUowing notes are in explanation of data given in this table: 

The figures giving area of test show the total area of the floor 
covered and do not count any area twice even though loaded 
twice as was done in the WenaJden Building test. It does include 
the area of separate singly panel tests which were made in the 
Wenalden and Franks tests. 

The maximum test load in lb. per sq. ft. is given in the column 
under that caption. In some cases this was over only a part of 
the test area. The proportional parts of the test area having the 
maximum load applied were as foUows: Wenalden 80 per cent, 
Powers 50 per cent, Franks 40 per cent, all others 100 per cent. 

The column giving the amount of load handled includes the 
rehandling due to change of position of loads. The proportionate 
parts of the loads rehandled in this way were Wenalden 40 per cent. 
Powers 60 per cent, Franks 80 per cent. In all the other tests no 
load was rehandled. 

The column giving the number of observers includes only 
those reading deformations. In the Wenalden and Powers tests 
another observer took deflection readings. In the Powers test 
and the Barr tests, almost all the deformation readings were taken 
by each of two observers, giving a larger number of gauge lines 
per observer than in the other tests. 

III. Instruments and Observations. 

Exiensometers. — ^In the past the great obstacle to the measure- 
ment of deformations in building tests has been the difficulty of 
attaching the measuring instruments to either the reinforcement 
or the concrete on the flat surface of a floor, and recent tests show 
the necessity of making measurements of reinforcement deforma- 
tion directly on the reinforcement. A satisfactory method of 
accomplishing this has been provided by the introduction of the 
extensometer invented by H. C. Berry of the University of Penn- 
sylvania and adapted to this work by improvements made at the 
University of Illinois. This instrument is similar in some respects 
to the strain gauge designed and used as long ago as 1888 by 
James E. Howard, then Engineer of Tests at Watertown Arsenal, 

The great value of this instrument in building tests lies in the 
following facts: (a) Its use makes it possible to take measurements 



Digitized by 



Google 



194 Slater on Testing Reinforced Concrete Buildinos. 



directly upon the surface of the reinforcement and concrete. 
(6) With its use there is no apparatus left in place to be damaged 
or disturbed during loading, (c) Due to the fact that it is portable, 
measurements may be taken in a large number of places with a 
single instrument. Measurements have been taken at as many 











-S^et^ Lirrm. 







Ecc&ntnc hole Central hole 

Finishinq 
tool, 

FIG. 15. — UNIVERSITY OF ILLINOIS EXTENSOMETER. 



as 104 points in a single test. This would call for an outlay of 
from 11200 to $2500 for instruments if fixed instruments were used. 
Fig. 15 shows the Illinois extensometer in its present form. 
Any movement of the point B due to a change in the length of the 
gauge line is transmitted to the Ames gauge through vertical 
movement of point C, by means of the leg BD and the arm DC 
pivoted at D. The Ames gauge is sensitive to a movement at C 
of .0001 in. The ratio of the length CD to the length BD is 



Digitized by 



Google 



Slater on Testing Reinforced Concrete Buildings. 195 

approximately 5 and the Ames gauge is thus sensitive to a move- 
ment at B of .00002 in. (.0001 rD.-r-S). However, this must not 
be taken to mean that the extensometer possesses this degree of 
accuracy in measuring stresses, since some movement of the point 
of the leg at B is certain to result from variation in the handling 
of the instrument. 

To obtain the exact ratio between movements at points B 
and C the instrument is calibrated by means of a Brown and Sharpe 
screw micrometer. For known movements of the point B readings 
of the Ames gauge are taken and a calibration curve plotted for the 
entire range of the instrument. 

The first instrument of this type built by the Engineering 




FIG. 16. — BERRY EXTENSOMETER. 

Experiment Station of the University of Illinois was made by 
arrangement with Mr. Berry for the Deere and Webber test. 
It was designed by H. F. Moore and A. E. Lord, and was like the 
instrument in use at present except that it had a 15-in. gauge 
length and was made entirely of steel. Later on in making the 
instrument for general use aluminum was substituted for steel in 
order to reduce its weight and the gauge length was made variable 
from 6 to 11 in. Since then several minor changes have been 
made. The legs have been made stiflfer in order to reduce the 
error due to unconsciously applied longitudinal thrust and the 
points have been made sharper in order to reduce the pressure 
required in seating the instrument. These improvements have 
reduced the probable error of observation considerably. 



Digitized by 



Google 



196 Slater on Testing Reinforced Concrete Buildings. 

The extensometer loaned by Mr. Berry to the University of 
Illinois in 1910 and used in the Deere and Webber test is shown m 
Kg. 16. It differs from the Illinois instrument in that the move- 
ment of the multiplying arm is measured by means of a screw 
micrometer instead of the Ames gauge head, the point of contact 
of the micrometer plimger and the lever arm being determined 
by means of a telephone apparatus. The screw micrometer and 
the frame of the extensometer are insulated from each other and 
are connected with the poles of a small battery by means of copper 
wires. A contact between the plimger of the screw micrometer and 
the multiplying lever completes the circuit and the current set up 
produces a vibration of the diaphragm of the telephone apparatus 




PIG. 17. — LATEST TYPE OF BERRT EXTENSOMETER. 

carried on the head. This method of observation is very slow and 
the apparatus gets out of order very easily, 

The use of the Ames gauge head instead of the screw 
micrometer and telephone apparatus adopted by Mr. Moore in 
the instrument used in the Deere and Webber test has greatly 
facilitated the use of the extensometer. The legs of this instru- 
ment also were made longer in order to adapt it to the measure- 
ment of deformations of reinforcement imbedded in concrete. 
Both of these modifications have later been used by Mr. Berry in 
instruments which he has put on the market. 

The extensometer more recently designed by Mr. Berry is 
shown in Fig. 17. It is not different in principle from the one 



Digitized by 



Google 



Slater on Testing Reinforced Concrete Buildings. 197 

just described. It differs from the IlUnois instrument in the 
following details: (a) Instead of having a uniformly variable 
gauge length ranging from 6 to 11 in. it has two fixed gauge 
lengths of 2 in. and 8 in. respectively. (6) The instrument shown 
here has a multiplication ratio of two between leg and arm, and 
in order to make this ratio five (as in the Illinois type) it is nec- 
essary to use a leg which is only one inch long. With this 
arrangement the instrument can not usually be used for meas- 
uring deformations in reinforcing bars, owing to their depth of 
imbedment. (c) This instrument is put out with framework of 
Invar steel or aluminum. While Invar steel makes the weight 




FIG. 18. — BERRY EXTENSOMETER AS MODIFIED BY TRELEASE. 

somewhat greater than that of the aluminum instruments, it has 
the great advantage that so great dependence on an Invar steel 
standard bar is avoided and the study of the temperature changes 
in the reinforcement and concrete of the structure is accomplished 
with greater ease. 

F. J. Trelease of the Corrugated Bar Company has designed 
an instrument of the Berry type and has used it in at least one 
test. This instrument, shown in Fig. 18, also has as its main 
feature a multiplying lever which actuates the plunger of an Ames 
gauge head. The principal difference between this instrument 
and the one shown in Fig. 15 is that the multiplying lever is 



Digitized by 



Google 



198 Slater on Testing Reinforced Concrete Buildings. 

vertical instead of horizontal. Results have been obtained with 
it which do not differ much as to accuracy with those of the 
Illinois type of instrument. 

Use of Berry Extensometer. — In obtaining good results with 
this extensometer, a great deal depends upon careful manipula- 
tion of it. Two things which are of great importance in this 
respect are (a) the preparation of the gauge holes, and (b) care 
and experience in the use of the instrument. 

The exact gauge length is best secured by the use of some 




FIG. 19. — INSTRUMENTS AND TOOLS. 

kind of gauge marker such, for instance, as is shown in Fig. 19 
used for marking points where gauge holes are to be drilled. In 
the work of the Illinois Engineering Experiment Station the holes 
are drilled with a No. 54 drill (.055 in. in diameter). At the 
beginning of the use of the Berry extensometer a number E 
countersink drill (approximately 3/32 in. in diameter) was used, 
but a smaller one seems to be better, because it is easier to get the 
properly finished hole, and because a slight eccentricity of the 
gauge holes on the reinforcing rod causes less error in manipula- 



Digitized by 



Google 



Slater on Testing Reinforced Concrete Buildings. 199 

tion of the extensometer when a small drill is used. In the case 
of measurements on small rods also, the 3/32 in. drill cuts away 
a large percentage of the metal in the rods. Up to the present 
time, for drilling these gauge holes a breast drill has been used 
which is geared so that one revolution of the crank gives about 
4^ revolutions of the drill. In the hands of a skilled workman 
very satisfactory work can be done in this way, but where, as 
quite frequently will be the case, the drilling has to be done by 
persons not familiar with this kind of work, something better is 
needed. A drill driven by a flexible cable attached to a small 
electric motor giving a speed of rotation of 400 r. p. m. and 
upwards probably would be much better. Where high carbon 
steel has been encountered many drills have been broken and 
even when a hole was drilled a poor job has often been the 
result. After drilling the holes, the edges should be finished to 
remove the burr and to round off the sharp comers. The tool 
shown in Fig. 15 is designed to accomplish this purpose. Such 
a tool should not be a cutting tool but rather a wearing or polish- 
ing tool. A pointed magnet to remove steel dust and small frag- 
ments of steel torn off in drilling would be of use. It is hard to 
place too much emphasis on the proper preparation of gauge 
holes. 

Standard Bar. — While the careful preparation of gauge holes 
is important, not less so is the use of a standard bar. The neces- 
sity for it was first found in the test of the Deere and Webber 
Building. Variation in temperature was sufficient to cause a 
change in the length of the instrument as great in many cases as 
that in the reinforcement due to the applied load. Hence it was 
found necessary to make observations on an unstressed standard 
bar showing any temperature changes in the length of the instru- 
ment. In this test a bar of about |-in. steel was used as a stand- 
ard. It was protected from rapid temperature changes by 
imbedment in plaster of Paris, but kept on the floor where the 
test was being made. In this way it was expected to make the 
change in the length of the standard bar due to temperature 
variations about equal to the change in length of the reinforce- 
ment due to the same cause. To some extent this purpose was 
accomplished, but as the plaster covering was thin and not very 
dry the change in the standard bar must have been much more 



Digitized by 



Google 



200 Slater on Testing Reinforced Concrete Buildings. 

rapid than that in the reinforcement. In the test of the Wenal- 
den Building precautions were taken to imbed the standard bar 
in concrete. This practice has been kept up in tests made since 
then, and in addition standard gauge lines have been established 
in parts of the floor not affected by the load. These latter have 
been placed both on the reinforcement and in the concrete. Fig. 
20 shows the taking of an observation on a standard gauge line 
in the Turner-Carter test. It can be seen that it is located in a 




Fia. 20. — TAKING AN OBSERVATION ON STANDARD GAUGE LINE, TURNER-CARTER 

BUILDING. 

part of the floor entirely away from the loaded area. The great- 
est development in the use of the standard bar has been in the 
frequency of reference to it and in the development of an exact 
system for the calculation of temperature corrections. It was 
previously noted that a steel instrument was used in the Deere 
and Webber test but that in the subsequent tests an aluminum 
instrument was used. Since the coefficient of expansion for 
aluminum is almost twice that for steel, it is apparent that 
dependence on the standard gauge line must have been of still 



Digitized by 



Google 



Slater on Testing Reinforced Concrete Buildings. 201 

greater importance in the later tests. Difficulty was found in 
interpreting the notes taken on the Wenalden test, but the 
greater dependence on the standard gauge line and the more 
systematic use of it observed since then has very largely overcome 
this difficulty. Subsequent to the completion of the last building 
test participated in by the writer, standard bars of Invar steel 
have been secured. Invar steel has a coefficient of expansion 
only about one-sixth that of ordinary steel and its use as a stand- 
ard bar makes it possible to eliminate from the results almost all 
the eflfects of temperature variation. If it is desired to determine 
how great are the temperature eflfects, a standard gauge line can 
be placed in the floor as before in such a position as not to be 
aflfected by the floor load. 

It has been the practice in the more recent building t^sts for 
each observer to make observations regularly on two standard 
gauge lines. This is done so that one may form a check on 
the other. If only one were used, a large accidental change in 
the readings due for instance to sand in the gauge holes might be 
mistaken for a temperature effect. If two standards are used, 
any such accidental change as the above would seldom be the 
same in both, and the error would be detected. An accident to 
the instrument would probably cause the same change on both 
standard gauge lines and the use of the two standards would not 
help to detect this kind of an error. However, such errors are 
usually so large as to be apparent in any standard reading and 
are infrequent as compared with errors due to filling of the gauge 
holes. 

Deflection Instruments. — In the building tests described in 
this paper deflection instruments of two types have been used, 
one being that used by the Illinois Enghieering Experiment 
Station and the other that used by the Corrugated Bar Company. 
The former, shown in Fig. 21, consists of a screw micrometer head 
of 1 in. travel, connected in tandem with an Ames gauge head 
micrometer of | in. travel. The screw micrometer is designed 
to cover large variations in deflections, and the Ames gauge 
head, small ones. Fig. 21 shows also the method of using this 
deflectometer. A plate, having a ^-in. steel ball attached, is 
plastered to the surface, deflections of which are to be measured. 
A |-in. bolt, which has a steel ball inserted into its upper end, 



Digitized by 



Google 



202 Slater on Testing Reinforced Concrete Buildings. 

is set into a wooden block (part of the deflection framework) in 
such a way that its elevation can be adjusted to give any desired 
zero reading of the deflectometer. Thus at the beginning of a 
test all the zero deflection readings can be determined so that for 
a considerable length of time all the change in deflection will be 
shown on the Ames gauge without any change of the screw 
micrometer. As larger changes take place, a second setting of 



Concr^fm Floor 




kVooden ■■ 




FIG. 21. — DEFLECTOMETER, UNIVERSITY OF ILUNOIS. 

the screw micrometer will probably be necessary. The great 
advantage of this instrument is the rapidity with which it can be 
used. It has been found to work very satisfactorily in most 
respects. A shortcoming, however, has been the lack of a revolu- 
tion counter on the Ames gauge so that in case of large changes 
of deflection it is possible to make an error of as much as 0.1 in. 
in interpreting the readings, though this is very unlikely. This 
instrument was last used in the Turner-Carter test and since then 



Digitized by 



Google 



Slater on Testing Reinforced Concrete Buildings. 203 

an Ames gauge head, which has a revolution counter, has been 
provided for it, so that the diflSculty here mentioned is not likely 
to occur in the future. 

The deflectometer used by the Corrugated Bar Company is 
shown in Fig. 22 and consists of a screw micrometer depth gauge 
by means of which distances for varying loads are measured 
between the stationary frame and a point on the beam or floor 

/ XT' '^ ' i//////P//////^//i//P/ ' ^'V^ii'f^FA c* - V ^ ■ '/ 



tnakh Micrometer'^ 




Fia. 22. — DEFLECTOMETER, CORRUGATED BAR COMPANY. 

slab. It has the advantage over the one previously described 
that actual distances are measured instead of changes in dis- 
tance, so that if the complete reading is taken each time, there is 
no possible way of misinterpreting results. It has also the advan- 
tage of a much larger range of measurement. In the Barr panel 
test a gross deflection of more than 3 in. took place. As the 
Illinois type of deflectometer has a range of only \\ in. it could 
not have been used in this test. This, however, is more than 



Digitized by 



Google 



204 Slater on Testing Reinforced Concrete Buildings. 

would often if ever occur in the test of a building. Its disadvan- 
tage is that it requires a longer time to make an observation than 
does the deflectometer previously described. 

Observers, — Observers should be experienced in the use of the 
Berry extensometer before imdertaking work on a field test. The 
chances of error in the manipulation of the instrument are large 
and as a rule the deformations measured are small, so that the 
error is likely to be quite a large proportion of the total measure- 
ment; hence it is important to reduce errors to the lowest pos- 
sible limit. 

Extensometer Observations. — ^If the observations at zero are 
equally as good as other observations, a curve may be drawn 
through all the points of any load-deformation diagram after the 
test is completed, weighting the zero observations equally with 
the others and the zero point shown by the most probable curve 
should be used as the origin. This method involves waiting until 
the completion of the test to draw these curves. It would be 
better to spend much more time on the zero observations, in order 
to make them reliable, than is paid to any other series. By this 
means a check can be had upon the action of the structure as the 
test progresses and the construction of the most probable curve 
will be made more simple. To do this it is essential that several 
complete series of zero observations should be taken with no 
load on the floor, and it would be weU to repeat this through 
considerable range of temperature to study temperature efifect 
on the reinforcement and on the concrete. This study was 
attempted in the Deere and Webber test, but the changes both in 
instruments and in reinforcement were included in the measure- 
ments and could not be separated, so no definite conclusions 
could be drawn. However, with an Invar steel standard bar or 
with an instrument made of Invar steel these two kinds of 
changes can be separated and to some extent at least the effect 
of temperature determined. 

In taking an ordinary observation about five readings should 
be averaged. In all of the building tests which have been made, 
individual extensometer readings were recorded, but in laboratory 
tests the practice of averaging the results mentally has been 
adopted. This gives very satisfactory results for laboratory tests 
and saves a great deal of time. It is possible that this practice 



Digitized by 



Google 



Slater on Testing Reinforced Concrete BmLDiNGS. 205 

could be adopted for field tests also. It would save time on a 
test and with a good recorder the calculations could be kept up 
with the observations. In the more recent building tests the 
practice followed in obtaining readings for any observation has 
been to reject all readings until 5 consecutive ones have been 
obtained which agree within .0004 in. These five consecutive 
readings then are averaged to form an observation. 

Deflectometer observations have been suflSciently discussed 
in the description of the deflectometer and will not be taken up 
again here. 

Observation of Cracks. — ^Up to very recently the observation 
of cracks has been considered one of the most important features 
of a test, and if carefully done it may yet add considerable to the 
confidence in the results. These observations should be made 
and recorded for zero load and at each increment of load. This 
is one of the most tedious parts of the test, and to carry it out 
faithfully requires a great deal of patience. The examination 
should be minute and very thorough. One who is not familiar 
with this kind of work will be likely to miss important indications 
and careful supervision should be maintained over this part of 
the investigation. 

Special attention has been called to observation of cracks 
because of incorrect ideas which apparently prevail with regard 
to them. It seems to be the idea of some engineers that the type 
of construction advocated by themselves is immune from cracks. 
When it is remembered that plain concrete fails in tension at a 
unit deformation of about .0001, it is apparent that cracks must 
form when the stress in the reinforcement is such as to correspond 
with this deformation, or at about 3000 lb. per sq. in. At this 
stage the cracks are often too small for detection with the naked 
eye, but almost always very fine cracks are foimd at stresses 
ranging between 3000 and 10,000 lb. per sq. in. Thus to report 
for a floor loaded to twice the designed load that no cracks were 
observed is to admit one of three things, namely, that an excess 
of reinforcement was used, suflScient care in taking observations 
was lacking, or that not all the facts of the case were reported. 

It should be borne in mind that the cracks referred to in 
this pAper are often extremely minute and usually are not visible 
to a casual observer. Frequently cracks have been traced with 



Digitized by 



Google 



206 Slater on Testing Reinforced Concrete Buildings. 

a lead pencil to make them distinct for the purpose of sketching, 
and it seems apparent that some persons visiting the test have 
mistaken these pencil marks for large cracks. At any rate reports 
have been circulated as to the existence of large cracks in a test 
where to the writer's personal knowledge there were none. 

ACCURACY OF DEFORMATION MEASUREMENTS. 

Probable Error. — The ratio of multiplication in the Berry 
extensometer is not exactly equal to the ratio of the length of the 
arm to the length of the leg, the error being due to the fact that 
the plunger of the Ames gauge head does not always travel in a 
line perpendicular to the multiplying lever. However, calcula- 
tions show that this approximation results in an error in the 
measurement of reinforcement stresses equal to only about one- 
quarter of one per cent for an extreme case. It may be seen later 
that errors of observation are large enough in proportion that this 
error can be neglected. 

In forming a basis for a conclusion as to the accuracy of the 
figures given out as results of tests, use has been made of the check 
readings taken by two observers on the same gauge lines and of 
calculated probable error of the means of five readings. While 
it is possible to calculate with some accuracy the probable error 
of replacing the instrument on the same gauge line time after 
time at one sitting, it is very difiScult to determine the error caused 
by gradually cramping the quarters of the observer as the loading 
material piles up. A determination of errors based on independent 
checking by a second observer should be expected to eliminate 
to a large extent errors of all kinds and the greatest dependence 
should be placed on this kind of results. 

In the test of the Powers Building most of the observations 
taken were checked by a second observer and some of the results 
are shown in the load stress curves of Fig. 23. The values shown 
in solid circles were observed by F. J. Trelease and those in open 
circles, by the writer. The zero reading for the latter is in each 
case at a load of 50 lb. per sq. ft., and in order to make a direct 
comparison of results, all these curves must be set over so that 
their zeros coincide with the stress values at 50 lb. per sq. ft. of 
Mr. Trelease's curves. Having made this correction the average 
variation between all the comparable points is about 670 lb. per 



Digitized by 



Google 



Slater on Testing Reinforced Concrete Buildings. 207 

sq. in. (.0000223 unit deformation), which amounts to a probable 
error of approximately ±340 lb. per sq. in. (±.000011 unit 
deformation). 

Fig. 24 shows the results of a series' of measurements taken in 
the same way on the upper and lower surfaces of a 4 x 4 in. timber 
beam loaded with sacks of sand on a 12-ft. span. The points in 
open circles represent measurements on the top surface and those 
in crosses on the bottom surface. Determined in the same way, 





S S S S ^ 

lo lo ^ "o o 

Si'ee/ Sfress mlh per S(f /n. 

FIG. 23. — LOAD-STRESS DIAGRAMS OF TWO OBSERVERS, POWERS BUILDING. 

these measurements show an average probable error of approxi- 
mately =*= .000017 unit deformation. As previously stated, these 
check measurements must be taken to give results more applicable 
than calculations of probable error of the mean of a group of 
readings. However, it may be expected that where an increase 
in accuracy of setting the instrument is found, a decrease in error 
due to cramped quarters, etc., will also be found. In Fig. 25 is 
given a curve which shows for each of four building tests the 
probable error of the average of five readings. Each plotted point 



Digitized by 



Google 



208 Slater on Testing Reinforced Concrete Buildings. 

is the average of the probable errors calculated for six different 
gauge lines at a given load. What this diagram may be expected 
to show is the improvement in results with increased experience 
rather than the actual value of the probable error. The marked 
improvement in results shown here is due in part to increased 
skill in the observer and in part to improvement in the instrument 
itself. Fig. 26 gives a curve showing deformations in a bottom bar 
of the Barr test panel as shown in the sketch. The points shown 




ZOO AOO 600 

L oad in Pounds 

FIG. 24. — LOAD-DEFORMATION DIAGRAMS OF TW'O OBSERVERS; TEST OP A 
4 X 4-IN. TIMBER BEAM. 

as open circles are for a load of 590 lb. per sq. ft. and solid circles 
are for a load of 615 lb. per sq. ft. This is the best curve the 
writer has been able to obtain on any building test and it can not 
be taken as representative, but rather to illustrate what may be 
obtained under the best conditions. The regularly varying differ- 
ences for a small difference of loads indicate that the stresses 
must^have been determined correctly within a very small range. 
A study of probable error was made in the Turner-Carter 
test by the use of a series of 100 observations taken by each of the 



Digitized by 



Google 



Slater on Testing Reinpobced Concbetb Buildinos. 209 



^/OOO 



^ 






eoo 



eoo 



^ 400 























- 








^, 




























> 


iw 






}J 








ki 












1 
1 


N 


V 




1 








1 








^ 


— 




N 


Nv 


$ 








8 








1 






^ 


N 








^ 








Hi 
Is 








^ 


N 






s 








$ 














s 


s^ 


1^ 








K 
















> 


k 








1 






















V 


























^ 





OrvfQr 



3 
of Tests 



PIG. 25. — AVERAGE PROBABLE ERROR FROM FOUR BUILDING TESTS. 




JO0O5 
\.0003 

\jOOOB 
^.0003 



Load'd/L '^ It per t "cf Ft- 



iK^ 7^t3g0Z \pffr S(^/- P" 



1§ % I ^ 8" 

Distance in Inches from Edge of Beam 



FIG. 26. — DEFORMATION ALONG BOTTOM REINFORCING BAR, BAR PANEL TEST. 



Digitized by 



Google 



210 Slater on Testing Reinforced Concrete Buildings. 

two observers on two gauge lines selected as likely to give the most 
and the least accurate results. The results of this study are given 
in Table II. 



TABLE II — ^Probable Error of the Average op Five Consecutive 

Readings. 





Obsenrer 


Gauge Line. 




1 


2 


Average. 


Unit deformation 


/ H. F. Moon 
\ W. A. BUter 

/ H. F. Moore 
\W.A.SUt6r 


.00000687 
.0000043 

206 
180 


.0000106 
.000014 

818 
436 


.00000873 


Stren in reinf oroement in lb. periq. in . 


.0000001 

262 
282 



While these measurements were not all on reinforcement, 
the probable error has been reduced to terms of stress in reinforce- 
ment for convenience of interpretation. It is very interesting 
to note that the average probable error of =^282 lb. per sq. in. 
agrees very weU with that for the Turner-Carter test as shown in 
Fig. 25. The same observer took the data in both cases, but the 
data for the value shown in Fig. 25 are taken directly from the 
records of the test and r^resent the condition on six typical 
gauge lines. The method of obtaining the values given in Table 
II is explained above. 

From the data in hand it seems safe to conclude that for 
ordinary conditions stresses in reuiforcement can be measured to 
the nearest 1000 lb. per sq. in., though in the past there have been 
some glaring failures to obtain as great a degree of accuracy as this. 
The advantage of further increase in accuracy of results lies in the 
determination of the relation of parts of the structure. 

Effect of Changes in Temperature on Accuracy of ResvUs, — 
Changes of temperature will give measureable changes of length 
in reinforcement, in concrete and in instruments made of ordinary 
materials. In most of the building tests corrections have been 
made for the changes in the instrument due to changes in tempera- 
ture by means of observations on standard unstressed gauge 
lines chosen to represent as nearly as possible the conditions of 
the reinforcement and the concrete in the part of the structure 
tested. The method of calculating this correction will be described 



Digitized by 



Google 



Slater on Testing Reinforced Concrete Buildings. 211 



below. It is there mentioned that in distributing the corrections 
found by reference to the standard bar, a linear variation from the 
time of one standard observation to the time of the next standard 
observation was assumed. Some observations have been made 
to determine the correctness of this assumption. 

To determine the amoimt of change in length of an aluminum 
extensometer covered and imcovered, a test was made in which 
the two instruments were suddenly exposed to a change of tempera- 
ture of 60 deg. F. A covering which consisted of a double layer 
of rather heavy felt protected one of the instruments from too sud- 



I 



005\ 

I 



.00/ 




(a) /nstrument 

f^) /nsfrument- 
Profcctod 



FIQ, 27.- 



/O 20 30 40 

Length of Exposure //? Minutes 

-DIAGRAM SHOWING CHANGE IN LENGTH OF INBTRUIIENTS DUE TO 
CHANGE IN TEMPERATURE. 



den a change in temperature. The other instrument was entirely 
unprotected. The results of this test are shown in Fig. 27 with 
the change of length of the instrument plotted as ordinates against 
time as abscissas. For these measurements a standard bar of 
Invar steel was used. The coeflScient of expansion of this being 
very small, the change of length measured must have been almost 
entirely that in the instrument. The curve shows that for an 
instrument not insulated from temperature changes only about 
five minutes is required for the instrument to come to the tempera- 
ture of the air. For the insulated instrument about 20 minutes 
was required. This may be interpreted to mean that if an unpro- 



Digitized by 



Google 



212 Slater on Testing Reinforced Concrete Buildings. 

tected instrument is used, readings on the standard bar should 
not be more than five minutes apart. With an instrument pro- 
tected as was this one, intervals of 20 minutes would not be too 
much. The amount of change for the case shown here is extreme 
as the instrument was suddenly exposed to a change in tempera- 
ture of about 60 deg. F. This range would seldom be found, 
and the length of time required to make the change for a smaller 
difference of temperature may be less but probably would not 
vary much for other ranges of temperature. It may he concluded 



f 



jOOSO 



^ jOO/C 
X)OOS 



/DOCO 





flX" 


T" 




o 


> 




/ 


/^ 


u 


(6) 
\ 










\. 




' «^ 


/' 









o /o ao 3o ^o so 
Time /n Minutes. 

Fia. 28. — DIAGRAM SHOWING CHANGE IN LENGTH OF A STEEL BAR DUE TO 
CHANGE IN TEMPERATURE. 

(a) |-in. square bar exposed. 
(6) |-in. round bar embedded in concrete. 

that the method used for distributing the correction is justifiable, 
since the instrument was protected from sudden change of tempera- 
ture and the observations on standard bars were usually at inter- 
vals not greater than 20 minutes. 

Temperature Effect on Reinforcetnent. — The above test shows 
the effect of change in temperature on the instrument. Another 
test was made to determine the effect of change in temperature on 
reinforcement imbedded in concrete and also exposed to the air. 
A f-in. square bar of steel entirely unprotected from temperature 



Digitized by 



Google 



Slater on Testing Reinforced Concrete Buildij^gs. 213 

changes and a f-in. round bar imbedded in 1 in. of concrete were 
exposed to a sudden change of temperature of about 43 deg. F. 
Measurements were taken on a 6-in. gauge length of each bar at 
very short intervals of time. The results are shown in Fig. 28. 
The results of this single test must be used with caution as the 
total measurements were very small and a small error would 
show up very plainly. However, the curve for the imbedded bar 
agrees in its general characteristics with some of the results 
obtained by Professor Woolson on '* Effect of Heat on Concrete" 
reported in 1907.* The test indicates that for this range of tem- 
perature rather rapid changes may be found in the reinforcement, 
corresponding with stresses of about 9000 lb. per sq. in. and 3000 
lb. per sq. in. respectively for exposed reinforcement and that 
protected as in this case. The range of temperature is extreme 
and the size of bars smaller than is often found in floor con- 
struction, therefore the results found in tests would probably be 
less extreme. However, this indicates the necessity of attempting 
to eliminate from the results of the test the effect of tempera- 
ture changes, especially if the stresses measured are small. 

IV. Records and Calculations. 

Since the beginning of the use of the Berry extensometer 
for testing purposes, as much development has been made in the 
keeping of notes as in the use of the instrument. Because of a 
lack of completeness of notes the advantages of the use of the 
standard bar were not fully realized for some time. Only after 
the method of keeping notes had been highly systematized was 
it possible to properly make the corrections which observations 
on the standard bars indicated- should be made. During the 
time of placing an increment of load the recorder will have con- 
siderable time in which to be working up results of the series of 
observations taken at the previous increment of load, and as the 
method of making these calculations is quite intricate, a man is 
required for this work who has ability to do more than merely 
record. It is important that calculations should be kept up as 
the work progresses, because it can be done with less labor then 
than at any other time and because it will be of value to know 
as the test progresses what results are being secured. 

• See Proceedtnga, Vol. VII. Am. Soc. for Test. Mala. 



Digitized by 



Google 



214 Sla-bpr on Testing Reinforced Concrete Buildings. 

RECORDS. 

It is very important on account of the great number of 
observations taken (about 12,000 in the Turner-Carter test) 
that all records be arranged systematically. The following 
points are mentioned as being important in this connection, 
(a) In the field test individual readings should be recorded and 
their average used as a single observation. The proposed abridg- 
ment of this procedure (see p. 204) should be considered as a sug- 
gestion for later development. (6) Recording readings in the 
order of their size will assist the recorder in obtaining the correct 
readings and in rapidly obtaining the average, (c) The exact 
sequence of observations should be maintained in the records 
as the calculation of corrections depends largely on this. 

Fig. 29 shows a form for the recording of original readings 
and the results calculated from them. 

calculations. 

The calculations of corrections and applying them to the 
results makes the reduction of data rather intricate. This work 
has been reduced to a definite form shown in Fig. 30. In this 
form the zero length of instrument (see definition, p. 172) is assumed 
as correct and is used as a standard of reference. The correc- 
tions are distributed among the gauge lines as though the change 
in the length were a linear function of the time from one standard 
bar observation to the next one. These assumptions do not 
entirely accord with the facts but have been satisfactory as a 
working basis. Any other standard bar observation than the 
zero length would do as well for a standard of reference except 
for matters of convenience. 

V. Cost of the Tests. 

An attempt was made to get information by which the cost 
of the tests could be estimated, but it is found that from the 
data on hand no finely drawn conclusions are warranted. The 
costs of the tests enumerated here range from about $50 to as 
much as $2000, depending on the nature of the test and the expenses 
for railroad fare, hotel bills and pay for expert assistance. In 
the case of $50 the cost is only that in excess of what would have 



Digitized by 



Google 



Slater on Testing Reinforced Concrete Buildinos. 215 



Digitized by 



Google 



216 Slater on Testing Reinforced Concrete Buildings. 



«Vj 



^ 



I 



P007 



^ 

^ 



^<^ 



^ 



^ 



^ 



^ 



<^ 



<^ 









1:^ ^ 






^H5 



II 



'u 



"1^ 



T 

^ 









v%' 



2^^ 

^ 



H 



/fao J0007 po9a 



't> 






.? 



«^ 



Vr 



^5 






vj ? 



^ 









*> N>ft 






3? 









3 



."8 3 



?2 



•5 a 



«^ 



^ 



i 



9 
■^ 



o 



(^ fiooj 3H/7 /(Ufr 






3 



P 

H 

« 

O 
§ 



o 
o 

OQ 

I 

T 



I 



\ 



Digitized by 



Google 



Slater on Testing Reinforced Concrete Buildings. 217 

been necessary for the acceptance test of a building. It does 
not include the salaries of the testing engineers and as there were 
no hotel expenses this cost is perhaps $150 less than what could 
be ordinarily expected for such a test. The data in hand indi- 
cate that a test as extensive as several in this series have been 
would be likely to cost from $1500 to $2000^ but that one of the 
type represented by the Carleton test may be made at an expense 
slightly above that of the acceptance test so frequently required. 

For a very slight additional cost, measurements of stresses 
in a building floor may be made at points of especial interest 
during the progress of the load test which is often required as a 
condition of acceptance. 

The stage has been reached in the investigation of reinforced 
concrete where building tests may be expected to contribute 
information of great value to the designer and builder in rein- 
forced concrete. The main feature of such tests should be the 
measurement of stresses, but information as to the location and 
size of cracks will be of great value in checking the results if the 
examination for cracks is conducted with sufficient care and 
minuteness. There is need for increasing as much as possible 
the accuracy of deformation measurements and experience in the 
use of the instrument is gradually accomplishing this. All the 
confirmatory evidence possible on the correctness of results should 
be obtained. 



Digitized by 



Google 



THE DESIGN OF CONCRETE FLAT SLABS.* 

By F. J. TRELEASE.f 

The rapid introduction and development of flat slab floors 
of reinforced concrete is sufficient evidence of the great importance 
of this type of construction. In fact, the demands of owners 
and builders for flat slabs are so insistent, that in spite of the lack 
of reliable methods of design, many acres of such floors are built 
every year. The insistence of this demand has led many engineers 
to advance theories of the structural action of flat slabs, and these 
theories yield perhaps the most striking contrasts and disagree- 
ments to be found in modem engineering practice. The reason 
for thib is largely because the flat slab is an extreme example of a 
redundant structure, and its mathematical analysis has so far 
been based upon certain arbitrary assumptions which vary in the 
different analyses. The differences in these assmnptions are so 
great that the practical application of the theories founded upon 
them to flat slabs of reinforced concrete yield results varying by 
about four hundred per cent. This variation leads one to class 
the flat slab as beyond the range of pure analysis and one must 
look to experimental engineering for a satisfactory solution of the 
problem. 

The usual load tests of completed structiu*es may be dismissed 
at once as being entirely inadequate as a basis of design. In such 
tests usually but one panel of a structure is loaded, which does not 
give maximum stress conditions in a flat slab floor. Even if several 
panels be loaded and the test carried to destruction it will at best 
only roughly indicate the stresses at the weakest point of the 
structure under the scheme of loading employed and cannot give 
any information as to the economy of the design. 

A very recent and more adequate form of test on completed 
structures is that in which several panels of a building are loaded, 
not necessarily to destruction, and in which the actual elastic 
deformations of both reinforcement and concrete are measured 



* Advance Review of a Thesis presented to Washington University for the degree of Civil 
Engineer. 

t Engineer, Corrugated Bar Company, Buffalo. N. Y« 

(218) 



Digitized by 



Google 



Trelease on Design of Concrete Flat Slabs. 219 

by extensometers. The results obtained in such a test would 
form a satisfactory basis for the design of similar structures if one 
could eliminate from the results the effects of tension in the con- 
crete, arch action, annular slab action and many other indeter- 
minate factors, which enter largely into all tests of reinforced 
concrete. These effects are so indeterminate that they cannot 
be satisfactorily eliminated, thus rendering results of tests on 
completed structures more or less imsatisfactory as a basis of 
design. The average engineer hesitates to coimt on these factors 
in designing and even if he had no personal objections to their 
use, none of the building laws or regulations will permit of any 
allowance being made for them. 

It would seem, then, that for satisfactory solution of the 
problem there must be obtained an empirical analysis of the struc- 
tural action of flat plates derived from experiments on plates 
of homogeneous material. 

Such an analysis could be then used in the design of flat slabs 
of reinforced concrete, using any desired combination of unit 
stresses ^nd making such allowances for tension in the concrete, 
etc., that personal judgment or various building regulations might 
dictate. 

It is the purpose of this paper to describe and give the results 
of such an empirical analysis based on experiments conducted 
by the author in the Research Department of the Corrugated 
Bar Company, under the general direction of Mr. A. E. Lindau, 
Chief Engineer. This work was started early in 1910, and has 
continued almost iminterruptedly to the present time. 

A very interesting and exceedingly simple little experiment 
was made on a sheet of heavy cardboard fastened over twelve 
spools representing columns. By pressing the fingers on this 
little model at various points much was learned as to the general 
action of flat plates. For instance, one could press upon the center 
of several panels and note that the surface of the model was convex 
upwards at right angles to the lines joining the columns and form- 
ing the sides of the panels, showing that tension in the top face 
existed at that point, although none of the systems of reinforce- 
ment then in use provided resistance to these stresses. 

The work most productive of results was a series of experi- 
ments on rubber models of flat slab floors. Rubber was chosen 



Digitized by 



Google 



220 Treleasb on Design op Concrete Flat Slabs. 

for the models as best fulfilling the requirements of reasonable 
homogeneity, low modulus of elasticity, and the ease with which 
all its physical properties could be determined in the laboratory. 
The models used consisted essentially of a plate of pure gum rub- 
ber, fixed so as to form the top of an air-tight box containing the 
proper number of 1 J in. round columns to divide it into nine panels, 
each 8 in. square. 

The method adopted for applying load to the model was 
rather unique and extremely satisfactory, as it not only permitted 
the intensity of the load to be easily and accurately read, but 




FIG. 1. — BOX AND ABPIRATOR FOR RUBBER MODEL OF FLAT SLAB FLOOR. 

insured absolute uniformity of distribution, and at the same time 
left the entire upper surface of the slab unobstructed and free 
for observations and measurements. A partial vacuum was 
formed in the box, thus obtaining on the face of the plate the pres- 
sure of the atmosphere which was read by a simple U-tube manom- 
eter. The box and aspirator are shown in Fig. 1. 

In the first series of experiments the rubber plate used was 
0.34 in. thick. A stress strain diagram obtained from a strip 
cut from this plate is shown in Fig. 2. At first the scope of \his 
series was limited to the measurement of the shape of the elastic 
surface of the plate under stress. With this end in view, deflection 



Digitized by 



Google 



Trblease on Design of Concrete Flat Slabs. 



221 



readings were taken at numerous points in the various panels. 
In some cases these were taken at points only one-twentieth of 
the span apart. 

The values of these readings were then averaged, grouping 
those which through symmetry should be alike, and the results 
plotted. Equi-deflection lines were then drawn, giving contour 



30 


"^ 




~~ 






■~" 




"~" 


""" 


"~ 


~~" 








""" 








~~ 












"~ 


""■ 




y 


7- 


" 


20 






















































? 




























































/ 










fA 
















































. 


/ 


























































4 


/ 














A^ 












































/ 


/ 
























































/ 




















r? 








































/ 




























































/ 
























^iU 




































/ 




























































/ 




























i 
































/ 


























































/ 


























































/ 


























































/ 




































\. 
























i/ 


























































/ 


ri 


1 




































\n 




















J 


r 


























































/ 


























































/< 




























































/ 


























































/ 




























































J 


















































^6 










i 


r 


























































/ 




















































4 








/ 




























































J 


r' 






















































r 




7 


f 




























































/ 


^ 
























































o 


I 




J 








_j 


_j 















































^ 6 a /o /e M /6 /3 so K Z^ ,» ZS ^30 
FIG. 2. — STRESS STRAIN DIAGRAM FOR RUBBER. 



maps of the surface. Fig. 3 shows such a map for one-quarter 
of the model, while Fig. 4 shows the lines of equi-deflection for 
the central panel. The approximate location of the lines of 
inflection for imaginary beams radiating from the column center 
have been plotted in Fig. 5. 

Many interesting conclusions can be drawn from the deflection 
maps. They show at once the general nature and intensity of the 



Digitized by 



Google 



222 Trelease on Design op Concrete Flat Slabs. 

stresses existing in the plate. They show beyond doubt that 
tension exists in the upper face of the slab at right angles to the 
lines joining the columns and forming the sides of the panel, being 
a maximum at the colunms and decreasing to a minimum at the 
mid-point, although even this minimum value of the tension in the 
top face is approximately equal to the tension in the bottom face 
at the center of the panel. Inspection of Fig. 6 will make this 
more clear. The top curve is a section along the side of a panel 




WQ. 3. — CONTOUR MAP OP ONE-FOURTH OF MODEL. 

and the lower curve a section through the middle of the panel 
parallel to one side. From the radii of curvature of these curves, 
it will be seen that the maximum stress at the side occurs over the 
colmnn, while the section through the middle shows that the stress 
is of practically equal intensity at the center of the panel and at 
the mid-point of the panel edge. Tension in the top surface 
at the edge of the panel has never been provided for in the rein- 
forcing of flat slabs, although the need for it has been clearly 
pointed out time and again by the formation of cracks in the 



Digitized by 



Google 



Tbblease on Design of Concbete Flat Slabs. 223 




FIG. 4. — CONTOUR MAP OF INTERIOR PANEL. 




FIG. 5. — USE OF INFLECTION FOR IMAGINARY RADIAL BEAMS. 



Digitized by 



Google 



224 Trelease on Design of Concrete Flat Slabs. 

concrete. These cracks, which run from column to cokmm 
along the sides of the panel, have formed in nearly every multiple 
panel load-test of a reinforced concrete flat slab which has been 
made, and have even formed in several floors under their dead 
weight only. It was hoped that the radii of curvature of the 
elastic surface could be accurately measured and the bending 
moments deduced from them. It was foimd, however, that 
results so obtained were too indefinite to be considered reliable. 
It was evident that to obtain exact information as to the 
distribution and magnitude of stresses existing in the plate, it 




5ecf^/o/7 a^ ^ of Co/(//7?/?^. 




^ecf/on fhroi/^^ d of Panel 

FIG. 6. — SECTIONS AT EDGE AND CENTER OP INTERIOR PANEL. 

would be necessary to measure the actual elastic deformation of 
the material caused by these stresses. This was attempted in 
many ways; by measuring distortion of squares ruled on the 
slab; by measuring cracks formed in plaster of paris coatings and 
lines, and by various forms of extensometers. After trying many 
devices, it was decided to use a microscope fitted with an ocular 
micrometer and to measure with it the deformations occurring 
over a gauge length 0.5 in. long. The ocular micrometer is a 
disc of glass, engraved with a fine scale, which is inserted in the eye 
piece. The disc lies in the plane of the image formed so that the 
engraved lines appear to be upon the image itself, and its length 



Digitized by 



Google 



Trelease on Design of Concrete Flat Slabs. 225 



( 




,4f 




'i. 


A^^J 


mM^^^^* 




f 


>- Jl 




-C ^- 


r 






iiip^ 


fcj^i.^ 


^1 




^ 


1^^^^^^^^ 


^^^^^^H 






'^^^^^^^^^^^^ 


^^^^^H 






•^^^^^^^^^h 


^>^ ^^^ 




• ^^^^^^F 




^^^^^^ 


^^^^j 


^ 


f 


^^^^^1 


lilu 


^ 


n 


^^^^^^I^^^^^H 






y 








'i 




^ 


r 




ria. 7. — MODEL AND EXTEN80METERS. 



Digitized by 



Google 



226 Tbelease on Design of Concrete Flat Slabs. 

can be easily read. The absolute value of one division of the scale 
depends upon the magnification used, and was approximately 
0.00048 in. in these experiments. The total field was about one- 
fortieth of an inch. 

Several devices were tried for bringing the deformations 
occurring over the gauge length into the field of the micrometer. 
At first a strip of heavy tinfoil was fastened at each end of the 
gauge length by passing a fine needle through it into the rubber. 
The pointed ends of these strips nearly touched at the middle of 
the gauge length and the distance between them was measured 
under no load, and again imder the various loads employed. This 
arrangement of tinfoil strips was not satisfactory, as readings 
taken with it were not consistent with each other, caused probably 
by slight play of the strips around the needles, or, perhaps, by 
the strips themselves buckling as the gauge length shortened. 

After many trials, the following method was adopted and 
foimd entirely satisfactory. A small piece of very fine needle 
was placed point up in the rubber at one end of the gauge length 
and at the other end was placed, point down, a piece of needle 
with the exposed end ground to a triangular shape and highly 
polished. A small strip of brass was gfoimd to a knife edge at one 
end, and at the other a small conical depression was made. The 
point of the first needle entered this depression, and the knife 
edge was very close to the triangular end of the other needle. 
The distance from the end of the strip to one point of the triangle 
was read very easily, and the arrangement gave per^pct satisfac- 
tion. Fig. 7 shows a general and a detail view of the extensometers 
in place. 

In this first series, these extensometers were arranged on 
lines radiating from the columns so that the readings both along 
the lines and perpendicular to them were obtained. Full results 
of these readings are not given, because it was found that the 
deformation at a given point increased more rapidly than the 
loads, indicating catenary action or pure tension throughout the 
cross-section of the plate. If such action existed, the tensile 
deformations read in the top fibers of the plate would be higher 
than the true values caused by bending alone and likewise the 
compressive deformations of the top fibers would be less than they 
should be. To test for this, short colmnns were clamped on the 



Digitized by 



Google 



Trelease on Design of Concrete Flat Slabs. 227 

top of the slab directly over those on the under side and the load 
reversed. Zero readings for the reversed load were taken under a 
pressure equaling the dead weight of the slab. Fig. 8 shows read- 
ings obtained in this way and under direct load, for points along 
the line forming the side of the panel. The values of the deforma- 
tion due to bending moment alone are the arithmetical averages 
of the direct and reversed readings, and the true zero line has been 





















































4C 




































































































\11 






\ 












































\ 








\ 










































\so 








\ 


















































\ 










































\r^ 








\ 








































\ 










\ 








































^ZO 










\ 


















































\ 








































'%/3 












I 






































\ 












\ 


















^«M 


, 




























> 


\ 








.«/ 


i> 


22 


rJJ 


522 


■^ 










> 








\' 














\ 




e 


i^ 


;J5 


























\ ^ 














-^ 


N 


< 
































Si o 














> 




"- 


-., 


^ 


Ty<6 


-^ 


S2 


i^ 


^ 


c 






























N 


^, 


















































\ 


s. 


















































s 


^< 


?-r; 












































( 


► 


^ 


S5 


^ 




SS 


a 












/o 
































( 













2 '4 e 



/O 



FIQ. 8. — DEFORMATIONS UNDER DIRECT AND REVERSED LOADS. 

plotted showing the amount of catenary action existing. This 
method of load reversal was too tedious to be adopted and it was 
decided to use a thicker plate in the next series of experiments. 

The results obtained in the first series of experiments have 
been discarded in favor of those in the second series and the first 
series has been regarded as merely preliminary and as having 
served its purpose in enabling the many mechanical difficulties 
to be overcome and in indicating many points to be covered. 



Digitized by 



Google 



228 Trelease on Design of Concbbte Flat Slabs. 

In the second series of tests the apparatus used was the same 
as that employed in the first series, except that the rubber plate 
was 0.485 in. think. This greater thickness was used because 
it was thought that the effect of catenary action would be less 
on a heavy plate than the lighter one used in the first series. The 
correctness of this assumption was later proved by a series of 
readings of deformations at various points under direct loads 




6 3 /o /jt /^ /3 /a 2o rz ^ 

FIG. 9. — propobhonalitt of btress to load for second model. 

equivalent to from 1 to 6 in. of water which showed proportion- 
ality of deformation to load. The results of these readings are 
plotted in Fig. 9. The lack of catenary action was also confirmed 
by reversing the load and obtaining curves identical to those 
obtained under direct load. 

Deflection readings were taken at points over the interior 
panel spaced one-tenth of the span apart. Fig. 10 is a plan of 
the interior panel upon which have been plotted contour lines 



Digitized by 



Google 



Trelease on Design of Concrete Flat Slabs. 229 




FIG. 10. — CONTOUR MAP OP INTERIOB PANEL. 




FEG. 11. — UKEB OF INFLECTION FOB STRIPS PARALLEL TO SIDES OF PANEL. 



Digitized by 



Google 



230 Trelease on Design of Concrete Flat Slabs. 

representing elastic deflections of 0.01 in. caused by a load equiva- 
lent to 5 in. of water. Sections of the surface so defined were 
taken parallel to the sides of the panel and the points of inflec- 
tion plotted in Fig. 11. The lines joining these are lines of inflec- 
tion for imaginary beams or strips parallel to the sides of the 
panel. These lines agree very well with the results obtained 
later from deformation readings. 

Fig. 12 is an exaggerated sketch of the elastic surface upon 
which have been marked coefficients for deflection. To apply 
these to concrete flat slabs reinforced parallel to the sides of the 
panels a strip of unit width is assumed along the edge of the panel, 




FIG. 12. — DIAGRAM OF ELASTIC SURFACE AND DEFLECTION COEFFICIENTS 

and another at right angles to this along the center line of the 
panel. Knowing the total load per square foot on the panel, 
as well as the eflfective depths and steel percentages of each strip, 
it is an easy matter to obtain the deflections. Instead of work- 
ing out the values of the moment of inertia of the strips one may 
follow the method outHned by Mr. Eli White in the Engineering 
Record of November 9, 1907, and elaborate<l by Mr. G. F. Dodge 
in his Diagrams for Designing Reinforced Concrete Structures. 
To simplify the computations the strips may first be treated as 
simply supported beams, and the deflections so obtained multiplied 
by the ratio of the deflection coefficients. 

The results of the first series of tests indicated that tensile 
reinforcement would have to be provided in the top of the slab 



Digitized by 



Google 



Trelease on Design of Concrete Flat Slabs. 231 



at right angles to the lines fonning the sides of the panel, and 
it was decided that a two-way system of reinforcement would be 
most adequate, both in providing this reinforcement and in tak- 
ing care of the stresses existing in other parts of the construction, 
and consequently the arrangement of the extensometers was as 
shown in Fig. 13. It will be seen from this figure that deforma- 





-.s \ ^ ^ ^r 


\ 

^ ^ N ^ /- 












s., 
















\. 
























■* 












L 


























r 
















\^ 








r, 


k 














/: 


J K 


^^ 








K 


y 












































































t 






















































































J 


t 




















\ r 


\ 








iC 


-\ 














z 



^ ^5 (J ''i S <& s 

PIG. 13. — ARRANGEMENT OF EXTENSOMETERS 



tions were read parallel to each side of the panel at points spaced 
one-eighth of the span apart, and so distributed that deformations 
in other portions of the plate were by the symmetry of the con- 
struction obtained from readings on these. 

It was found that a load equivalent to 5 in. of water gave 
deformations large enough to be easily read, and this load was 
therefore adopted. The extensometers at all the points were 



Digitized by 



Google 



232 Trelease on Design of Concrete Flat Slabs. 



to 
/3 


\ 


1 































\, 
































s 


s 


k_ — 






























o 




\, 










^ - 




" 


' ^_^ 










^ - 










^ . 






^^ 










^ , 






^^ 






/5 


v 








"^^ 


J>efor/y?af/o^3 aiy^ //ne ^-/l 












AO 


\ 
































S 


\ 


^ 






























o 




\ 










^ 




■^ 


^ 










^ 


4 






X, 






y 










''^^ 






.^^ 














" — 


"-"^ 


DeA 


or/na; 


MU7J 4 


f/or,^ 


//-v ^.a 














^, 








■ 
























^ 




N, 










^ 




^ 














, 


r 




N 










/ 






N 










y 










S 






y 










^ 




















*"**-^ 


— " 


/>eA 


'orma^ 


J'OW < 


^^ 


///JK? CC 












m 




































N 














,^ 


"N, 














j^ 


/O 














/ 


\ 














/ 


3 




S, 










y 






\ 










/ 




o 




\ 










/ 






\ 










/ 




^ 






s^ 






y 










k 






^ 






/o 






V 


— 


- 


^ 










^v^ 






=^^ 






to 












Z>e/a 


rma/A 


0/7J <^ 


3^/ 


//re Jj 


-z? 












/3 


-\ 














j 


\ 














/ 


JO 


\ 














f 


\ 














{ 






V ■ 










J 






\ 














o 




\, 










/ 






\ 










/ 




a 




s 


^, 






y 


r 






\ 


»v 






> 


/ 




JO 






"^ 


.^ ;_ 


. 


y 










^ 






^ 








?'«^ 










2?e/b 


r/7?af/ 


0/7J <i 


/^^ 


O/fe J. 


rjT 












V 4 


^N 


v 










> 




"^ 


V 










^ 


^ — ' 


V o 




S 










/ 






N 










/ 




V 






X 






y 










V 






y 












""^- 


^-"^ 


Z^e/b 


rmo// 


ao? dM 


OJ^l 


//7^ O 


■a 












^. 


V 
































^. 


s 




































X 










^ 




( . 


h-*^ 












^,„»- 


K 






""•^ 


^ 


^___^ 




'' 








^ — . 








""^^ 
















Z?e/l 


vmaA 


^0/73 a 


/to^^ 


'//7e / 


■/- 












?^ 


































5^ 


. 
















■ 


, 










^ 





%l 






■^ 






"^e/o 


'-ma// 


1/rj a/ 


'<y^ /.//?e A-'/C 












pj 


































N 
















































^ 




"^ 


*-^>^ 










^^^ 


-"^ 




^ 

^ 






1 


1 


^/c 


VYTfO// 


0/f3 


^0/7£ 


Z//?e 


z-z 


1 


1 






V 



Pia. 14. — MOMENT CURVES. 



Digitized by 



Google 



Trelease on Design of Concrete Flat Slabs. 233 



o 


^-^m 








^ 








1 




























""^->. 


^ 






^ 






^-^..^ 
















I>/j/r//)t//xv7 acrajj lf/?e ^-A 










^^ 


o 












































, 


\ 










"*^ 




^ , 




















""^ 





1 ^ 












"""^^i 


r-. 














D/j/r/itHz/fo/? ocroJ3 l/ne S3 












Q 


































s 
















" , 


^ ' 












.--*ij 


^..^^^ 
















































D/sfr/tk/^/on acroxs A//>e 


cc 












































J 












^^ 


^-^^ 














^>-^ 


-^ 


o 


" ' 








^— -^ 


^ 






L 


^^ 




^ 


















J)/jfr//>t// 


/au7 ocrosj l/ne 


DD 












K 




















































^— V 
















^ 














y 


/' 


^S 


< 










^ 


/^^ 














/ 






\ 










/ 




^ 


^.>^ 










^ 


\ 








^ 


L 










o 










•""^ 














r— 


►— — 


















/>. 


fjfr/bi 


///on 


aero. 


w l/ne £^%JL 












Q 


































3 














' 


^ 


y-^ 












■ 


" 












I>/3/r//>t///o/? across One OO 












































o 
° 


































3 















































^""^ 


•-"— .^ 








>- H 








r- 
















!>/ 


3/r/i)o 


//on t 


7cro33 l/ne 


'/■/ 










X' 













-A 








Z!^ 


























" H 


"— — 




— — "^ 








""-"^ 






r-*— -. 












/?. 


'jmbi/f/o/? 


acro^ 


^3 l//fe A'/r 












































^49 
























— 




















== 


Z> 


/j/r/A 


u//on 


f7cra. 


^S l/A 


V li 


\ — ' 










































*> 


































V" 






















^ 




' ~- 


-. — ^ 






K 












, 


, 






---' 


' 








*'*"**^ 




















' 


■^ 


1 












' 


x^J 












/)/j/r/d<///on 


oe/vj3 Z/ne /^-/f 












1 






^ 


I 






1 








> 


» 






1 



FIG. 15. — ^DISTRIBUTION CURVES. 



Digitized by 



Google 



234 Tkeleasb on Design op Concrete Flat Slabs. 

read under no load, and again, under load; the differences giving 
the elastic deformations caused by the load. The load was 
applied and released time and again, as many as forty readings 
being taken on some of the critical points by each of two observers. 
The readings obtained at any given point agreed with one another 
to within a very few divisions of the micrometer. Table I gives 
the averages of all the readings taken. It will be seen upon inspec- 
tion of Fig. 13 that two sets of curves may be plotted from these 
readings. One set will show moment diagrams for imaginary 
strips, the axes of which lie along the zero lines of the curves. 
Curves of the other set show the distribution of moment among 
various strips of this kind. Since a two-way system of reinforce- 







Table I. — Averages of Deformation Readings. 




Point. 
Line. 


Bos. 


A 


A 


C. L. of 
Exterior 
Panel. 


A. 


Tane 

of 
ColB. 


pI 


A. 


A 


C. L. of 
Interior 
Panel. 


A 

B 
C 
D 

e 

G 

I 
K 
L 


18.8 

9.17 
14.0 


6.6 

0.0 
4.5 


-2.26 



-3.0 
-2.0 
-0.83 
-0.67 


-1.83 

-6.0 

-0.5 
-6.5 
-4.1 
-3.0 


0.5 

-1.0 

-3.0 

-1.17 

-0.25 

-1.67 

-0.6 


3.82 

6.5 

0.33 

17.33 

21.25 

10.33 

2.54 

2.0 


2.67 
4.25 
6.33 
7.0 
10.61 

6.0 


0.5 
-0.5 
-0.25 
-1.0 
-4.0 

-0.6 
-1.33 


-3.0 

-3.76 

-4.75 

-«.o 

-8.0 


-4.03 
-5.6 
-6.0 
-10.0 
-0.66 
-7.6 
-3.33 
-2.0 
-0.75 



ment was to be used, it was decided to design each set of bars 
independently. For square panels the reinforcement would, of 
course, be the same in each direction. 

Fig. 14 shows the moment curves for strips parallel to one 
side of the panels, while Fig. 15 shows distribution curves for 
these same strips. These two sets of curves can be combined to 
form a surface showing both moments and distribution for one 
set of strips. A photograph of this surface for four panels in one 
comer of the model is shown in Fig. 16. 

Fig. 17 is a sketch showing the surface for the interior panel. 
In this figure the ordinates from the plane A-B-C to the surface 
are proportional to the deformations of the top fiber of the plate 
parallel to side B-C at the point at which the ordinate intersects 
the surface. Sections parallel to B-C will be the moment dia- 



Digitized by 



Google 



Tbelease on Design of Concrete Flat Slabs. 235 

grams similar to those in Fig. 14, and sections parallel to A-B 
will be distribution curves similar to those shown in Fig. 15. 

In Figs. 14 and 15 the ordinates to the ciu^es are shown in 
terms of divisions of the micrometer. In order to co-ordinate 
divisions of the micrometer with bending moment, a strip of the 
rubber was cut from the plate, and the stress strain diagram shown 
in Fig. 18 obtained. Both the axial and lateral deformations of 
the strip were measured, from which the modulus of elasticity 
was found to be about 1,000 and Poisson's ratio about .44; one 




FIG. 16. — MOMENT SURFACE FORMED BY COMBINING MOMENT CURVES AND 
DISTRIBUTION CURVES. 

division of the micrometer corresponding to a unit elongation of 
0.001. 

The value of one division of the micrometer in terms of 
bending moment was, however, obtained directly in the follow- 
ing manner: 

A strip of rubber cut from the plate was supported over two 
knife edges, and loaded as shown in Fig. 19. An extensometer 
exactly like those used on the plate was placed on top of the strip 
between the supports. Known weights were applied, and the 
accompanying deformations read. Fig. 19 shows deformations 
so obtained, plotted against bending moments in in.-lb. per inch 



Digitized by 



Google 



236 Trelease on Design of Concrete Flat Slabs. 

width of strip. From this curve the moment equivalent of any 
number of divisions of the micrometer can be obtained. The 
unit load on the panel being known^ together with the span, 
what might be termed the "gross bending moment" or M=wS^, 
may be obtained. In this case the clear span S was 6.75 in. and 
the load equivalent to 5 in. of water, or .1808 lb. per sq. in., so 
that M=wS^ = 8.2377 in.-lb. If the net moment at any point 
be ilf == wS^/z, the value of the z may be obtained from the moment 
curveB in Fig. 14 and the calibration curve in Fig. 19. With S 




FIQ. 17. — ^DIAGRAM OF MOMENT BURFACB FOR CONTtNUOUB FLAT PLATS 
SUPPORTED AT CORNERS. 

equal to the clear span from edge to edge of column heads, z » 
215/D, where D is the ordinate to the moment curves in divisions 
of the micrometer. 

On Fig. 20 the moment expressions obtained in the above 
manner have been marked at the critical points. 

Having now obtained an empirical analysis of a homogeneous 
plate it will be applied to flat slabs of reinforced concrete. This 
may be done without hesitation because the whole history of the 
development of reinforced concrete has been along such lines. 
It was found that once having solved the problem of internal 



Digitized by 



Google 



Tbelease on Design op Concketb Flat Slabs. 237 

stress distribution in a concrete beam, there could be safely- 
used the equations for external moments applying to homogeneous 
beams, and so on. 

In this empirical analysis the components in two directions 
of the stresses existing in a homogeneous plate have been measured 
and reinforcement will be supplied in the concrete flat slab to 
resist these components where they show tensile stresses. In 











A 


















"H 




~^ 


























/v 








t 






















'j 


































1 






















/ 


























n 








\ 






. 
















f 


































h 


















• 


J 


• 


























m 








R 




















J 




























\o 








\\ 
















< 


/ 






















































/ 




















































•i 


r , 






























\^ 


































































\- - 














f 
































•w 










\. 












•> 


































^, 










\ 












/ 




















































i 














































I 










I 




































\/o 










• 








• 


f 












































\ 








} 






































%^ 










\ 








/ 














































1 






1 i 


r~- 
















































1 






J 
















































1 






f 
















































\ 




y 






















































n 


/ 










































r 












I , 
























































\ / 












































o 












f 












































i 


ia* 


4, 


S/ 




nac//a^ 


r 


J 




/ 




to/ 


^9 


s 


'A 


V 
►/7 


I 


a 


3i 





Jt 


5 


^ 


o 


•4 


s 




k> 





no. 18. 



^/i^/j/o/fj o/' Af/cra/7?efer 

FROM PLATE LOADED FOB MOMENT CALIBRATION OF BXTEN- 
80METEB8. 



laying out the reinforcement a departure is made from the old 
diagonal or four-way system and the two-way system, Fig. 21, 
employed. 

One of the greatest advantages of the two-way system is 
the location of tensile reinforcement in the top of the slab between 
the columns, which tends to prevent the formation of cracks 
extending from column to column and stiffens the whole struc- 
ture, giving all the advantages of continuous construction. 



Digitized by 



Google 



238 Tbelease on Design op Concrete Flat Slabs. 

Another point is that the tensile reinforcement over the columns 
is in but two layers and hence the effective depth at this critical 
point is much greater than would be possible with the systems 
in which four or more layers are used. 



' ^x/>e/9SomeAfr 




1\ 



1\ 




Z'/4' 



/fk' 



£'/■*' 



^ 1 



A3 


^ 








^~ 




■"*' 




"~" 




"" 




~~' 


""" 


~~ 


"*" 




"~~ 




~~" 








■"" 


"~* 




71 




"^ 


~~ 


At 


















































f 


/ 


% 




} 






















































/ 












/./ 
















































/ 




























































/ 
















5 AO 












































/ 


























































( 


/* 




















X" 








































/ 


























































/ 
























\a6 




































/ 


























































/ 




























\^7 
































/ 
























































4 


/ 
































\a6 


























J 


/ 


































\a.6 
























J 


r 
























































/ 


r 
























































> 


W 








































^04 


















y 


r^ 












































' 














> 


r- 












































\o.5 














/ 




























































/ 


















































^AZ 








1 


;/ 


'■w 


























































^ 






















































O.A 






y 




























































/ 




























































z 












































1 














i 


> 








5 






/ 


O 






A 


5 






/ 


*o 






i 


a 






J 


\o 






J 


^ 





FIQ. 19. — MOMENT CALIBRATION CURVE FOR EXTENSOMETER. 

The ideal reinforcement would be such that its area at any 
point be directly proportional to the ordinate of the moment 
surface shown in Fig. 17, but such reinforcement is obviously 
impossible in practice. The panel is, therefore, assumed to be 
divided into ten strips or imaginary beams, five parallel to each 



Digitized by 



Google 



Tbelease on Design of Concrete Flat Slabs. 239 

side, and such moments assumed for their design as will give 
reinforcement areas conforming most closely to the theoretical 
values. 

The widths of these imaginary beams together with the 
moment factors applying to each are shown in Fig. 22. ^ To 
avoid confusion in the drawing, the widths and moments for but 




FIG. 20. — ^MOMENTS PER UNIT WU>TH DERIVED FROM EXPERIlfENTB. 

one set of strips are shown, those for the other set being the same 
but turned through ninety degrees. The numbers in the circles 
on the diagrams are moment coefficients entering into the denomi- 
nator of the moment equation. Thus, the moment at the center 
of the panel in the middle strip or beam is Af = i6JjSV40, where 
w is the total unit live and dead load on the panel in lb. per sq. ft., 
« is the clear span in feet, and the 40 is the moment coefficient 



Digitized by 



Google 



240 Trelease on Design op Concrete Flat Slabs. 




H 



p 



o 



Digitized by 



Google 



Trelease on Design op Concrete Flat Slabs 241 

shown in the circle at this point. M is the bending moment in 
ft.-lb. per ft. of width, and is positive, requiring reinforcement 
in the bottom of the slab, as indicated by the fact that the arrow 
showing the direction of the reinforcement is dashed. The 
moments of other points of the panel are obtained in the same 
manner. The sections, at the bottom of Fig. 22 show how the 
theoretical distribution curves are replaced by more practical 
stepped lines. In making this change an excess of reinforcement 
has been introduced at the center of the strips to provide for 
single panel concentrations of loading. 

Since the maximum stresses in a flat slab occur at. the col- 
umns, the use of a slab of uniform thickness is wasteful of concrete 
and adds mmecessary dead load. The imnecessary concrete 
can be done away with, decreasing the weight and cost of the 
structure and at the same time adding to its efficiency, if a rec- 
tangular cap of concrete is left over the columns, extending to 
about the fifth point of the span. 

The general method of procedure in designing a concrete 
flat slab in accordance with this method is to first find the areas 
of reinforcement required in the different bands at the center 
line of the panel. These areas are supplied by rods of the proper 
size and spacing, every other one of which is bent up in the top 
at about the quarter point and extends over into the adjacent 
panels, while the remainder are straight and extend about 6 in. 
past the edge of the panel. This arrangement gives the necessary 
amounts of reinforcement in both the top and bottom of the slab, 
except in the areas over the columns where the necessary extra 
reinforcement in the top is supplied in the form of short, straight 
bars, remembering that at these points the effective depth of the 
slab is increased. To avoid bimching the bars in the top of the 
dab, a bent bar in one panel should be in line with a straight 
bar from the next panel. 

The line of inflection may be safely taken at the fifth point 
of the clear span, or at the quarter point of the center to center 
spacing of the columns, using the higher value in every case. The 
point of bend in the bent bars, and also the length of the extra 
short rods in the top, will be governed by this dimension. 

In the end panels the area of the reinforcement perpendicular 
to the wall should be increased by the usual 20 per cent. The 



Digitized by 



Google 



242 Trelease on Design of Concrete Flat Slabs. 



A 



-^)-r-^ 



■-^-^ 



-P^ 



V^ 



tl 



^^ 






~f ^ 



^ 



-^o 



-^H-^ 



■^-^ 



4 



■^v 



"Ol 



1 



^ ^y 



D/jfr/Jbt/f/o/7 across l//?e XX 



\ 



\ 



/o 



T 
_L 



N. 



:^ 



/ 



7^ 



7^ 



^/s^r//>u//o/? across Z//?e Y-y. 

FIQ. 22. — ^MOMENTS PER UNIT WIDTH USED IN DESIGN. 



Digitized by 



Google 



Trelease on Design op Concrete Flat Slabs. 243 

easiest way to do this is to use round bars in the interior panels, 
and square bars of the same nominal size and spacing in the end 
panels. In order to facilitate the placing of the reinforcement, it 
will be found necessary to provide light bars, preferably | in. 
round, along the sides of the panels, so that they may be used to 
support the bent bars. Of course, these spacing bars may be 
figured as effective in tension over the top of the colunm, and thus 
replace some of the short bars used there. 

Table II gives recommended dimensions for flat slab floors 
to suit various live loads and panel sizes. The table is based on 
a maximimi theoretical concrete stress of 750 lb. per sq. in. The 
slab thicknesses are such that the deflections under a superimposed 
test load equal to once the dead load plus twice the live load 
will not exceed one five-hundredth of the span for a theoretical 
working stress in the reinforcement of 18,000 lb. per sq. in. Ten- 
sion in the concrete, arch action, etc., all tend to reduce the actual 
stresses below the theoretical values, but no building code will 
permit these factors to enter into the design of concrete structures. 

Flat slab designs showing smaller column head diameters, 
or thinner slabs over the supports, than those given in the table 
should be carefully checked for shear at the edge of the column 
heads. 

To illustrate the application of this new method for the 
computation of flat slabs, the detailed design of a typical panel 
will be given. To enable comparison of the results with those 
obtained by other methods described and tabulated by Mr. Angus 
B. McMillan in a paper* before the Association, a panel 20 ft. 
square will be designed for a live load of 200 lb. per sq. ft., using a 
steel stress of 16,000 lb. per sq. in. and a ratio of the moduli of 
steel and concrete of 15. 

In Table II under the above span and loading a slab thickness 
of 8 in. is given, with a 2 in. cap 8 ft. square, and a column head 
50 in. in diameter. The average weight of the slab, including the 
cap, is 104 lb. per sq. ft. 

The clear span from edge to edge of colimMi heads is 15.83 ft., 
and the "gross bending moment" or wS^ is 12X304X15.83^ or 
914,150 in.-lb. per ft. of width. Referring to Fig. 22, it will be 



* Proeudingg, Vol. VI, p. 248.— Ed. 



Digitized by 



Google 



244 Trelease on Design op Concrete Flat Slabs. 



z 

< 



•J 
o 
O 



^a 



Q.9 



2:c;S5S?5!5?35S$gSS 



^^^SiftSssSSSh 






P 
O 

< 

z 
o 

M 



i 






^^ 



oO'^NMrteO'^So© 



0.S ^$$ss;sssss3 



r^ d r*oooooaoaOO-^»^NN 



^1 



g§822g^g§§5l$ 



9^^^^iOt/3U 



K.S 



O9<-i<-i«-4C0C0^^i4C0 



^ d t*t^ooooooaao>ooo^ 
^^ i>xoo>ooO'-4f-ioeo 



Q.S 


S?§^S5SS25S2§§ 


tt.S 


aoooftaaoo^NNM 




(O (O t^ b- tt'oo 00 00 oa o o 


Q.S 


SSS$3$SISSSS 



^ ti aoooxooo)a>oco<->'^ 





Hi 


«0 «0 (O O t<«tt'oo 00 00 oa 9 




^^ 


^^g^^rs^sssss 


5 


Q.S 


SSg^^gS^fSS^^ 


"* 




t<.b-i^i^t<«t«r«xooa)0 




•OiAiOiOiOiO>O«D«DIN00 



•-^d 



ocooacooaoeDo»coa»o I 
(O(0(0t*t*ooooooasaso 



Digitized by 



Google 



Trblease on Design op Concrete Flat Slabs. 245 

seen that the net positive moment for the central strip of the slab 
is 1/40 of the gross moment, or 22,850 in.-lb. per ft. 

To enable the comparison with Mr. McMillan's table, the 
efifective depth of the slab will be taken as 7 in., so that the area 
of reinforcement per foot width of this strip — 

^«= - — J?^-^^, or 0.237 sq. in. 
7 X. 86 X 16,000' ^ 

This will be supplied by using |-in. corrugated round bars 
spaced 15 in. on centers. The negative moment for this strip is 
of the same magnitude as the positive, so that it can be properly 
taken care of by bending up every other bar in this strip and in 
the corresponding strips of the adjacent panels. 

At the center of the next strip, there is a positive moment 
equal to 1/30 of the gross moment, or 30,470 in.-lb., requiring an 
area of reinforcement per foot — ^il3o = 0.316 or f-in. roimds, 11 J in. 
on centers. 

Over the edge of the panel the negative moment for this 
strip is 1/15 of the gross moment, or 60,940 in.-lb. — 

An^ ^^^^ or 0.492 sq. in. p3r ft. 

9X.86X 16,000' ^ ^ 

If every other bar from the bottom of thi?, band is bent up, 
there are 0.316 sq. in. per ft. and the rema'nder, 0.172, can be 
supplied in the form of short, straight bars, using f-in. rounds 
spaced 21 in. on centers. 

For the strip or band at the edge of the panel the net positive 
moment is 1/20 of the gross moment, or 45,710 in.-lb., calling for 
ilM = 0.474 sq. in. per ft., or f-in. rounds spaced 7.5 in. on centers. 
At the ends of this strip over the column head the net negative 
moment is 1/10 of the gross moment, or 91,415 in.-lb., requiring 
-4io = 0.738 sq. in. per ft. The bent bars will supply 0.474 sq. 
in. per ft., so that the remaining 0.264 sq. in. will be taken care of 
by short straight bars, using f in. rounds spaced 13.5 in. on centers. 

The length of the short straight bars will be such that they 
will reach the fifth point of the clear span, so that they will be 
10 ft. 6 in. long for this panel. 

Having determined the size and spacing of the bars they can 
be laid out on a drawing, or the number required estimated in 
the following manner: 



Digitized by 



Google 



246 Trelease on Design of Concrete Flat Slabs. 

The balanced average of the moment factors in Fig. 22 for the 
positive moments at the center of the panel is 30, so the net moment 
is 30,470 m.-lb. per ft., or m the 20 ft. width of panel, 609,400 
in-lb. To resist this moment will require 6.33 sq. in. of reinforce- 
ment, or twenty-one f in. romid bars. The average area of the 
short top bars required is 0.218 sq. in. per ft., and the total width 
in which they must be supplied is 10 ft., so that 2.18 sq. in., or say 
seven f-in. round bars are required. 

This gives the total number of bars in one direction; in the 
whole panel there are 21 straight bars 21 ft. long, 21 bent bars 
31 ft. long, and 14 straight bars 10 ft. 6 in. long, all f-in. round, or a 
total of 1300 lb. of reinforcement per panel. Table III gives a 



Table III. — Quantity of Reinforcement Required Per Panel Under 

Various Designs with a Unit Reinforcement Stress of 

16,000 Lb. per Sq. In. 



Method. 


Thickneae 
of Slab, in. 


Pounds of 
Reinforcement. 


Ca&tQeyar 


8 

12 
8 
8 
8 
8 

V 


2,189 


TiimfMiira and Mauror. 


i;931 


Grashof 


784 


Mensch 


2,120 


Turner. 


649 


McMillan 


1,084 


Brayton. 


1.900 


TrelmlM) r r . . r r 


1,300 







comparison of the quantity of reinforcement required under various 
designs. 

Fig. 23 shows the arrangement of the reinforcement in this 
t3^ical panel. To avoid confusion in the drawing, the bars from 
the adjoining panels are not shown. In designing this slab, it 
has been assumed that the slab is fixed at the edge of the column 
head, as it is the general practice to flare the head at a small angle, 
making it very rigid. The head is not counted on to resist bending 
moment, and is flared merely to take care of the shear and to 
somewhat reduce the clear span. 

To compute the deflection of the typical panel designed, the 
deflection of a strip of unit width at the side of the panel will be 
computed and also of a unit strip at the middle of the panel, and 
the total deflection of the plate taken as the sum of these. It will 



Digitized by 



Google 



Treleasb on Design op Concrete Flat Slabs. 247 

first be assumed that these strips are freely supported, and the 
deflections transformed later by the ratio of the deflection coeflS- 
cients given in Fig. 12 to those for simply supported beams. The 



4-1 



;:i 



*=:'"!:=!!;SSSSSSSE5=!!==:=-!f! 




FIQ. 23. — ARRANGEMENT OF REINFORCEMENT IN TTPICAL PANEL. 

computations may be very quickly made by using the graphical 
solution of Mr. White's method published in Dodgers Diagrams. 
In this way the deflection at the working reinforcement 



Digitized by 



Google 



248 Trelease on Design op Concrete Flat Slabs. 

stress of 16,000 lb. per sq. in. is found to be 0.15 in., or 1/1600 of 
the panel side. For higher stresses the deflection would, of course, 
be greater. 

The unit stresses which have been employed for this panel 
are very conservative, and could be safely increased to 18,000 lb. 
on the reinforcement and 750 lb. on the concrete. In this particu- 
lar panel the maximum concrete stress is at the colunm head and 
is only 600 lb. per sq. in., while at the center of the side band 
eth concrete stress is about 550 lb. With the 50-in. column head 
used the shear over jd is about 100 lb. per sq. in. and as this is 




MG. 24. — TEST LOAD ON PANEL IN ST. LOUIS. 

pimching shear, such as in footings, a higher value could be used 
if desired. 

Floors designed in accordance with the methods outlined in 
this paper have been built in several cities, and a few have been 
thoroughly tested. In these tests, both reinforcement and concrete 
stresses have been measured by extensometers. The floors tested 
were designed for reinforcement stresses of 20,000 lb. per sq. in. 
with the exception of the building in Minneapolis, where 18,000 
lb. was used. These theoretical stresses were, of course, not 
realized in the tests as tension in the concrete, arch action, etc., 
were not taken into account in the designs. 

Fig. 24 shows a panel in a factory in St. Louis under a test 



Digitized by 



Google 



Trelease on Design op Concrete Flat Slabs. 249 

load of 400 lb. per sq. ft. The panel was 20 ft. i in. by 22 ft., 
designed for a safe live load of 150 lb., with an 8-in. slab. The slab 
was inverted in this case, giving a perfectly flat ceiling. Two inch 
caps were to be cast on the top and hidden in the cinder fill, and 
the drawings showed the bars bent up into them. As actually 
constructed, the reinforcement in this panel had an effective 
depth over the column, the critical point of the structure, of but 
60 per cent of that shown on the drawings. Even under these 




FIG. 25. — ADDITION TO THE FORD MOTOR COMPANY'S FACTORY, DETROIT, MICH. 

adverse conditions, the maximmn reinforcement stress over the 
column head under the full test load was but 17,000 lb., and the 
concrete stress at this point 950 lb. The reduction of the effective 
depth over the column caused a large reduction of the moment of 
inertia at that point, giving conditions approaching those in a 
freely supported panel, as the stresses and deflections clearly 
indicate. The deflection increased from the normal amount for a 
fixed panel to 0.68 in. imder full test load, somewhat less than the 
normal amount for a free span. The average reinforcement stress 



Digitized by 



Google 



250 Trelease on Design of Concrete Flat Slabs. 

at the middle of the panel was also increased in like manner to 
25,000 lb. per sq. in. 

Perhaps the most severe test of a floor was one made in 
Minneapolis in which four panels of a warehouse floor were loaded. * 
The building as laid out was not well adapted to flat slab floor 
construction, because of the number and shape of the panels. 
There was one row of colunms through the building, dividing the 
floor into panels 13 ft. 6 in. by 19 ft. 11 in.; with the short side 
resting on brick bearing walls. An 8-in. slab was used for the 
200 lb. live load required. Four panels of the floor were loaded 
with cement piled to prevent arching, and very complete readings 
of stresses and deflections were made. Under the maximum 
superimposed load of 400 lb. per sq. ft. the deflection at the middle 
of the panels averaged 0.39 in., a very good figure remembering 
that the panels were free at one end. The reinforcement stress 
over the column head was 23,000 lb., and at the mid-point of the 
span 13,500 lb. The concrete stress had a maximum value at 
the column head of 1050 lb. per sq. in. These stresses show the 
design to be conservative. 

Fig; 25 is a view of a factory in Detroit having floors and 
roof of this type of flat slab. The panels are 25 x 20 ft. and the 
floors were designed for 150 lb. per sq. ft., using an 8-in. slab, 
2-in. cap and 48-in. colunm head. One panel of the floor was 
loaded with gravel to 300 lb. per aq. ft. Under this load, the 
deflection was found to be 0.60 in. The reinforcement stress 
over the column head was 13,500 lb. per sq. in. and at the center 
of the slab 13,850 lb., showing a well-balanced design. 

These tests seem to show quite conclusively that the method 
of design outlined can be used for almost any condition of span 
and loading with confidence as to the resulting strength of the 
structure. 



* For complete data of test see Report of Committee on Reinforoed Conorete and Buildiag 
Laws. p. IOS.—Ed. 



Digitized by 



Google 



DISCUSSION. 



Mr. Alfred E. Lindau. — ^The object of the test was to gain Mr. Lindaa. 
some knowledge of the action of an elastic slab, supported on 
points, under a uniform load. At the time the tests were under- 
taken it was not expected to obtain a great deal of information 
that could be used in actual designing of concrete structures, but 
rather some idea of the nature of the deformations; by plotting 
the deformations obtain the deflection of the surface and perhaps 
determine points of inflection, rate of change of curvature at the 
various points and perhaps some idea of the relative deformation 
for various points of the slab. 

Many suggestions were made as to the material to be used 
in making this preliminary investigation. It was suggested that 
a steel plate be polished and by some mirror apparatus obtain 
those deformations. Hard rubber was also suggested, plaster 
plates were tried as well as a cement coated wire screen — ^all of 
which gave some valuable information but did not give the 
information desired and the results would not have been of any 
particular interest. 

Mr. Arthur N. Talbot. — ^I have been very much interested Mr. T«iboc 
in the experiments made by Mr. Trelease. They show the dis- 
tribution of bending moment in a homogeneous slab of rubber in 
a way which could not be obtained by mere computation. The 
negative moments at and between the supports, the relative 
values of the positive bending moments at the center of the 
panels are well brought out and there are many features of the 
results which will be helpful in the application to design problems. 
Of course, we have here a material that resists tension all through 
it, as well as compression and one having a modulus of elasticity 
much the same in tension as in compression. In the case of 
ordinary flat slab construction the concrete is in compression on 
one side and has a reinforcement of steel on the other to take 
most of the tensile stresses. The conditions in the test must be 
kept in mind because they are different from those of reinforced 
concrete. Rubber is a material with a very low modulus of 

(251) 



Digitized by 



Google 



262 Discussion on Concrete Flat Slabs. 

Mr.T«ibot elasticity, a material that even under small deflections would 
bring in the question of catenary action, rather more than beam 
or slab resistance. 

I wish to ask whether in the experiments an eflfort was made 
to determine the eflfect on the elasticity or on the deformations 
occurring when the rubber was stretched or compressed in two 
directions. This has a bearing on slab action. 

Mr. Lindaa. Mr. Lindau. — There was an attempt made to obtain such 

results, but some difficulty was encountered in getting satisfac- 
tory compression tests on the strips used. The results on the 
lateral deformation in tension were quite satisfactory and seemed 
to give very uniform results, but the compressive test did not 
show up quite so well. The rubber was only stretched in one 
direction. The question of obtaining lateral deformation was 
discussed at considerable length but tests were not carried out, 
largely on account of lack of time. 

Mr. Talbot. Mr. Talbot. — The matter of compressive strength is one of 

considerable importance in flat slab construction, for with a 
rather small percentage of reinforcement in each of four direc- 
tions, or even two directions, the combined calculated stress may 
nm high, unless reinforcement in compression is used, which to 
my mind is not a very satisfactory arrangement with this type of 
construction. It may be expected that when concrete is sub- 
jected to compressive stresses in two directions, its resisting 
strength will not be the same as when the pressure is applied in 
only one direction. It is true — ^and this has a bearing upon the 
distribution of stresses between the reinforcement and concrete 
in such construction — that the amoimt of compressive deforma- 
tion resulting from applying a given load is less when the load is 
applied in two directions than when it is applied in one; that 
seems to be the result of some experiments which we have made 
and which agree with tests made elsewhere. Specimens were 
made up in the form of a cross and a load applied in one direction 
and also in two directions and measurements taken on the con- 
crete in the cross portion and in the arms of the cross; and, 
making a comparison, there seems to be less shortening in the 
concrete where it is stressed in two directions, and of course a 
greater expansion in the direction at right angles than when the 
compressive stress is applied in only one direction. 



Digitized by 



Google 



Discussion on Concrete Flat Slabs. 253 

We have attempted to make tests of cubes by appljdng com- ^r. T«ibot. 
pression against four faces instead of two. The first tests showed 
considerably greater strength for loads applied in two directions 
than in one, but we felt that a large part of this difference was due 
to the restraint of the loading faces themselves, the friction along 
the face, which kept the concrete from expanding laterally. We 
then made other tests by lubricating the faces of the bearing 
blocks very carefully and the values for compression in two 
directions were approximately the same as those in one. 

I may add that measurements were made on cantilever slabs, 
slabs composed of what you may think of as the column capital 
and the part of the floor slab out to the line of inflection, support- 
ing them at this line of inflection, just as the load of the remain- 
der of the panel would be suspended in an analysis of the flat 
slab — omitting, you will see, the uniform load for the central 
portion immediately surrounding the column — ^measurements of 
the deformations of the reinforcement were made. We find that 
in the middle line of the band the maximum stress is the greatest 
at the edge of the column capital; that the stress decreases pretty 
regularly from there to the edge of this cantilever slab, the line 
of inflection; that through the middle portion immediately over 
the column capital it decreases rapidly, although it does not 
become zero with the ordinary size of bars even directly over the 
center of the column; that along the edges of the bands the 
maximum stress in the rods is somewhat further in, nearer the 
center of the length of the reinforcing bars. The maximum 
stress in the bar at the middle of the band does not differ far from 
the maximum stress in the bar at the edge of the band, even 
though its position differs considerably. 



Digitized by 



Google 



THE PRACTICAL 
DESIGN OF REINFORCED CONCRETE FLAT SLABS. 

By Sanford E. Thompson.* 

The purpose of this paper is to present material covering 
the practical task of designing flat slab floors for reinforced con- 
crete structures. The requisite thickness of slab, amount of 
reinforcement and size of column head for different loadings and 
different spans are given in Tables I-IV; and the theories and 
assumptions involved in the computation are briefly discussed. 
Values not included in the tables may be worked out from the 
formula, finding the desired values of C^ and C, from the dia- 
grams.! Curves are given also for the constants used in the 
design of members with reinforcement in top and bottom, and 
apply not only to flat slabs, but to any beam or slab reinforced 
both in compression and tension. 

For reinforced concrete buildings, the flat slab, or girderless 
floor, — ^as it is sometimes called, — ^is as cheap, and frequently 
cheaper, than beam and girder construction. The smooth ceil- 
ings with no intersecting beams allow better distribution of the 
light. The expense and complication of installing sprinkler 
systems are reduced. The clear headroom for the same story 
height is increased, or else, on the other hand, the story height 
may be made less without reducing the effective headroom. This 
last consideration alone is often important enough to dictate flat 
slab floors. 

With flat slab floors the entire load is supported directly on 
the columns, which are usually spaced about equally in both 
directions. The column heads are enlarged so as to give increased 
resistance in shear and bending at the points where this is most 
needed. The reinforcing bars run through the slabs over the 
colunm heads in four directions, two rectangular and two diagonal. 

The simplest way of considering the flat slab is to assmne 

* Consulting Engineer, Newton Highlands, Mass. 

t For an example of flat slab design worked out in detail see Taylor and Thompson's 
''Concrete, Plain and Reinforced." 2d edition, 1011. pages 487 and 488. 

(254) 

Digitized by VjOOQIC 



Thompson on Design of Concrete Flat Slabs. 255 

that a portion of the slab extending a certain distance out from 
the column is a flat, circular plate, similar to a Japanese parasol, 
but with no slope to its surface. This plate is fixed to the colunm 
and is assumed to extend out from it on all sides like a canti- 
lever as far as the line of inflection of the slab, which line, — as 
in other forms of monolithic construction, — ^is about one-fifth of* 
the net span away from the support. The rest of the slab may be 
considered as entirely separate from the flat circular plates but 
simply supported from their outer edges or circumferences. 

This is no new theory but is somewhat similar in effect to 
that of a imiformly loaded, fixed or continuous beam. To illustrate 
this in practical fashion, an ordinary beam uniformly loaded 
and fixed at both ends will be considered. This illustration does 
not in any way show the methods of determining a bending 
moment in the flat slab, since, as stated below, the actual bending 
moment is dependent upon the elastic tjieory. It does, however, 
show quite clearly the justification of assuming the slab to be cut 
through on the line of inflection. 

It is known from simple mechanics that the moment at the 
support of an ordinary fixed or uniformly loaded continuous beam 
is Wl/12* and, at the center, is TFZ/24. Now, suppose at the 
points of inflection, which also by mechanics are known to be 
located at a distance 0.21 13Z from each support, the beam is cut 
completely through so as to have a cantilever at each end with 
a simply supported beam between. The bending moment of 
the cantilever at its support, due to the load upon it, is 0.2113 
TrX0.2113Z/2, and the moment at its support due to the load on 
the supported beam between cantilevers, is [1—2 (0.2113)]2 
WX0.2n3l. The sum of these two moments is 0.0223 
m+0.0610 FZ= 0.0833 Wl or Wl/12. In other words, while 
this analysis is not that which can be used for a flat slab, because 
of the extra strength of the flat slab due to the multiple reinforce- 
ment, the division into sections corresponds to our assumption 
in the flat slab theory. In the same way it might be shown that 
the center moment of the simple beam supported by the two 
ordinary cantilever beams is Wl/24. 

Tests of the flat slab construction at Minneapolisf indicate 

* W *■ total live plus dead load. I >" distance in feet between supports. 
t See paper on "A Test of a Flat Slab Floor in a Reinforced Concrete Building, " by Arthur 
R. Lord, ProeeeditHf, VoL VII. page 166. 



Digitized by 



Google 



256 Thompson on Design of Conckbtb Flat Slabs. 

that the line of inflection of a flat slab floor is substantially the 
same as in a fixed beam, or about ^/s the net distance between sup- 
ports, although, as would be expected, the bending moment is 
entirely diflferent. 

Problem of Design. 

The problem of the design of the flat slab, then, resolves it- 
self into (1) a determination of the proper thickness and reinforce- 
ment required at the support for the cantilever circular plate 
supporting its own load and also the load of the rest of the slab, 
and (2) a determination of the thickness and reinforcement at the 
center of the span required for the simply supported section lying 
between the circular plates. 

Various Methods of Design of Slab. 

Various methods have been advanced for the design of the 
flat slab. Some are based merely on deflection tests, which give no 
true basis for computations; others compute the reinforcement 
carefully at the center of the slab, which is not the critical part; 
others consider the construction to consist of beams between 
columns with a slab between, thus obtaining ultra-conservative 
results; while a plan still more common is to take the moment 
at the supports arbitrarily without regard to the size of the col- 
umn head. The shear or diagonal tension near the column head 
is frequently disregarded altogether. 

Shear at the Support. 

The direct shear at the support, as in any mechanical con- 
struction, is equivalent to the total load supported by the column. 
This shear is readily borne by the concrete and reinforcement. 
The diagonal tension, however, which, as in a beam, may be 
considered as measured by direct shear, must be carefully con- 
sidered. To reduce the diagonal tension and also to increase 
the resistance to bending of the slab, the column head is enlarged. 
To still further increase the resistance, a part of the bars in the 
top of the slab over the supports may be bent down just outside 
of the supports and then carried along in the bottom of the slab. 

In either case, the shearing stress should be limited to definite 



Digitized by 



Google 



Thompson on Design op Concrete Flat Slabs. 267 

units, although it seems permissible to use a somewhat higher 
stress than in a beam. 

The diameter of the enlarged column head, which is the actual 
support of the slab, should be governed by the shearing stress 
either at its circumference or at a short distance outside of it. 

Bending Moment at Support. 

The theory of flat plates, which must be used in designing a 
circular plate, is not yet clearly established. By the use of what 




rr-rXL M Ada'^med iine of I f/ 

II 'n* — 'f^iAy mctxJmum bending i (( 

\\ \\ / 7 .' / moment . \ 

t r^- — ^ ^f ^Theoretical t!ne \ 

\\ Column head^/y of inflection \ 
\ >v . / / ^Assumed iine v 

^^>-r-j— j y<-^^ ^ of inflection ^^_ 

FIG. 1. — PLAN OF FLAT SLAB. 

is termed, in mechanics, the elastic theory, we have a fairly good 
working hypothesis. The analysis solved by Prof. H. T. Eddy* 
offers, in the writer's judgment, the most rational solution of the 
problem yet advanced. 

In the design of the flat slab, therefore, the authorf has started 
with Prof. Eddy's analysis of stresses in a homogeneous circular 
plate, and from his general formulas has deduced by mathematics 

* EDgineers' Society, University of Michigan, 1899. 

t The author i« indebted to Mr. Edward Smuloki for the oomputationB involving intricate 
analyses by higher mathematics; also to Mr. John Ayer for further studies in the practical 
design. 



Digitized by 



Google 



258 Thompson on Design of CoNCRfiTfi Flat Sla^s. 

other formulas applying to circular plates free on their edges and 
clamped around the columns. In a flat slab thus supported there 
are horizontal stresses at right angles to each other. The eflfect of 
these lateral stresses has been taken into account, this being ex- 
pressed by Poisson's ratio, which is the ratio of the lateral deforma- 
tion to the deformation in the direction of the stress. The value 
of this ratio is taken as 0.1, which has been shown by experiments 
to be a fair value for concrete of 1 : 2 : 4 proportions. 

It has been found possible to reduce the complicated formulas 
derived by the Eddy analysis into four formulas which are com- 
paratively simple although still rather complicated for practical 
use. These formulas are for four bending moments and can be 



2t 



u^ 



1%0 In pouftds ptraifuare foot 
1 1 1 I 1 1 iiiiiil li i 1 11 



il 





^Isi^ 



r$etangular itett in 
two tayera 



q In pounds par linear foot 

no. 2. — 8EcnoN of flat slab. 



applied not merely to the slab at the support, but to any point in 
the circular plate surrounding the column. The four moments 
are as follows: 

Ml = moment produced by the loading that is uniformly distrib- 
uted over the circular plate and causes circumferential 
fibre stress. 

Jlf 2— moment produced by this same loading but which causes 
radial fibre stress. . 

Mo = moment produced by the loading from the rest of the slab 
that is distributed along the outer edge of plate and 
causes circumferential fibre stress. 

Mb « moment produced by the latter loading but which causes 
radial fibre stress. 



Digitized by 



Google 



Thompson on Design of Concrete Flat Slabs. 259 

A study of the analysis, however, shows that the two cir- 
cumferential moments are a minimum at the support and may be 
safely disregarded. The two formulas for the radial moment may 
be combined and still further reduced to the following simple 
form which can be used for a circle of any radius, r, within the 
circular plate. The meaning of the symbols is made clearer by 
reference to Figs. 1 and 2, which show the plan and the section 
of a flat slab. 

Let 
q = uniformly distributed load around the outer edge of the 

plate in lb. per ft. of length. 
w « uniformly distributed load on surface of plate (including 

dead load) in lb. per sq. ft. 
r« = radius in feet to line of maximum bending moment (which 

is within the column head). 
fi = outer radius of assumed plate in feet, 
r = any radius in feet where moment is to be computed; for 

critical section, r is radius of column head. 
Ci, Ce = constants given in Figs. 3 and 4. 
Mr s total radial bending moment to be used ordinarily. 
h » distance in feet between lines of inflection. 

Then total radial moment at any point of plate is 
Mr=wrlCi + qr^^ 

For convenience in computation, values of the constants 

T T 

C5 and C„ for various values of the ratios -^ and - are plotted 

in the curves given in Figs. 3 and 4.* 

With q expressed in lb. per ft. of length, w in lb. per sq. ft., 
and To in ft., the moments are in ft.-lb. per ft. or in.-lb. per in. 

Position op Maximum Bending Moment and op 
Maximum Stress. 

As commonly constructed, the column head flares at the 
top and is therefore more or less flexible. For this reason the 
line of maximum bending moment will be located, not at the 

* TheM Are drawn up from values in tables in Taylor and Thompson's ** Concrete, Plain 
•Bd Rsiaforoed." Sd edition. 1911, pate 618. 



Digitized by 



Google 



260 Thompson on Design op Concrete Flat Slabs. 




1.0 1.1 La 13 M- 1.5 1.6 1.7 18 L9 2X) 
Values of£ 

FIQ. 3. — ^DIAGRAM GIVING VALUES OW Cf IN VORMULA 

Mr- w To* C^ + qr^ C$ 
foi- radius in feet to line of maximum bending moment. 
n « outer radius of assumed plate in_feet. 

r—any radius in feet where moment ia to be computed for actual aeotkni 
ordinarily r is radius of column head. 



Digitized by 



Google 



Thompson on Design op Concrete Flat Slabs. 261 




1.0 If . L2, 13 lA L5 I.B IT L8 1.3 iO 
Valuer of-| 

FIG. 4. — DLIGRAII GIVING VALUES OF Cg IN FORMULA 
Mr^W To* Ct + q fo C# 

r«" radius in feet to line of maximum bending moment. 

n » outer radius of assumed plate in feet. 

r""any radius in feet where moment is to be computed for actual section , 

ordinarily r is radius of column head. ^^ j 

uigiiized by VjOOQIC 



262 Thompson on Design of Concrete Flat Slabs. 

extreme edge of the column head, but a little within it. The 
maximum stress, on the other hand, will not be on the line of 
the maximum bending moment because the strength there (since 
it is within the head) is increased due to the greater depth of 
concrete. It is fair to assume, therefore, that the maximum 
stress is at the edge of the column head, and we may assume 
the "critical section" as on this line. The exact location of the 
liile of maximum moment is indeterminate. Under ordinary 
conditions it appears fair to assume its location as within the col- 
umn head, a distance equal to the thickness of the slab. There- 
fore, Mf is figured for a value of r^Vo+L In figuring this mo- 
ment, values of the constants Cg and C« should be taken from the 
curves in Figs. 3 and 4. As in an ordinary fixed beam, this bend- 
ing moment is negative, so that the upper side of the slab is in 
tension and the lower in compression. Having found the moment, 
the design of the reinforcement and the thickness of the slab may 
be worked out as for an ordinary beam. 

The curves in Figs. 5 to 8 inclusive will be found of assistance 
in working out the design. 

Reinforcement in Column Head.* 

The slab at the column head might be designed with the 
reinforcement all in the top of the slab running in four directions 
provided the slab is thick enough so that the concrete will not be 
overstressed in compression. In order to reduce the thickness of 
the slab and therefore save the additional cost and weight of con- 
crete over the entire floor, it is economical to place reinforcement 
in the bottom of the slab as well as the top, and figure it as assist- 
ing the concrete to take compression. Since a portion of the 
bars need to extend only far enough beyond the column head to 
furnish suitable bond, the cost of this additional reinforcement 
will be much less than the cost of an additional thickness of con- 
crete over the entire slab. The tensile reinforcement must not 
sag over the column head. 

To make it easy to place the concrete and also to bring the 
center of gravity of the reinforcement as near to the surfaces of 
the slab as possible in order to give the longest moment arm and 

* Certain features of flat slab reinforcement are covered by patents of C. A. P. Turner. 



Digitized by 



Google 



Thompson on Design of Concrete Flat Slabs. 263 

thus a thinner slab, two layers of reinforcement may be placed 
in the top of the slab and two layers in the bottom. The relation 
of the quantity in the top and bottom must be determined by the 
design. If a thin slab is desired, even more reinforcement may 
be placed in the bottom than the top. In the tables, three ratios 
of reinforcement are given and the percentages selected are those 
that will give the required working stresses in the concrete and 
the reinforcement. 

The Minneapolis test already referred to shows that not 
only the remforcement directly over the column head, but the 
reinforcement for a considerable distance each side takes tension. 
In view of this test and of the tests made at the University of 
Illinois* it is safe to assume that the reinforcement may be spaced 
over a distance at least equal to the diameter of the column head 
plus three times the thickness of the slab. 

The determination as to whether the diagonal or rectangular 
reinforcement should be placed at the top is governed by the 
relative quantities of each. More reinforcement is required for 
the diagonal direction through the slab, hence the layers which 
are largest in section may be run diagonally. 

Agreement with Minneapolis Tests. 

By our theory it is possible to compute the stresses not only 
next to the column head but at any point in the slab. In several 
cases, knowing the exact location of the points where the deform- 
ations were measured in the Minneapolis tests the stresses at 
these points have been computed. Using 5.6 in. as the moment 
arm, and including the radial bars as assisting to take tension, 
the maximum stress in the reinforcement over the edge of the 
column is 26,000 lb. per sq. in. under the normal load of 225 lb. 
per sq. ft. as compared with 20,700 lb. per sq. in. given by Mr. 
Lord as the actual maximum stress in the floor. This is no greater 
diflference than there ought to be between design and test and 
shows our method to be slightly more conservative than the 
actual test. 

The compression in the concrete is more difficult to check 
since the exact locations of the test points are not given. Com- 

♦ See paper on *' A Test of a Flat Slab Floor in a Reinforced Concrete Building." by Arthur, 
R. Lord, Proceedingt, Vol. VII. page 182. 



Digitized by 



Google 



264 Thompson on Design of Concrete Flat Slabs. 

putations, however, show unquestionably that our methods are 
conservative enough to allow for the irregularities in concrete 
mixtures, and the danger of not having perfect concrete at the 
critical section. 

Moment at Center of Slab. 

It is possible to adapt the Eddy theory to the design of the 
center of the slab as well as to the supports. In practical design, 
however, as has been indicated, the thickness of the slab is deter- 
mined by the thickness at the support, which is always the greater. 
But, in order to avoid too wide spacing of the bars and to adapt 
the center reinforcement to that over the supports, more bars 
are generally run through the slab than the results of tests would 
show to be necessary. Consequently, instead of considering this 
from a theoretical standpoint alone, safe values for the bending 
moments may be selected, based on general principles of mechanics 
and qualified by actual tests. 

Let Ii« distance between lines of inflection. This distance 
will be about 7 of the net span between column heads. 

For the rectangular reinforcement, if the slabs between the 
points of inflection were simply supported, we should have a 
moment of wli/8. However, the bending moment in the Minne- 
apolis tests, based on the maximum stresses under imif orm work- 
ing load, is about wli/ZZ. It would appear amply safe, therefore, 
to adopt a value of M^wli/\2, 

For the diagonal reinforcement, the bars run in two direc- 
tions, and considering both theory and test, a value of M=wli^/24t 
is conservative to use for the reinforcement in each direction. 

Cross Reinforcement Between Columns. 

In flat slab floors cracks are apt to occur between colimms on 
rectangular lines, because, since the span is shorter, the deflection 
is less than in the center of the slab. To prevent these cracks, it 
is advisable to place cross reinforcement of small bars in the top of 
the slab. 

Tables for Design of Slabs. 

Tables I-IV give thicknesses of slab, reinforcement and 
size of column head for various column spacings and loads. 



Digitized by 



Google 



Design of Concrete Flat Slabs. 265 




OJO ai5 020 025 030 035 040 QA5 0,50 055 060 
Valued of Constant Cc 
a>0.10 




040 0.15 0^0 ass 0,30 0-35 040 0.4-S 0-50 a55 050 
Valuer of Cons+anf Cc 
a. 0.15 

FIG. 5. — ^DIAGRAM GIVING VALUES OF CONSTANTS IN FORMULA 
/c"^-r^ FOR a =0.10 AND 0.15 

Depth of Reinforcement in Compression. Area of Reinforcement in Tension. 

Q Ml pas 

Depth of Reinforcement in Tension. Area of Concrete above Reinforcement. 



Digitized by 



Google 



266 



Thompson on Design of Concrete Flat Slabs. 




OiO OJS OJZO a25 O^Q 035 040 045 0^0 055 a60 
Vdlues of Con3+cint Cc 
a*0.ao 



0.04a 




aiO OJS 0^0 0.^5 030 0.aS ft40 a45 050 055 0.60 
Vdlues of Cons+cinf Cc 
a=0.a5 

FIG. 6. — DIAGRAM GIVING VALUES OP CONSTANTS IN FORMULA 

/c"^,^ FOR a =0.20 AND a =0.25 

Depth of Reinforcement in Compression. Area of Reinforcement in Tension. 

Depth of Reinforcement in Tension. Area of Concrete above Reinforcement. 



Digitized by 



Google 



Thompson on Design op Concrete Flat Slabs. 267 

Three arrangements for reinforcement over the column head 
are chosen: The first where the area of reinforcement in the top 
is twice the area of reinforcement in the bottom; the second 
where the two are equal; and the third where the area of rein- 
forcement in the bottom is one and a half times that in the top. 
This gives the designer a variety of thicknesses of slab. The 




aio ais 020 0.25 oao 035 C40 045 050 055 ojbo 

Votued of Cone+Gin+ C© 

FIQ. 7. — DIAGRAM GIVING VALUES OF CONSTANTS IN FORMULA 

M 



/*-; 



FOR a "0.30 



a"> 



P- 



Cch^ 

Depth of Reinforcement in Compression. 
Depth of Reinforcement in Tension. 
Area of Reinforcement in Tension. 
Area of Concrete above Reinforcement. 



percentages of reinforcement selected are those which produce, 
with the given conditions, a compressive stress of 800 lb. per 
sq. in. in the concrete and 16,000 lb. in the reinforcement. In 
order to allow 800 lb. in the concrete, it should be mixed in pro- 
portions as rich as 1 part cement to 2 parts fine aggregate to 4 
parts coarse aggregate. Poisson's ratio is assumed as 0.1, which 
from recent tests appears to be a fair value. 



Digitized by 



Google 



268 Thompson on Design of Concrete Flat Slabs. 

The size of column head has been figured for a shear of 60 lb. 
per sq. in. on a circle a distance, t, (the thickness of slab) outside 
of the column head. This shear is used simply as a measure of the 
diagonal tension. The value is somewhat larger than is permitted 
in beam design but appears to be warranted in the case of flat slabs. 

a040r 




ooos 



0,04(1 



Valuer of Conatan+ C^ 



no. 8. — ^DIAGRAM GIVING YALUBS OF C0N8TANTB IN FOBlfTTIiA 

Depth of Reiaforcement in Compression. 
Depth of Reinforcement in Tension. 
Area of Reinforcement in Tension. 



P = 



Area of Concrete above Reinforcement. 



The reinforcement in the center of the slabs has been figured 
for a stress of 16,000 lb. per sq. in. 

Diagrams for Designing Slabs. 

To provide for cases not covered by the tables, curves for 
values of Cg and C, are given so that the moment under various 
conditions can be readily figured from the formula for the bending 
moment given above. 



Digitized by 



Google 



Thompson on Design op Concridtb Flat Slabs. 269 

Diagrams for Determining Reinforcement in Top 
AND Bottom of Beams or Slabs. 

In Figs. 5 to 8 curves are plotted for finding the values of 
the constants C^ and C, in the formulas for the reinforcement and 
concrete stresses in beams or slabs with reinforcement in top 
and bottom. The curves are drawn for diflferent values of a, 
the ratio of depth of reinforcement in compression to depth of 
reinforcement in tension, and for diflferent values of pVP> where 
p = ratio of cross-section of reinforcement in tension to concrete 
above it*, and p^ ^ ratio of crossHsection of reinforcement in 
compression to this same area of concrete. 

EXAMPLB. 

Example. — ^For a warehouse floor with a live load of 150 
lb. per eq. ft. and a column spacing of 20 ft. each way, what is 
the necessary thickness of slab, size of column head, and amount 
of reinforcement? 

Sohdian. — ^From Table II the thickness of slab is given as 
8} in., the size of column head as 5.5 ft., and the area of rein- 
forcement as 24.7 sq. in. at top of slab and same amount at bottom 
of slab over column, using ratio of area of reinforcement in ten- 
sion to area of concrete below as 0.017. Dividing these values 
by 4, as each end of the bands is effective, we have 24.7/4 = 6.2 
sq. in. as the area of reinforcement in each band. For this may 
be used twenty |-in. round bars spaced 5 in. center to center for 
both tension and compression reinforcement. 

The amount of reinforcement required at center of rectangular 
band is 0.17 sq. in. per ft. of width. Placing a f-in. round bar 
every 10 in. gives more than the necessary area, but ease in plac- 
ing the reinforcement makes up for the extra amount. The 
amount required at center of diagonal band is 0.35 sq. in. per 
ft. of width. |-in. round bars every 10 in. will thus give neces- 
sary amount. 



* Where the tenaile reinforoement ia at the top, aa over a support of a flat slab or beam, 
the concrete area is taken below the tensile reinf oroement. 



Digitized by 



Google 



270 Thompson on Design op Concrete Flat Slabs. 

Table I. — Design of Flat Slab. 

Thickness of Slab, Areas of Reinforcement and Sizes of Column 

Head are Given for Different Spans and Percentages of 

Reinforcement. 

LIVE LOAD 100 LB. PER SQ. FT. 



ft. 





t 



go 

< 

it) 
in. 



Q 
ft. 



•a.9 

i 

|i 



8q. in. 




aq. in. 



aq. in. aq. in. 



12 
12 
12 

14 
14 
14 

16 
16 
16 

18 
18 
18 



22 
22 
22 



0.014 
0.017 
0.022 

0.014 
0.017 
0.022 

0.014 
0.017 
0.022 

0.014 
0.017 
0.022 



20 0.014 
20 0.017 
20 0.022 



0.014 
0.017 
0.022 



24 0.014 
24 I 0.017 
24 0.022 



0.007 
0.017 
0.033 

0.007 
0.017 
0.033 

0.007 
0.017 
0.033 

0.007 
0.017 
0.033 

0.007 
0.017 
0.033 

0.007 
0.017 
0.033 

0.007 
0.017 
0.033 



5J 

5 

4i 

5 
5i 

7i 
6i 
5i 

8i 
7i 
6i 

9i 

8i 
7i 

lOi 
9 
8 

m 

10 
8J 



2.00 
2.00 
2.50 

2.25 
2.75 
3.00 

3.00 
3.25 
3.75 

3.50 
3.75 
4.50 

4.00 
4.50 
5.00 



4.50 
4.81 
7.26 

5.94 
7.93 
9.96 

9.51 
10.95 
14.01 

12.48 
14.42 
18.67 

16.36 
19.47 
23.89 



4.50 
5.00 
5.75 I 31.05 



20.21 
I 24.05 



5.00 25.10 
5.75 I 30.41 
6.50 I 37.80 



2.25 

4.81 

10.90 

2.97 

7.93 

14.95 

4.76 
10.95 
21.05 

6.24 
14.42 
28.00 

8.18 
19.47 
35.80 

10.11 
24.05 
46.60 

12.55 
30.41 
56.60 



0.16 
0.17 
0.18 

0.19 
0.20 
0.21 

0.22 
0.23 
0.24 

0.26 
0.27 
0.28 

0.30 
0.31 
0.32 

0.34 
0.34 
0.35 

0.38 
0.39 
0.40 



0.09 
0.09 
0.09 

0.11 
0.11 
0.11 

0.12 
0.12 
0.12 

0.14 
0.14 
0.14 

0.16 
0.16 
0.15 

0.18 
0.17 
0.16 

0.20 
0.20 
0.19 



* Area of reinforcement over column head » circumference of column head in inchea 
Xd Xp or p' depending upon whether the reinforcement is in tension or compression. This 
reinforcement is assumed as distributed over the entire widths of the bands. Thus if a band 
of reinforcement has 2 sq. in. in section the area, efTective, for two bands will be 4 aq. in. (See 
example.) 



Digitized by 



Google 



Thompson on Design op Concrete Flat Slabs. 271 



Table II. — Design op Flat Slabs. 

Thickness of Slab, Areas of Reinforcement and Sizes of Column 

Head are Given for Different Spans and Percentages of 

Reinforcement. 

LIVE LOAD 180 LB. PER SQ. FT. 



1^ 



12 
12 
12 

14 
14 
14 

16 
16 
16 

18 
18 
18 

20 
20 
20 

22 
22 
22 

24 
24t 



d a ^ 

■ffi 

(p) 



0.014 
0.017 
0.022 

0.014 
0.017 
0.022 

0.014 
0.017 
0.022 

0.014 
0.017 
0.022 

0.014 
0.017 
0.022 

0.014 
0.017 
0.022 

0.014 
0.016 



ml 



(pO 



0.007 
0.017 
0.033 

0.007 
0.017 
0.033 

0.007 
0.017 
0.033 

6.007 
0.017 
0.033 



007 
017 
033 

007 
017 



0.033 

0.007 
0.013 



iff 

ID. 



4J 

u 

5J 

5 

4i 

4} 

7i 
6i 
5i 



8i 10 

7 



:i 



< 
(0 



6 

5J 

4i 



8 

7i 

6 

8} 
7} 

61 



lOi 

8i 



n 11* 

8i lOit 



ft. 



2.25 
2.50 
3.00 



S o 



^1 

it 

aq. in. 



5.64 
6.81 
8.72 



3.00 , 8.72 

3.50 I 11.22 

3.75 I 13.22 

3.50 I 12.03 

3.75 13.83 

4.50 I 17.75 

4.00 I 15.32 

4.50 18.05 

5.50 ' 24.00 

5.00 21.80 

5.50 ! 24.70 

6.25 I 31.14 

5.50 I 26.15 

6.25 I 31.07 

7.00 39.24 

7.00 36.05 

7.00 36.95 



I- 

g1 



aq. in. 



IP 

hi 



aq. in. 



2.82 ' 0.18 

6.81 0.19 

13.09 , 0.21 

4.36 0.22 

11.22 0.23 

19.82 0.24 

6.02 ' 0.26 

13.83 0.27 
26.80 0.28 



7.66 
18.05 
36.00 

10.90 
24.70 
46.70 

13.08 
31.07 
58.80 

18.03 
29.57 



0.31 
0.32 
0.33 

0.34 
0.35 
0.36 




aq. m. 

0.10 
0.10 
0.10 

0.12 
0.11 
0.11 

0.14 
0.13 
0.13 

0.16 
0.15 
0.17 

0.17 
0.17 
0.16 



0.38 I 0.18 
0.39 0.17 
0.40 0.17 



0.42 
0.43 



0.21 
0.20 



The values printed in black type are figured for a column head 7 ft. in diameter and the. 
thickncM of the alab is increased to withstand the shear. 

* Area of reinforcement over column head — circumference of column head in inchea Xd X P 
or pi dei>ending upon whether the reinforcement is in tension or compression. This reinforce- 
ment is assumed as distributed ovei the entire widths of the bands. Thus if a band of reinforce- 
ment has 2 sq. in. in section the area, effective, for two bands will be 4 sq. in. (See example.) 

t The thickness of slab for the 24-ft. span may be decreased to 8} in. by using 0.022 and 
0.033 ratios of reinforcement and bending the bars to resist diagonal tension. 



Digitized by 



Google 



272 Thompson on Design op Concrete Flat Slabs. 



Table III. — Design of Flat Slabs. 

Thickness of Slab, Areas of Reinforcement and Sizes of Coltthn 

Head are Given for Different Spans and Percentages of 

Reinforcement. 

LIVE LOAD 200 LB. PER 8Q. FT. 



I 

P 

GQ 
ft. 



h 

.- P 2 « 
.23 !£ 

(P) 



i 



5 

.2 

i 
i 

1 



(lO 



Us 

m 






t 



So 

-< 

(0 
in. 



ft. 



2 2 



is 



0) Q. 

as 



£a 



sq. in. 



ill 

in 

la I 



sq. in. 




aq. in. 



12 
12 
12 

14 
14 
14 

16 
16 
16 

18 
18 
18 

20 
20 
20 

22 
22 1 

24t 



0.014 
0.017 
0.02!^ 

0.014 
0.017 
0.022 

0.014 
O.017 
0.022 

0.014 
0.017 
0.022 

0.014 
0.017 
0.019 

0.014 
0.016 

0.014 



0.007 
0.017 
0.033 

0.007 
0.017 
0.033 

0.007 
0.017 
0.033 

0.007 
0.017 
0.033 

0.007 
0.017 
0.029 

0.007 
0.012 

0.006 



5 

4J 

3i 

6 
6 
4i 

6 
4i 

7i 

II 

9i 
8i 

lOi 



6i 
5i 
5 

7J 
6i 
6i 

8i 
7i 
6 

9 

8 
7 

lOi 
8i 
7i 

Hi 

lot 

12t 



2.50 
3.25 
3.76 

3.25 
3.75 
4.50 

4.00 
4.50 
5.50 

4.75 
5.50 
6.50 

5.50 
6.25 
7.00 

6.25 
7.00 

7.00 



6.60 

9.38 

10.37 

10.31 
12.02 
15.88 

14.26 
16.60 
21.70 

18.80 
22.92 
29.70 

24.71 
29.08 
31.35 

31.38 
34.90 

37.90 



3.30 

9.38 

15.40 

5.16 
12.02 
23.80 

7.30 
16.60 
32.55 

9.46 
22.92 
44.60 

12.36 
29.08 
47.85 

15.69 
26.15 

16.24 



0.20 
0.21 
0.23 

0.24 
0.25 
0.26 

0.28 
0.30 
0.32 

0.33 
0.34 
0.35 

0.37 
0.38 
0.39 

0.42 
0.42 

0.46 



0.10 
0.10 
0.10 

0.13 
0.12 
0.11 

0.14 
0.14 
0.14 

0.17 
0.16 
0.15 

0.18 
0.17 
0.16 

0.20 
0.19 

0.22 



The values printed in black type are figured for a column head 7 ft. in diameter and the 
thickness of the slab increased to withstand the shear. 

* Area of reinforcement over column head— circumference of column head in inches 
Xd Xp or p' depending upon whether the reinforcement is in tension or compression. This 
reinforcement is assumed as distributed over the entire widths of the bands. Thus if a band 
of reinforcement has 2 sq. in. in section the area, effective, for two bands will be 4 sq. in. (Seo 
example.) 

t The thickness of slabs for the 22- and 24-ft. spana may be decreased to BH u^* i^d 9H 
in. respectively, by using 0.022 and 0.033 ratios of reinforcement and bending the bars to i 
diagonal tension. 



Digitized by 



Google 



I^OMPSON ON Design of Concrete Plat Slabs. 273 

Table IV. — ^Design of Flat Slabs. 

Thickness of Slab, Areas of Reinforcement and Sizes of Column 

Head are Given for Different Spans and Percentages of 

Reinforcement. 

LIVE LOAD 300 LB. PER SQ. FT. 



•s 




)n of re- 
ompres- 
! below 
tension. 






1 
n head. I 


rcement 
tension. 

1 


int over 
ireasion. 


IP 
III 




8^ 




lag-: 
his 




1^ 


1 


is 

6 


a 

is 


2*- 1 


em 

2 -^ a 


CQ 






!•§:§ 

(d) 


< 


1 

eS 

s 


1^ 


°.2 

< 

• 




He! 


ft. 


(p) 


(p') 


in. 


in. 


ft. 


sq. in. 


sq. in. 


sq. in. 


sq. n. 


12 


0.014 


0.007 


5i 


6} 


3.50 


9.72 


4.86 


0.24 


0.12 


12 


0.017 


0.017 


4i 


5i 


4.25 


12.27 


12.27 


0.25 


0.11 


12 


0.022 


0.033 


31 


5 


5.00 


15.56 


23.30 


0.26 


0.10 


14 


0.014 


0.007 


61 


7} 


4.25 


14.03 


7.02 


0.29 


0.13 


14 


0.017 


0.017 


5i 


6i 


5.00 


16.84 


16.84 


0.30 


0.12 


14 


0.022 


0.033 


4i 


5i 


6.00 


21.18 


31.72 


0.31 


0.11 


16 


0.014 


0.007 


7 


8i 


5.00 


18.48 


9.24 


0.33 


0.15 


16 


0.017 


0.017 


6 


7J 


6.00 


23.10 


23.10 


0.35 


0.15 


16 


0.022 


0.033 


5 


6i 


7.00 


29.05 


43.50 


0.35 


0.12 


18 


0.014 


0.007 


7i 


9i 


6.00 


24.59 


12.30 


0.39 


0.17 


18 


0.017 


0.017 


6i 


8 


7.00 


29.18 


29.18 


0.40 


0.16 


18 


0.015 


0.015 


61 


8i 


7.00 


26.74 


26.74 


0.39 


0.15 


20 


0.014 


0.007 


81 


10} 


7.00 


32.32 


16.16 


0.43 


0.19 


20t 


0.015 


0.011 


8} 


lOit 


7.00 


33.65 


25.36 


0.43 


0.18 


22t 


0.012 


0.003 


10} 


12it 


7.00 


32.61 


8.23 


0.47 


0.22 


24t 


0.010 


0.000 


13i 


ISit 


7.00 


35.64 


0.00 


0.51 


0.25 



The values printed in black type are figured for a column head 7 ft. in diameter and the 
thickness of the slab increased to withstand the shear. 

•Area of reinforcement over column head = circumference of column head in inches 
X<f Xp or p' depending upon whether the reinforcement is in tension or compression. This 
reinforcement is assumed as distributed over the entire widths of the bands. Thus if a band 
of reinforcement has 2 sq. in. in section the area, effective, for two bands will be 4 sq. in. (See 
example.) 

fThe thickness of slabs for the 20-. 22- and 24-ft. spans may be decreased to 8>i in.. 9H 
in. and 11 in. respectively, by using 0.022 and 0.033 ratios of reinforcement and bending the 
bars to resist diagonal tension. 



Digitized by 



Google 



DISCUSSION 



Mr. Lindau. Mr. Alfred E. Lindau. — Regarding this method of design- 

ing a flat slab or plate of concrete it might be of value to note 
that the analysis is based on the assumption that there exists 
a line of contra-flexure around the colmnn substantially circular 
in form and near the quarter point of span, that the balance of 
the span is supported along this line and that consequently there 
is no stress in any direction along this line. There is some ques- 
tion as to this line of contra-flexure, in fact there is some evidence 
to show that instead of a line of contra-flexure there exist per- 
haps only four points where there is no stress in any direction. 
If such is the case the analysis would be entirely diflFerent from 
that which has been outlined. 

Mr. Andenon. Mr. W. P. Anderson. — One of the important things in 
designing a flat slab has not been touched — ^the bending moment 
on the exterior column. I think that has an effect on the thick- 
ness of the slab. The thicker the slab, it appears to me, the less 
that bending moment would be. I have not gone into the matter 
thoroughly enough to determine how the bending moment of 
the exterior column is to be figured, but it seems to me that it is 
the critical point in the design of the flat slab and one that has 
not yet been touched upon enough by those who have given the 
subject thought. This point in the design of the flat slab has 
not been covered in the tests made on the flat slabs. They have 
been made in the center of the building and the exterior colimm 
has not been tested at all to see what the moment is. It is some- 
thing that ought to be covered. I am not prepared to say how 
it ought to be taken care of, but I believe it ought to be looked 
into. 

Mr. Lindau. Mr. Lindau. — The test covered in Part II of the report of 

the Committee on Reinforced Concrete refers to an end or wall 
panel. In fact, it was a building with one line of colimms making 
all panels end panels, so to speak. 

Mr. Green. Mr. HERBERT P. Green. — I have been experimenting a 

little bit this last year on flat slabs of a different construction 

(274) 



Digitized by 



Google 



Discussion on Concrete Plat Slabs. 275 

from any others I have seen. Realizing that the shear around Mr. Green, 
the head of the column is the greatest thing to be considered, I 
endeavored to work out a way of thickening the slab at this point 
and have succeeded in obtaining a flat ceiling without the flar- 
ing head. We have built one building by this system and tested 
it and it is a perfect success. 

The reinforcing bars run directly and diagonally from col- 
umn to column. The bending moment on each beam is considered 
as coming from the load of one-fourth of the panel, which posi- 
tively occurs because triangles are formed in between the direct 
and diagonal girders which are filled with ordinary stock tile. 
Between these tile are concrete joists, supported by the beams. 
This construction can be figured by any engineer without em- 
pirical formulas. 

At the center of the slab between the columns on the diagonal 
girders in the bottom or top of the slab, supplementary rein- 
forcement can be placed, but it is generally not necessary. Every 
beam and every girder, or every joist and every beam, as they 
may be called, is of a T-section, the concrete above the tile form- 
ing the flange of the direct and diagonal beams and the supple- 
mentary joists in the triangles, so that the analyses of the stresses 
in this slab are very easy, according to the ordinary theories of 
design. 

Mr. Lindau. — I would like to ask Mr. Green whether in Mr.Lindau. 
his method of flat slab construction he has made joints along 
these triangles, so that the tile filler is disconnected absolutely 
from the balance of the slab thereby making separate beams, 
or whether it is all concreted in and becomes a portion of the 
slab as a whole. 

Mr. Green. — ^The tile actually becomes a portion of the Mr. Green, 
slab, but — according to the methods generally used for figuring 
the efficiency of tile joist construction — the tile is considered as 
nothing but a core, so that the concrete construction is the prin- 
cipal thing considered. If we should remove the tile, in this 
construction it would be the same thing as removing it in any 
tile joist construction, L e., we would have a floor with ribs or 
stiffeners below, and in this case they would be in four directions 
from column to column, with supplementary stiffeners in the 
triangles. The tile is simply a form. 



Digitized by 



Google 



276 Discussion on Concrete Flat Slabs. 

Mr. xjadaii. Mr. Lindau. — I take exception to the assumption that 

you have a series of beams that are free and independent to act 
as beams; because just as soon as you connect up the slab so as 
to have anything resembling a homogeneous structure you have 
changed the stress condition from a slab to a beam. The loads 
are transmitted by bending moment or transverse stresses, or a 
combination of those two; you cannot draw imaginary diagonals 
or anything on that slab and say that this is the direction in 
which the load will travel. The load will travel along the lines 
of relative rigidity of the structure and you cannot make it go 
in one direction rather than another unless the slab is built accord- 
ingly. This is the difficulty with most theories and methods of 
strip analysis, where the freedom of the slab is tied up in various 
ways and consequently cannot be separated into the various 
elements it may be considered to be made up of. 

Mr. orMn. Mr. Green. — ^That may be so, but in the two-way reinforced 

concrete floor with tile fillers one figures the loads are carried 
through the joists, which are supported on the beams, to the 
beams which run directly between the columns. Now I simply 
make two more beams or headers in the slab. I have run them 
diagonally across from column to column and the joists in the 
triangles between the direct and diagonal beams are supported 
by these direct and diagonal beams. It is the same principle 
that would occur in a wooden framed structure with diagonal 
beams, except that we have a T-section in all directions and 
the stresses in the concrete above and below the neutral axis 
at any point can be figured just as readily as in the two-way 
reinforced concrete floor. 



Digitized by 



Google 



THE DESIGN OF CONCRETE GRAIN ELEVATORS. 

By E. Lee Heidenreich.* 

During the International Congress of Engineers in 1893, at 
Chicago, III., I was appointed by the American Society of Civil 
Engineers to present a paper on '* American Grain Elevators," 
having at that time had about ten years' experience in the con- 
struction and erection of grain elevators in North and South 
America. Elevators at that time had the same functions as they 
have today and may as a rule be divided into certain classes: 

1. Farmers' elevators or small station elevators with a 
capacity from 5000 to 50,000 bushels — ^where the farmers would 
deliver their grain in wagon loads, up to 5 tons capacity, have 
it weighed, elevated and shipped out in railroad cars. 

2. Mill elevators or cleaning elevators, built as adjuncts 
to flour mills, within varying capacities from 50,000 to 500,000 
bushels and containing, besides receiving and shipping apparatus, 
scales and cleaning machinery. 

3. Terminal elevators, divided into storage and working 
houses. The working houses would receive the grain, weigh 
and ship it into the storage elevators or cars. The storage 
elevators would merely be for the purpose of storing grain. 

4. Transfer elevators located at prominent points where 
different grains are graded and changed from one grade to 
another and shipped from western to eastern cars. 

5. Marine elevators, where the grain would either be received 
from canaJ boats or vessels or shipped into them. These 
elevators are either dock elevators or floating elevators. 

Up to 1893 there were no reinforced concrete grain elevators 
in the United States. At that time two large grain elevators 
were imder construction at Galatz and Braila, Roumania, and 
Mr. Herman 0. Schlawe, a representative of the Roumanian 
government, sent over to see how we handled grain and took care 
of oiu" grain elevators, showed me the plans of the same elevators 
designed by Luther of Braimschweig, Germany. These elevators 

^Consulting Engineer. Kansas City, Missouri. 

(277) 



Digitized by 



Google 



278 Heidbnreich on Concrete Grain Elevators. 

were of unit construction, somewhat similar to the one described 
by Mr. Darnell,* excepting that the bins were honeycombed 
hexagonal cells, built of flat slabs 3 ft. square, and comer pieces, 
all interlocked. They were all built on the ground, erected and 
interlocked somewhat similar to tile construction. The difficulty 
with flat slab walls was, of course, that they were subject to 
direct flexion, and therefore became quite cumbersome, owing to 
the lateral, rather heavy pressure in grain bins, and were only 
suitable for bins of small dimensions. 

Steel tanks have been tried in this country, but there was 
more or less sweating of the tank, which to some extent damaged 
the grain. Circular reinforced concrete tanks were then designed, 
using the space between the tanks whereby considerable economy 
of ground area was effected as compared with isolated tanks and 
thereby increasing their combined strength and carrying capacity 
on the soil or the sub-structure. The introduction of the new 
design was, however, exceedingly difficult. 

The owners thought the tanks would sweat and damage the 
grain and builders thought the tanks would surely burst. From 
1896 to 1899 I designed some scores of cluster tank elevators and 
presented them to millers and grain elevator owners entirely in 
vain, imtil in 1899, Frank H. Peavey, of Minneapolis, as an 
experiment built a tank 23 ft. in diameter 130 ft. high, filled it 
with grain and found that it held grain without spoiling. He then 
sent some representatives over to Europe to see what effect 
grain storage in reinforced concrete had in Europe; they examined 
the tanks and grain elevators at Galatz and Braila, in Roumania, 
and came back with a report which resulted in the construction, 
in 1901, of a large one million and a half bushel elevator in 
Duluth. In the meanwhile I had built, in 1900, a cement storage 
elevator at South Chicago, consisting of four tanks 25 ft. in 
diameter and 56 ft. high, clustered in such manner as to utilize 
the space between them. 

It must be remembered that wheat weighs 50 lb. per cu. ft. 
while cement weighs 100 lb.; and the fact that a cement tank with 
5-in. walls at the top and 7-in. at the bottom, 25 ft. in diameter, 
would hold cement, went a long step towards convincing people 
that it was safe to store grain in circular tanks and also in cluster 

* See p. 4G4.— Ed. 



Digitized by 



Google 



Heidenr£ich on Concrete Grain Elevators. 279 

tanks. The tanks in South Chicago were erected on top of 
reinforced concrete girders, which in turn rested on columns, 
thus forming a working house underneath where the bagging of 
the cement took place. 

Since that time cluster tanks have grown up like mushrooms 
all over this country, Canada, South America and Europe. 
There are upwards of 60,000,000 bushels capacity built in the 
United States alone to-day and they are being built in every 
part of the country for terminal elevators and largely for milling 
elevators, giving safe and good storage and eliminating the ques- 
tion of insurance and the calamity due to fire or burning up of 
the storage plant. 

The design of reinforced concrete grain elevators, consisting 
of circular tanks, has been carefully studied. It is of course 
entirely a function of the. grain pressure in rest and in motion. 
Experiments and calculations have been made by a number of 
well-known engineers. Janssen, Wilfred Airy, Tolz, Prante, 
Jameson of Canada and Milo B. Ketchimi, have all made experi- 
ments and developed formulas whereby the grain pressure may 
be determined at the different heights of the bin. Grain is not 
like a liquid — ^Professor Ketchum calls it a semi-liquid, — the 
pressure line forms a curve. 

Inasmuch as Janssens' solution tallies very closely with 
Jameson's experiments, we niay write the lateral pressure 



and the vertical pressure 



Cm' 



^(■-" * ) 



Where ty= the weight per cu. ft. of grain or 50 lb. for wheat, 
jB=bin area divided by its perimeter (hydraulic radius), 
ft' = the coefficient of friction of grain against bin surface; 
A=the height of the bin, and 

C=the ratio between the lateral and the vertical pressure; 
e = being the base of the Naperian logarithm or 2.71828. 

According to Jameson for wheat C=0.6 and for grain on 
concrete ft' =0.4 to 0.425. 



Digitized by 



Google 



280 



Heidenbeich on Concrete Grain Elevators. 



For quick and easy calculation the lateral pressure per sq. 
ft.^qwh, where q is the ratio of grain pressure to liquid pressure 
and 

F--^ -1.667 gwA 
Fig. 1 shows the value of q for different ratios of height 
WHEAT PRESSURES IN GRAIN BINS- 




\23^5ST6B\0 



PIG. 1. — VALUES OP Q POB DIPPBRENT RATIOS 
OP HEIGHT TO DIAMETER OP BIN. 

divided by the diameter or width ^-r. The maximum bottom 

h 
pressure occurs when t =3.6. 

Fig. 2 shows a series of circular bins with intervening spaces. 
The lateral pressure per sq. ft. of the grain in a circular bin at a 
depth h is equal to 

L^qwh 



Digitized by 



Google 



Heidbnbbich on Concrete Grain Elevators. 



281 



and the tension in the reinforcement reqxiired for 1 ft. in height is 



The reinforcement area in sq. ins. 
where /« is the working stress of the reinforcement per sq. in. 




WIQ. 2.-HSEBIBB OF CIRCULAR BINS. 

If, however, -the circular bin is empty and the two opposite 
interstices A and B filled, we have a condition as shown in Fig. 3 
where it is quite apparent that instead of single reinforcement of 
the bin, there should be reinforcement both at the intrado and 
extrado of the ring. 

Compression in the direction A B clearly causes tension at 



Digitized by 



Google 



282 



Heidenreich on Concrete Grain Elevators. 



the extrado at C and Z>, and at the intrado A and B, From 
experiments in loading culvert pipe firmly supported at the two 
lower quarter points corresponding to E and F, the maximum 
moment may approximately be written (Fig. 4) : 



.s^^t- 




FIG. 3. — PLAN OF CIRCULAR BIN. 

The usual formula for the resisting moment is 

M«/C6d«andX«iM(l-|)=p/,(l.|) 

where 

/, and /c the unit stresses in reinforcement and in concrete. 



Digitized by 



Google 



Heidenreich on Concrete Grain Elevators. 



283 



The stress in the concrete or in the reinforcement may be 
expressed by 



-5'('-l) 



bd* 



■('-!) 



bd* 



By adding 1 in. to di or (is, the ring thickness is found. 

Example: The 25 ft. diameter bins are 80 ft. deep and sur- 
romided by interstices, find the dimensions of concrete wall and 
reinforcement required at 60 ft. from the top. 



h 25 



H A— - 



hence 
Then 

and 



no. 4. 



9-0.23 and L=0.23X60X60=6901b. per sq. ft. 



T.m^^^^,^, 



8625 
"^•"Tfinno ** '^^ sq. in. per each foot high 

The inner and outer reinforcement from the eflfects of grain 
in the center bin is 

-~ =0.27 sq. in. or say §-in. rounds 8§ in. on centers 
Vertical rods J-in. rounds 24 in. on centers tied to horizontals. 



Digitized by 



Google 



284 Heidenbeich on Concrete Grain Elevators. 

For external pressure we have 

^ - 59 -6 hence g "0.1 andL^O.lX 50X 60-300 lb. per eq. ft. 
10 

(^ -o) L'D' C-?^, -6) 300X 26 X 12 

jif . W2 L = Vl._41_/ 18700 in. lbs. 

64 64 



^ ^ / 18700' / 18700 __^. 
*"*'\ibXl2 "\ 110X12" ^•^^'"• 

hence the wall should be 5 m. thick. 

18700 
'^•*' 0r8fiy^l6o6oS^4 ""^-^^ ^^- ^^' ^^^ ^^^» ^^^^ extrado and intrado, 

showing that the stresses from the filled interstices are greater 
than those resulting from the filled circular bins. Some con- 
tractors using single reinforcement in the center bins have rodded 
the interstices, others increase the thickness of the bin walls and 
the connection between contiguous circular bins. 

The logical method, would, however, seem to be to use double 
reinforcement, a practice the author invariably prefers for circular 
culverts. 

As to the calculation of square bins, this is a simple matter, 
after the lateral pressure L has been found. 

The bending moment will be in in.-its. 

LP 
Jl/ « — 12, where Z= width of bin in feet. 



The resisting moment as before 

■V" 



M^UxJP and 4= ^,,, 



Area of reinforcement required 

A« = 17-; as before 

Interior bin walls having alternately pressure on either side 
are, of course, reinforced on both sides. The connection of the 
reinforcement at the intersections and the comers of the bin walls 
become of the greatest importance. Where possible the author 
prefers to employ a strong wire fabric as a part of the reinforce- 



Digitized by 



Google 



Heidbnreich on Concrete Grain Elevators. 



285 



ment for the outer walls, nmning this reinforcement horizontally 
atoiind the entire building. 

Fig. 5 shows a common country elevator of 15,000 bushels 
capacity. The structure rests upon a slab or raft extending 
under the entire building and connected with the side walls so as 
to make the pit absolutely waterproof. The concrete is mixed 




nte 



30-0- 



EE 






\ , 



P""^ 



FIG. 6. — CONCRETE COUNTRY ELEVATOR. 

1 : 2 : 4 to maximimi density and to each bag of cement is added 
5 lb. of petroleum residuum oil which has been found an excel- 
lent and cheap method of making the walls and roof impervious 
to moistiu^e. The roof and cupola walls are reinforced with 
fabric to prevent cracks by shrinkage. 



Digitized by 



Google 



286 



Heidenreich on Concrete Grain Elevators. 



Fig. 6 shows an outside wall connection and the reinforce- 
ment, illustrating how the horizontal and hooked members are 
tied to the vertical rods, to insure their proper position until 




FIG. 6. — OUTSIDE WALL CONNECTION OF REINFORCEMENT. 

the concrete has been poured. The hooks of the comer rods 
(Fig. 7) are placed by springing the vertical rods until the comer 
rods can be hooked in and thereby form a support for the outside 
carrying rods. 




FIG. 7. — METHOD OF HOOKING CORNER REINFORCEMENT. 

In 1911 the com crop of the United States was. . 2,531,488,000 bu. 
and of wheat 631,388,000 " 



Total of 3,152,786,000 " 



Digitized by 



Google 



Heidenreich on Concrete Grain Elevators. 287 

Most of this crop was handled through farmers' elevators, 
averaging, say 150 cars a year, or 150,000 bu., meaning a total 
of approximately 20,000 farmers' grain elevators in the country. 
Most of these are built in wood and will ultimately have to be 
replaced with concrete construction, which in turn shows one 
immediate channel for the use of Portland cement. The cost 
of a 150,000-bushel grain elevator in concrete is about $5;000.00 
exclusive of machinery and millwrighting. Considering the 
thousands of wooden elevators rotting away and burning up along 
our western trunk and grain lines — ^the almost irreparable loss 
to a town or community of farmers when its elevator bums down, 
the ajonual cost of insurance and maintenance of these wooden 
elevators — ^the great doctrine of conservation brings into the 
limelight the impending want of reinforced concrete grain 
elevators in every nook and comer of the land. 



Digitized by 



Google 



DISCUSSION; 



Mr.Lindau. Mr. Alfred Lindau. — In the design of circular bins for 

exterior pressure of the space between the bins, is account taken 
of the support that is afforded by the adjacent bins? As I under- 
stand, the worst condition is where the two opposite spaces would 
be filled with grain and others would not be filled with grain; 
but the bin, I understand, is supported on the other diameter by 
the adjacent bins. Is this taken into consideration in the strength 
of the structure, or merely left as an added factor of safety; 
What are the general principles that govern the economy of grain 
storage? Bins are built high and you can build them wide, 
circular or square. Has an investigation been made to show the 
design or adopt a bin that would attain economy for storage 
pmposes? 

Mr. Heidenreich. Mr. E. Lee Heidenreich. — ^Thc location of the adjacent 
bins is taken into consideration, otherwise we could use Profes- 
sor Talbot's formula divided by 16, vxl/lQ, in place of dividing by 
64 — ^an empirical factor that I brought in. 

There are many items, many functions, which come into 
play in the determination of a grain elevator. First, the storage 
capacity required. It is possible to store grain in very large 
bulk, requiring very large bins. I have built them up to 30,000 
and 40,000 bushel bins for winter storage. In other cases there 
are what we call pocket bins, carload bins, which are required 
where certain shippers want their grain individually, a practice 
quite common. Then again comes the floor space. In a city the 
ground space is very expensive and it is important to economLee, 
and elevators are built as high as possible, as high as the soil will 
carry them. If the soil will not carry them they are put on 
timber piling under all piers. In other places, again, you can 
spread the bins considerably, but, as a general rule, to utilize the 
gravity of the grain in handling it throughout the elevator is 
considered the best economy. For instance, in a cleaning elevator 
you build it so high that from the top of the elevator head the 
grain runs first into the gamer, then into the scales, then into the 

(288) 



Digitized by 



Google 



Discussion on Concrete Grain Elevators. 289 

cleaning bin; from there through the cleaners and separators to Mr. Heidenreich. 

receiving bins and then into the loading apparatus. But you gain 

by it, because the grain goes by the gravity from the top in one 

process to the bottom in the place of being re-elevated, which 

consumes power and of course adds to the cost of handling of the 

grain. 

There are a great many other elements entering into shape of 
grain elevators, but these are a few of the functions that must be 
considered. 

Mr. Wm. M. Kinnet. — In bms holding cement, there is Mr. Kinney 
often a sweeping action, similar to a wave action. Mr. Heidenreich 
spoke of the semi-fluid condition of grain, or an action like a 
semi-fluid. In a cement bin the cement will hang up on one side 
and all at once give way and sweep clear to the other side, across 
to the opposite side wall of the tank, and put a strain there that 
would have to be figured on. I was wondering whether there was 
anything like that in the bins. 

Mr. HEmENREiCH. — This only happens with wet oats; they Mr. Heidenreich. 
act very much like cement in bulk. On account of that action 
instead of building the hoppers converging to the center in a 
cement storage elevator at South Chicago, they were built diverg- 
ing to the sides, like over a peak in the center of the bin, whereby 
the hoppers diverge downward toward the circumference, so that 
the cement in sliding down the hopper would, as it were, spread 
itself around to the edges. That seemed to help considerably 
against the wedging proclivity of the cement at the top of the 
hopper. 

Mr. F. L. Williamson. — In connection with the storage Mr. wiuiamson. 
of cement it might be interesting to note that at our plant, cement 
is being stored in circular tanks, in interface tanks and also in 
square tanks, all of reinforced concrete construction. The storage 
tanks are 80 ft. deep and 30 ft. in diameter. The interspace 
tanks, of course, are of corresponding size and three square tanks 
are 8 ft. square and 64 ft. deep. They have been in use for two 
years and have proven satisfactory. The strain Mr. Kinney 
speaks of has no doubt been exerted many times and with no 
damaging effects. 



Digitized by 



Google 



REPORT OF THE COMMITTEE ON MEASURING 
CONCRETE. 

Engineers and contractors for reinforced concrete structures 
have long felt the need of some standard method of measuring 
quantities in contracts for concrete work. This is especially 
the case in drawing up and adjusting unit price contracts. At 
present, although mo^t engineers and contractors measure and 
estimate on the same general principles, they nearly all differ 
in the details of their work, with the result that disputes occur 
and sometimes serious loss is occasioned or injustice done. A 
feeling exists that some of the principles or fundamental rules 
of measurement are founded on a wrong basis and need careful 
consideration and revision. With this object in view this Com- 
mittee was appointed last year to consider the matter and now 
submits its report. 

In submitting these Proposed Standard Methods the Com- 
mittee wishes to call attention to the purpose for which they are 
intended and the principles by which they were guided in fram- 
ing them. 

The purpose is, first, to establish a correct method for award- 
ing unit price contracts and for measuring up work performed 
under the same; second, to inform concrete contractors and 
engineers of the best methods of estimating their work and work- 
ing up unit costs; third, it the above two objects are attained 
there is likely to be greater uniformity in published cost data. 

The principles which have guided the Committee in draw- 
ing up these rules are of great importance and are as follows: 

First, — All work shall' be measured net as fixed or placed 
in the building or structure, and therefore material cut to waste, 
voids, temporary work, etc., shall not be allowed for in measure- 
ment but in price. 

Second. — In no case shall non-existent material be measured 
to pay for extra labor in different parts, but such diflScult or 
expensive parts or extra labor shall be separately measured 
and described. 

(290) 



Digitized by 



Google 



Report of Committee on Measuring Concrete. 291 

Third, — That all the chief items of labor and material entering 
into the cost of concrete work shall be separately measured and 
described by units that correctly represent the labor and material 
involved. 

It follows from the first and second of these principles that all 
work shall be measured as it stands, and that forms or extra labor 
in placing concrete work shall not be allowed for by doubling or 
trebling the measurements of concrete, and that laps in reinforce- 
ment should be measured. 

The acceptance of the third principle involves the recognition 
of the several items entering into the cost of a mass of concrete 
as separate and distinct operations, viz: 

(1) Concrete mixing and placing. 

(2) Forms. 

(3) Reinforcement. 

(4) Surface finish. 

The separation of forms from concrete is not in opposition to 
the first principle of omitting incidental work, as forms should be 
considered as an item of labor. The labor of supporting wet 
concrete by means of forms is entirely distinct from the labor 
of mixing and placing concrete and is done by different men at a 
higher rate of pay. It is a distinct operation in the progress of 
the work and not a labor incidental only to the placing of the 
concrete. 

The committee realizes that the proposed methods if generally 
adopted will be used and administered by inspectors, superin- 
tendents and foremen, as well as by engineers, architects and 
contractors. There are some who will search for opportunities 
for taking imfair advantages of the methods for their own profit 
while many will catch the spirit and use them fairly. It is, there- 
fore, very necessary that care should be taken to provide no oppor- 
tunity for fraud or unfair dealing. 

In framing methods of measurement the Committee has felt 
that it is not necessary to give instructions as to how to measure, 
but only as to what units should be used in measuring and what 
items should be included or left out of such measurements. For 
instance, it is necessary to say whether I-beams shall be deducted 
or not deducted from the mass of concrete work, but it is not 



Digitized by 



Google 



292 Report of Committee on Measuring Concrete. 

necessary to give a method as to how to measure the irregular thick- 
ness of a tunnel lining or of the backing to a masonry wall. The 
latter are matters of mensuration and outside the scope of this 
report. 

The Committee has endeavored to keep to methods of measure- 
ment only as distinct from methods of cost, and in general no 
instructions are given as to the way to fix the price or cost of any 
item, or as to how to make up any prices for work measured under 
these methods. For instance, the cost of plant is added by some 
contractors to the per yard cost of concrete while by others it 
is taken as a lump sum at the end of their estimate Nothing 
is said about this and in the committee's judgment should not be. 

The Committee wishes to lay very special emphasis upon the 
classification of concrete, forms, reinforcement and surface finish 
into separate items. It is realized that especially in the case of 
forms a radical departure from the present method of measuring 
imits is recommended, but after careful consideration the Committee 
feels convinced that the present method of including forms and 
finish and sometimes steel in the cubic measurement of mono- 
lithic concrete is fundamentally wrong and should be altered. 
A few instances will be cited to make this point clear. 

Supposing a contractor takes a contract for a building with 
unit prices per cubic foot for concrete in floor, columns and foun- 
dations, the said unit to include the cost of forms and granolithic. 
If the engineer decides to reduce the thickness of the floor slab 
from 5 to 4 in. the contractor has to put in just as much form work, 
but is paid for only 80 per cent of the form work he estimated 
upon. If the building was increased from two stories to three 
stories in height and floors were finished in granolithic he would 
have to finish twice as many floors with granolithic as he estimated. 
Or vice versa, suppose a wall shown 8 in. thick is increased to 12 in. 
thick, the owner has to pay for 50 per cent more forms and surface 
finish than before, although the amount done is the same. In a 
dispute recently settled in the courts a contract provided for a 
certain price per cubic yard being paid for concrete foundations, 
this price covering the cost of concrete, forms and steel. In the 
execution of the work it was found necessary to make the footings 
deeper than shown, but most of the steel was omitted. The 
contractor claimed that though the steel was omitted his contract 



Digitized by 



Google 



Report op Committee on Measubing Concrete. 293 

price per yard of concrete was the one he was entitled to be paid 
on even if some of the steel was left out. The court upheld his 
view and he received judgment accordingly. 

These instances are t3rpical of the gambling nature of the pres- 
ent system of letting and taking contracts for concrete at an 
inclusive unit price per cubic foot, and the Committee feels con- 
fident that if the methods suggested are adopted they will be 
of real value to the community, eliminating such uncertainties 
and inequalities. Very few contractors in the present day like 
to take these gambling contracts. 

In other respects the Conunittee has endeavored to conform as 
far as possible to the general practice in this country and desires 
to make it standard. 

Some of the detailed points involved in the methods submitted 
will now be considered. 



I. Monolithic Concrete. 
(a) Concrete. 

The rule stating that concrete in diflferent parts of the build- 
ing or structures shall be measured and described according to 
its accessibility and location and the purpose of the work would 
indicate that there should be separate measurements appearing 
in the bill of quantities for: 

Concrete in footings. 

Concrete in columns. 

Concrete in floors, beams and girders. 

Concrete in paving. 

Concrete in basement walls. 

Concrete in curtain walls and partitions. 
And so on according to the nature of the work. 

Concrete in mass foundations. 

Concrete in abutments. 

Concrete in arch ribs. 

Concrete in spandrel walls. 

Concrete in bridge floors and beams. 

Concrete in parapet walls and cornices. 
And so on according to the nature of the work. 



Digitized by 



Google 



294 Repobt op Committee on Measuring Concrete. 

In the event of any of these items, such as columns, foot- 
ings, etc., being of different mix on different floors or places, 
these also would be measured separately (see Appendix I). 

The matter of placing bolts, inserts, pipes, etc., in the con- 
crete was considered. It is not covered by any rule, as the Com- 
mittee held that it was not strictly a concrete item. The placing 
of such items should be taken by the number or the lineal foot 
as the case may be and not allowed for in the price of concrete 
or forms. The cost of placing is part of the cost of the item 
itself, and as such would come under the heading of plumbing, 
steel work and so on. 

(6) Forms. 

After careful consideration the Committee feels that the 
square foot of the surface supported is the correct unit for the 
measurement of forms. In their judgment the item which should 
be measured and paid for is the operation of supporting the wet 
concrete imtil it is set. This item is practically an item of labor 
although material enters into the cost of same. 

It is the practice of some firms to estimate forms by the 
amount of lumber used, but this is not a correct unit of measure, 
not only on the theoretical ground stated above, but on the practi- 
cal ground that no two contractors would use the same amoimt 
of lumber in erecting a piece of form work, and if lumber on imit 
price contracts were measured and paid for by the board foot 
there would be an incentive to a contractor to use more lumber 
than was necessary to do the work. 

It is necessary to give a definite unit which everyone knows 
how to measure. It is not possible to determine beforehand 
just how much lumber will be used in any piece of form work. 
It is not possible to accurately measure the amount of lumber 
used during the work, and it is not possible to take this amount 
from the lumber bills because all lumber bought on a job does 
not go into forms, some of it going into temporary buildings, 
sheeting and shoring and many other necessary parts. For 
instance, when the sheet piling of trenches is done, the lumber 
often goes into forms, and of 1,000 board feet of roofers delivered, 
half may go to the forms and half to temporary buildings. Fur- 
ther, much lumber is used two and three times. Other parts 



Digitized by 



Google 



Report of Committee on Measuring Concrete. 295 

are used but once only and some pieces are sawed off as waste 
and not used at all, and if a rule were to be framed that would 
measure lumber, how should all these difficulties be provided for. 

Paying for lumber on a unit price contract gives to an un- 
scrupulous man an opportunity to claim a higher payment from 
the owner, or in the case of cost data, for a foreman to deceive 
his employers as to the cost of the form work he is doing. A 
superintendent can put in twice as much bracing and posts as 
he actually needs and make his costs look low, and if the con- 
tractor is paid by the board foot he also would receive payment 
for unnecessary work. 

In contrast to these difficulties the square foot measure- 
ment of form work does not alter and is easily and quickly meas- 
ured; and further, even those who estimate by the board foot 
have to determine the number of square feet to be supported 
before computing the number of board feet to be required in 
the building. 

In the measurement of iSoor forms the sides of beams are 
added to the measurement of under side of slab and beams, 
although some contractors take a flat measurement of floor sur- 
face. The latter method would work out unjustly in the case 
of a change being made in the depth of the beams, such change 
working to the detriment of either the owner or the contractor, 
while if the method that the Committee recommend is adopted 
such a change will adjust itself. 

Some discussion may be raised on the omission of any allow-, 
ance for angle fillets to columns and girders, but the Committee 
as practical concrete men realize that such items are a very small 
part of the cost of forms and it is not a usual practice on the part 
of contractors to estimate them separately. They believe that 
to measure angle fillet by the lineal foot, as has been suggested, 
would prove a possible source of misunderstanding in the carry- 
ing out of a contract. 

The separation of forms to floors, columns, footings, etc., 
follows the lines laid down for the separation of concrete items. 

In the case of forms to the concrete walls poured as backing 
to granite or other facing, the correct interpretation of the rules 
would be to measure forms to one face only, the stone facing 
doing the work of forms on the other side. 



Digitized by 



Google 



296 Repobt or Committeb on Measttbing Concbete. 

(c) Reinforcement. 

Some may question the correctness of making an allowance 
for laps or passings in steel rods or fabric for reinforcement. The 
Committee after careful consideration felt that these ought to 
be allowed. Laps are for two purposes, one as in beam steel to 
take up negative moment, and the other to provide suflScient 
bond to take up the tensile strength in two bars where one bar 
cannot be obtained long enough for the purpose. In the first 
case no question would arise as to measuring the full length of 
each rod; in the second, we think that as the specification usually 
specifies the lap to be a certain number of diameters, usually 
forty or fifty, there would be little difficulty on this head. 

There is, of course, an opportunity for unfair dealing by a 
contractor in putting in steel in short lengths. A rule not allow- 
ing lap would sometimes work hardship the other way if an in- 
spector insisted that steel should go in in short lengths instead 
of long, and then only pay for a rod the net length of the build- 
ing. An example would be in a floor of a building laid out in 10 ft. 
bays where rods could be put in one length over one, two or three 
bays as desired. 

The CJommittee feels that the rule they have formulated 
should stand. It is in accordance with the first fundamental 
principle laid down and all work should be measured net as fixed 
in place. They believe that any engineer would use reasonable 
discretion in refusing to measure laps if it was apparent that too 
many had been made for an unfair purpose. 

III. Structural Cast Concrete. 

The use of structural cast concrete is growing rapidly, is of 
an entirely different nature than monolithic concrete, and the 
Committee has endeavored to treat this part of the subject in 
a way similar to which structural steel is measured and estimated. 

It is, therefore, not suggested that forms be measured sepa- 
rately, but recommended that erection should be separate from 
making and that the unit of weight be the correct unit for measur- 
ing erection rather than the unit of volume. 

In measuring or computing erection quantities, it is recom- 
mended that an arbitrary weight of 150 lbs. per cu. ft. be adopted 



Digitized by 



Google 



RePOBT op COMMITTEk ON MEASURING CONCRETE. 297 

(similar to the rule for steel). This method will save the ex- 
pense of weighing each piece and a good deal of dispute. This 
rule would not apply to cinder concrete, for which another weight 
should be agreed upon. 

In buildings built partly of cast concrete and partly of mono- 
lithic concrete, monolithic concrete would be measured under the 
rules laid down for same and the cast concrete under the structural 
cast concrete rules. 

The Committee doesnot recommend that grouting be measured. 
This is looked upon as a small labor item incidental to the erec- 
tion of the concrete members and similar in character to riveting 
on structural steel work or mortar in laying cut stone. 

IV. Cast Concrete Trim and Ornamental Work. 

Cast concrete trim and ornamental work is more nearly 
akin to cut stone work than to any other trade. The Committee 
has followed the custom of the cut stone trade and based the rules 
on stone mason rules, viz: to measure the smallest rectangular 
solid out of which a piece can be taken for any piece of trim, in- 
stead of measuring the net volume of the finished block. 

Surfacing is generally done in the mold and therefore should 
not be separated. 

Temperature reinforcement is a very small part of the cost 
and for that reason is not separated. 

Trim serving any structural purpose such as lintels with 
reinforcement in same to take tensile strength should be classi- 
fied as structural cast concrete and measured accordingly. 

As the stone mason's unit of erection is the same as the unit 
by which he supplies the stone, the Committee does not suggest 
that erection be separated from making. 

The Committee has not at present drafted rules covering 
plastering, waterproofing, concrete blocks, concrete piles, etc., 
except in so far as they are covered by the general rules governing 
surface finish, concrete trim and structural cast concrete. 

Plastering is understood to be any surface coating of cement 
or lime mixed with fine aggregates on soffits or vertical surfaces 
which is put on by hand without the use of forms. 



Digitized by 



Google 



298 Report of Committee on Measuring Concrete. 

The rules of structural cast concrete would apply to con- 
crete piles cast and driven, except the driving is usually measured 
by the lineal foot and not by the pound, and the Committee does 
not suggest that this practice should be altered. 

Respectfully submitted by the Committee on Measuring 
Concrete. 

Robert A. Cummings, Chairman. 

L. H. Allen, 

Chas. Derleth, Jr., 

H. H. Fox, 

Thos. M. Vinton. 



Digitized by 



Google 



Repobt of Committee on Measuring Concrete. 299 



Appendix I. 

BILL OF QUANTITIES FOR A BUILDING. 

Measured by Proposed Methods. 



Concrete 1 : 2| : 5. 



CJoncrete 1:2:4. 



Concrete 1 
Cinder 

Concrete 1 
Forms. 



1J:3. 



;3:6. 



Footings 600 cu. yds 

Basement walls: 220 '' 

Base to paving 90 " 

Floors and beams 2,100 " 

Colunms 650 " 

Partitions and curtain walls 200 '' 

WallSi roof and pent house 30 " 

Cornice and parapet 100 " 

Columns 110 " 



Reinforcement. 



FiU between screeds (2 in. thick) 225 

Footings l.OOOsq. 

Floor slabs and beams 15,500 

C(dumns 3,500 

Curtain walls and partitions 3,350 

Pent house 900 

Basement walls 3,000 

Parapet and back of cornice 1,200 

Face of cornice 36 in. girth 400 lin. 



ft. 



ft. 



Plain round bars, in- 
cluding cutting, wir- 
ing, placing 



li in. dia. 
1 " 



Plain round bars, in- 
cluding cutting, wir- 
ing, placing, but also 
including fabricating 
Square twisted bars, 
including cutting, 
wiring and placing. 

1 in. thick, laid integral with paving 

1 in. thick, laid on floors after concrete 

has set 15,000 

On cornice and parapet 600 

On window sills 1,200 

Picked face to concrete smiace 2,000 

Rubbed face and cement wash one coat on concrete surface. . . 3,000 



Granolithic Finish. 



i 
11 

1 

i 

I 

1x1 

ix} 



5,000 lbs, 

7,000 •' 

12,000 " 

24,500 " 

8,000 " 

17,800 " 

28,500 " 

6,000 " 

3,000 " 

8,000 " 

4,000 " 

5,000 sq. 



ft. 



Digitized by 



Google 



300 Report op Committee on Measuring Concrete* 

Appendix II. 

BILL OF QUANTITIES FOR A BRIDGE. 
Measured by Proposed Methods. 

Concrete 1:3:6. Cyclopean masonry with 30 per cent 

rock in foundation 2,000 cu. yds 

Abutments 1,500 " 

Piers 1,000 " 

Abutments (upper part) 500 " 

Piers (upper part) 500 " 

Arch ribs 1,200 " 

Columns above same 150 " 

Spandrel walls and wing walls 250 '' 

Bridge floor and beams 600 *' 

Forms Abutments (below finished grade) 12,000 sq. ft. 

Abutments (above finished grade) 6,000 '' 

Piers (below finished grade) 10,000 " 

Piers (above finished grade) 8,000 " 

Arch ribs 15,500 '* 

Columns 1,200 " 

Spandrel walls and wing walls 5,000 '' 

Bridge floor and beams 12,000 " 

Parapet wall 2,500 " 

Coping of same 300 lin. ft. 

Reinforcement. Plain round bars 1 Jin 30,000 lbs. 

" " " 1 " 50,000 " 

" i'' 75,500 " 

" " " 1" 12,000 " 

" " " I" 10,000 " 

Crandalled Finish. Piers and parapet 2,500 sq. ft. 

Rubbed Finish. Arch ribs (face and soffits) 40,000 " 

Granolithic Finish. 1 in. thick to sidewalks to bridge floor 

laid integral with slab 2,000 " 

Curb and gutter of same 300 lin. ft. 

Sidewalk. To approaches with 4 in. base 1 : 21 : 5 
and 1 in. top and includes cinder foun- 
dation 800 •' 

Curb and gutter to last, curb 10 in. high, 
gutter 12 in. wide including forms and 

finish done in one operation 200 " 

Extra for rounded corners to same 3 ft. 

girth 4 

Cast Concrete. Balusters 8 in. x 8 in. x 2 ft. high, includ- 
ing forms and steel and setting in place 50 
Coping to same 18 in. x 9 in. with molded 
edges, including forms, steel and finish 
and setting in place 200 lin. ft. 



Digitized by 



Google 



PROPOSED STANDARD METHODS FOR THE 
MEASUREMENT OF CONCRETE WORK.* 

The following divisions are recognized as separate and dis- 
tinct items in the construction of concrete work for which separate 
modes of measurement are necessary. 

I. Monolithic Concrete: 

(a) Concrete. 
(6) Forms. 

(c) Reinforcement. 

(d) Surface Finish. 

II. Sidewalks. 

III. Structural Cast Concrete: 

(a) Concrete. 

(6) Reinforcement. 

(c) Erection. 

IV. Cast Concrete Trim and Ornamental Work. 

The following general rules shall govern the measurement 
of the above items (with the exceptions where specifically 
noted) : 

(a) All work shall be measured net as fixed or placed in the 
structure. 

(6) In no case shall non-existent material be measured to 
cover extra labor. 

(c) No allowance shall be made for waste, voids, or cutting. 

I. Monolithic Concrete. 
(a) Concrete. 

1. The unit of measure for all concrete shall be the cubic foot. 

2. In no case shall the measurement of concrete be held to 
include the forms. 



*'Tbe proposed method* ■ubmitted by the Committee were disouwed. refened baok to 
the Committaa aod appear here u amended. — Ed. 

(801) 



Digitized by 



Google 



302 Standard Methods for Measuring Concrete Work. 

3. AH concrete shall be measured net as placed or poured in 
the structure. 

4. In no case shall an excess measurement of concrete be 
taken to cover the cost of forms or extra labor in placing. 

5. All openings and voids in concrete shall be deducted with 
the following exceptions: 

(a) No deduction shall be made for reinforcement, I-beams, 
bolts, etc., embedded in concrete except where a unit has a sec- 
tional area of more than 1 sq. ft. 

(6) No deduction shall be made for pipes or holes in concrete 
having a sectional area of less than 1 sq. ft. 

(c) No deduction shall be made for chamfered, beveled or 
splayed angles to columns, beams and other work, except where 
such chamfer, bevel or splay is more than 4 in. wide measured 
across the diagonal surface. 

6. Each class of concrete having a different proportion of 
cement, sand or aggregate shall be measured and described 
separately. 

7. Concrete in the different members of a structure shall be 
measured and described separately according to the accessibility, 
location or purpose of the work. 

8. Concrete with large stones and rocks embedded in same 
(cyclopean masonry) shall be measured as one item and described 
according to the richness of the mix and the percentage of rock in 
same. 

9. Concrete in stairs shall be measured by the cubic foot 
and shall include surface finish when the mixture is the same 
throughout. 

(6) Forms. 

10. The unit of measure for form work shall be the square 
foot of actual area of the surface of the concrete in contact with 
the forms or false work. 

11,' Forms shall in every case be measured and described 
as a separate item and in no case shall the measurement of concrete 
be taken to include forms. 

12. No deduction shall be made in measurement of surface 
of concrete supported by forms, because of forms being taken 
down and re-used two or three times in the course of construction. 



Digitized by 



Google 



Standard Methods for Measuring Concrete Work. 303 

13. The unit price for superficial measurement of forms 
shall be deemed to include the cost of struts, posts, bracing, 
bolts, wire ties, oiling, cleaning, and repairing forms. 

14. No distinction shall be made between wood and metal 
forms. 

15. Forms to different parts of a structure shall be measured 
and described separately according to the position in the structure, 
accessibility, purpose and character of the work involved. 

16. No allowance shall be made for angle fillets or bevels 
to beams, columns, etc., but curved moldings shall be measured 
and described separately as hereinafter provided. 

17. No deduction in measurement of forms shall be made 
for openings having an area of less than 25 sq. ft. 

18. No deduction shall be made in floor forms for heads 
of columns of any shape. 

19. No deduction shall be made in column and girder forms 
for ends of girders, cross beams, etc. 

20. No allowance shall be made for hand-holes in column 
forms for clearing out rubbish. 

21. The measurement of column forms shall be the girth 
of the four sides or circumference multiplied by the height from 
the floor surface to the under side of floor slab above, 

22. Forms to octagonal, hexagonal and circular columns shall 
be measured and described separately from forms to square col- 
umns. 

23. Caps and bases to columns and other ornamental work 
shall be measured by number and fully described by overall 
dimensions. 

24. The measurement of beam forms shall be the net length 
between columns multiplied by the sum of the breadth and twice 
the depth below the slab, except for beams at edge of floor or 
around openings which shall have the thickness of floor added 
to the sum of the breadth and twice the depth. 

25. Wall forms shall be measured for both sides of concrete 
wall. 

26. Allowance shall be made by number for pockets left 
for future beams. 

27. Moldings in form work shall be measured by the lineal 
foot. 



Digitized by 



Google 



304 Standard Methods for Measuring Concrete Work. 

28. Forms to circulax work shall always be measured sepa- 
rately from forms to straight work. 

29. No measurement or allowance shall be made for con- 
struction joints in slabs, beams or arch ribs, to stop the day's 
concreting. 

30. Construction joints or expansion joints to dams and 
other large masses of concrete shall be measured by the square 
foot as they occur. 

31. Forms to cornices shall be measured by the lineal foot 
and the girth stated. (The term girth shall be taken to mean 
the total width of all curved and straight surfaces touched by 
the forms.) Plain forms to back of cornice to be measured 
separately. 

32. Forms to window sills, copings and similar work shall 
be measured by the lineal foot. 

33. Forms to the upper side of sloping slabs such as saw 
tooth roofs shall be measured whenever the slope of such slab 
with the horizontal exceeds an angle of 25 degrees. 

34. Forms to the under side of stairs shall be measured by 
the superficial foot. 

(a) Forms to the front edge of the stairs shall be measured 
by the lineal foot. 

(6) Forms to the ends of steps shall be measured by number. 

(c) Reinforcement, 

35. The unit of measure of reinforcement shall be the weight 
in pounds. 

36. The weight shall be calculated on the basis of a square 
rod 1 in. X 1 in. x 12 in., weighing 3.4 lb. 

37. Steel rods for reinforcement shall be measured as the 
net weight placed in the building. 

38. Deformed bars shall be measured separately from plain. 

39. No allowance shall be made for rolling margin. 

40. No allowance shall be made for cutting or waste. 

41. No allowance shall be made for wire ties, spacers, etc. 

42. No separation shall be made according to accessibility, 
location and purpose of reinforcement except in special cases. 

43. In measuring reinforcement the rods shall be measiured 
by the lineal foot as laid. All laps shall be allowed for. 



Digitized by 



Google 



Standard Methods for Measuring Concrete Work. 305 

44. The rods of each dififerent size shall be measured and 
described separately. 

45. Bent bars shall be measured separately from straight 
bars. 

46. Pipe sleeves, tumbuckles, clamps, threaded ends, nuts 
and other forms of mechanical bond shall be measured separately 
by number and size and allowed for in addition. 

47. Wire cloth, expanded metal and other steel fabrics 
sold in sheets or rolls shall be measured and described by the 
square foot. The size of mesh and weight per square foot of steel 
in tension shall be stated. No allowance shall be made for waste, 
cutting, etc., but all laps shall be measured and allowed for. 

(d) Surface Finish. 

48. The unit of measure for finish of concrete surfaces shall 
be the square foot. Finish shall always be measured and described 
separately. 

49. No measurement or allowance shall be made for going 
over concrete work after removal of forms and patching up voids 
and stone pockets, removing fins, etc. 

60. Granolithic finish shall be measured by the square foot 
and shall include all labor and materials for the thickness speci- 
fied. 

51. Finish laid integral with the slab shall be measured 
separately from finish laid after the slab has set. 

52. No allowance shall be made for protection of finish with 
sawdust, sand or tenting. 

53. Grooved surfaces, gutters, curbing, etc., shall be meas- 
ured separately from plain granolithic and shall be measured by 
the square foot or lineal foot as the case may require. 

54. The following shall be measured by the square foot: 

Cement wash. (State how many coats.) 

Rubbing with carborundum. 

Scrubbing with wire brushes. 

Tooling. 

Picking. 

Plastering. 

Etc. 



Digitized by 



Google 



306 Standard Methods for Mbasumng Concrete Work. 

II. Sidewalks and Pavements. 

55. Sidewalks and pavements shall be measured by the 
square foot. 

56. The one measurement shall include concrete, finish, lin- 
ing in squares and cinder or stone foundations. 

57. Curbs and curb and gutter work shall be measiu'ed by 
the lineal foot and separated according to character and size, 
and shall include foundations, forms, finish and cost of special 
tools if any. 

58. In measuring curbs the full height and width or thick- 
ness of same shall be taken, but the measurement of sidewalks 
shall also be taken the extreme width of horizontal surface. 

69. Circular comers to curbs and gutters shall be measured 
separately by number, stating radius and length measured on 
the curve. 

60. Vault lights shall be measured by the square foot, the 
measurement to include glass, forms, steel and finish. Beams 
under vault lights shall be measured by the lineal foot. In measur- 
ing vault lights the measurement shall go at least 4 in. beyond 
the outside line of the glass in each direction. 

III. Structural Cast Concrete. 
(a) Concrete. 

61. The term structural cast concrete is taken to mclude 
unit construction by the various systems. 

62. The unit of measurement for structural cast concrete 
shall be the cubic foot, and shall be measured net as provided 
for monolithic concrete. 

63. The various members shall be measured on the ground 
before erection. 

64. No measurement shall be taken of forms. 

(6) Reinforcement. 

65. Reinforcement shall be measured separately as pro- 
vided in Paragraphs 35 to 47, inclusive. 

(c) Erection. 

66. The unit of measure for the erection of structural con* 
Crete shall be the weight of the finished member in pounds. 



Digitized by 



Google 



Standard Methods for Measuring Concrete Work. 307 

67. In measuring the erection of structural cast concrete 
having a crushed stone or gravel aggregate, the concrete shall 
be assumed to weigh 150 lb. per cu. ft. 

68. No measurement shall be taken of the grouting in 
structiu-al cast concrete. It shall be deemed to be covered in 
the price of erection. 

IV. Cast Concrete Trim and Ornamental Work. 

69. Cast concrete trim shall be measured by the cubic foot, 
but the measurement shall be the smallest rectangular solid 
that will contain the piece measured and not its actual content. 

70. No allowance shall be made for forms. 

71. No allowance shall be made for reinforcement in trim 
and ornamental work. 

72. No allowance shall be made for surface finish in trim 
and ornamental work. 

73. Circular work shall be measured separately from other 
work. 

74. Mitre blocks and end blocks for cornices, etc., shall 
be measured separately from straight molded work. 

75. Vases, seats, pedestals, balusters and other similar 
items shall be measured by number and description with over- 
all dimensions. 



Digitized by 



Google 



CONCRETE RETAINING WALLS. 

By J. M. Meade.* 

Concrete walls are now mostly built of two types, viz., plain 
and reinforced. They are very popular for track elevation, depres- 
sion and dike work, especially where ground is valuable and they 
replace earth dikes to economize space, etc. 

PLAIN MONOLITHIC CONCRETE RETAINING WALLS. 

The plain concrete walls are designed for what is known 
as gravity section, being heavy enough so that their weight and 
stability will stop them from overturning. The best authorities 
figure the width of base of such walls as 0.45 up to 0.65 of their 
height, varying the width according to circumstances. Where a 
retaining wall of this type is built up close to the end of ties, as 
on an elevated road, it becomes a surcharged wall. When designed 
as 0.45 1 have seen them fail by pushing over and would recommend 
not less than 0.65 of the height for a surcharged wall, finishing 
18 in. wide at the top. The common practice in railroad work 
of using arbitrary ratios of width of base to height of walls tends 
to cause a neglect of the study of the proper distribution of the 
pressure on the foundation and it seems to be difficult to get away 
from such practice. It is a well-known fact that movement from 
the original alignment, due to unequal settlement, is the most 
common cause of failures or defects. The writer has in mind some 
flagrant cases of this kind in. the City of Chicago that have caused 
:he owners a very heavy expense. This question is one of great 
importance and each particular case should be carefully investi- 
gated and studied, so the amount and distribution of the pressure 
on the foundation may be accurately determined. 

Many walls of poor design have come to the attention of the 
writer, there being entirely overlooked, due to a lack of analysis 
of the design, the most effective section and minimum amount 
of material for an economic design. 



* Engineer, EaBtern Linee, Atchison, Topeka and Santa F6 Railway, Topeka, Kan. 

(308) 



Digitized by 



Google 



Meade on Concrete Retaining Walls. 



309 



In constructing retaining walls, it is of the greatest importance 
that serious thought be given to the matter of earth filling and 
embankments behind the walls. The drainage is quite easily 
accomplished by filling or placing close up to the back of the wall 
some open or porous material, such as crushed or refuse stone, 
large size gravel, brick bats, etc. ; cinders will also do considerable 
good. Weep holes should be placed in the wall of 3 or 4 in. drain 




'A 






f3 



FIG. 1. — REINFORCED CONCRETE TYPE OF RETAINING WALL AS USED IN RAIL- 
ROAD WORK. 

tile, vitrified, about 15 or 20 ft. apart, according to conditions, 
extending the blind drains to a point near the top of the wall if 
circumstances seem to warrant. It is also quite important in 
plain concrete retaining walls, to use expansion joints about 30 ft. 
apart, of the dovetail pattern. If this is not done temperature 
stresses will crack the wall. Such expansion joints are a good 
investment. 



Digitized by 



Google 



310 Meade on Concrete Retaining Walls. 

reinforced concrete retaining walls. 

Reinforced concrete retaining walls consist of either a thin 
vertical plane attached to a horizontal base and well braced by 
counterforts on the back and buttresses in front or they may 
be designed as cantilevers, in which case the wall is connected to 
a wide base resembling an inverted T. 

It has been found in actual practice that reinforced concrete 
walls are more economical than the plain monolithic gravity walls, 
as the material in the latter type cannot be fully utilized for the 
reason that the section must be made heavy enough so that the 
dead weight and size prevent overtmning. On the other hand, 
in reinforced concrete walls a part of the retained material is 
used to prevent overturning and the wall only need be made strong 
enough to withstand the moments and shears due to the earth 
pressure. The wall is lighter and exerts less pressure on the ground, 
which, with the opportunity of extending the base of the wall, 
often enables the builder to use ordinary foundations instead of 
piles. 

Reinforced walls allow the use of a more scientific design 
than the gravity walls and have been known to be more reliable 
than the plain concrete. It is quite common practice to make the 
base of these reinforced or cantilever type of walls about 0.60 of 
their height and then reduce in size about as shown by Fig. 1. 



Digitized by 



Google 



DISCUSSION. 



Mr. Willis Whited. — ^I would like to ask Mr. Meade if ui. wwted, 
the long dike at Topeka is built open, exposed to the action of 
floating ice and, so far as he can judge by the experience so far 
had, whether it is necessary to thicken the wall to resist the impact 
of floating ice. We have some work in Pennsylvania on some of 
our highways where it is necessary to use a very long dike which 
will be exposed almost to the top in flood time and there would 
be a good deal of impact from floating ice against it. 

Mb. John M. Mbadb. — ^In a great many instances the Mr. Meade, 
reinforced concrete wall has proven very economical in cities. 
At Topeka, where the floods of 1903 were so disastrous, the 
river was diked on one side with an earth dike and on the other 
side with a reinforced concrete wall, which was largely planned 
on account of the valuable right of way. It is one of the longest 
concrete dikes in this country, being about a mile in length. 
The wall was finished late last fall and of course we have not had 
any tests yet. It is 18 in. thick at the top. The river has only 
been up to above normal or high water a few feet, and the wall is 
set back well from the bank, in some places 30 or 40 ft., so it 
would take a big flood to get above the banks and reach this wall. 
With a heavily reinforced wall the ice would shear off so rapidly, 
going parallel with the wall, as to cause little pounding. 

This work was especially watched with reference to its cost 
of construction, etc., being a new departure in dikes. Another 
departure was the use of Joplin chatts. There have been some 
misgivings as to the reliability, for heavy work, of chatts, a by- 
product in the manufacture of zinc. The chatts can be had for 
about the cost of loading, which is about 15 cents a cu. yd. at 
Joplin and of course railroads do not figure the freight. I had 
occasion to look at a pier built over the Chikaskia River in the 
southwestern part of Kansas on a line used jointly by the Santa F^ 
and the Frisco Raiboads. A stone pier had failed about eight 
years ago and was re-built with Joplin chatts. There was a dam 
below the bridge, so that the pier has been in water from 6 to 8 ft. 
deep all the time. I was very favorably impressed with the results 
of the use of chatts at that place. 

(311) 



Digitized by 



Google 



REINFORCED CONCRETE PILES. 
By Robert A. Cummings.* 

It is well established that concrete piles are a satisfactory 
substitute for wooden piles in increasing the stability of founda- 
tions, whether they are of the cast-and-driven type or the casi- 
in-place type. The cast-and-driven type is made of reinforced 
concrete, molded to the desired shape and cured before being 
driven. The caat-dn-place type is made by forming a hole in 
the ground and filling it with concrete. 

The history of the cdstrand-driven type is intimately con- 
nected with the development of reinforced concrete and in the 
same manner has been embarrassed by patent litigation. In 
March, 1907, a final decision was reached in the British High 
Courts of Justice that the fundamental idea of the reinforced 
concrete pile was covered by Brannon's patent of 1871 and that 
subsequent improvements must be limited to the details of 
reinforcing. 

At the Chicago Convention of this Association in 1909, 
the writer described certain tests and methods of reinforcing 
for increasing the unit value of concrete in compression. It is 
the application of one of these methods and other practical 
improvements that form the subject of this paj)er. 

It is nearly ten years since the writer began the design 
and manufacture of concrete piles. During this period, by a 
process of elimination, an efficient method of reinforcing has 
been developed, which is correct in design and economical in 
cost. 

METHOD OF REINFORCEMENT. 

All piles of the cast-and-driven type must be reinforced with 
longitudinal rods, because the pile is hoisted and handled by a 
line from the pile driver which is fastened at or near the butt. 
Consequently, the pile must sustain its own dead weight while 
being raised, as well as shocks and impact against obstacles, 

* Consulting Engineer, Pittsburgh, Pa. 

(312) 



Digitized by 



Google 



CuMMiNos ON Reinforced Concrete Piles. 313 

before reaching its position in the leads of the pile-driving machine. 
.With the point on the ground and the other end being elevated 
the pile must act and be designed as a beam supporting its own 
dead weight and shocks. The limiting proportions of depth or 
thickness of pile to the unsupported beam length produces heavy 
tensile and compressive stresses, with considerable deflection. 

It has been observed that when handling a pile of this type, 
it rarely fails in compression in the concrete, but cracks are 
usually discovered on the tension side. These cracks can be 
accounted for by the slipping of the usual longitudinal rods used 
in the concrete while the pile is being hoisted. While such cracks 
are not suflSciently serious to condemn the pile, they may affect 
the permanency of the reinforcement. 

In order to overcome this defect all longitudinal rods should 
be anchored at the ends, those at the butt opposite each other 
being bent over into a loop and welded together, while those at 
the point are all brought together and electrically welded into 
one piece. Twisted and deformed rods are . advantageous for 
longitudinal reinforcing, as the allowable bond stress is higher 
than for plain rods. 

The uniform circumferential spacing of longitudinal rods is 
very important, because any side of the pile may be subjected 
to tensile stresses and whichever side is in tension there must 
be sufficient reinforcement in position to take the strain. The 
circumferential spacing of the longitudinal rods can be secured 
by means of a special spacing device placed at intervals of about 
5 ft. throughout the length of the pile. 

The hooping of concrete adds greatly to its ability to resist 
axial loads. Therefore, longitudinal reinforcing should have a 
helical wrapping of wire throughout the length of the pile, the 
pitch of which must not exceed 3 in. This wire wrapping will 
assist in taking care of diagonal stresses resulting from the 
handling of the pile. 

Practical experience indicates that the butt end of the pile 
which receives the impact of the hammer should be especially 
reinforced. This has been done by means of a special reinforce- 
ment consisting of a unit cage of flat bands, held 2 in. on centers 
by a spacing bar for a distance of 2 ft. In the plane of each 
band a flat wire spiral is fastened to the cage. The embedment 



Digitized by 



Google 



314 



CxTifMiNGs ON Reinforced Concrete Piles. 



of this unit cage in the butt of the pile forms a resilient cushion 
to receive the impact of the hammer. In no case, in the driving 
of thousands of piles with this cushion, has the butt itself been 
broken. 

The general design of the pile is shown in Fig. 1. 

METHODS OF DRIVING. 

Piles of the aistrandrdriven type are handled and driven by 
means of an ordinary pile driver. The ability to vary the fall 
of the drop hammer is of great utility in overcoming the variable 
resistances to be met with in the driving. It has been found 



r-^*~-^ 



snoops I'x^' 



XSoffSkelSpirai 
mlpper4Ringi 




Na 5 Spirvl mrt S 'hfch 
FIG. 1. — ^DESIGN OF REINFORCED CONCRETE PILE. 

advisable to increase the weight of the ordinary drop hammer 
in ratio of weight of hammer to weight of pile of from 2 or 3 
to 1, so that the weight of the drop hammer for driving the 
concrete piles will vary from 7000 to 12000 lb. 

Steam hammers are not as efficient or desirable for driving 
concrete piles as are drop hammers. This was shown last fall 
on a contract when a test was made between a steam hammer 
and a heavy drop hammer, under the same conditions, using 
the same kind and size of concrete pile. The heavy drop ham- 
mer did not break a pile, whereas the steam hammer broke sev- 
eral below the cushion. Further, the steam hammer did not 
drive as many piles in a given time as did the drop hammer. 



Digitized by 



Google 



CuMMiNGs ON Reinforced Concrete Piles. 315 

The following explanation is offered as it applies also to 
the driving of the heavy steel casings or core used in the making 
of the cast-dn-place type of pile: The limited fisdl (3 ft.) and light 
ram (3000 lb.) of the steam hammer^ while 'delivering twice as 
many blows per minute as the drop hammer, loses a large part 
of its energy in overcoming the proportionately heavier weight 
of the pile, steel casing or core. 

The following analytical treatment by Mr. Barton H. Coffey, 
of New York, confirms and justifies the practice of the writer 
in using hammers as heavy as 12,000 lb. for driving concrete 
piles and it may be asserted with some feeling of confidence that 
if penetration is to be the gauge for measuring the supporting 
power of a pile, the ratio of the weight of the hammer to the 
weight of the pile or core that is driven must be taken into con- 
sideration. 

The advantage of using a heavy hammer is evident from the 
following analysis: 

w 

Let M * Mass of hammer=— 

9 
" TT = Weight of hammer. 

" X = Weight or mass of pile in per cent of hammer. 

" y «8 Velocity of hammer on striking pile. 

" Vi = Common velocity of hammer and pile. 

Momentum of hammer = ilf 7 (on striking pile. 

" " " and pile at common velocity = Af (i-fX) Vi. 

These are equal, provided no external force acts, which we 
will assimie for the present is the case. 

Then ilf F = M (1+X) Vi] therefore Fi = fx v ^^^ 

The energy in hammer on striking pile is 

»JJIfF» (6) 

The energy in both hammer and pile at common velocity is 
}ilf (1 +X) V* which upon substituting (a) becomes } JIf r-r^ (c) 

The difference between (6) and (c) represents the loss of 
kinetic energy in the system at point of common velocity or 
greatest compression. In other words, the percentage of the 



Digitized by 



Google 



316 CuMMiNGs ON Reinforced Concrete Piles. 

original energy of hammer that has gone into compressing pile 
and hammer^ distorting or crushing them and into heat, or 
broadly, internal work done upon each other by the striking 
bodies. Obviously, the smaller this percentage of internal work 
the less liability there is of crushing or distorting the pile. To 
resume 

(c)-(6)- } ilf V« - i JIf ^ - J ilf F« (l-jij^) W 

This equals the internal work. 

The following table gives internal work for various values 
of X in per cent of the total kinetic energy of the hammer. 

X Internal Work. 

0.25 20 per cent 

0.50 33 

0.75 43 

1.00 50 

1.50..... 60 

2.00 67 

3.00 75 

If the hammer weighs 4 tons and the pile 1 ton, there will 
be 20 per cent internal work at maximum compression, whereas 
if the pile weighs 8 tons the internal work will be 67 per cent. 
There is an external force acting against the pile, t.e., the fric- 
tional and displacement resistances of the earth. 

Two extreme cases may be assumed limiting all others. 

1. There is no external force. There the internal work 
will be simply that necessary to overcome the inertia of the pile 
and put it in motion. In this case the table is rigorously accurate. 

2. The external force is great enough to prevent any move- 
ment of the pile. In this case the entire kinetic energy of the 
hammer goes into the internal work and the relative weights of 
hammer and pile are immaterial. 

All intermediate cases where movement occurs are a com- 
bination of (0 and (2), where obviously it is advantageous to 
employ a heavy hammer. 



Digitized by 



Google 



CuMMiNGS ON Reinforced Concrete Piles. 317 

MANUFACTX7RE. 

The procedure in constructing cast-andrdriven piles com- 
mences with the preparation of the molding bed. This, of course, 
will vary with the site, but it is desirable to select a flat and 
convenient location near the place of driving. It is very important 
that the bed shall be stable, so that settlement due to the weight 
of the piles is avoided. Where the ground is soft and yielding, 
pine stakes 2 in. x 4 in. x 3 ft. long, pointed at the ends, are 
driven to a solid bearing. These stakes are located at intervals 
of about 4 ft. in each direction and the tops cut to a uniform 
level. Then, 4 x 4 in. pine sills are toe-nailed to the top of the 
stakes in longitudinal rows about 4 ft. on centers. Upon these 
sills a 2-in. solid wooden floor is placed, which forms the molding 
bed. It is desirable that this bed shall be uniformly level to 
receive the forms for the piles. 

The forms are made of two pieces of 2 x 8-in. dressed pine, 
battened together and placed on edge to form the sides of the 
pile. The bevels or angles for tapered or octagonal piles are 
made by placing loose pieces of bevelled wooden strips at each 
comer of the form. The reinforcement is delivered on the work 
in factory-made-units, so that it can be placed in the forms at 
once. When a reinforcing unit is suspended and centered in 
position, the concrete of a wet consistency is deposited and care- 
fully puddled. As soon as the concrete of the pile has solidified, 
the forms are stripped and used for making other piles. The 
number of forms required will vary with the quantity of piles to 
be made and the prospective salvage in the lumber. 

The curing of concrete in the normal manner delays the 
driving of the piles for a period of not less than 3 weeks, although 
a greater length of time is desirable, especially in cool and damp 
weather. Therefore, unless a stock of cured piles is always on 
hand, it frequently occurs that this type of pile cannot be used 
at all and resort is had to the use of the cast-dn-place type. This 
practice is open to question on account of the inability of plain 
concrete to resist even moderate tension. In fact, it is almost 
axiomatic that all concrete piles must be reinforced. Every 
concrete pile is subjected to strains that induce very serious 
tensile stresses in the pile. Such stresses may result from super- 
imposed loads or a lateral strain from the soil. Further, the 



Digitized by 



Google 



318 



CuMMiNGS ON Reinforced Concrete Piles. 



making and storing of concrete piles for use at any time necessi- 
tates a large financial investment. 

In order to avoid the above-mentioned objections, the writer 
has adopted the method of steam curing of cast-and-driven piles. 
This enables such piles to be made and driven within 3 or 4 days 
and places the speed of driving on the same basis as that of the 
cast-in-place type. 

The means used for steam curing will vary with circum- 
stances, location of the work, speed required for delivery of the 
piles, the number needed, etc. During the past winter the 
writer has used concrete piles made and driven within 10 days, 
and, as a result of this experience, confidently recommends that 
such piles can be made, cured and driven immediately. 

On the work with which the writer was connected, the piles 



* — ^*^^..> 







PJl ^ ^p^^ ^^-^^^^0^^ 



III 



h -"iJ'-i?'"- - 



_J J a D 






\s>'^ '2> J •: 







EN^Nt«4 



FIG. 2. — SHED FOR STEAM-CURING PILES. 



were allowed to set in a normal manner for 5 or 6 days — and 
were then gently hoisted^ from**^ the molding bed by*a derrick, 
using an equalizing spreader and bridle, the chains of the bridle 
being fastened so that the pile was balanced. They were then 
placed in stacks of 25 or 30 and separated from one another by 
wooden blocks, particular attention being given to securing a 
solid bearing for each pile. 

A light wooden shed, practically steam tight, was built 
entirely around the stack of piles. Fig. 2. The steam was con- 
ducted direct from a boiler through a 1-in. pipe to 3 branch 
openings inside the shed. The steam pipe valve was opened and 
the piles were exposed to live steam for 2 or 3 days, when they 
were found to be ready for driving. On being first exposed to 
steam, the moisture condensed on the surface of the piles and 
remained until absorbed by the concrete when the temperature of 



Digitized by 



Google 



CUMMINGS ON ReINFOBCED CONCRETE PiLES. 319 

the steam was reached. This steam treatment should be distin- 
guished from heat applied indirectly or baking, in accelerating 
the set and hardening the concrete. The writer sees no reason 
why boiling water should not be used for the same purpose if 
the conditions are favorable for adopting this method. Pre- 
caution should be exercised in making sure that the concrete 
has solidified and that it has received its initial set, before expo- 
sure to steam treatment. 

The prospective field opened up by the steam treatment 
for the rapid curing of concrete seems to the writer to solve the 
diflBculty incidental to the present methods of procedure in the 
construction of all classes of concrete structures. 

Attention is directed to the publication of the tests of the 
Structural Materials Laboratory of the United States Geological 
Survey, wherein it is conclusively shown "that a compressive 
strength considerably (in some cases over 100 per cent) in excess 
of that obtained normally after ageing for six months, may be 
obtained in two days by using steam pressure for ciu*ing mortar.'* 

Practicing engineers will have little difficulty in modifying 
the writer's methods and using improved schemes for quick cur- 
ing of the piles with steam; and in this connection it may be of 
interest to state that the writer has already under way the con- 
struction of steel molds and appliances for the steam treatment 
of concrete piles. 

A list of references on Concrete Piles to 1908 is given in the 
appendix. 



Digitized by 



Google 



APPENDIX. 
L18T OF REFERENCES ON CONCRETE PILES. 

Concrete Piles for Sandy Ground. Engineering News, v. 49, p. 275. 
(March 26, 1903.) (Description of the Raymond system of sinking piles 
with water jet. Illustrated.) (1 column.) 

A Concrete-Steel Pile Foundation. Engineering News, v. 49, p. 173. 
(February 19, 1903.) (Description of piles used for the Court House, 
Berlin, Germany. Driven by steam hammers.) (1 column.) 

Concrete Pile Foundations, Carnegie Public Library, Aurora, III. 
Engineering News, v. 48, p. 495. (December 11, 1902.) (Description 
of the Raymond method of using steel core for making holes for concrete 
piles. Illustrated.) (§ column.) 

A New System op Concrete Pile Construction. Engineering News, v. 
45, p. 450. (June 20, 1901.) (Description of the trial of the Raymond 
steel core system, made at Chicago, May 16, 1901. Illustrated.) (1 
column.) 

Concrete Pile Foundations. Engineering News, v. 46, p. 75. (August 1, 
1903.) (Description of the foundations of the nine-story apartment house 
built for W. J. Bryson, Lake Avenue, Chicago. Holes bored by water- 
jet system before Raymond steel core was dropped. Material largely 
sand and quicksand.) (| column.) 

The Hennebique System of Armored Concrete Construction. By 
Leopold Mensch. Journal AssodcUion of Engineering Societies, v. 29, 
p. 108. (September, 1902.) (Contains three pages on concrete-steel 
piles.) 

Abstract of same. Concrete-Steel Piles of the Hennebique System. 
Engineering Record, v. 46, p. 618. (December 27, 1902.) 

Construction in Concrete and Reinforced Concrete. By F. C. Marsh, 
1902. Minutes of Proceedings of the Institution of Civil Engineers, 
V. 149, p. 297. (Gives a short description of piles made after the Henne- 
bique System.) 

Neuere Bauwesen und Bauwerke in Beton und Eisen, nach dem Stands 
bei der Pariser Weltausstellunq 1900. Fritz v. Emperger, Zeit- 
sckrifl des Oeslerreichischen IngenieW' und Architekten-Vereines, 53. 
Jahrgang, pp. 713 and 765. (October 25 and November 15, 1901.) 
(General description of concrete piles and their use.) (Illustrated. 7 
pages.) 

Abstract of same. Concrete-Steel Piles and their Driving. Engineering 
Record, v. 46, p. 560. (December 13, 1902.) (Gives extracts from Mr. 
C. F. Marsh's paper read before the Institution of Civil Engineers and 
from Mr. F. Von Emperger's description of a pile driver from the Zeit- 
schrift des Oeslerreichischen Ingenieur- und Architekten Vereines, Novem- 
ber 7, 1902.) (2 columns.) 

(320) 



Digitized by 



Google 



CuMMiNGS ON Reinforced Concrete Piles. 321' 

Concrete Piles. Railroad GazeUe, v. 34, p. 645. (August 15, 1902.) (De- 
scription of the Raymond System with illustrations from the J. I. Case 
Plow Works, Racine, Wis.) (1 column.) 

The Raymond Concrete Piles. Cement and Engineering News, v. 13, p. 
22. (August, 1902.) (Illustrated description of the Raymond System.) 
(1 page.) 

Concrete Pile Foundations op the Hallenbeck Building, New York. 
Engineering Record, v. 47, p. 377. (April 11, 1903.) (Piles sunk by 
means of water jet. Material, gravel and sand. Illustrated.) (1 page.) 

QUAIMAUERN UND FuTTERMAUERN AUB BeTON UND EiSEN (SySTEM HeN- 

nebique). Zeitschrift des Oesterreichischen Ingenieur- und Architekten" 

Vereines, 53. Jahrgang, p. 539. (August 9, 1900.) (Short descriptions 

of quay walls in Southampton, Paris and Nantes.) (Illustrated. H pages.) 
Ueber Beton-Eisen-Piloten. Zeilachrift des Oesterreichischen Ingenieur- 

und ArchUekien^Vereines, 54. Jahrgang, p. 746. (November 7, 1902.) 

(Illustrated. 2 pages.) 
Bbtoneisen-Ppahlrost vom Neubau des Amtsgerichtes-Weddinq in 

Berlin. Deutsche Bauzeitung, 36. Jahrgang, p. 582. (November, 1902.) 

(1 page.) 
Pfahlrobtkonbtruktionen in Beton-Eisen. Deutsche Bauzeitung, 36. 

Jahrgang, p. 411. (August, 1901.) (Short description of use of concrete 

piles according to Hennebique system for a building for Holland-American 

line in Rotterdam.) (i column.) 
Beton-Pfbiler der SANGAMON-FLUSS-BRtJcKE. Thonindustrie Zeitung, 

26. Jahrgang, 1. Halbjahr, p. 83. (January 17, 1901.) (Description of 

the use of concrete piles for the St. Louis, Peoria and Northern Railroad 

Bridge over the Sangamon River.) (f column.) 

StaHL-BeTON-PfEILER Fto DIE CLYBOURN-PLACE-BRtJCKE IN CHICAGO, IlL. 

Thoninditsirie-ZeUung, 25. Jahrgang, 1. Halbjahr, p. 631. (April, 1901.) 
(Short description of steel-concrete piles for a drawbridge over Chicago 
River.) (i column.) 

EiNRAMMEN VON Betonpfahlen. ThonindiLstrie Zeitung, 27. Jahrgang, 1. 
Halbjahr, p. 296. (February 21, 1903.) (Short description of method 
of driving concrete piles.) (Illustrated.) (i column.) 

Ueber Betonpfahle. Thonindustrie Zeitung, 27. Jahrgang, 1. Halbjahr, 
p. 1106. (June 13, 1903.) (Short description of construction and 
method of driving concrete piles.) (Illustrated.) (§ column.) 

Recent Developments in Pneumatic Foundations for Buildings. By 
D. A. Usina, Associate American Society Civil Engineers. Proceedings 
American Society of CivU Engineers, v. 34, p. 220. (March, 1908.) 
(Contains two pages on the comparison of concrete piles and caissons for 
foundations of buildings.) 

Foundations; an Informal Discussion at the Annual Convention, 
July 10, 1907. Proceedings American Society Civil Engineers, v. 33, p. 
812, 816. (September, 1907.) (Contains some data on concrete piles.) 

CoNCBBTE and Concrete-Steel. Transactions American Society CivU 
Engineers, v. 54, Pt. E., pp. 436, 461, 469, 548, 615. (International 



Digitized by 



Google 



322 CuMMiNGs ON Reinforced Concrete Piles. 

Engineering Congress, St. Louis, 1904.) (Very short references to 
concrete piles.) 

Cyundrical Foundations for a Quay Wall in the Harbor of Dblfzyl. 
By W. F. Druyvesteyn. Transactions American Society Civil Engi- 
neers^ V. 54, Ft. E. (International Engineering Ck)ngres8y St. Louis, 
1904.) (Contains 1 page on concrete piles.) 

Use of Reinforced Concrete in Buildings. By Frank C. Schmitz. 
Proceedings Brooklyn Engineers^ Club, 1906, p. 67. Brooklyn, N. Y. 
1907 . (Contains a short table comparing the price of wooden and concrete 
piles.) 

Driving Concrete Piles Below the Battery Tunnel, New York. 
Engineering Record, v. 65, p. 678. (June 8, 1907.) (The construction 
consists of a series of transverse pile bents at irregular intervaLs of about 
50 feet supporting the undersides of the tubes for two lengths of about 
600 feet each. In each bent there are two reinforced concrete piles 20 
inches in diameter and about 7 feet apart on centers.) 

Grade Correction and File Foundations in the East River Tunnel 
OP THE New York Rapid Transit Subway. Engineering News, v. 57, 
p. 718. (June 27, 1907.) (An illustrated description of the construction 
of the reinforced concrete piles.) 

Concrete Pile Foundations for a Tower 7(X) Feet High. Engineering 
Record, v. 55, p. 531. (April 27, 1900.) (The piles are of the Raymond 
concrete type and are made in the standard manner by ^t driving with 
a solid steel core a thin conical steel shell, which excludes water and sand 
and is filled with concrete after the core is withdrawn.) 

Steamship Terminal with Concrete Pile Piers at Brunswick, Ga., 
Atlantic and Birmingham Railway. Engineering News, v. 56, p. 654. 
(December 20, 1906.) (Gives specifications for concrete piles, illustra- 
tions and method of construction.) 

Cement Piers. Scientific American Supplement, v. 63, p. 26241. (May 25, 
1907.) (The cement cylinders for use in piers at San Francisco are made 
of three wooden piles enclosed in reinforced concrete.) 

Improved System of Concrete Piling. Journal of tke Franklin Institute, 
v. 160, p. 455. (December, 1905.) (A report of a committee on the 
merits of the concrete pile invented by Frank Shuman, illustrated.) 

The Simplex System op Concrete Piling. By Constantine Shuman. 
Proceedings Engineers* Club of Philadelphia, v. 22, p. 347. (October, 
1905.) (lUustrated.) 

Concrete Piles, Description of the Methods of Manufacture and 
Usages of the Two Leading Types op Concrete Piles which are 
Replacing the Wooden Products. By David Lay. Cement Age, 
V. 2, p. 626. (February, 1906.) 

Reconstruction of the Atlantic City ** Steel Pier" in Reinforced 
Concrete. Engineering News, v. 56, p. 90. (July 26, 1906.) (The 
reinforced concrete piles were molded on small pile platforms adjacent 
to the location of the piles in the piers; after hardening the piles were 
lifted from the platforms, set in position and sunk into the sand by 
means of a water jet, having a pressure of 65 lbs. per sq. in.) 



Digitized by 



Google 



CuMMiNGs ON Reinforced Concrete Piles. 323 

Cost of Making and Placing Reinforced Concrete Piles at Atlantic 

City, N. J. Engineering News, v. 56, p. 252. (September 6, 1906.) 
The Manufacture and Use of Concrete Piles. By Henry Longcope. 

Proceedings National Association of Cement Users, v. 2, p. 277. (1906.) 
Abstracts of same. Scientific American Supplement, v. 61, p. 25375. (May 

12, 1906.) Municipal Engineering, v. 30, p. 106. (February, 1906.) 
Reinforced Concrete Pile Foundation for the Lattemann Building, 

Brooklyn, N. Y. Engineering News, v. 54, p. 594. (December 7, 

1905.) (The piles used were of the corrugated form invented by Frank 

B. Gilbreth.) 
The Making and Driving of Corrugated Concrete Piles. By Frank 

B. Gilbreth. Association of American Portland Cement Manufacturers, 

Bulletin No. 7. (1906.) (Illustrated.) 
L'Emploi des Pieux en B£ton pour les Fondations. Le Genie Civil, v. 

49, p. 104. (June 16, 1906.) (An illustrated description of different. 

types of concrete piles.) 
Translation of same by George L. Fowler. The Use op Concrete Piles. 

RaUroad Ga^tU, v. 41, p. 238. (September 21, 1906.) 
Appontement M£tallique de Lome (Afrique occidentale). Le Genie 

Civil, v. 47, p. 178. (July 15, 1905.) (Contains a description of the 

concrete piles.) 
Corrugated Concrete Foundation Piles. Engineering Record, v. 52, p. 

548. (November 11, 1905.) (Describes the method of constructing 

the foundations for the Lattemann Building. Brooklyn.) 
Building and Machinery Foundations in Quicksand. Engineering Record, 

V. 53, p. 248. (March 3, 1906.) (For the Knickerbocker Building, 

New York City, the foundations of the column and wall piers consist 

of clusters of tubular steel piles 12 in. in diameter and { in. thick sunk to 

bed rock and filled with concrete.) 
Concrete Piling. Scientific American, v. 90, p. 248. (March 26, 1904.) 
Die GrI^ndung des Amtsgerichtsgebaudes auf dem Wedding in Berlin 

MIT BetoneisenpfXhlen. By Hertel. Beton und Eisen, v. 2, p. 246. 

(October, 1903.) 
Reinforced Concrete Piling. By A. R. Galbraith. Proceedings, Incor- 
porated AssocicUion of Municipal and County Engineers, v. 31, p. 356. 

(1904-5.) Spon & Chamberlain, 123 Liberty Street, New York. 
Abstract of same. European Reinforced Concrete Piles. Engineering 

Record, v. 52, p. 99. (July 22, 1905.) 
Concrete Piles at the United States Naval Academy. By Walter R. 

Harper. Engineering Record, v. 51, p. 277. (March 4, 1905.) (Gives 

the comparative cost of wood and concrete piles; test of concrete pile 

and methods of construction.) 
The Strength of Pile and Concrete Foundations. Engineering Record, 

V. 50, p. 358. (September 24, 1904.) (Results of experiments made to 

determine the adhesion of timber piles to concrete, when imbedded in 

that material.) 



Digitized by 



Google 



324 CuMMiNGS ON Reinforced Concrete Piles. 

Concrete Pile J'oundations at Washington Barracks, D. C. By Captain 
John Stephen Sewell. Engineering Record, v. 50, p. 360. (September 
24, 1904.) (2 pages.) 

Reinforced Concrete Piles with Enlarged Footings for Underpinning 
A Building. By J. Albert Holmes. Engineering Record, v. 51, p. 567. 
(June 16, 1904.) (Describes the construction of piles for the foundation 
of a building in Boston, Mass.) 

Des Applications du Ciment Arm£. By G. Liebeaux, Reime Generate 
des Clieynins de Fer, v. 24, Pt. 2, p. 525. (December, 1901.) (Contains 
illustrations of concrete piles for foundations.) 

Notes on European Reinforced Concrete Structure. Engineering 
Record, v. 51, p. 38. (January 14, 1905.) (Contains one column on 
ferro-concrete piles.) 

Reinforced Concrete. (Pt. 162, 375.) By Albert W. Buel and Charles 
S. Hill. Ed. 2, N. Y., 1906. Engineering News Publishing Company, 
220 Broadway. $5 net. (Contains data on concrete piles.) 

Abstract of same. The Construction and Use of Concrete Steel Piles 
IN Foundation Work. Engineering News, v. 51, p. 233. (March 10, 
1904.) 

Concrete-Steel Piles. Cement, v. 4, p. 16. (March, 1903.) (A descrip- 
tion of heavy pile drivers designed for driving concrete-steel piles.) 

A Treatise on Concrete Plain and Reinforced. (P. 477.) By Fred- 
erick W. Taylor and Sanford E. Thompson, Associate Members, American 
Society Civil Engineers, N. Y., 1905. John Wiley & Sons, 43 East 
Nineteenth Street. $5. 

Concrete and Reinforced Concrete Construction. (P. 428.) By 
Homer A. Reid, Associate Member, American Society Civil Engineers, 
N. Y. 1907. Myron C. Clark Publishing Company, 13 Park Row. 
$5 net. (Gives data on concrete piles.) 

Reinforced Concrete. (Pp. 41, 181, 200, 449.) By Charles F. Marsh, 
Member American Society Civil Engineers, and William Dunn. Ed. 3, 
London, 1906. Archibald Constable & Co., Ltd., 16 James Street, 
Haymarket. $7 net. (Very short references.) 

Cement and Concrete. (P. 485.) By Louis Carlton Sabin, Member Amer- 
ican Society Civil Engineers. Ed. 2, New York, 1907. McGraw 
Publishing Company, 239 West Thirty-ninth Street. $5. (Contains 
2 pages on concrete piles.) 

Concrete Steel. (P. 169.) By W. Noble Twelvetrees, New York, 1905. 
Whittaker & Co., 64 Fifth Avenue. $1.90. (Contains 10 pages on con- 
crete-steel piles.) 

Reinforced Concrete Piles; Their Making and Driving. Cement, 
V. 4, p. 331. (November, 1903.) 

A New System op Concrete Piles. By W. P. Anderson, Engineering Record, 
V. 50, p. 494. (October 22, 1904.) (Piling for the Dittman factory 
building, Cincinnati, Ohio.) 

Concrete Piles for Building Foundations. Engineering Record, v. 49, 
p. 596. (May 7, 1904.) (2 columns.) 



Digitized by 



Google 



CiJMMiNGs ON Reinforced Concrete Piles. 325 

A Practical Treatise on Foundations. (Pp. 467, 477.) By W. M. 
Patten, Ed. 2, New York, 1906. John Wiley & Sons. $5. (Contains 
information on concrete piles.) 

EiNE NEUERB KaIMAUER MIT ElSENBETON-PFAHLGRt^NDUNG. ZeUschrifi 

fUr Baurvesen, v. 57, p. 550. (Pt. 10-12, 1907.) (On the reinforced 
concrete piles for the foundation of a new quay wall at Dusseldorf.) 
Methods and Cost of Driving Raymond Concrete Piles for a Building 
Foundation. Engineering-CoTitracting. (February 13, 1907.) (Gives 
figures of cost computed from records obtained in constructing the pile 
foundations for a building in Salem, Mass.) 

Kt^NSTLICHE BeFESTIGUNG DES BaUBODENS MITTEL8T SCHWEBENDER PILO- 
TAGE. By Ottokar Stern. Beton und Eisen. (January, 1907.) (Calcu- 
lations and dimensions for concrete piling.) 

Concrete Piles. By Charles R. Gow. Journal Associated Engineering 
Society. (October, 1907.) (Illustrated.) 

The Simplex System of Concrete Piling. By Thomas MacKeller. Jour- 
nal Associated Engineering Society. (October, 1907.) 

PiLOTis "Simplex." Le Ciment. (December, 1907.) (Illustrated.) 

Concrete Piles — Forms, Advantages and Cost as Compared with 
Wooden Piles. By C. W. Gaylord. Proceedings, v. 5, 1909. National 
Association of Cement Users. (21 pages. Illustrated.) 



Digitized by 



Google 



THE HANDLING OF CONCRETE IN THE CONSTRUC- 
TION OF THE PANAMA CANAL. 

By S. B. Williamson.* 

The estimated amount of concrete that will be used in the 
construction of the Panama Canal aggregates 5,416,645 cu. yd. and 
is distributed as indicated in Table I. 

Table I. — Status op Concrete Work on the Panama Canal on 
January 1, 1912. 



Location. 



Terminal Docks at Cristobal 

Gatun Spillway Dam 

Gatun Locks 

Culebra Cut: Revetment 

Pedro Miguel Ix>ck8 

Miraflores Locks 

Miraflores Spillway Dam 

Terminal Docks at Balboa 

Municipal Roservoirs. Dams, Power HoufM>, 
Bridges, etc 

Totol I 



Amounts in Cubic Yards. 


Placed. 


To be Placed. 


Total. 


500 

167.500 

1.761.345 

786."696 
671.860 

'6."535 

22.000 


51.800 
57.580 
238.895 
400.000 
111.993 
840,876 
75,000 
330.065 


62.300 

225.080 

2.000.240 

400.000 

892.689 

1.412.736 

75.000 

336.600 

22.000 


3.310.436 


2.106.209 


6.416.645 



While, no doubt, it is generally known that concrete plays a 
leading part in the building of the Canal, a summary of the 
quantities emphasizes its importance, which becomes still more 
significant when one realizes that the estimated cost of the con- 
crete structures represents 23 per cent of the entire estimate for 
construction and engineering and that, had it been necessary to 
adopt stone masonry, the cost of these structures would have 
narrowly escaped a prohibitive figure, as there is no building stone 
within a reasonable distance from the Canal Zone. It does not 
seem out of the way, therefore, to advance the claim that the 
lock type of canal, now under construction, and acknowledged 
to be the most preferable by all engineers who have given the 
subject careful consideration, became feasible largely through the 
use of concrete. 



* Engineer, Pacific Division. Panama. 



(326) 



Digitized by 



Google 



Williamson on Handling Concrete at Panama. 327 

The locks comprise a large proportion of the total quantity 
of masonry and in order to place the amount of concrete involved, 
economically and at a rate to accomplish the work within the 
allotted time, it was necessary to devise especial and unusual 
handling appliances; and it is the purpose of the writer to describe 
the plants and methods adopted. 

GATUN LOCKS. 

General Description. — ^At Gatun the difiference of 85 ft. between 
sea-level and the lake surface will be overcome by a flight of three 
locks, that is, the locks are directly connected, without inter- 
vening basins, and form a continuous structure. They are built 
in duphcate, which requires the construction of two side walls 
and a center wall, each 79 ft. high. The side walls are 4,038 ft. 
long; the center wall is extended in both directions to provide 
guide walls for vessels entering the locks and is 6,330 ft. in length. 
An outline plan and typical cross-sections of the Gatun flight of 
locks is shown on Pig. 1, Plate I, and 2,000,240 cu. yd. of 
concrete will be used in their construction. 

The lock sites are about 3,300 ft. east of a channel that was 
dredged by the French Company from Colon to Gatun, and as 
the concrete aggregates were to be obtained from points on the 
coast and transported by water to Colon, it was considered advis- 
able to continue this method of transportation to Gatun by utiliz- 
ing the French Canal; the latter, therefore, at once became a 
controlling element in designing the handling plant. The banks 
of the French Canal are composed of an alluvial material, with 
a decided tendency to slide, and were considered unsafe for the 
heavy structures and storage piles required at the unloading point. 
A boat slip was therefore dredged in firmer ground between the 
canal and lock sites and a channel excavated to a connection with 
the French Canal — ^incidentally the movement of the unloading 
point shortened the distance between the storage piles and mixing 
plant. 

Handling Plant. — Referring to Figs. 2 and 3, Plate II, it is seen 
that the entire plant is composed of the following units, each 
having a separate and distinct function: 

1. Facilities for unloading and storing cement, sand and stone. 



Digitized by 



Google 



346 Williamson on Handling Concrete at Panama. 

Berm Cranes, — Two berm cranes only were used at Pedro 
Miguel, but the booms were replaced by the cantilever arms of the 
other two so that, as erected, they were balanced cantilever cranes 
as shown in Fig. 13, Plate VII. Their runway tracks were so located 
between the trestles that each cantilever extended well over the 
respective storage piles; both cantilevers were equipped with trolleys 
operating a 2i-cu. yd. excavating bucket. The bins from the other 
cranes were also placed on these two, and with these and two 
cantilevers the facilities for handling and storing sand and stone 
were doubled. Aside from the above changes the previous 




FIG. 12. — MIXING CRANES AND STORAGE TRESTLES, PEDRO MIGUEL LOCKS. 

description of these cranes applies, except also that the swinging 
platforms were omitted and the mixers emptied into buckets on 
cars. 

Narrow-Gauge Road. — The track system of the narrow-gauge 
road used for transporting concrete from the berm to chamber 
cranes is shown in Fig. 11. The difference of 30 ft. in elevation 
between the forebay and lock floors was overcome by means of an 
inclined trestle having a 2.5 per cent grade. The tracks were laid 
with 70-lb. rails which enabled the locomotives to attain a greater 
rate of speed than would be safe on the lighter rails usually 
employed. The equipment included twelve llj-ton Porter 



Digitized by 



Google 



Plate VII. 

Proc. Nat. Assn. Cement Users. 

Vol. VIII, 1912. 

Williamson on the Handling of Concrete 
IN THE Construction of the Panama Canal. 



Digitized by 



Google 



1 I Miiffi 



Digitized by 



Google 



Williamson on Handling Concrete at Panama. 347 

locomotives and 24 steel-framed flat cars, all equipped with air, 
each car being large enough to hold two 2-cu. yd. buckets. 

The trains were composed of 2 cars and each car carried a 
bucket, so placed that when alongside a berm crane each bucket 
was filled from the corresponding mixer without moving the train. 
Usually the trains alternated in going into the respective lock 
chambers and stopped under the first chamber crane; the crane 
placed an empty bucket and picked up a loaded one from the 
same car; the train then moved to the next chamber crane where 
the operation of exchanging an empty bucket for a loaded one 




FIG. 14. — GENERAL VIEW OF LOCKS AND CHAMBER CRANES, PEDRO MIGUEL 

LOCKS. 

was repeated, after which the train of empty buckets returned to 
the mixers. 

Chamber Cranes. — Two chamber cranes, Fig. 14, were erected 
in each lock at Pedro Miguel — they placed the concrete in both 
side and center walls. Except that the long cantilever arms 
extended over the side walls and the short ones over the 
center wall, the description of them given for the Miraflores plant 
applies. 

Concrete Forms, — The same type of forms was used here and 
at Miraflores, in fact, all of the steel forming and some of the wood 
forms were used fiirst at Pedro Miguel and later at Miraflores. 



Digitized by 



Google 



348 Williamson on Handling Concrete at Panama. 

Auxiliary Mixing Plant, — Delays in the delivery and erection 
of the permanent plant, combined with the desire to increase 
the rate of placing concrete led to the erection of auxiliary mixing 
plants for the Pacific locks. At Pedro Miguel a 2-cu. yd. mixer 
was first temporarily set up at the lower end of the west wall and 
later ipoved to a similar position as regards the east wall, where it 
continued work until December 16, 1911. Two 2-cu. yd. cube 
mixers were installed under the south end of the storage trestle 
in the forebay and have continued in operation since. In all of 
the above cases the mixers were on hand, having been ordered for 
the permanent plant as applied to Miraflores where 8 mixers are 
required as against 4 in the application of the plant .to Pedro 
Miguel. In each location the mixers were charged from bins 
to which material was delivered from tracks overhead, by standard 
railroad dump cars, and the product was transported to the walls 
by narrow-gauge equipment. Half-yard portable mixers are also 
used for floor construction and certain portions of the walls where 
they may be set up so as to pour concrete directly into the forms. 

The stationary mixers located in the east wall at Miraflores 
have been previously described. As they are for the purpose of 
feeding the chamber cranes, and it was necessary to purchase 
additional mixers, they really constitute an addition to the per- 
manent plant, made for the purpose of increasing its efficiency. 
Aside from these there is no auxiliary plant used at Miraflores, 
except the half-yard portable mixers. 

PERFORMANCE OF PLANT. 

For the Pacific locks, crushed stone is delivered into cars 
directly from the quarry bins and dumped from the cars on the 
storage trestles. Sand is dredged, loaded into barges and trans- 
ferred from the latter to bins at Balboa with electric cranes, it 
then flows by gravity from the bins into cars for transporting to 
the storage trestles. There is nothing corresponding to the unload- 
ing plant at Gatun, therefore, unless it is the crane for handling 
sand at Balboa. Their performance is given in Table VII. 

Pedro Miguel Locks. 

Auxiliary Plant, — The placing of concrete in the lower guide 
wall at Pedro Miguel began on September 1, 1909, with a mixing 



Digitized by 



Google 



Williamson on Handling Concrete at Panama. 349 



o 

S 
> 

Q 
o 

h 

I 

CO g 

a g 

o -I a 



.S 



a 
eg I 

8 ^ 



<£ 



n 






1^^ 



II 



.s » 

1^ 



ooeo«oaaoo»oo^o»^o 



t^f^cooo6h«t^c0(Ot^tov} 



iQe>«iQe>«i-HQot^^QO-«co 



sggsss 



g^^^r^ss^g^^^^^ss 






^5 



si 

hi 



8$SS88S8SS88 



Q«OQ«O)Ci-iC0Q0C 



3:22§gS8S88888 



85 



8S8 : : : :8 : 



S2«S8SS888888 






O u 



t« 



32^888 



8388SS88 : :88 



o>-«oa>raoc<4eM • -e^oo 



g^^8S8S88888 









S55 



5 <^ .-4 ic .^ oi X S 






e<)C9NC4e4c^oic4MMo«e4 i n 






• a 

o S 
HO* 






ao90 0o>o^^ 



S3$$3$SS 



8888888 



SSS8888 



88 






8888888 



-«Q0 



8 8 



.S 



88S8888 



ssi^g'sii 



C4C^^C4OlO«04 1,^ I 






Hum I ! 



Digitized by 



Google 



350 Williamson on Handling Concrete at Panama. 



o 



I 

H 

P 
C 



s 

o 



o I. 

O »- 

►< & 

< ^ 



< 
o 

8 

I 
I 



9 



si 



^ 




o'^ 


1 




1:^ 


>l 




1= 


i 


s 


^° 


t 


H 




< 




ea 


0) 




Sao 


^ 




04 






•Sfi 


? 


^1 


o 


aiaa 


^ 


C 


^00 



.T3 



> a 



a 






1 




:8S8S8S888S888S88S88 




1 

"Sri 




:Si:iS3!;S!^SSSSSSSSi;S&SS 


:2SgJ2SS8SSS?SJ5S2GS2S 


:$S3i:S3;3$:;S33S:8S3SSS 


;assss|s|sgg§g|§Sgg| 



g> t-" o « -* i-i lO d d e<i « ci •* CO i-J eo CO 



S88SS8S8§8§S8§888 

5 ^ « »o c« t^ »H « ^ M ts. oj 00 55 ti cft <© 






• ir)t<idoddtxidoddod»^odh<iad*oi»(0<ot>ii 



°s3sssssssss;ss;sss;s3!; 






— I 






. ■QQQ«OQtxC<ICOaQC<)'HP|<DeoOOOe>«Q 

• ■ooo^o^i-ir*oO'HO«*o>'^«>o«o 



I I I § § E V .-^i 



00 ^ S iS '4* N> a» ? ra CI S S S o5 o r» <o 



8 8 S S ^ 3 S S m 8 ? tx S lO Ss o 00 

SSS|2:S|S§i§;2l88gi:iS 





S8S{2f:S82222g5^825? 


«a>-«ec>o^ic>oc4»or^iO>o>ctoioio 



S^*ss§s5s:55;ssssss5t: 



:f :i! ::R :« :?; 

C4NNCIC4C4e<)C4 CI CI 09 04 



QQQO4QO4t»-«Q00QQt^tOOia0a» 

^ ,^ ,^ ^e>io) ^04 f-ioi ^o< ^ ^ i-ioi ^ 



^l?;b 




Digitized by 



Google 



Williamson on Handling Concrete at Panama. 351 



mil 



III 



gnssssj; 




S«SS!oSSS 


o 


•coO'-iwo*^© 


s 


c5«?3c5?iSn 


ss 









■ ■S S «^ S c ^ 

ilis ii I 



Digitized by 



Google 



352 Williamson on Handling Concrete at Panama. 

plant consisting of two 2-cu. yd. and three i-cu. yd. mixers. The 
performance of this equipment to Jmie 30, 1910, working on an 
8-hr. day basis, is detailed in Part 1 of Table VIII. 

During the fiscal year ending June 30, 1911, the auxiliary 
plant consisted mainly of three 2-cu. yd. mixers, one located at 
the south end of the east wall and two in the forebay, the half- 
yard having been transferred to Miraflores. It placed 121,530 
cu. yd. and the detailed performance, on an 8-hr. day basis, is 
shown in Part 2 of Table VIII; and the detailed performance, 
on an 8-hr. day basis, of this plant for the first seven months of 
the fiscal year ending June 30, 1912, is shown in Part 3 of the same 
table. 

Permanent Plant, — One berm and two chamber cranes, or 
one-half of the plant, began operating on April 4, 1910 (a chamber 
crane placed some concrete from auxiliary mixers in March); 
the other half began on July 15, 1910. The portion of the per- 
manent plant in operation laid 73,083 cu. yd. prior to June 30, 

1910, on an 8-hr. day basis, as detailed in Table IX. 

The plant worked as a whole from July 15, 1910, to January 
31, 1911. On the latter date the dismantling of one berm crane 
began preparatory to erecting it at Miraflores. The dismantling 
of two of the chamber cranes for the same purpose began on 
April 20 and May 9, 1911, respectively, and that of the remaining 
berm crane on May 19, 1911. Two chamber cranes remained at 
Pedro Miguel until December 12, 1911, and January 31, 1912, 
respectively, being used in the meantime for placing concrete 
from auxiliary mixers, setting iron work and backfilling the middle 
wall. The plant placed 379,190 cubic yards during the fiscal year 

1911, and the detailed performances of the berm and chamber 
cranes respectively are given in Tables X and XI. 

Miraflores Locks, 

AiLCiliary Plant. — Placing concrete in the floors and lateral 
culverts of the upper locks at Miraflores was begun on June 1, 
1910, with a plant consisting of two ^-cu. yd. mixers, and 1630 cu. 
yd. were placed before June 30, 1910. 

In the fiscal year ending June 30, 1911, the auxiliary plant 



Digitized by 



Google 



Williamson on Handling Concrete at Panama. 353 



525 
O 

S 
> 

o 

M 

g 



o 

a 

Q 



Z K 

as 



5 
2 






I 



3 
o 



l« 









•s . 






1 






II 






•-1 00 uSa 












:s§^g; 



illtii 




ss3;s:ss;^SoSSS|s^ 



OQ 04* to 00 t*' CO ^' CO CO 9 1^ 

a lO lO 55 »o ■* «o iS -* CO »-i I 



g : 



iQ CD e^ Q a» CI CO CO :«> 00 Q 



oSS'^'-iioSSoSo 

^Neoeocieocoodcowco 
t^O>0<0-H00<-*C4C0S 
ooa»oo»^io^to^ra 



88 
§8 



(Ooaoeoo^cD^coo 
SSS«^oQ>oa»Scot* 



^'codi-Io6ci«o«o2i> • 



to 

^8 



S^SBSiiSSSSSSS 



Nor^iHio^Neotoc 

iOr^iooot*cocooo« 



oeo 



S M ra^9»^ CO S 0(000 

oi :* lO CO »0 00 N to CO £1 -^ 



Vicococit>^r<^'-4«p«dto 
N.3Sco>*'*eo»cSir^ 



?25 



SS 






! 

Concrete 
Mixed, 
cu. yd. 


34.890 
50.808 
49.852 
55.152 
49.568 
31.156 
35.640 
23.770 
25.704 
17.188 
5.462 


8 : 


Number 

of 

Mixere. 

(2ou.yd.) 


888S8t3Se8888 § : 

^'^^eo'^eocic4 0«C4e<« co • 



I 1 



o 



888S8?^^8888|S 




ltlilii-» 



iill^ 11 



Digitized by 



Google 



354 Williamson on Handling Concrete at Panama. 



55 
O 









^ g 

s s 

« o 



I 



5 s 
(« s 

8 I 



I 

9 



Raht 



UN 
11 |S 






|.2l 



oj? o S 5,0 



c 

o 



I 



11 









o 
H 



10 ^ CP <«' <0 (0 to A 00 01 94 -< 
00Cnt^t«t«iO«OtO<AtCiQ<4' 



(D t» r» t^ « o SS to 10 10 iS 



xot*coeo 






r«'^9t0C4aa9ixoS^i- 



QDC0C4t<«-<a»tO^tOt»t>«« 



:SSS255S23gS 



•O CO 



^^ lO c« N N .-• -^ ^ 



Dt<«CPt»(O^tO'^t0C0<-<'-4 



9 5 



.Sg 



Slo^-^aoS-^owN-eo o 5 



Dt^«t^»^t0^iB«i-i,-4 



^11 



£J 






ooOQOcoeor'.MCooNgo^' 

O)kO00«OiO^QOli-iQWC0 



S8SS8SSS8^^§§|^ 

CO <« 09 09 ^ CO CO ^ ^ CO -4 f-i I w 






i 






«o d to ci CO »^ CO 



ooieooe^to^ 



^^'^•tO^COM 



^S:;:S^S8 



tOtOQQ 



ocoSt»tora^ 



S^S8S^S 



e«^®Q^Nto 
t^eoototo^pi 






»-i-^C5N»-»i^ 



s:§SSS^S^ 



SS888^$iS 



^itlfli I 



9 



I 



o 

-a 

► 

o 

i 



i 

OB 

I 



Digitized by 



Google 



WlLUAMSON ON HANDLING CONCBSTE AT PANAMA. 355 

consisted of two 2-cu. yd. mixers located under the north, end of 
the east storage trestle, and four i-cu. yd. mixers. The 2-cu. yd. 
mixer plant was moved to the east wall for supplying the chambers 
cranes in May, 1911: previous to this it supplied a berm crane 
that was sufficiently complete to place concrete with the boom, 
though its mixers and cantilever arm were still in use at Pedro 
Miguel. The auxiliary plant handled 205,255 cu. yd. during the 
year, as detailed in Table XII; and a similar statement giving the 
detailed performance of the 2-cu. yd. auxiliary mixers at Mira- 
flores, on an 8-hr. day basis, for the first seven months of the fiscal 
year ending June 30, 1912, appears in Table XIII. 

Permanent Plant. — One berm crane without cantilever arm 
and mixers began, in its uncompleted condition, to place concrete 
supplied by the auxiliary mixers on September 2, 1910, and 
continued placing until February 15, 1911, when it was taken out 
of commission for completion. It began operating again, as a 
complete machine, on March 22, 1911, and a second berm crane 
began work on April 7, 1911. With these 67,774 cu. yd. were laid 
previous to June 30, 1911. 

Other units of the plant were placed in commission as follows: 
2 chamber cranes on July 13 and August 3, 1911, respectively, 
and 2 additional berm cranes on July 25 and October 28, 1911, 
respectively. There are 2 more chamber cranes under erection as 
follows: Table XIV shows the performance of berm cranes, 
Miraflores Locks, to January 31, 1912, and Table XV the perform- 
ance of chamber cranes, Miraflores Locks, for first peven months 
of fiscal year ending June 30, 1912. 

The performances of the Pacific Division plants for the fiscal 
year 1911 are given by the Cost Accountant as follows: In Pedro 
Miguel locks 497,802 cu. yd. of concrete were placed, average 
division cost $4.70 per cu. yd. and 385 cu. yd. of reinforced con- 
crete at $17.74 or a total of 498,187 cu. yd. at $4.71 per cu. yd. 
In Miraflores locks 272,933 cu. yd. were laid at an average 
division cost of $4.68 per cu. yd. The lowest average cost for 
any one month was for Pedro Miguel in November, 1910, 
when 64,248 cu. yd. were placed at $4.20 a cu. yd. and for Mira- 
flores in May, 1911, when 36,154 cu. yd. were placed at $4.05 per 
cu. yd. 



Digitized by 



Google 



366 Williamson on Handling Concrete at Panama. 



o 



> 
Q 
u 

I 



O 

2 - 

S s 

;^ 2 



^ 



00 J 

i 










! 0^ 



OS 

b 5 



II 

QqO 



«e^®--'fM»odM«M.-<ddh*gd^ddx-HMgo-^fdt«^diO--dec^5 

«S5««c©o>wicoswr-'-<o6pi««.So^ioi««.^w?iNr*'*^e5NO»PO 
bJt^ts!h«iio«odeod^«t«rao>ciodd^ddcoi^ice>4t^t^c>9io<d<o 






a 3 






:j::^:?::^ :f:j! :^ :?::« :f!:R :s:s :s:« :j!:f :s:f :j;xj! 

C4 C4 04 C4 es 04 d 04 « 






Ill 

^ I i I 

5* -2 JJ 4s 
»^ < 00 O 



\ 



I 









••2 

< 



Digitized by 



Google 



Williamson on Handling Concrete at Panama. 357 







Digitized by 



Google 



368 Williamson on Handling Concrete at Panama. 



o 

> 
Q 

M 
U 






g a 

" I 

It. ^ 
S & 

U 

> 

3 



ri^lij^lll^ 






N 






I 






1-^ 



^6sB 



SS8S 



I 



<Oaot^O 



ll 



SSSJIS 



n 
.a 



Is* 






ss 



8 8 






: :S : 



S S 






ii- 



:2 :a '°g ^ 



8 to COM 



I 



dSSS 

^>ouS 



lulil^i 



iilli 






25 8 






c««eo<o 
Mcoooeo 



€4 C4 






8S36S 

cieoeoeo eo 



OooSS S 








63.55 
59.87 
51.20 
56.56 
50.17 
66.69 
67.06 


8 




945.00 49.40 26.80 
1,278.00 49.70 26.40 
1,188.00 44.70 21.30 
1,368.00 30.70 19.10 
1.672.0029.70 12.80 
1,575.34133.50 19.80 
1,742.25 38.10 21.30 


8 

S 




i 




512.68 
678.85 
566.70 
706.28 
653.62 
929.85 
971.54 


to 

8~ 

to 




604.00 
729.00 
648.00 
711.00 
864.00 
949.00 
1,063.00 




183.68 
292.51 
316.25 
316.99 
429.50 
316.15 
350.26 




Ok 


30.00 
56.60 
46.60 
45.84 
40.46 
63.03 
83.46 


i 


s 

tf 


33.43 
67.36 
62.87 
75.17 
78.55 
62.19 
73.26 


i 


1 

1 

8 


61.17 
97.65 
158.84 
148.34 
262.75 
136.32 
147.36 


^ 
M 

3 


::::::: 


''- 


^ 








5.60 
4.00 
3.00 
1.00 
2.60 
6.92 
4.18 


o 


1 


320.321 53.58 
436.49' 67.90 
331.75 44.94 
395.01 45.64 
434.50 45.26 
632.85 47.69 
712.74 42.00 


8 

eo 






9 




273.45 
387.21 
309.07 
367.09 
373.60 
560.15 
692.76 


i 




26,326 
33,726 
26,349 
26,200 
19.427 
31,180 
37,054 


i 




^' >o to id r«' «D «D 


to 




2^8Sg3888 

N eo c«' CO ^ ^ ^ 


eo 




July 

August.... 
September. 
October... 
November. 
December. 
January... 


\ 



Digitized by 



Google 



WiLUAMSON ON HaNSUNQ CONCRETE AT PaNABIA. 359 



o 

CO 



e 
<2 



S S5 

§ 8 

h g 

5 



8 



> 
« 

s 

5 



•^e 



p< o 















31 









aooooooot«>oio 1 1^ 



SSt^h-S^o 






88SS8S^8 






8 : :8 :5;S 






S 3 









S !<• 00 a> 00 a> to 



S.3 3 I 



^Op'*« »««-"< 



I 



I 






$8S8mSS 



«D<^CD^tOQ 












M O «C O C«l 00 -^ 



•s^sss 









sssssssm 






III 



Digitized by 



Google 



358 Williamson on Handling Concrete at Panama. 



2 

o 

> 
Q 

< 

« 
U 



^ . 



M O 

'^ I 
o & 



o 

> 

5 






I 






■Si 



if 

M 
I 



ssss 


3 


:;SSS 


s 


2SS2 


s 


2^SS 


3 


sssss 


s 


;;^55 


5 



It 



^oas 



'888S 



3l 
^2 



» 

.a 

i 



Is 






cm' t>^ I— d 






ss 



8 8 

i § 









^1^1 



S S 



2 'Sl]9j^ 

© U3 — 






I 









11^ 
ass 



li"s 

•5 us 



sqi^s 



C9t00<0 

*^'2SS 






8336^ 

esicoeoeo eo 



1 



8SS88 S 



•H«-IMC^ *^ 



Jtii il^ 



•Q 00 S to *-< S O 



00 ^ ro ^ 00 00 w 



945.0049.40 
1,278.00 49.70 
1,188.00,44.70 
1,368.00'30.70 
1,572.00 29.70 
l.S?^ 34 33.50 
1.742 25 38.10 
1 


i 


512.68 
678.85 
566.70 
706.28 
653.62 
929 H5 
B7i M 


CD 1 


183.68, 504.00 
292.51' 729.00 

316.25 648.00 
315. 99| 711.00 
429.50 864.00 
316.15 949 on 

350.26 1,063 jwi 


a 

Is 






rt iS * !^ I-, rs PS 

I-' h- SC< 1^' -^1 ts t^ 

i£, a k^ -* «5 rt ^ 













S888SSS2 

MS '^ CO "-• ci w ^ 
ioa33e4«8 

8' CD ^ to ^' esl ci 
CO CO a» CO CO •-) 
CO ^ CO CO ■^ «o tN. 



ooevi«a8r»S 
«o o» ci i>i -i,^ ci ot 

^ ^ CS| d «0 l>. 1-1 

•«•• es o S i35 ^ t- 

0COCOU»M3 



gsi 



^' u3 id to 1^ «c> «d lo 



^SSS888|P^ 



In-^i^ 






Digitized by 



Google 



WiLUAMSON ON HaNSUNQ CONCRETE AT PaNABIA. 359 



o 



^ o 



111'- 



u 



PI 



mM 



1^ 









;$s 



SSSS3$§ 






SsoSocoS 






8 : :8 :^S S 

*4 I le* •d«o 



■4 a> CO <-• oo M CO 



*-4«D 'CDiOi 



:2g5S 



•5 w. 9-2 



SSSS8SS 



•i 1 






1 

o 









Sa« 



^'StSS 












lijroi 



<"z 









sssssss I s 



■Mm 



Digitized by 



Google 



360 Williamson on Handling Concrete at Panama. 

The detailed division cost for the year is given for Pedro 
Miguel as follows: 

Concrete (495,037 cu. yd.): 

Cement $1 .5365 

Stone 8242 

Sand 3729 

Mixing 1771 

Total cost of concrete $2.9107 

Large rock (2,765 cu. yd.) $1 . 1483 

Masonry (497,802 cu. yd.): 

Concrete $2.8945 

Large rock 0064 

Forms 4387 

Placing 3118 

Reinforcements 0367 

Pumps 0343 

Power 0454 

Maintenance of equipment 1723 

Plant arbitrary 6847 

Division expense .0792 

Total division cost $4.7040 

The division costs per month since June 30, 1911, have been 
as follows: 





Cost 
Pedro I 


per Cubic Yard of Concrete. 


Month. 1911. 


^iigucl. 


Miraflores. 




Plain. 


Reinforced. 


Plain. 


Reinforced. 


Julv 


$5.82 
5.63 
6.08 
6.26 
5.80 
6.27 


$6.26 
8.74 

11.91 
8.85 
9.94 
8.98 


$4.93 
4.45 
4.41 
4.50 
4.89 
5.06 




August r r T T . T 


$11.17 


September 

October 


16.32 
21 39 


November 

December 


23.76 
14.45 



It is obvious from the tables that a large percentage of delays 
in placing concrete at both Gatun and the Pacific locks is charge- 
able to forms. At times the forms are filled so rapidly that it is 
difficult to keep them ahead of the placing, but the greatest loss 
of time is occasioned by the amount and complication of forms for 
the electrical tunnels, conduits and machinery rooms near the 
tops of the walls. The records given for the Miraflores plant are 
not representative, as the entire plant is not in operation. 



Digitized by 



Google 



USE OF CONCRETE IN THE FOURTH AVENUE 
SUBWAY. 

By Frederick C. Noble.* 

The Fourth Avenue Subway, Brookljna, is an example of the 
use of remforced concrete on a large scale. The portion now 
under construction extends about four miles from the Brooklyn 
end of the new Manhattan Bridge, over which it is intended to 
connect with a subway system in Manhattan. The route, Fig. 
1, lies through Flatbush Avenue, Fulton Street, Ashland Place 
and Fourth Avenue, to Forty-third Street; in a direction generally 
south. From here it is proposed to extend it in future by two 
branches to the southern limits of the borough. The subway 
is being built for the City of New York, under the supervision 
of a state commission, with Mr. Alfred Craven as the chief 
engineer. The work was divided into six contract sections, which 
were let in November, 1909, at an aggregate price of about $15,- 
000,000. Construction is now nearly finished. 

The structure normally has space for four tracks; two for 
local and two for express service. These are increased to seven 
and eight in places where connection spurs, for future extensions, 
are provided in two levels to avoid turnouts at grade. There 
are six local and two express stations, with platforms long enough 
to accommodate ten-ear trains. 

The excavation was principally in a moraineal deposit, con- 
sisting of sand with more or less gravel and boulders. No rock 
was found anywhere on the line. Some quicksand was met 
near the middle of the route, where the subway traverses ground 
filled in over the bed of an old salt marsh. Ground water was 
encoimtered at or near tide level and was controlled by pumping. 
Excavation was usually carried on under covered roadways, but 
in the extension of Flatbush Avenue open excavation was per- 
mitted. 

In connection with the work it was necessary to underpin 



* Division Engineer. Public Service CommiMion, Broolclyn. N. Y. 

(361) 



Digitized by 



Google 



362 Noble on Use of Concrete in Brooklyn Subway. 

about forty buildings with masonry piers to sub-grade in advance 
of the excavation, where it approached close to and below their 
foundations. It was necessary also to temporarily support the 
elevated railway on Fulton Street and to drift a crossing under 
the present subway in operation in Flatbush Avenue. Since 
Fourth Avenue lies along the foot of a considerable drainage 
area, sloping from Prospect Park, it was necessary to depress or 
intercept many cross-sewers carrying a heavy storm-flow, and 
to provide under-grade crossings at intervals. About six miles 




Fia. 1. — PLAN OF FOURTH AVENUE SUBWAY, BROOKLYN, N. Y. 

of sewers, of all sizes up to 9i ft. diameter, were thus built or 
rebuilt. 

The typical four-track section is shown by Fig. 2. The roof, 
sides, intermediate walls and floor, where below ground water, 
are reinforced with bars ranging between li and f in. square. 
The design provides for a uniform live load of 300 lb. per sq. ft. 
at the street surface. Waterproofing is used at stations on the 
roof and sides and generally below ground water on the sides and 
floor. Elsewhere the concrete is relied on for sufficient protec- 
tion against seepage. 

The amount of concrete required is somewhat over 400,000 



Digitized by 



Google 



Noble on Use op Concrete in Brooklyn Subway. 363 

cu. yd. The usual proportions are 1 to 2J to 4J. Much of the 
excavated sand is suitable for concrete and is used accordingly. 
The remainder is washed or dredged sand from the north shore 
of Long Island. The aggregate is mostly washed or dredged 
gravel from the same source, usually under 1 in. and well graded 
in size. At times when gravel is not readily obtainable, cobbles 
and fragments of boulders are crushed and used. The concrete 
is mixed rather wet to facilitate its flowing around the reinforce- 
ment, as this is generally spaced quite close. 

Reinforcing rods were required to be deformed. Trans- 



OMiriM IOO^Iam 




SMtioii Ml DryGrMBd. 



TkiekMSS •! concrete and spocin^ 
of rods ore ^ivtn for nlnimwiB cover, ond 
ciiaajo whort cofor aCMtfo 7 ft. 






^ 



FIG. 2. — TYPICAL POUR-TRACK SECTION OP SUBWAY. 

verse and longitudinal rods were wired at their intersections 
(Fig. 8). For this purpose, special devices of bent wire were 
used; these served at the same time to space the rods, or to hold 
them at the proper minimum distance from the forms. Where 
imusually long spans or concentrated loads occurred, the roof 
was of girder and jack-arch construction. To hold the thin 
covering of concrete imder the lower flanges, special clips or 
hangers of twisted wire were attached at intervals by bending 
them over the upper sides of the flanges. Sometimes a strip of 
coarse wire mesh was wrapped loosely around the flanges to 
serve the same purpose. In a few instances, where the concrete 



Digitized by 



Google 



364 Noble on Use of Concrete in Brooklyn Subway. 

under the flanges tapped hollow, it was chipped off around the 
edges of the flange, and a strip of wire mesh was attached and 
covered with mortar blown in place by means of a cement gun. 

Concreting was permitted throughout the winter, except 
on the very coldest days. Materials were heated by fires or 
steam coils, and the freshly laid concrete was covered with salt- 
marsh hay and tarpaulins. Salamanders were placed under the 
roof forms in some places. On accoimt of the varying condi- 
tions of mixing and degree of exposure, it was impracticable to 
devise a set of rules to fit all cases; but the criterion was held to 
be the temperature at which the mixture could be deposited in 
the forms. 

Roof forms were struck as soon as the temperature condi- 
tions and setting qualities of the particular brand of cement 
used would permit. As this was a factor limiting progress, espe- 
cially in the case of large steel forms, it was desirable to strike 
them as soon as practicable. In summer they were struck after 
80 hours, and in exceptional cases even at 50 hours, but in winter 
it was sometimes necessary to wait a week or 10 days before 
striking. 

The methods of mixing and handling concrete vary between 
the diffei;pnt sections and are described briefly as follows: 

Section Af. B. Ex, i, extending from Nassau Street through 
Flatbush Avenue to Willoughby Street, is being built by Smith, 
Scott & Co., contractors. 

Materials for concrete are brought on the work over a tram- 
way extending along each side of the cut from a dock \mder the 
Manhattan Bridge. Sand and gravel are loaded at the dock in 
side-dump cars, drawn by 20-ton dinky locomotives. The cars 
are run up an incline and dumped into bins over the mixer, which 
is located about centrally on the section. This is a li-yd. machine, 
operated by a 30-h.p. electric motor. The mixture is discharged 
at the track level into bottom-dump cars of 1-yd. capacity, in 
trains of 3 or 4 cars, hauled to the point of deposit and dumped 
in frame chutes through which it slides to the forms. 

As the sides of the cut were almost self-sustaining and as 
the contract terms allowed open excavation on this section, very 
little' cross-timbering was necessary; a condition that greatly 
facilitated the erecting and moving of forms. The forms were 



Digitized by 



Google 



Noble on Use of Concrete in Brooklyn Subway. 365 

of the collapsing pattern, with stiffened steel sides and wood top 
to suit the varying widths of roof. The floor having been laid 
for the entire width of structure, and the reinforcement placed 
for the sides and intermediate walls, the forms were moved into 
place, one 20-ft. section at a time, traveling on tracks laid on the 
floor. In this way it was possible to lay concrete for the walls 
and roof over four tracks in sections 20 ft. long at a continuous 
operation. These forms were very similar to those used on one 
of the lower sections, in which connection they will be more fully 
described. 

Where structural roof framing was substituted for rod con- 




no. 3. — COLLAPSIBLE JACK ARCH FORMS. 

struction, centering (Fig. 3) was used to turn the jack arches. 
These were of angle-stiffened plate, with an adjustable center 
strip of wood for varying the span, and were set in position on 
falsework. The span was maintained by tumbuckle ties, which 
also served to draw in the sides on striking the forms. After 
striking, each section was lowered with a winch. 

Section Q-C-l, extending from Willoughby Street along 
Flatbush Avenue and Fulton Street to Ashland Place, is being 
built by William Bradley, contractor. 

Materials for concrete are delivered at a large yard along- 
side the subway between Third and Sixth Streets, where the 
contractor has a cement storehouse and a stable for his horses. 



Digitized by 



Google 



366 Noble on Use of Concrete in Brooklyn Subway. 

Teams are employed for moving all materials. Concrete was 
mixed principally by a gravity mixer located in the cut. Storage 
bins for sand and gravel were placed just below the street level. 
These fed 4 loading hoppers, each of one-bag-batch capacity. 
After moistening, the contents were dropped successively through 
three mixing hoppers, receiving the full dose of water in the first 
one. The middle hopper was slightly offset from the other two. 
The mix was discharged at the bottom level into cars which 
were nm onto a cage, hoisted above the street surface, and dumped 
into an overhead bin. From the bin it was almost immediately 
discharged through a gate into tight rear-dumping steel carts 
of about If-yd. capacity, which were teamed to the point of 
deposit. The mixture was dumped on hopper platforms and 
conveyed through 8-in. telescopic chutes to the forms. The 
opening at the bottom of the hopper was closed by a spatula until 
the entire load was cleared. 

Most of the structmre on this section is of steel bent and 
jack-arch construction, so that no unusual form-work was re- 
quired. The reversion to this older type of design was at the 
request of the contractor, and partly because of the concentrated 
loadings brought by the elevated structure and building foimda- 
tions to be supported permanently on the roof. 

Sections 11-E-l and 11-A-l, which together form one con- 
tract section, extend from Fulton Street, through Ashland Place 
and Fourth Avenue, to Sackett Street, and are also under con- 
tract with William Bradley. 

The methods of mixing and placing concrete are the same 
as those described for this contractor's adjoining section. The 
form of construction is also similar, except where the line passes 
by a curve under the present subway in Flatbush Avenue. Here 
the design is of massive concrete arches, one for each track, 
without reinforcement (Fig. 4). 

Section ll-A-^, from Sackett Street along Fourth Avenue 
to Tenth Street, is being built by the E. E. Smith Contracting 
Company. 

Materials for concrete are stored in a yard beside the Gowanus 
Canal at Third Street, and are distributed to the work by mule 
teams. Concrete is mixed in the roadway, directly over the 
point of deposit, in five small readily-shifted drum mixers of 



Digitized by 



Google 



Noble on Use of Concrete in Brooklyn Subway. 367 

i-yd. capacity, each operated by a 15 to 25-h.p. electric motor. 
The mixture was discharged and conveyed to the forms through 
10-in. sectional chutes of 12-gauge iron, with a funnel at the top. 

Generally, the floor and walls of the westerly two tracks 
were built first, then the floor and walls of the easterly two; after 
which the roof was carried across all foiu- tracks in sections about 
30 ft. long. Five such sections would be in progress at one time. 

The necessity for maintaining the cross-bracing of the trench 
precluded the use of large forms such as were used on certain of 
the other sections. Forms were made of 2-in. lumber, dressed 
all sides, and with dapped edges. They were assembled in well- 




FIG. 4. — TYPICAL SECTION THROUGH SUBWAY AT TUNNEL UNDER PLATBUSH 

AVENUE. 

braced panels and could be used many times. Jack-arch forms 
(Fig. 5) were made of 16-gauge iron, nailed on wood ribs and 
were suspended in position by long bolts. 

On this section the usual waterproofing on sides and bottom 
below ground water level is omitted. In such situations the con- 
crete is locally made richer, 1-2-4, and the longitudinal rein- 
forcement is doubled (Fig. 2). 

Section 11- AS, from Tenth Street along Fourth Avenue 
to Twenty-seventh Street, is under contract with the Tidewater 
Building Company and Thomas B. Bryson. 

Materials for concrete are brought on scows to a dock at 



Digitized by 



Google 



368 Noble on Use of Concrete in Brooklyn Subway. 

the foot of Nineteenth Street, from which a 3-ft. tramway leads 
to the cut and through it in both directions. The sand and 
gravel are transferred from the scows to overhead bins by a crane 
and clam-shell bucket. The bins discharge through gates into 
4-yd. side-dump cars. These are hauled by a 20-ton locomotive 
to the main storage bins, of about 3,000 cu. yd. capacity, situated 
in the cut near Nineteenth Street near the mixing plant. Belt 
conveyors run under the gates of the bins and take the material 
to a bucket elevator which raises it to bins over the mixer. The 




FIG. 5. — METAL JACK ARCH FORMS. 

overhead bins feed into measuring hoppers, where the cement is 
added, and thence into the mixer, which is a 1-yd. machine mixing 
a 5-bag batch. Conveyors and mixer are electrically driven. 

Taking advantage of the firm character of the ground and 
the great width of the avenue, a method of excavation was adopted 
that dispensed with cross-bracing; thus leaving the cut (Fig. 6) 
imobstructed and making it feasible to construct large sections 
at one time. The floor of the two middle Iracks was laid firet, 
and this was followed by the floor and duct bench of each of the 
two outside tracks. Concrete was brought from the mixer for 
this part of the work in 2-yd. side-dump cars. On the eomple- 



Digitized by 



Google 



Noble on Use of Concrete in Brooklyn Subway. 369 

tion of a section of the bottom, steel forms were moved into place, 
and the sides, intermediate walls and roof were concreted con- 
tinuously in a long section. 

The steel forms (Fig. 7) were of the collapsible type, made 
of iV"ii^' plate and stiflfened with shapes. They were composed 
of 5-ft. units, bolted together in sections 40 ft. long, and traveled 
on rails. Four such sections, one for each track, made up a set 
of forms; of which two were in use generally on different parts 





i 




* ^ 




p 




z. 


^^^^^^HS^MlBkbf 'i^9Sl^^^^3Bp^H^^V^ 


\ 




i 


HKS^ jIHbis^ 


A 



FIG. 6. — METHOD OF OPEN CUT EXCAVATION. 

of the work. The forms were struck, each section separately, 
by means of a hand tackle arrangement that swimg the top 
leaves down around hinges at each side, and drew in the tops of 
the side-panels automatically. The vertical reinforcement 
(Fig. 8) for the next section being set up, the forms were then 
pulled ahead, one 40-ft. section at a time, by a locomotive, into 
its new position; an operation that could be performed in a 
very few minutes. As each section was advanced it was adjusted 
to exact position and connected to its neighbor by long bolts and 



Digitized by 



Google 



370 Noble on Use of Concrete in Brooklyn Subway. 

pipe separators. The adjustment was effected by the jack- 
screws and sliding axles of the trucks; affording a range of move- 
ment both vertically and horizontally. The reinforcing bars 
of the roof were then laid in position. 

For concreting the roof, the mixer discharged into trains 
of 4 or more 2-yd. bottom-dump buckets on flat cars, which were 
pushed by a locomotive over the tramway through the cut to the 
forms. Here the buckets were lifted by a locomotive crane 




FIG. 7. — COLLAPSIBLE STEEL FORMS FOR ROOF AND SIDES. 

(Fig. 6), and dumped on a hopper raised about 12 ft. above the 
forms. From the hopper the mixture flowed through shallow 
troughs to the points required. When both sets of forms were 
joined together, an 80-ft. section, requiring over 600 cu. yd., 
could be concreted in about 15 hours. 

Section ll-A-4, from Twenty-seventh Street along Fourth 
Avenue to Forty-third Street, is being built by the E. E. Smith 
Contracting Company, who are also the contractors for section 
ll-A-2. 



Digitized by 



Google 



Noble on Use of Concrete in Brooklyn Subway. 371 

Materials are delivered into bins in the contractor's yard 
at Thirty-first Street and Second Avenue, where there is also a 
cement storehouse of 30-cars capacity to serve both sections. 

The methods of mixing and distributing concrete and the 
use of forms, are the same as those described for section ll-A-2. 

All cement is inspected at the mills and shipped in sealed 




no. 8. — VERTICAL REINFORCEMENT. 

bags. The testing laboratory is in Allentown, Pa., in the Lehigh 
Valley district. Over half a million barrels are required. Under 
the specifications, preference is given to brands whose records 
show continued increase in strength over long periods. As a 
criterion of this quality, mortar briquettes are required to show 
a gain of at least 50 lb. between the 7 and 28-day tests. 



Digitized by 



Google 



THE USE OF REINFORCED CONCRETE IN HYPO- 
CHLORITE WATER PURIFICATION WORKS. 

By Walter M. Cross.* 

During the year 1911 an experimental installation of the 
hjT)ochlorite process for the approximate sterilization of the 
entire municipal water supply of Kansas City was so remarkably 
successful in diminishing the sickness and death rate in the city 
on account of typhoid fever as well as other forms of intestinal 
disease, that the Kansas City Fire and Water Board undertook 
the construction of a permanent building and apparatus for the 
application of this purification process to the water supply. 

A separate building was constructed to make possible the 
satisfactory storage, handling and making of the solution of hypo- 
chlorite ready for mixing with the sedimented water. The build- 
ing itself was designed by W. C. Root, an architect, and the appa- 
ratus for use in connection with the sterilization process was 
installed under the direction and supervision of Burton Lowther, 
engineer in charge, and S. Y. High, superintendent of the Water 
Works Department. 

The apparatus for the handling of the hypochlorite and the 
supports for it are of reinforced concrete. It is to be observed 
that no other material is so well suited for use in connection with 
this sterilizing agent as good concrete for the reason that all other 
materials that are capable of oxidation are promptly attacked 
by the hypochlorite solution and become rapidly deteriorated. 
The prime consideration with regard to this class of installation 
is to employ such methods of construction and to use material 
that is so permanent in character as to obviate the necessity 
of repairs which would force the discontinuance of the application 
of the sterilizing agent even for an hour. 

The basement of the building is used fcr storage of the reagent 
that is kept in reserve. The main floor is used to house the dilu- 
tion tanks and the feeding devices, while on the floor above is 
placed the tank in which the hypochlorite is reduced to paste 

•City Chemist. KanBas City. Mo. 

(372; 



Digitized by 



Google 



Cross on Concrete in Hypochlorite Plant. 373 

of a creamy consistency before being delivered to the dilution 
tanks beneath. This pasting tank, 3 ft. in diameter and 4 ft. 
high, is provided with a stirring device carrying two rather heavy 
rollers, disposed horizontally at its lower end. The rollers clear 
the bottom of the concrete tank only by a fraction of an inch, 
thus insuring the mashing and disintegration of all of the small 
lumps that are invariably present in commercial calcium hypo- 
chlorite. Owing to the fact that the action of the reagent on 
bronze is to form on the surface of it a fairly insoluble and pro- 
tective coating of metallic carbonate and oxychloride, that metal 
appears to be the most available for use on all bearings and stirring 
or disintegrating devices that come in contact with the solution. 
Leading from the concrete pasting tank are pipes so arranged 
that the contents of the tank may be discharged into either of 
the large dilution tanks on the floor beneath. The outlet of the 
pasting tank is placed at a considerable distance above its bottom 
so as to avoid the possibility of drawing off with the paste any 
fragments of considerable size. The pipes carrying the paste 
are so arranged as to be readily cleaned in a few minutes in the 
event that they become clogged. Ultimately they are sure to 
become clogged if they are not occasionally cleared because of the 
formation in them of carbonate from carbon dioxide absorbed 
from the air. 

The dilution tanks are hexagonal in form, 9 ft. in maximum 
diameter and 7 ft. high; the walls are 6 in. thick. Although 
the difficulty experienced in properly disposing the reinforcing 
metal in the construction of a hexagonal tank is much greater than 
is the case in the building of a round one, the hexagonal tank is 
to be preferred on account of the fact that in a round tank a 
rotary stirrer does not produce nearly such thorough agitation 
and mixing of the solution of hypochlorite as the same stirrer can 
do in the hexagonal tank. The paste is mixed with water in the 
dilution tanks until a uniform solution of a strength of 2 per cent 
occurs. The use of the two tanks makes it possible to accurately 
adjust the strength of the solution in one dilution tank while the 
contents of the other are being utilized. The dilution tanks are 
placed on supports high enough to permit the use of a gravity 
feed to the orifice box which is placed on the floor of the room 
housing the big tanks. 



Digitized by 



Google 



374 Cross on Concrete in Hypochlorite Plant. 



H 



17 



I 



I I 

1 



h- 



j£^ 1 




4-1-* 






i 



^J 



"I 




FIG. I. 



Digitized by 



Google 



Cross on Concrete in Hypochlorite Plant. 375 

Bronze pipes, 1^ in. in size, so arranged as to be readily cleaned 
in the event of stoppage, connect the dilution tanks with a gauging 
tank, 4 ft. in diameter. This gauging tank contains a float, 
scale and pointer so arranged that the man in charge can accu- 
rately check the speed of outflow of solution from the orifice box 
into the big water main carrying the entire city water supply 
from the settling basins to the pumps. The solution passes through 
the gauging tank to the orifice box. Each division on the gauge 
represents 1 gallon of the hypochlorite solution. 

The orifice box is oblong in shape and carries a float of about 
250 cu. in. displacement. The float operates a valve which, 
by either opening slightly or closing, maintains the hypochlorite 
solution in the orifice box to a constant level. One end of the 
orifice box is of plate glass to enable the operator to see at a glance 
that the solution is filling the box to the proper height. Attached 
to the plate glass and covering a hole in it, is a hard rubber disc 
having near its periphery several slits, the adjustment of which rep- 
resents the size of a stream of the 2 per cent hypochlorite solution 
that will be the proper amount to treat the quantity of water 
passing through the main. All movements of the hypochlorite 
solution after its preparation are by gravity. Ample opportunity 
for the hypochlorite after its addition to the water to react with 
any putrescible organic matter and germs, is afforded during, the 
time in which the water passes through the centrifugal pumps, the 
flow line and a small storage basin at Turkey Creek before it is 
pumped to the domestic water users. 

All of the stirring devices are run by an electric motor belted 
to a line shaft carrying clutches so placed as to make possible 
the running of any one of the stirrers whether or not any of the 
others are running. 

The principle involved in the construction of practically 
all hypochlorite installations for the purification of water by the 
oxidation of germs and putrescible organic matter in municipal 
water supplies is substantially the same as that in Kansas City. 
Concrete, usually reinforced, is universally used in the construc- 
tion of all permanent apparatus for the preparation and solution 
of the hypochlorite for mixing with the water to be purified. 

A fairly good idea of the disposition of the various parts of 
the purification installation is given in Fig. 1. 



Digitized by 



Google 



DESIGN AND CONSTRUCTION OF THE ESTACADA 

DAM. 

By Hermann V. Schreiber.* 

Located in the northwestern section of the country, where 
fuel costs are high, it is natural that the Portland electric com- 
panies should have early appreciated the abundant stream flow 
characteristic of the region, which results from the high precipi- 
tation on the western slopes of the Cascade Mountains and in the 
Willamette Valley and which was rendered available for their 
power purposes by the successful development of high tension 
electric transmission for the distances involved in delivering such 
hydro-electric power to their markets. 

The first plant developed to serve Portland, and one of the 
first water-power transmission plants on the coast, is located at 
the falls of the Willamette, adjacent to Oregon City on the 
Willamette River, which stream flows through Portland and 
empties into the Columbia River about six miles below the center 
of that city. Some years later, as the community rapidly grew 
in size and industrial activity, a combination plan was projected 
by interested parties for the construction of a railway and power 
development on the Clackamas River, based upon an available 
power site some forty miles east of Portland, and this develop- 
ment has since been completed and, with the other railway and 
lighting interests, has been included in a consolidated property 
known as the Portland Railway, Light and Power Company, 
which at present supplies practically the entire electrical market 
in Portland, including the local and interurban railway systems. 

Following the consolidation, the rapid increase in the power 
requirements of the community indicated the need for immediate 
additions to the generating capacity, and Sellers and Rippey, 
Consulting Engineers of Philadelphia, were retained by the finan- 
cial interests in control of the consolidated property to investi- 
gate and report upon certain features of the existing plants and 

* Sellers and Rippey, Consulting Engineers, Philadelphia. 

(376) 



Digitized by 



Google 



SCHREIBER ON ESTACADA CONCRETE DaM. 377 

the power possibilities and costs of certain new developments 
which were suggested for consideration from time to time during 
the investigations. Because of the exceptional water-power 
possibilities of this region, the number of alternative locations 
presented for consideration was greater than would ordinarily 
come within the radius of economical transmission to a large 
city, but attention was chiefly directed to the Clackamas River, 
upon which the company already owned the Cazadero develop- 
ment, the electric railway system and the transmission lines 
connecting the generating station with Portland. 

This stream has its source in the forest-covered western 
slopes of the Cascades and the snow-covered peaks adjacent to 
Mount Hood. The run-oflf resulting from the high precipi- 
tation peculiar to this section is distributed in a remarkably 
uniform stream flow which, because of the favorable geological 
formation, is maintained even during the long dry summer sea- 
son. Consideration was primarily given an available site for a 
large new development upon property already owned by the 
company on this river above the existing Cazadero plant, but 
investigations extending over the thirty miles of the river down- 
stream to its mouth revealed a site capable of economic develop- 
ment, which, though it did not offer as large capacity as the 
upper site suggested, was foimd to be worthy of recommendation 
for the company's immediate consideration, because of the con- 
siderably lower cost involved and sufficient size to meet the 
immediate power requirements. This site, by far the most 
attractive on the lower river, is a short distance below the com- 
pany's town of Estacada, about 3^ miles below the original 
development at Cazadero, and offered the advantages of a rea- 
sonable head for direct development without long flumes, acces- 
sible railway connections and several incidental advantages from 
a construction standpoint. After some delay, the necessary 
property was acquired and executive decision given that the 
construction proceed without delay. Within a distance of a 
few hundred feet several possible dam sites were available, giving 
the same head, with advantages and disadvantages peculiar to 
each, making the proper selection one of considerable difficulty. 
Fig. 1. gives a general view of the Estacada dam, a phn of which 
is shown in Fig. 2. 



Digitized by 



Google 



378 



SCHREIBER ON EsTACADA CONCRETE DaM. 




< 

< 

< 

& 

u 

O 



O 



Digitized by 



Google 



SCHREIBER ON ESTACADA CONCRETE DaM. 



379^ 




o 

X 
O 



Digitized by 



Google 



380 SCHREIBER ON ESTACADA CONCRETE DaM. 

It is of considerable interest at this point to consider further 
the character of the country, its geologic history and bed-rock 
formation as affecting the selection of site, preparation of founda- 
tions and design and construction of dam. 

The mountains which form the source of the water supply 
include several extinct volcanic peaks, the eruptive discharge 
from which has covered the country for miles around with various 
forms of volcanic debris of such porous and uncertain nature that 
it introduces serious obstacles to satisfactory hydro-electric 
developments of any considerable magnitude, even when they are 
limited to the use of a low dam and extended canal such as that 
which has been installed at Cazadero. In order to properly study 
the situation and secure all possible information relative to the 
formation of this material which now constitutes the "bed rock" 
a careful examination of this section was made on request by Mr. 
J. S. Diller of the U. S. Geological Survey, who in his reports on 
the subject made the following statements: 

The volcanic breccia (bed rock) is made up of unassorted angular frag- 
ments of lava andesite and basalt of various colors ranging in size from dust 
particles and grains of sand to large rock fragments many feet in diameter. 
This fragmental material was blown by explosive eruption from the volcanic 
craters higher up on the range and fell upon the mountain slopes where it 
became so saturated with water from the copious rains accompanying the 
eruptions that it flowed in great steaming sheets from the Cascade Range to 
the gentle slope of the plains, in much the same way as similar material 
flowed down the old stream channels on the western slope of the Sierra 
Nevada in California and covered the early and often rich deposits of aurif- 
erous gravels. 

Sheets of solid nonfragmental lava forming part of the bed rock and out- 
cropping on the slopes of the canyon occur within and between the great 
sheets of volcanic breccia. Some of the lava sheets are basalt, others are 
andesite and they are usually less than 30 ft. in thickness. The basalts are 
generally very porous and gray or dark. The andesites are often reddish and 
porphyritic with white crystals of feldspar. 

The depth to which these sheets of volcanic breccia and lava extend 
cannot be readily determined but it is certainly hundreds of feet and may be, 
as it is along the Santiam and McKcnzie River canyons, over a 1000 ft. in 
thickness. 

Nearly vertical dikes of basalt cut up through the sheets of volcanic 
breccia and lava and outcrop on the surface. These dikes in some places 
have a well developed columnar jointing which divides the rock into columns. 
In the case of the dikes the columns lie horizontally and extend across the 
dike. In the lava flows the columns are vertical, but in all cases the columnar 



Digitized by 



Google 



SCHREIBER ON ESTACADA CONCRETE DaM. 381 

joint cracks are limited to the dike or lava sheet and do not extend into the 
adjacent rock nor make an opening of great extent. There is, however, 
another set of parallel joints, the open cracks of which cut up through the 
volcanic breccia and sheets of lava about vertical in a direction approx- 
imately parallel to the course of the canyon. Such joints may be of consid- 
erable extent and form important openings for the circulation of water; 
Such joints may be expected and should be carefully looked for where the 
rock b covered with soil or gravel. It is especially significant that the dikes 
are approximately parallel to these joint cracks and suggest that the joint 
cracks may extend to great depths. 

The conditions that confront the engineer along the Clackamas River 
in the volcanic breccia plain region are very much the same as will be found 
all along the western foot of the Cascade Range from the Colxmibia River in 
Oregon to Feather River in California, one of the most import,ant water- 
power belts in the United States, and the successful solution of the problem 
which it presents at one point will greatly facilitate the work elsewhere. 

The original development at Cazadero included a rock fill 
timber crib dam covered on the up-stream slope with a large 
quantity of surface soil and gravel, used to reduce to a minimum 
the leakage beneath the structure. The dam creates less than 
one half the operating head, the balance resulting from the exten- 
sion of a head race canal and flume about If miles long to the 
generating station, the tail water of which is much lower than 
the base of the dam because of the intervening slope of the river. 
This represents a conmion method of development in the West^ 
where first cost, rather than low operating, maintenance and 
depreciation charges, has in the past largely governed such pro- 
jects. This development was well advanced when taken over by 
the present owners and as completed by them utilized conditions 
advantageously, although it possesses in a degree the inherent 
disadvantages of such construction in that its peak load capacity 
is limited by the forebay pond capacity and the leakage under the 
dam and from the canal tends to reduce the dry season plant 
capacity. The present owners have however succeeded in reduc- 
ing the leakage under the dam to a trifling amount. 

Considering the magnitude and permanence- of the consol- 
idated property, it was felt desirable in planning extensions to 
endeavor to provide for direct power developments, creating the 
entire head by a concrete dam affording large storage reservoir 
capacity and locating the plant directly at the dam without inter- 
vening flume or canal, thus permitting almost indefinite peak 



Digitized by 



Google 



382 SCHREIBER ON ESTACADA CONCRETE DaM. 

Joad capacity to be developed. This method of development 
involves higher dams and consequently demands greater assur- 
ance concerning the security of foundations and before asking 
Mr. Diller to report upon the geological formation core drill 
investigations were started to determine the characteristics of 
the underlying material. His report, as will be noted, strongly 
confirmed our own conclusions as to the importance of this prob- 
lem and we continued our extensive explorations with core drills, 
going as deep as 250 ft. at times, and not content with a partial 
examination, we made investigations at considerable cost and 
trouble in the river bed as well as upon both banks. 

SITE. 

At the first new site considered for the development of power 
•on the Clackamas River above Cazadero it was proposed to 
create 135 ft. head at a direct connected plant, providing large 
pond storage which would be of considerable value to all the 
plants existing or constructed at any future time on the river 
below this point. 

At the Estacada site below Cazadero it was proposed to 
develop power under 83 ft. head, utilizing all the available fall in 
the river in the 3J miles below the original Cazadero power sta- 
tion and providing pond storage which would be of great value 
in carrying the daily peak load. 

The investigations for the higher head development at the 
upper site were started before the lower property was acquired 
and were under the direct supervision of the field engineer, Mr. 
Shirley C. Hulse, under our direction. The very thorough study 
of the situation there gave considerable data for use in develop- 
ing a method of treatment to insure as far as it might be prac- 
ticable an absolute cut-oflf across the canyon which would prevent 
any undue leakage or erosion of the river bottom after completion 
of construction, and the results of this work were immediately 
applied to the Estacada construction, saving much time in the 
preliminary engineering there. 

When the company was ready to proceed with this construc- 
tion the work was very urgent. The general preliminary inves- 
tigations covered both sides of the river for a distance of several 
hundred feet; well drillings were made without finding any 



Digitized by 



Google 



SCHREIBER ON EsTACADA CONCRETE DaM. 383 

radical difiference in the character of the bed rock, except that 
the first and the last of the several sites considered were found to 
be underlaid with a soft clay formation which might have neces- 
sitated a reinforced mat or piling to support part of the struc- 
tures. Of the four sites here considered, the third one was selected 
as offering an island which it was expected would considerably 
reduce the amoimt of material required and at the same time 
make the construction much more convenient by facilitating the 
diversion of the stream through the rainy season. 

Further investigations were then made of the actual condi- 
tions on the island with the result that a large ravine was foimd 
to extend parallel to the thread of the stream, which being filled 
with debris necessitated its entire excavation. As the construc- 
tion proceeded the left end of the island adjoining this ravine was 
found to contain a layer of clay sloped in such a way as to endan- 
ger its stability if loaded with any of the dam superstructure. 
This was also removed and with the removal of another section 
of the island on account of the great uncertainty of the effective- 
ness of any device against leakage there was little left of the 
island and no expense was saved by its use in the construction. 



CUT-OFF WALL. 

The study of the foundation conditions on the Clackamas 
River was evidenced by the investigations and test pits, together 
with rough pressure tests, Mr. Diller's expressions relative to 
the geological formation, and the general hesitancy among engi- 
neers with respect to constructing high masonry dams upon such 
foimdations confirmed the original belief that somewhat novel 
methods must be adopted to give reasonable assurance concern- 
ing the integrity of the work to be constructed. Moreover, it 
was evident that the nature of these methods should be such as 
to permit demonstration of their efficacy before any large invest- 
ment in the construction should be made, and this involved some 
experimental work and expense which would not be involved on 
more substantial or satisfactory foundations. 

It will be evident that impermeable foimdations are desirable 
for the following reasons: 

(a) To minimize the possibility of upward pressure under the 



Digitized by 



Google 



384 SCHREIBER ON EsTACADA CONCRETE DaM. 

base of the dam superstructure. (This applies chiefly to solid 
dams.) 

(b) To prevent percolation under the dam which might lead 
to sufficient erosion to involve undermining of the structure. 

(c) To overcome any structured weakn^ess due to the original 
geological formation and properly provide a support for the 
superimposed load. 

(d) To avoid waste of water (the chief power asset of the 
company) from the reservoir, through, under or around the dam, 
instead of through the turbines, thus providing K.W.H. for sale. 

To satisfy these conditions with existing foundations of the 
character described in Mr. Diller*s reports required a departure 
from any methods heretofore used of which we had knowledge 
and in studjdng the problem it appeared that the only safe pro- 
gram must practically provide for actually changing the struc- 
tural character of the underlying formation. It occurred to our 
chief engineer, Mr. S. Howard Rippey, who had spent some time 
on the ground, that the most promising method would be the 
solidification of the porous material by the introduction of cement 
grout under pressure, but that the success of the work must be 
susceptible of demonstration by actual test before the super- 
structures should be started. Thorough inquiry failed to disclose 
any precedent for the use of grout for the general treatment of 
foundations, although it was foimd that cavities in limestone rock 
under the New Croton Dam had been filled with grout, much as a 
dentist would fill a cavity in a tooth. The grouting method had 
also been used in filling back of lining walls in tunnels, etc. 
Notwithstanding the absence of precedent, it was decided to pro- 
ceed with a grouting scheme and a program was outlined for 
preventing leakage of water from the reservoir created by the 
dam, under or around the dam, to the low tail water level below 
the dam and the experiments which should be made to properly 
demonstrate the efficacy of the method before the complete 
development should be imdertaken were prescribed. The general 
idea provided for drilling a double line of holes of an average 
depth of say 50 ft. under the heel of the dam across the entire 
valley to and under the shore abutments and the subsequent 
forcing into each of these holes of grout of such consistency as to 
percolate through the entire substructure and so permeate it as 



Digitized by 



Google 



SCHREIBER ON EsTACADA CONCRETE DaM. 385 

to solidify it absolutely throughout the entire length of the super- 
structure, thus making the foundations absolutely solid and the 
equivalent of a deep cut-off wall. 

It was recognized that experiments would be necessary to 
determine the proper spacing of the grout holes and their depth, 
which would ensiu^e sufficient diffusion of the grout through the 
varying material encoimtered to create a continuous impermeable 
barrier, thus preventing seepage of water from the reservoir imder 
the hydrostatic head which would be created by the dam. 

After drilling the double line of holes, the program con- 
templated the test of each hole with water pressiu^e, a record 
being kept of the quantity escaping and the pressure applied to 
each hole. This water pressure test also provided for washing 
out the interstices ready for the reception of the grout. Upon 
the completion of the tests, the cement grout was to be pumped 
into the holes imder pressure and after allowing time for harden- 
ing, a third line of holes was to be drilled midway between the 
first two or outer lines and tested with water pressure. The idea 
was that if water pressiu^e be applied to the center holes at or 
slightly above the hydrostatic pressure to which this rock would 
be subjected by the water in the reservoir after completion of the 
dam and no appreciable leakage occiured, we should feel rea- 
sonably certain that the cement grout had proven effective in 
making the entire foimdation impermeable. 

Before purchasing apparatus to handle the thin cement 
grout communication with several pump manufactiu*ers showed 
that in so far as the manufacturers' guarantees were concerned, 
choice was limited to two standard makes of hand-operated 
diaphragm pumps which would require 8 men on the handles to 
develop a pressure of 100 lb. per sq. in. Only one power dia- 
phragm pimip was offered by the manufacturers who would 
assume no responsibility for its successful operation. It was 
foimd that in both this and foreign coimtries plimger, diaphragm 
and centrifugal pumps had been successfully used, but after care- 
ful study of all the commercial pumps available it was decided 
that the use of compressed air would be more effective, flexible 
and economical than any mechanical pump. Having eliminated 
the question of pumps it was foimd that several different types 
of compressed air tanks were available; some being provided with 
different numbers and shapes of blades which were revolved in 



Digitized by 



Google 



386 



SCHREIBER ON EsTACADA CONCRETE DaM. 



the tank for the mechanical mixing of the cement grout while in 
others the grout was mixed entirely by the circulation of air. 

Interviews were had with a number of contractors and engi- 
neers who had had occasion to use the different types of machin- 
ery and it was finally decided to purchase the Canniff Pneumatic 
Grout Mixer and Injector. This machine, Fig. 3, consists of a 
plate steel cylinder with a conical shaped bottom and flat plate 



Blow off when nixing 




FIG. 3. — CANNIFF PNEUMATIC GROUT MIXER AND INJECTOR. 

steel top which is provided with a smaller hinged circular lid 
opening inward through which the water, cement and sand are 
introduced. The tank is provided with a blow-off valve on top 
of the lid, also an air inlet pipe tapped into the side near the top, 
a grout discharge pipe tapped into the bottom and a by-pass pipe 
between the air and grout pipes. The valves are so arranged that 
when the cement, sand and water are placed in the tank the com<< 



Digitized by 



Google 



SCHREIBER ON EsTACADA CONCRETE DaM. 387 

pressed air is forced into the bottom of the tank through the 
grout pipe and having free escape through the blow-off valve at 
the top, the grout is thoroughly mixed by the passage of air 
bubbles through it. 

When it is desired to discharge the grout the valves are 
so manipulated that the air is introduced near the top of the 
tank and the grout forced out through the bottom pipe, the press- 
ure being limited only by the capacity of the compressor and 
the strength of the tank. It is possible to connect any number 
of these tanks on to one main grout discharge pipe, thereby insur- 
ing a continuous flow of grout and by a proper system of pipe 
connections any one tank may be cut out for repairs without 
interfering with the operation of the remaining tanks. 

After experimenting on the Estacada foundations with both 
the shot and diamond core drills, an equipment of 7 Calyx shot 
drills was finally installed and operated at this development 
night and day for a period of about one year. These drills gave 
a hole varjdng from 2f to 8 in. in diameter depending upon the 
character of the bed-rock material. 

Where possible, in advance of the core drilling 3-in. wrought 
iron pipe casings were set to a depth of from 4 to 6 ft. in the rock 
and grouted, the upper ends projecting about 1 ft. above the 
surface and being threaded to permit the attachment of the 
testing and grouting apparatus. While the original plan called 
for grouting in advance of any construction, the failure to start 
this portion of the work at once tended to retard the actual con- 
struction work and it therefore became necessary to construct a 
concrete cut-off wall 7 to 10 ft. deep and set the casings in this 
wall and drill the holes through the casings into the rock below 
where this was necessary. The cut-off wall was placed upstream 
from the foot of the dam and the construction of the dam super- 
structure was continued without interference. 

The final report of Mr. Frank R. Fisher, resident engineer, 
for the Light and Power Department, Portland Railway, Light 
and Power Company, imder whose supervision the Estacada con- 
struction work was completed, gives the following additional 
detail information relative to the conditions which existed and 
methods used in the grouting: 

The rock mass is traced throughout with Beams, very irregular in shape 
and size, extending in all directions with but slight continuity. They vary 



Digitized by 



Google 



388 SCHREIBER ON EsTACADA CONCRETE DaM. 

from an almost imperceptible cleavage joint up to those having an approx- 
imate width of from one to two inches, the large majority observed meas- 
uring but a fraction of an inch. The seams are more or less choked with 
sand, gravel, small particles of rock, and other debris. No large crevices or 
faults were found at the site, nor observed in the vicinity of the dam. 

On the whole, the rock gives indication of possessing fair bearing value, 
and while not what would be classed as hard, would probably offer consid- 
erable resistance to the erosive action of water, except under high velocities. 

The plan was adopted of distributing drills over a wide area and each 
hole drilled was tested and grouted immediately on completion before putting 
down any other nearby holes. In grouting, the method usually followed was 
to make connection between the grout tanks and the casing by means of a 
flexible copper hose, and introduce the grout at the top of the hole; but in 
order to prevent, if possible, the rapid choking of the hole with cement, which 
frequently occurred, the method of introducing a pipe into the hole, and dis- 
charging the grout at various depths was tried out. For this purpose a 2-in. 
pipe, made up in sections, was used, and the operation was started with the 
same inserted to within a few feet of the bottom of the hole. When the hole 
gave evidence of tightening with the pipe in this position, one section was 
detached, usually 10 ft. in length, thus raising the outlet, and the operation 
repeated. At intervals a charge of water was shot in, to keep the pipe from 
plugging, and also to loosen up the cement that settled in the hole. This 
method of grouting through the pipes was given a thorough trial, but so far 
as could be observed it had very little advantage as to the amount of grout 
the hole would take, over the less laborious operation of introducing it 
directly at the top. The changing of the position of the pipes also interrupted 
the continuous flow of the grout, which it was desirable to maintain in order 
to accomplish the best results. 

The consistency of the grout was varied to meet the different conditions, 
1 part of cement to 5 of water appearing to give the best results, but the 
proportions tried out varied from 1 cement and 2 water to 1 cement and 
15 water. 

The grout was forced in under air pressure ranging from 50 to 200 lb. 
per sq. in., depending on the tightness of the hole, but the material at the 
lower depths would not tighten up beyond a certain degree. While it was 
desirable to use the higher pressures, in order to accomplish the greatest 
diffusion, it was not always practicable; to do so, on account of it blowing 
out at the surface. 

As the general nature of the rock had been investigated through the 
cores from the preliminary exploration holes, no special effort was made to 
preserve the same when drilling the primary holes. That from the proving 
holes, however, was carefully examined for any traces of grout, and while 
some very fair specimens of cement core were obtained, the amount was not 
large. This was probably due to the fact that most of the seams were small, 
and the cement entering therein would be ground to powder under the action 
of the bit. It is also probable that in many instances, at the time the prov- 
ing holes were put down, the cement had not set sufficiently hard to core. 



Digitized by 



Google 



SCHREIBER ON ESTACADA CONCRETE DaM. 389 

as the conditions arising from the various stages of the construction work, 
as well as the rate of progress to be maintained, made it necessary to follow 
on with the proving, within a short time after the grouting of the primary 
holes. 

The pressure testing of the proving holes was important and the final 
plan adopted gave a direct pressure from tanks located above the proposed 
pond water level as best approaching the final conditions. After the piping 
was filled, a test run of ten minutes was taken, the rate of seepage per minute 
being averaged and recorded. Besides testing individual holes, severs! 
combinations were tested in order to determine if possible the extent of inter- 
communication beneath the surface. The combined seepage in this was 
found to be considerably less than the sum of the separate tests and in some 
instances was only about half. An additional refinement was introduced by 
the testing of each proving hole in some sections every 10 ft. in depth as it 
was being treated, thereby giving an indication of the effectiveness of the 
grouted cut-off wall at different depths. In some instances communication 
was found to exist between holes located as far as 70 ft. apart. 

After the completion of the dam, and subsequent filling of the pond, two 
test holes were put down inside of the dam, approximately 30 ft. from the 
cut-off. One of these was located between buttresses Nos. 10 and 11 and the 
other between Nos. 12 and 13, opposite the weakest points in the grouted 
cut-off. After drilling to a certain depth, backfiow was obtained from each, 
and the maximum height to which the water rose in an extended pipe, due to 
upward pressure, was within 13 ft. of the elevation of that of the pond. The 
results of these tests indicate that some upward pressure exists in the founda- 
tion material in this locality, which the mass of overlying rock is resisting. 

The excavation of a supplementary cut-off trench at one point where 
grouting had previously been done offered excellent opportunity for observing 
its effects, as many seams were exposed and all of them proved to be well 
caulked with cement. 

The data of quantities and cost of this portion of the work is also of 
interest. 

Drilling 555 Holes, 34,038 Pr., Costino $1.55 per Pt. for Drilling 

AND Grouting. 
Labor — 

DrilUng $0.68 

Grouting 18 

Repairs 17 

Plant (drills and grout tanks) 30 

Cement 12 

Power 05 

Overhead (incorporated general plant) 32 

Total $1.72 

Less salvage 17 

$1.55 



Digitized by 



Google 



390 SCHREIBER ON EsTACADA CONCRETE DaM. 

In one section of the work the original pressure tests on the 
ungrouted holes frequently developed a seepage of over 100 
gallons per minute per hole and communication existed between 
holes located over a wide area and surface leakage also appeared 
in many places. Where the original primary holes gave an 
average leakage before grouting of 80 gallons per minute, thirteen 
proving holes when drilled showed an average leakage of 7 gallons 
per minute. Another section, for the first 30 ft. of depth below 
the concrete cut-ofif the treatment appeared to be fairly success- 
ful, but at the full depth a test on a group of thirteen proving 
holes showed an average seepage of 6.4 gallons per minute. In 
still other sections the proving holes showed seepage varying 
from 2.2 gallons to 3.6 gallons per minute. 

A total of 1942 bbls. of cement were used in grouting. The 
average depth drilled per hour was 1.32 ft., including time for 
moving, etc. The cement pumped into the holes varied from 
3.32 bbls. to 50 bbls. 

This treatment indicated the possibility of constructing what 
may be termed an effective cut-off without being literally imper- 
vious. There is however no assurance that this degree of success 
would result from similar work on another Qite with this kind 
of foundation. 

DESIGN OF DAM. 

Previous experience in the construction of hollow dams 
together with numerous studies of comparative costs and advan- 
tages inherent in hollow and solid dams prompted us to recom- 
mend the hollow dam for use in this instance particularly because 
of the character of the foundations. WhOe several other forms 
of hollow dams are available and have been carefully studied and 
some have been built, the distance at which the work had to be 
directed together with the urgency of the construction strongly 
favored the adoption of the reinforced deck and buttress form of 
hollow dam, on account of the existence of an experienced con- 
struction organization which could be started at once on the work 
and push it to completion without delay. 

The materials and stresses specified for use in this design 
were as follows: 

Concrete — 

Deck and apron 1:2:4 

Buttresses 1:3:6 



Digitized by 



Google 



SCHREIBER ON ESTACADA CONCRETE DaM. 391 

Reinforcement — Corrugated square bars 50,000 lb. elastic limit and 80,000 
to 100,000 lb. ultimate strength. 

Stresses — 

Modulus of elasticity of concrete 1,500,000 lb. 

Modulus of elasticity of steel 30,000,000 " 

Compressive stress in concrete 500 lb. per sq. in. 

Shear in concrete 75 " " 

Tension in concrete " " 

Tension in reinforcement 15,000 " " 

Base pressures 100 " " 

The hollow dam of the design selected consists of a series of 
parallel walls or buttresses running parallel to the thread of the 




FIQ. 4. — SECTION OP DAM THROUGH SPILLWAY. 

stream, with an upstream covering or deck on one side and, on 
the spillway section, a downstream covering or apron which ter- 
minates at its base in a heavy curved bucket section designed to 
divert and discharge the falling water downstream parallel to the 
river bed. A section of the dam through the spillway is given in 
Fig. 4, and through the intake in Fig. 5. 

The deck, which is inclined 45 deg. with the vertical, has a 
thickness that varies from 20 to 48 in. depending upon the depth. 
The apron which makes an angle of 30 deg. with the vertical also 
varies in thickness, being thickest at the crest and bucket but 
reduced to a thickness of 18 in. on the straight slope section. 
The reinforcement in deck and apron is laid horizontally 2 in. 



Digitized by 



Google 



392 



SCHREIBER ON EbTACADA CONCRETE DaM. 



from the under side of the slab with vertical bars on 24 in. spac- 
ing as an extra precaution against weakness in joints and to avoid 
temperature cracks. Hydrated lime, 30 lb. per cu. yd., was added 
to the deck, crest and apron to make the material less pervious. 
The extra cost for this material averaged 23 cents per yd. The 
buttresses spaced on 18 ft. centers are built in horizontal lifts of 
12 ft. and vary in thickness from 15 in. at the top to 38 in. at the 
bottom, being also tapered along their length for equalization of 
pressure. Haunches with 18-in. seats for the deck slabs are pro- 




FIG. 5. — SECTION OF DAM THROUGH INTAKE. 

vided at the upstream end and tongues extended between the 
slabs support the deck forms during construction and provide an 
opportunity to get a tight joint between the slab and the buttress. 
Because of the uncertain character of the foimdations the foot- 
ings were spread to reduce the base pressure to 100 lb. per sq. in. 
Additional reinforcement was also provided in the bottom lift 
to permit bridging any spaces of uncertain bearing value. The 
spillway section of the dam has a maximum height of about 86 ft. 
and the bulkhead section a maximum height of about 101 ft. 



Digitized by 



Google 



SCHREIBER ON ESTACADA CONCRETE DaM. 



393 



As constructed at Estacada the usual features were embodied 
in the spillway construction with the addition of sluice gates and 
the usual closing device in the left or main river channel. The 
spillway end abutment was of solid retaining wall section, a large 
part of it being built in gravel and clay, serving principally as a 
cut-off or core wall to prevent wash around the end of the dam. 
The right-hand or bulkhead end abutment was considerably 
longer, of reinforced retaining wall design. 

The island which, as stated, was included in the development 




iL^jUt&W.J 



Fia. 6. — METHOD OF CONSTRUCTION AND FORM WORK. 

was found to be of such poor material that about one-third of it 
was entirely removed to better bottom about 10 ft. above the 
water surface and the balance was notched out parallel to the 
deck and faced with 2 ft. of concrete which formed an extension 
of the deck and tied in with the cut-off wall at the base and a 
low section of dam on the top of this portion of the island. 

The total length of the spillway or overflow section extending 
from the left abutment to the non-overflow section on the island 
is 404 ft. 10 in. and has a rounded crest and heavy bucket and 



Digitized by 



Google 



394 



SCHREIBER ON ESTACADA CONCRETE DaM. 



apron extending over the downstream side of the island far 
enough to fully protect this portion of the structure from imder- 
mining. The buttresses in this portion are pierced near the top 
with openings used for a runway extending entirely through the 
dam from the power house up through the left abutment. 

The non-overflow or bulkhead section in which the but- 
tresses extend 15 ft. higher and are covered only on the upstream 
side and top, is open on the downstream side. One bay in this 




FIG. 7. — METHOD OF CONSTRUCTION AND FORM WORK. 

section contains a large concrete fishway tank discharging into a 
concrete fishway. 

Figs. 6 and 7 show the dam in course of construction and the 
various types of forms are illustrated in Figs. 8, 9 and 10. 

Because of the necessity of providing adequate trash rack 
area, supporting penstocks and meeting the requirements imposed 
by the high tension transformers and wiring installed here, the 
power-house section of the dam is quite complicated and required 
considerable thought and skill to work out the situation satisfac- 
torily. 



Digitized by 



Google 



SCHREIBEB ON EsTACADA CONCRETE DaM. 



395 



To locate the penstocks at convenient distances and properly 
support them on the adjoining buttresses as they pass from the 
intakes through the dam to the turbines, the spacing between 
buttresses is here made 14 ft. instead of 18 ft. as in the balance 
of the construction. An extra thickness of deck with additional 
reinforcement is provided aroimd the penstock intakes and to 
carry the weight and prevent vibration several extra heavy 
struts are built against each penstock so as to tie across to the 
adjoining buttresses. 

To provide means for getting machinery into the power 
house, which is built against the downstream ends of the but- 
tresses of the power-house section, a railway connection is 




«1 

I 
I 






FIG. 8. — ^METHOD OF HOLDING TONGUE AND HAUNCH FORMS IN PLACE. 

extended along the top of the right end abutment so as to enter 
directly on the unloading platform where crane service is avail- 
able to quickly lower material to the power-house floor level and 
where, after being transferred by a suitable truck, it is handled 
by the power-house crane. 

The deck for the intake section of the dam is set back from 
the face of the deck below it to allow space behind the trash 
racks for head gates and stop logs and to permit easy access of 
the water from racks to penstock opening and is of special con- 
struction on this account. 

The racks, hoists and loading platform are housed by a con- 
crete structure extending the full length of the power house. The 



Digitized by 



Google 



396 



SCHREIBER ON EsTACADA CONCRETE DaM. 



power house has a solid concrete substructure with flood protec- 
tion for a height of 14 ft. above which is a concrete and steel 
superstructure. 

The initial installation consists of three 6000 h.p. 240 r.j).m. 
twin flumC; center discharge, volute casing turbines, of special 
design originally suggested by us, each connected to a 3300 k.w., 
3-phase, 10,000-volt generator, and penstock and power-house 
space is provided for fwo. more units. 

These turbines feed from below through a cast iron Y pipe 
connection to their penstocks, have substantial cast iron bed 
plates and casings, presenting a neat and compact appearance. 




m 






--^ _ / _ .-----^ _ _ Iv^'^ 



a^ *. io» ftevin 






FIG. 9. — METHOD OF ANCHORING DECK SLAB FORMS. 

The oil type governors are provided with the usual distant control 
and other auxiliaries as well as extra flyball head for emergency 
"shut downs" in case the regular head fails to act. 

The generators are so supported on the foimdations as to 
have ample ventilation in all parts. A direct connected exciter 
on the outboard end of each shaft and one reserve combined 
turbine and motor-driven exciter provide the necessary exciting 
current. 

The 3-phase, 60-cycle, 33,000-volt delta and 57,000-volt Y 
step-up transformers, each carrying the load of the adjacent 
generator, are located in the dam back of the turbines and inmae- 



Digitized by 



Google 



SCHREIBER ON ESTACADA CONCRETE DaM. 



397 



diately above them are the 57,000-volt high tension and the 
10,000-volt low tension oil switches, bus bars, electrolytic light- 
ning arresters, etc., connecting with the two 3-phafie transmission 
lines to Portland. 



CONSTRUCTION. 



The scarcity of suitable construction materials was at fii^t a 
matter of considerable importance but with the assistance of 
Mr. Robert S. Edwards, consulting chemical engineer, of Port- 
land, who was retained to investigate the quality and costs of 



s 



BoUt to boia 



flMM dott*« an MM to np- 

•ort ToacM roiw oBtll boia 

(7 Belt*. BmovM UUt* ft»apr«t«iv 




■ttHhtiw teok Strei ' 
*• » 4- « 4- VMtar 



1- ■ I" h»U ta CtnanU ter 
4ptk-4, iB« VMf to ioppoit 



FIG. 10. — FORMS FOR BUTTRESS. 

sand, stone and cement available, satisfactory materials were 
obtained. 

Sand for construction in this section is ordinarily pumped 
from the Columbia River, but on investigation of the possibilities 
along the railway right of way a large pit was discovered which 
provided suflScient sand and gravel of good quality at consid- 
erable saving for not only this job but for much of the other 
construction work which the company had on hand. 

The only rock quarry opened up was also along the railway 
right of way, and though far from satisfactory produced a good 
quality of basalt rock at reasonable cost. 



Digitized by 



Google 



398 SCHREIBER ON ESTACADA CONCRETE DaM. 

Because of the fact that the cement manufacturing business 
on the Pacific Coast was still in a rather non-«ystematized state 
both as regards the standard of uniform quality and the matter 
of regular supply, the purchase of cement was considered only 
after a careful personal investigation and report on the available 
mills on the coast to determine the history and quality of the 
product and the probability of securing satisfactory shipments. 
Considerable time and expense were devoted to this investigation, 
but with these data in hand an advantageous contract based on 
satisfactory specifications was closed for sufi^cient cement to meet 
the company's requirements for several years at a considerable 
saving. 

The time for construction being quite limited with a large 
amoimt of excavation to be made, the construction plant invest- 
ment was considerable. The equipment for excavation consisted 
of 8 steam drills assisted by 5 derricks, 1 elipctric locomotive and a 
steam locomotive crane. For the core drilling and grouting there 
were required 6 Davis-Calyx core drills, pressure tanks for water 
testing and 2 motor-driven Peerless air compressors and 2 Canniff 
air-stirring tanks for grout supply. 

The sand pit was equipped with screen for eliminating the 
large stone and coarse gravel and with several bins which were 
arranged to empty directly into the railway cars which passed 
beneath them. The quarry equipment included a motor-driven 
crusher and screens together with bins for dumping directly into 
the cars. At the site liberal sand and stone bins were provided 
for each of the 2 electrically driven 1-yd. mixers first installed. 
A third steam-driven mixer was later installed and the three 
together were utilized to supply the two cableways which carried 
the concrete directly onto the work. 

Cement and lime storage sheds were also provided close to 
the mixers to meet the ordinary needs between receipt of ship- 
ments. 

The concrete mixing plant and material storage bins and 
sheds were located beyond the right end of the dam where they 
were fed by electric railway supply trains from the company's 
line to Cazadero. The mixed concrete was then fed to the cable- 
way buckets for distribution over the work. On the buttress 
forms tracks were arranged to carry a suitable concrete car which 



Digitized by 



Google 



SCHREIBEB ON ESTACADA CONCRETE DaM. 399 

received the concrete from the cableway bucket and distributed 
it as required in the buttress. 

The construction necessarily proceeded along lines dictated 
to a considerable extent by the large excavation necessary on the 
island and the time required for installation of machinery in the 
power house. 

Starting on the left channel during the beginning of the dry 
season the cut-off wall was installed, holes drilled and grouted 
and the first lift of a buttress constructed in midstream to serve 
in dividing this channel during the following dry season when 
completing the construction. During this time excavation and 
grouting were proceeding on the left bank and island above water 
line to permit concreting at these points to follow as soon as 
possible. Next the right or power-house channel was per- 
manently unwatered and the old river chasm found below the 
bed of the stream was cleaned out and work was pushed, on grout- 
ing and concreting to permit installation of racks, penstocks and 
machinery in advance of the completion of the spillway. The 
great amoimt of excavation on the island hindered this portion of 
the work to some extent, but fortunately no unusual floods were 
experienced and the entire structure up to the left river channel 
was completed early enough so that with the installation of two 
special openings in addition to the four sluice gates the stream 
flow could be passed imtil such time as it might be possible to 
discharge it over the crest. 

The construction work started in June, 1910, and the water 
first passed over the spillway November 7, 1911, which in view 
of the conditions to be met may well be considered quite rapid 
construction. The lar£:?st average force employed for any month 
was 655 men. The record excavation was 9476 cu. yd. per 
month. The greatest yardage of concrete placed any month was 
8325 cu. yds. 

Credit is due Mr. Robert S. Edwards, who had charge of the 
cement investigations and testing, and Mr. Frank R. Fisher, 
resident engineer on the later part of the construction, for data 
used in this paper. 



Digitized by 



Google 



UNIT COSTS OF REINFORCED CONCRETE 
FOR INDUSTRIAL BUILDINGS. 

By Chester S. Allen.* 

Unit costs are harmless when used with judgment and 
prudence, but likely to bring remorse and anguish when employed 
promiscuously. Rare and talented indeed is the man who pos- 
sesses the experience, judgment and intuitive sense to know 
when, where and how to properly modify any tables or state- 
ments of unit costs to meet the peculiar conditions of each indi- 
vidual case. While the figures given in this paper are all taken 
from structures erected during the past two years under the 
writer's supervision, the wide range of territory, local condi- 
tions, and different seasons of the year imder which the various 
pieces of work have been executed are so great as to render the 
information of value only in a very general way. 

As a general proposition it has been found that reinforced 
concrete is the lowest-priced fireproof material suitable for fac- 
tory construction and while it is true that its first cost will gen- 
erally run from 5 -to 20 per cent higher than first-class mill con- 
struction, recently in several instances, with lumber at a high 
price, reinforced concrete has worked out cheaper than brick and 
timber. It is especially adapted to heavy construction and for 
heavy loads of 200 lb. per sq. ft. and over where the spans are 
18 to 20 ft. centers, not even timber can compete with it. 

The unit costs of projected or completed buildings are com- 
monly figured either as so much per cubic foot or as so much 
per square foot of area occupied. Table I gives the unit costs 
both on the sq. ft. and the cu. ft. basis, together with a general 
description of a number of reinforced concrete industrial buildings 
of different types erected during the past two years. It will be 
seen from an examination of this table that the average cost per 
sq. ft. of these buildings, excluding the one-story structures, was 
$1.12; while the average cost per cu. ft. was 8.7 cts. The one- 
story structures both had reinforced concrete sawtooth roofs 

*■ Engineer, Lockwood, Greene & Company, Boston, Mass. 

(400) 



Digitized by 



Google 



Allen on Unit Costs op Reinforced Concrete. 401 



p 

< 
i 

OQ 

P 



o 
O 



o 
U 

M 



n 









5 






I ! 
fl 












HqqM^MM^MM 



-33 



8 



g gcQQQ'O'aQQGOCQ no 00 53 

S«^3SS||3 S 3 3 



|S2SS§|§ 



lOlQ 



^ 09 lO 00 



^e^^ioe^e^^^^ o ^ e« 






£3 






II 
.lgb|.||||i§i§ Sm 



j $ S Q 5 



3 . 

QQ OQ 



Digitized by 



Google 



402 Allen on Unit Costs of Reinforced Concrete. 

and the average cost per sq. ft. was $1.77, while 8.5 cts. was the 
average cost per cu. ft. The above costs are for the finished 
buildings, including plumbing, but do not embody heating, light- 
ing, elevators, sprinklers and power equipment. The cost per 
sq. ft. of floor area was obtained by dividing the cost of the 
building by the total number of sq. ft. of floor area exclusive 
of roof area but including basement floors; and the cost per cu. 
ft., by dividing the cubical contents into the cost of the structxire. 

While no coal pockets are included in Table I, it has been 
our experience that above 3000 tons capacity reinforced concrete 
elevator coal pockets cost from $5.50 to $7.50 per ton of capacity. 
Standpipes, exclusive of the foundations, average from 2^ to 3 cts. 
per gallon of capacity. 

On much of the reinforced concrete work which has been 
done imder our supervision it has been possible, owing to the 
contract being either on a percentage or cost plus a fixed sum 
basis, to obtain quite accurate and comprehensive cost data. 
This data, of course, is only of particular value when all the 
local color of each specific case is known, but the average results 
are at least interesting. 

The average unit cost of the 1-2-4 concrete in the floors 
including the beams, girders and slabs, was $6.10 per cu. >d., 
and for the columns $6.70 per cu. yd. Where 1-1^-3 mixture 
was used for the columns the average cost was $7.60 per cu. yd. 
This cost was made up of the items of cement, sand, stone or 
gravel, labor and plant. The cement of course varied greatly 
with the demand, but the average net cost was $1.35 per barrel 
including 3 cts. for tests. The sand averaged 80 cts. per 
cu. yd. and the crushed stone $1.25 per cu. yd. The cost of 
labor of unloading the materials and mixing and placing the 
concrete varied from 65 cts. to $2.90 per cu. yd. The cost of 
plant, consisting of freight, depreciation or rental of mixing and 
hoisting towers, erection of same, power and coal, and losses 
and waste on the small tools, ranged from 50 cts. to $1.50 per 
cu. yd. of concrete placed. 

Next to the proper design of the structural features of a 
concrete building, the economical design of the form work is of 
paramoimt importance. The truth of this statement is borne 
out by the fact that on the average job the cost of the forms 



Digitized by 



Google 



Allen on Unit Costs of Reinforced Concrete. 403 

amounts to about one-third the cost of the entire structure. On 
the buildings under consideration the average cost of the forms 
for the floors, including beams, girders and slabs, was 10 cts. 
per sq. ft., and for the columns 13 cts. per sq. ft. The lowest 
cost was in a building of the girderless or flat slab type of con- 
struction, where by the intelligent use of corrugated iron for 
the slab forms the cost of the floor forms, including wall beams, 
was 7 cts. per sq. ft. The highest cost was for an artistic but 
not elaborate overhanging cornice on a 12-story building, and 
was 32 cts. per sq. ft. This last item rather forcibly demon- 
strates that any attempt at architectural development is very 
apt to be a costly proposition. 

The cost of the labor of making, erecting and stripping the 
forms varied according to the price of lumber, design of the 
structure, method of forming, character of the supervision and 
the skill of the workmen from 4J to 12 cts. per sq. ft. The cost 
of lumber, nails and oil divided by the sq. ft. of forms 
averaged from 2J to ^ cts. per sq. ft. 

The cost of bending and placing the reinforcing metal, in- 
cluding the necessary wire, averages $10 per ton, the range being 
from $5.75 to $17.20 per ton. 

Granolithic floor finish l}-in. thick when laid before the 
concrete below it had set so as to form one homogeneous slab, 
cost on the average of 4^ cts. per sq. ft. When put on after the 
rough concrete slab, the cost averaged 7 cts. per sq. ft. 

Inasmuch as the only economical design of a reinforced 
concrete structure is one which closely resembles that of the 
steel skeleton type, the relative cost of the various materials 
commonly used for curtain walls under the windows may be of 
interest. The writer has used brick, vitrified tile, concrete blocks, 
cast concrete slabs and solid concrete walls for this purpose. 

The most common type of curtain wall has been either an 
8-in. or 12-in. brick wall resting on the concrete wall beam. The 
average cost of these walls has been 45 cts. per sq. ft. There 
is practically no difference in cost between the 8-in. and the 
12-in. brick curtain wall, as the saving in material is offset by 
the great amount of extra labor in culling and laying the thinner 
wall. 

An excellent and inexpensive curtain wall is constructed 



Digitized by 



Google 



404 All^n on Unit Costs op Reinforced Concrete. 

by using 8 x 12 x 18 in. vitrified tile. This is a non-absorbent 
wall and when properly laid in cement mortar makes a tight 
weather-proof curtain wall. The cost of this wall averages about 
25 cts. per sq. ft. If the tile is plastered both sides, the cost is 
about 38 cts. per sq. ft. 

Where 8-in. concrete curtain walls were cast in place after 
the skeleton frame was completed, the average cost was 40 cts. 
per sq. ft., and when poured simultaneously with the columns 
48 cts. per sq. ft. 4-in. cast concrete slabs cost about 35 cts. 
per sq. ft. 

While concrete blocks make a very cheap and light curtain 
wall, the price being about the same as for the 8-in. tile, the 
writer's experience with them has been rather unfortunate on 
account of the extreme porosity of the blocks used. 

Where the location of the buildings has demanded special 
treatment of the exposed surfaces, they have generally been 
specified to be rubbed with a block of carborundum. The aver- 
age cost of this work has been 4 cts. per sq. ft. In two instances 
portions of the structures have been bush hammered with a result- 
ing average cost of 7 cts. per sq. ft. 

Concrete piles were used on the foundations of several of 
the buildings and the average cost of the piles was $1.15 per 
lin. ft. 

The most common methods of waterproofing concrete struc- 
tures are by the introduction of foreign ingredients into the con- 
crete, by the application of a compound to the concrete surface, 
by the use of paper or felt waterproofing, and by accurately 
grading and proportioning the aggregates and the cement. 

Where an addition of hydrated lime in the proportions of 
10 per cent to the weight of the cement has been used, the added 
cost to a cubic yard of 1-2-4 concrete has been 50 cts. Patented 
compoimds have cost from 25 to 35 cts. per sq. ft. of surface 
covered. On horizontal or inclined surfaces, we have sometimes 
used a granolithic surface of rich mortar of Portland cement and 
sand or Portland cement and screenings in the proportions of 
1-1, laid at the same time as the base and troweled as in side- 
walk construction. The cost of this work has been about 5 cts. 
per sq. ft. 

Taken as a whole, the lowest possible cost on a reinforced 



Digitized by 



Google 



Allen on Unit Costs of Reinforced Concrete. 405 

concrete building can be obtained only by a careful study of 
each particular case to determine the cheapest type of construc- 
tion and most economical spacing of columns. As a general 
proposition it has been found that for light loads with ordinary 
beam and girder construction the most economical spacing of 
columns is 18 ft. each way and for flat slab construction 20 ft. 
each way. For heavy loads such as 300 lb. per sq. ft. and over, 
it has been our experience that the cheapest colimm spacing for 
beam and girder construction is 15 ft. by 15 ft., and for flat slab 
construction 17 ft. by 17 ft. In arriving at the most economical 
layout it is always well to bear in mind that the construction 
which allows the greatest simplicity of form units, together with 
the maximum number of repetitions of same, is invariably the 
one that will work out cheapest in the end. The fact that the 
actual amoimt of concrete or reinforcement required for a certain 
floor construction is less than that required in another by no 
means implies that this is actually the cheapest floor construc- 
tion, as the unit labor of the form work may easily have been 
increased out of all proportion. 



Digitized by 



Google 



REINFORCED CONCJRETE CONVENTION HALL AT 
BRESLAU, GERMANY.* 

By Dr. S. J. TRAUER.f 

There is being built at the present time in the City of 
Breslau, Germany, a large convention and exhibition hall (Fig. 1), 
surmounted by the largest concrete dome in the world. The 
building, of reinforced concrete throughout, has a seating capacity 
of 9000 persons and standing room for 12,000 people. 

The structure (Fig. 2) consists essentially of a main hall, 




FIG. 1. — CONVENTION HALL, BRESLAU, GERMANY. 

circular in form, connecting directly with four semi-circular halls 
called Apsiden, all of which are surrounded by a lower circular 
hall serving exhibition purposes. A dome of 213 ft. span with 
a rise of 65 ft. rests on the substructure, 65 ft. in height, and 
carries a light dome. 

The substructure (Fig. 3) consists of four main arches, (A) 
of 131 ft. span and 62 ft. rise, circular in plan and supported 
by four piers. Each main arch (^4) is supported from the out- 
side by four auxiliary arches (B) which cover a smaller semi- 



* Translated from author's notes by the Secretary. 
fCity Bridge Engineer, Breslau, Germany. 

(406) 



Digitized by 



Google 



Traubr on Reinforced Concrete Convention Hall. 407 

circular hall called an Apside. These auxiliary arches rest on 
individual piers and take the outward thrust from the main 
arches. The area of the cross-section of the main arch at the 
skew-back is 24 sq. ft. and at the crown 3.6 sq. ft., all reinforced 
with round bars of 1.18 in. diameter. The main arches are soUd 
and are subject to their own temperature stresses only. The 



Smfi^'^^p. 




FIG. 2. — PLAN OF CONVENTION HALL, BRESLAU, GERMANY. 

horizontal thrust amounts to 771 tons, the pressure at the skew- 
back is 1323 tons and in addition the main arches are subjected 
to torsion and bending. 

The auxiliary arches (JS) which receive about 220 tons thrust 
from the main arches, are connected with the main arches and 
the small abutments through ball bearings (C) of malleable steel 



Digitized by 



Google 



408 Trauer on Reinforced Concrete Convention Hall. 

castings so that the auxiliary arches receive only axial stress. 
These arches are 3.28 ft. in width and 5.28 ft. in thickness. 

The substructure is entirely independent of the dome itself 
and supports the latter by means of steel roller bearings (D) under 
each rib, the bearings having radial movement. So that in 
general only vertical stresses are transmitted to the substructure, 
which is, therefore, not subject to the temperature stresses in the 
dome. The wind pressure on the dome is not transmitted in a 




FIG. 3. — SECTION THROUGH MAIN AXIS. 

radial but in a tangential direction, that is, the direction in which 
the four main arches have the highest power of resistance. The 
frames (E) over the four abutments of the structures are calcu- 
lated to take up the total wind load on the dome. 

The dome consists of 32 half ribs (F), which bear on the 
top against the pressure ring ((?) and on the bottom against the 
tension rmg (H). The pressure ring of 47.2 ft. inside diameter 
is surmoimted by an upper light dome (•/)• The pressure ring 
with a cross-sectional area of 19.7 sq. ft. carries 551 tons in com- 



Digitized by 



Goo'gle 



Trauer on Reinforced Concrete Convention Hall. 409 






FIG. 4. — RIVETED STEEL TENSION RING AND HAUNCH OF ONE RIB. 




FIG. 5. — SUBSTRUCTURE AND CENTERING FOR HALF-RIBS OF DOM£. 



Digitized by 



Google 



410 Trauer on Reinforced Concrete Convention Hall. 




FI.J. (). — CONCRETING HALF-RIBS OF DOME. 







^''«'IIJIil iillllMII'tt'»^"' - 






FIG. 7. — VIEW OF INTERIOR. 



Digitized by 



Google 



Trauer on Reinforced Concrete Convention Hall. 411 

pression and the tension ring (Fig. 4), with a cross-sectional area 
of 29.9 sq. ft., carries 551 tons in tension. The ribs carry a hori- 
zontal thrust of 110 tons and 115 tons on the skew-back and 
have a cross-sectional area which increases from 3.4 ft. x 2.13 
ft. to 3.94 ft. X 2.62 ft. The ribs and the tension and pressure 
rings are reinforced with round rods. The ribs are strengthened 
by means of rings (M) and horizontal circular slabs (K). The 
tension ring is of riveted steel construction. 

The roof supports are almost horizontal, the windows (M) 
are vertical so that in the case of snow the interior illumination 
will not be affected. 

The architectural design was made by the City Building 
Engineer, Berg; the calculations by the City Bridge Engineer, 
Dr. Trauer, and the verification and construction by the firm 
of Dyckerhoff & Widmann, of Dresden, Germany. 



Digitized by 



Google 



THE SUITABILITY OF CONCRETE FOR GAS HOLDER 

TANKS. 

By Herbert W. Alrich.* 

In its purpose the gas holder represents the common feature 
of all manufacturing enterprises — ^it is the receptacle for the 
storage of the product. But in their design, gas holders are not 
resembled by any structure or equipment employed in any other 
line of industry. A prominent engineering writer has recently 
said that the modem gas holder is ''A magnificent achievement 
in engineering, and one of the wonders of it is the telescopic 
feature." As the majority of engineers are not familiar with 
the mechanical features of a gas holder, a description is given; 
for in discussing tank design, it is necessary to consider the 
structure as a whole. 

There are three principle parts to a gas holder; the tank, 
the telescopic sections or gas holder proper, and the guide frame. 
Each of these parts differs from the others in function and in 
the type of construction. The tank, resting upon the ground, 
is filled with water up to a level about 15 in. below the top. The 
gas holder proper consists of the telescopic sections, of which 
there are five in the case of the largest holders that have been 
built in this country. These sections consist of cylindrical steel 
shells, concentrically located with relation to each other and to 
the tank. When the holder contains no gas, these shells are 
nested together, resting upon the bottom of the tank and sub- 
merged almost completely in the water which the tank contains. 
(See Fig. L) The outer-most shell is usually about 3 ft. less 
in diameter than the tank, and each succeeding section is about 
2 ft. 9 in. less in diameter than the preceding one. All of the 
sections are open at each end, with the exception of the inner- 
most one, the upper end of which is enclosed by crown plating, 
having a spherical form. 

The inlet and outlet pipe connections enter the tank through 



* Engineers' Department. Coneolidated Gas Company of New York. 

(412) 



Digitized by 



Gqogle 



Alrich on Concrete''for Gas Holder Tanks. 



413 



the bottom and pass vertically upward through the water and 
termmate at an elevation above the water and just under the 
crown of the inner section. When gas is admitted to the holder, 
it first acts upon the crown, lifting that section gradually out of 
the water as shown by Fig. 2, until the cup, constructed around 
the lower edge of that shell, engages the upper edge of the next 




FIG. 1. — CROSS SECTION OF A GAS HOLDER GROUNDED. 

outside section. To accomplish this engagement, the upper edge 
of each outer section is constructed in the form of a continuous 
annular hook, called the grip, and correspondingly the lower 
edge of each section, except the outermost one is formed into a 
continuous annular cup. Both the grip and the cup are identical 
in construction, and differ only in the respect that the grip is 
inverted to permit interlocking with the cup as shown by Fig. 3. 



Digitized by 



Google 



414 Alrich on Concrete for Gas Holder Tanks. 

As the inflation of the holder continues, each succeeding section 
is lifted out of the water, and, in turn, picks up the next outer 
section. As each interlocked cup and grip pass upward out of 
the tank, there is carried along that quantity of water which is 
necessary for forming a hydraulic seal against the maximum 




FIG. 2. — HOLDER PROPER JUST BEFORE ENQAQINO AN ADDITIONAL SECTION. 

pressure of the holder. This action continues until the holder 
is filled to the limit of its capacity as shown by Fig. 4. 

While thus ascending, or conversely descending, the sec- 
tions are maintained in their relative concentric positions by 
two sets of rollers. First, the rollers spaced equi-distant around 
the lower edges of the shells, and second, the rollers mounted 
on brackets attached to the upper edges of the shells. The first 



Digitized by 



Google 



Alrich on Concrete for Gas Holder Tanks. 415 




Digitized by 



Google 



416 Alrich on Concrete for Gas Holder Tanks. 

set are known as the internal rollers, and travel upon vertical 
guides, located on the inner surface of the tank and also on the 
inner siurface of each shell, except that of the innermost section. 
The other set of rollers travel upon the guide rails, which are 
carried by the guide frame. The guide frame consists of vertical 
columns spaced equi-distant, and about 30 ft. apart around the 




FIG. 4. — HOLDER FULLY INFLATED. 

circumference of the tank. These columns are connected by 
horizontal and diagonal cross bracing. The frame is carried to 
such a height as will provide guidance for the inner section in 
its maximum upward travel. It will be apparent from this dis- 
cussion that the enormous pressure of the wind against the inflated 
sections is ultimately transmitted to the guide frame by the tank. 



Qigitized by 



Google 



Alrich on Concrete for Gas Holder Tanks. 417 

In the selection of materials in manufacturing, we are sub- 
ject to an inexorable law of suitability. In producing any manu- 
facture, there will be ultimately employed that material which 
possesses preponderant advantage. The use of any other cannot 
long be sustained. From the very beginning of the gas industry, 
100 years ago, the guide frames and holder sections have been 
built exclusively of iron or steel. It does not appear that any 
attempt was ever made to use any other materials. In the con- 
struction of the tanks, however, many different materials and 
combinations of materials have been employed. It is true, how- 
ever, that in this country, until recent years, all holder tanks 
were built either of brick or stone; but in England, the birthplace 
of the gas industry, there were constructed tanks of almost every 
conceivable type, many of the designs being more fanciful than 
practical. It was inevitable that those constructions without 
merit should disappear until the prevailing practice, throughout 
the world, during the last twenty years, may be stated as limited 
to steel tanks placed upon the ground, and brick or concrete 
tanks located below the ground. During the last ten years the 
preponderance of advantage in this country has been in favor of 
steel tanks for all holders, large or small. In England, though 
concrete tanks had been constructed as far back as the year 1870, 
the practice of the last decade has been quite imiformly to build 
small tanks of steel, and large tanks of brick, there being but few 
instances of concrete construction. In Continental Europe, 
practice for a decade has followed the English, with the excep- 
tion that during the last three or four years there has been a 
marked tendency toward the exclusive employment of steel, 
though a few concrete tanks of small size have also been con- 
structed. 

The competitive, economic and geological conditions exist- 
ing in England, all favor the brick tank for large holders. While 
the population of the Island of Great Britain is somewhat less 
than that of the United States, the number of concerns engaged 
in holder construction is very much greater, in fact the building 
of the largest holders in this country is confined to a very few 
companies, having special qualifications and known responsibility. 
Hence, the commercial conditions prevailing in this country have 
enabled these firms to develop extensive equipments, which have 



Digitized by 



Google 



418 Alrich on Concrete for Gas Holder Tanks. 

wonderfully facilitated the construction of steel tanks. When 
selecting sites for large holders in England, they appear to have 
generally encountered formations of clay possessing remarkable 
stability. This has permitted them to construct economical 
tanks by lining a cylindrical excavation in the clay with brick 
walls, which are thinner than our geological conditions will 
generally permit. Unskilled labor is relatively very cheap in 
England, and in some cases, the cost of construction has been 
partly defrayed by the commercial value of the excavated clay 
In the case of the 10,000,000-cu. ft. holder erected at Manchester 
in 1909 and 1910, the brick for the tank walls were manufactured 
upon the site from the excavated clay. It is a fact, however, 
that the time required for the construction of this particular 
holder would be regarded as economically impossible in the 
United States. The result of all of these conditions has been 
that the English holder builders have not developed the special 
equipment required for the construction of very large steel 
tanks, nor does it appear that the British steel plants produce 
plates suitable for the purpose. 

To confine the discussion now to American conditions the 
controlling requirements in the design of a gas holder are: 

1. Structural Stability. 

2. Economy. 

3. Rapidity of Construction. 

4. Durability. 

Certain physical phenomena having to do with the distribution 
of gas, require the holders to be located upon the lowest avail- 
able ground. In this coimtry such sites rarely consist of a very 
stable geological formation, and are frequently reclaimed land. 
Hence, if under the usual conditions there be undertaken the 
construction of an underground masonry tank, there will probably 
be encoimtered such diflSculties as to greatly enhance the cost 
of the work and prolong the period of construction. If it should 
be suggested that a concrete tank might be built above the ground, 
a moment's reflection upon its relation to the guide frame is suffi- 
cient to dispose of the proposition. 

During the last eight years the writer has participated in the 
design and construction of the three largest concrete holder tanks 
in the world. Two of these tanks are 300 ft. in diameter by 48 



Digitized by 



Google 



Alrich on Concrete for Gas Holder Tanks. 419 

ft. 3 in. deep, while the third is 189 ft. in diameter and 41 ft. 6 in. 
deep, all inside dimensions. In referring to these particular 
tanks, and also throughout the discussion, he wishes to be under- 
stood as speaking for himself alone. The work on these three 
tanks proceeded in such simultaneous relation as permitted any 
improvement in methods developed on one, to be employed to 
advantage on another. 

At the time the tanks were being designed, Mr. William H. 
Bradley, the Chief Engineer of the Consolidated Gas Company 
of New York, made a thorough inquiry as to the design of any 
other concrete tanks that had been constructed anywhere in 
the world. There was foimd little precedence to consider. As 
regards the two 300-ft. tanks, the undertaking was one of unusual 
n^agnitude, for the holders were the largest that have ever been 
built. The possibilities of nickel steel were not at that time 
fully understood, hence the construction of steel tanks would 
have been unthinkable, as there would have been required plates 
4 in. thick, connected by rivets 3 in. in diameter. It was evident 
that masonry construction of some kind must be employed. A 
number of different designs were worked out, analyzed and com- 
pared, resulting in the conclusion that the most advantageous 
construction would be a plain annular wall of reinforced concrete, 
and all three tanks were thus built. The work upon all three 
tanks was accomplished with entire success. While there were 
special conditions, the controlling factors in these instances, 
and determining the selection of concrete construction, the 
writer concludes that steel is far preferable, as a general proposi- 
tion. The smallest of these three tanks required a full year for 
its construction. An entire holder of the same size having a 
steel tank, can be constructed in seven months, without special 
effort. It has also been demonstrated to the satisfaction of the 
writer, that in no case where a steel tank is at all possible, will 
the cost reach 70 per cent of the outlay required for a correspond- 
ing concrete construction. 

The question as to whether steel or concrete tanks are the 
more durable, must be settled in another generation; but the 
writer believes that the steel tanks are the more expedient. He 
has knowledge of several brick tanks having been badly damaged 
by blasting in their vicinity. From his knowledge of other con- 



Digitized by 



Google 



420 Alrich on Concrete for Gas Holder Tanks. 

Crete structures having been badly cracked in a similar way, 
he must conclude that concrete tanks are not immune from the 
hazard. The durability of concrete tanks is also open to question, 
in view of the peculiar behavior of concrete observed by the 
writer, in the instance of two of the most prominent engineering 
undertakings in the United States, in which cases, several years 
after the concrete had set, masses of it became soft and pulpy, 
requiring replacement. A holder tank is not such a structure 
as one upon which such repairs can be easily made, but it is one 
of such importance as to forbid speculative constructions. 

The writer has examined and reported upon a large number 
of steel tanks, some of which had been in service for over twenty 
years. While these older tanks are doubtless constructed of a 
steel far inferior to that which is now produced, all of these tanks, 
with one exception, could survive the youngest man now engaged 
in the gas business. It has sometimes been urged against the 
durability of steel tanks, that all favorable conclusions had been 
formed from exterior examinations. The writer has had the 
opportunity of examining internally, some of the oldest steel 
tanks in existence, through which he concludes that, excepting 
at the water line, there is no perceptible deterioration. In one 
case, the holder had been employed for twelve years, in connec- 
tion with a process of gas manufacture now obsolete. As a part 
of this process, a non-luminous gas containing some sulphur 
was stored in this holder, which resulted in the sulphur being 
absorbed by the water in the tank. The writer found the interior 
of this tank to be but very slightly pitted. 

The writer would state it as his conviction, that while not 
possessing one single element of advantage over steel tanks, 
that concrete tanks are subject to the following comparative 
disadvantages: 

1. Increased cost, generally 75 per cent, 

2. Longer period required for construction, generally 100 
per cent for the entire holder. 

3. Liability to impairment from unforeseen or unavoidable 
causes, such as internal stresses, bad water, oil or alkaline soil 
and blasting. 

4. Difficulty of making repairs and obtaining the original 
strength in case of rupture. 



Digitized by 



Google 



AuacH ON Concrete for Gas Holder Tanks. 421 

5. Possibility of corrosion of reinforcement, the vital ele- 
ment of the tank's strength. 

6. Inaccessibility of metal to inspection. 

7. Greater load on fomidations. 

8. Requires the most rigid inspection and supervision to 
control the quality of the work during construction. 

9. Liability to serious dfficulties from storms during con- 
struction, due to the depth of excavation, the requirement of 
maintaining adjacent streets, requiring sheet piling 50 ft. deep 
and massive shoring timbers. Also liability of damage to the 
concrete work itself from the same cause. 

In submitting a discussion of this question to the American 
Gas Institute in 1910, the writer urged the following as the twelve 
distinct advantages of steel tanks when compared with those of 
masonry: 

1. Less cost. 

2. Shorter period required for construction. 

3. The ease with which the quality of the work may be con- 
trolled during construction. 

4. The high state of development in fabrication and erec- 
tion. 

5. Susceptibility to exact computation and greater reliability 
under stress. 

6. Accessibility for inspection. 

7. Tank may be placed at any elevation with relation to 
the ground line that may be desired. 

8. No liability to damage by storm durmg construction. 

9. No internal stresses from shrinkage or temperature that 
are serious. 

10. Possibility of rectifying an imequal settlement. 

11. No liability to cracking from undetermined causes. 

12. The ease of making repairs and obtaining the original 
strength. 

Every one of the enumerated disadvantages of concrete 
tanks, and advantages of steel tanks might be separately ex- 
tended and elaborated, but the writer will confine himself to 
items 2 and 4 under steel tanks. 

The bottom of a steel tank, excepting an outer course, con- 
sists of rectangular steel plates, generally about f in. thick, con- 



Digitized by 



Google 



422 Alrich on Concrete for Gas Holder Tanks. 

nected by single riveted lap joints. The outer course consists 
of heavy segmental plates, which conform the bottom to the 
circle. Riveted around the outer edge of this outer course, is 
the bottom curb, which, in the case of a large tank, will be an 
8 X 8 X IJ in. angle. The bottom course of curved plates, form- 
ing the shell of the tank, is riveted to the upstanding leg of this 
curb angle. The next course of plating is then attached to the 
lowest course by a single riveted lap joint running horizontally 
around the tank circumference. The other courses are connected 
in a similar manner, the entire number in the height of the tank 
usually being about eight or nine. It is necessary for the vertical 
joints, which occur about 30 ft. apart around the circumference, 
to be spliced in a manner capable of resisting the full circum- 
ferential tension. To accomplish this, the joints are covered 
by double butt straps, each such splice being quadruple riveted 
with double shear rivets, and triple riveted with single shear 
rivets. The total number of rivets in the cylindrical shell of the 
tank alone, may be as many as 25,000. 

After the bottom of the tank has been completed, a vertical 
steel post is erected at its geometric center. A circular rail is 
also laid concentrically upon the tank bottom. Mounted upon 
trunnions, upon the top of the central steel post, are two radial 
traveler arms, the outer ends of which are supported by legs 
having wheels running upon the circular rail. These radial 
travelers are equipped with such suitable attachments, as will 
permit one of them to be used for assembling the curved plates 
while the other supports the hydraulic-pneumatic riveting 
machine. Thus, as one traveler proceeds around the circumfer- 
ence lifting the plates into position, it is followed by the other 
with the riveter. Quite obviously, the first traveler is able to 
keep well ahead of the riveting, which permits that traveler to 
participate in the simultaneous erection of the five-holder shells. 
When the tank and the holder sections have been completed, 
the tall center post around which the travelers revolve, is replaced 
by a much shorter one, resting upon the center of the crown, 
while the outer leg of one of the travelers has its lower section 
removed, and thus shortened, it travels upon a circular rail which 
has been attached to the completed crown. The other radial 
traveler is entirely removed. After water has been placed in the 



Digitized by 



Google 



AuacH ON Concrete for Gas Holder Tanks. 423 

tank, the holder is gradually inflated with air, and as it rises the 
remaining radial traveler moves aromid the circumference, erect- 
ing the guide frame successively in tiers. It requires but little 
reflection to perceive that if the holder is to be provided with a 
concrete tank, any such co-ordination is impossible. As to the 
time required, it may be stated that in the instance of the largest 
steel tanks ever built, which were 251 ft. 3 in. diameter, the lower 
courses of which consisted of 2]^-in. plates and having 26 double 
butt joints in each course, the tanks were erected and riveted 
complete at the rate of 32 hours per course. One of these holders 
was entirely completed in 7j months. If there be a method of 
concrete tank construction which would have permitted the 
completion of that same holder in two years, it has not yet been 
found. 

In closing, the writer will condense into a few words, his 
opinion on the matter as stated in another discussion. "He may 
be charged with entertaining a strong prejudice against masonry 
tanks. He will admit that he is opposed to any type of tank, 
25 to 40 per cent of the cost of which may go into digging a hole 
in the ground instead of putting quality into the structure.'' 
To this he would add his conviction that a site which forbids a 
steel tank is not the place to put a holder. 



Digitized by 



Google 



PROTECTION OF STEEL IN CATSKILL AQUEDUCT 
PIPE SIPHONS. 

By Alfred D. Flinn.* 

Cement manufacturers and many cement users are already 
so familiar with the Catskill aqueduct which New York City is 
building to convey an additional supply of water from the Catskill 
mountains that a general description is not here necessary. This 
aqueduct is to have a nominal capacity of 500,000;000 gallons 
daily. In the 92 miles of its length, it crosses 14 minor valleys 
where metal pipe siphons were determined upon in preference to 
reinforced concrete pipes or deep pressure timnels in rock; three 
of these siphons are west of the Hudson river, and eleven east; 
in seven of them the diameter of the steel shell is 9 ft. 6 in.; in 
four this diameter is 9 ft. 9 in., and in the remaining three, 11 ft. 
3 in. These various diameters were determined by an economic 
distribution of the available fall or head. The length of the 
siphons varies from 608 to 6671 ft.; one of the U-ft. siphons 
is 5584 ft. long; the total length of all siphons is 33,031 ft. The 
thicknesses of plates are i^-, i-, xk-i H" a^^d i"^^- Th^ mairimum 
heads on the siphons range from 50 to 340 ft. With one excep- 
tion, each siphon rises to the hydraulic gradient at each end and 
is there connected by means of a concrete chamber to the adjacent 
portions of the cut-and-cbver or tunnel aqueduct. They are all 
of open hearth steel. Ultimately there will be 3 pipes in each 
siphon, in order that the siphons may have the full capacity of 
the aqueduct and also provide for one pipe being temporarily 
out of service for cleaning, repairs or renewal. Only the middle 
pipe of each siphon is being laid at this time. The 7 more northerly 
siphons are included in Contract 62 and the 7 southerly siphons 
in Contract 68. 

For years, both in connection with the work of the Board 
of Water Supply and in other engagements, several of the Board's 
engineers have been observing the results obtained by coating 
steel pipes with the asphalt, tar and other dips commonly em- 

* Department Engineer, Board of Water Supply of the City of New York. 

(424) 



Digitized by 



Google 



Flinn on Protection op Steel Pipe with Cement. 425 

ployed. Study had also been made of some special forms of 
protection of steel pipes. Excepting Portland cement grout 
or mortar used in one or two cases, in a small way, none of the 
coatings of which knowledge could be had gave evidence of real 
permanence, nor have they been fully satisfactory in other respects; 
therefore it was decided to jacket the steel pipes of the aqueduct 
outside with rich concrete, and to line them with Portland cement 
mortar. The outside concrete was to have a minimum thickness 
of 6 in., and the mortar lining a thickness of not less than 2 in.; 
after some experimentation, these dimensions were definitely 
adopted, as shown on Fig. 1, which is a reproduction of the draw- 
ing of the standard types of pipe construction. 

Means for applying the concrete and mortar to the pipes 
so as to attain intimate, complete and permanent adhesion, in 
spite of the unavoidable distortions due to handling, tempera- 
ture changes, and the water stresses, were studied, along with 
the most economical and feasible methods for the various steps 
in construction. Preparation of the steel was also carefully 
studied. Numerous experiments on a small scale, and finally 
on substantially full-size, were conducted. For the latter, a steel 
pipe 9 fib. in diameter and 12 ft. long, of |-in. riveted plates, was 
lined by plastering and by pouring grout or very thin mortar 
into the space between a cylindrical form and the inner surface 
of the pipe. In the plastering experiments several kinds of metal 
reinforcement were tried; also, terra-cotta and cement blocks 
or tiles were bedded on mortar and plastered. Briefly, it may 
be stated that no combination of plasterer's skill, with the various 
materials suggested, gave linings that were adequate and this 
method was expensive. The methods using sohd*tiles or blocks 
were more successful than those using any form of metal lath. 
But when removed, all the plaster coatings showed a tendency 
to separate at the surfaces between the successive layers. Grout- 
ing proved by far the most satisfactory and least expensive, 
and was adopted as the basis of the contracts. 

In fabricating the pipes, the plates were bevel-planed on 
their edges, then punched for the rivets and then each plate was 
bent to proper radius by bending rolls. This bending cracked 
the mill scale and removed a considerable portion of it. Pickling 
was resorted to for the removal of the remaining mill scale, the 



Digitized by 



Google 



426 Flinn on Protection of Steel Pipe with Cement. 




a 

OQ 

< 



< 

OQ 

H 

K 
o 
Z 

OS 

en 

g 



Q 



Digitized by 



Google 



Flinn on Protection op Steel Pipe with Cement. 427 

rust and the dirt. Two vertical wooden tanks of suitable diam- 
eter were placed side by side, one containing hot dilute sulphuric 
acid kept at a strength of about 5 per cent of oil of vitriol, which 
was approximately 93 per cent pure sulphuric acid; after 15 
minutes in this solution at a temperature of about 125 deg. F., each 
plate was of a uniform clear steel gray color all over and when 
removed was at once dipped into clean hot water in the adjacent 
tiank. Riveting followed pickling quickly, the pipes being made 
in 15-ft. sections of 2 rings of 7^-ft. net length each. Each ring 
Was one plate, except in the 11-ft. 3-in. pipes, and in those made 
of plates thicker than ^ in., for which two plates were used. It 
has been observed that where steel is exposed to corrosion with 
the mill scale on the surface, the corrosion tends to concentration 
at certain points and is accompanied with pitting; hence the 
care exercised to remove the mill scale thoroughly. 

Observation and such information as was available indicated 
that cement mortar adhered most strongly to steel and afforded 
the best protection when it could be applied directly to the clean 
surface of the metal. Having obtained a satisfactorily clean 
surface in the shop, there remained the problem of preserving this 
surface in as good condition as practicable until the pipes could 
be laid, covered with the concrete, and lined with the mortar. 
Various temporary coatings were suggested, but it was finally 
decided to use whitewash, and so each pipe was given a coat of 
heavy lime whitewash before it left the shop. To each barrel 
of whitewash, about 50 gal., there were added about 20 lb. of glue; 
the glue was dissolved in water before mixing in the whitewash. 
After this mixture had been used for a time, about 1 lb. of Port- 
land cement was added for each gallon of whitewash. This 
whitewash, applied with brushes, did not adhere very well, and 
through lack of care in handling, the pipes suffered more or less 
almost as soon as it was applied. They suffered more from expo- 
sure to the weather, however, than from abrasion; but even 
where the whitewash was not disturbed light rusting occurred. 
Only around the rivets and at joints where the whitewash had 
formed a very thick coating was there no sign of rusting. Hence, 
as delivered at the trench, the pipes had more or less complete 
coats of light yellow rust very uniformly distributed on the bared 
portions of the steel, without indications of any tendency to 



Digitized by 



Google 



428 Flinn on Protection op Steel Pipe with Cement. 

pitting. This light rust has been regarded as not seriously 
objectionable. 

Probably the greatest obstacle to securing the desired pro- 
tection of the steel between the time it leaves the shop and the 
time it is covered with the mortar or concrete is the tendency of 
the steel mill and fabricating shop to push their operations mu^h 
more rapidly than the pipe can be laid, tested, and covered^ and 
the uncertainties of the field work, which make it desirable to 
have pipe on hand ahead of the progress of laying. These cir- 
cumstances resulted in some cases in the gradual formation of 
heavier rust; this has been removed, as required by the specifica- 
tions, as is also the whitewash which still adheres to the pipes, 
just before the applications of the jacket or lining. 

Ease of removal and probable lack of objectionable effects 
upon the mortar, if small quantities should not be removed, 
were among the reasons for adopting whitewash. Unquestion- 
ably, Portland cement grout would have stuck more tenaciously, 
but it would have been correspondingly difficult to remove, and 
it was thought probable that the mortar of the lining and jacket 
would not adhere well to the old cement surface; these were 
considered sufficient arguments for forbidding its use. For 
removing the rust and dirt from the pipes, wire brushes are com- 
monly used, and in some of the worst places steel scrapers also. 
Inside some of the siphons the surfaces have been rubbed with 
empty cement bags after the wire brushing. This final cleaning 
is done in short stretches just in advance of the placing of the 
concrete or mortar. 

To support the pipe in the trench so as to permit the placing 
of the concrete jacket beneath it, and also aid in bringing the 
pipe to line and grade as it was being laid, concrete blocks called 
cradles were built in the bottom of the trench. Figs. 2, 3 and 4 
show the several styles. At first attempts were made to have 
these cradles fit the bottom of the pipe closely, but this gave 
trouble and the shape was modified. Some cradles were made 
about 24 in. square and extended about 6 in. below the ordinary 
sub-grade of the trench, with their tops at the proper grade for 
the bottom of the pipe, and, on the whole, this shape was satis- 
factory. Some of the longer cradles were cracked, due probably 
to unequal bearing or shocks in placing the pipe. Most of these 



Digitized by 



Google 



Flinn on Protection op Steel Pipe with Cement. 429 



cracks were due primarily to the uneven bearing of the pipe on 
the cradles. 

The pipe having been laid, riveted and calked was filled 
with water to hydraulic gradient, inspected, and leaks further 
calked. Bulkheads were placed in the open ends of the pipe and 
a small riser pipe carried up to the proper elevation to represent 




IS' 

4. 










-W' 



_j 



FIG. 2. — CONCRETE CRADLES FOR PIPE SIPHONS. 



the working head when the aqueduct would be in service. To 
maintain this pressure on the pipe constantly, a small reservoir 
or tank was attached to the top of the riser pipe and kept just 
spilling over. While the pipe was still full of water under this 
normal working pressure, the concrete jacket was placed about 
it by methods similar to those used in building concrete conduit, 
excepting, of course, that no inside form was needed, Figs. 5 and 



Digitized by 



Google 



430 Flinn on Protection of Steel Pipe with Cement. 

6. The water pressure was continued until this concrete had 
attained considerable strength, the period depending upon the 
weather, kind of cement, and other conditions. It was found 
undesirable to maintain the water pressure by direct pumping 
into the pipe while the concrete was being placed and was harden- 
ing, since fluctuations of head caused a few cracks. Maintaining 
this pressure by the small overflow tank and riser pipe mentioned 
above was more satisfactory. When the last concrete was suffi- 




(//? roc/r on/yj /or /79/€fyo//ff 



FIG. 3. — TYPICAL ARRANGEMENT OF CONCRETE CRADLES AND JOINT HOLES FOR 

FIELD RIVETING. 

ciently hardened, the water was slowly withdrawn from the 
pipe. 

Three variations in the procedure of placing* the concrete 
jacket were tried: (l) Monolithic, as specified; (2) first the invert, 
then the remainder, and (3) first invert and side walls to the hori- 
zontal diameter, and then the arch. The best contact between 
the jacket and the pipe seemed to have been obtained when the 
concrete was placed monolithically. It was found unwise to 
permit the concrete to be dumped from the buckets with a greater 



Digitized by 



Google 



Flinn on Protection of Steel Pipe with Cement. 431 

drop than 2 ft., because the vibration tended to crack the concrete 
recently placed. 

On one siphon the average rate of placing concrete jacket 
was about 130 ft. a week; on another, 170 ft.; and on a third, 




PIG. 4. — LAYING STEEL PIPE ON CONCRETE CRADLES. 

190 ft., the maxima on these three siphons being, respectively, 
223 ft. per week, 297 ft. and 260 ft. On one siphon the progress 
of 30 ft. of full section per 8-hour shift, using about 50 yd. of 
concrete mixed by machine was attained. Many factors influ- 



Digitized by 



Google 



432 Flinn on Protection of Steel Pipe with Cement. 

enced progress, such as the method of mixing, method of trans- 
portation whether by cableway, wheelbarrows or cars, and the 
style of outside forms, whether steel or wood, and the conve- 
niences for moving and setting the forms. Earth covering was 
placed in many cases immediately after the completion of the 
concrete jacket, but, in a few instances, considerable time elapsed 
. before the concrete was covered. So far as observed, there have 
been few cracks in the concrete jackets built to the end of last 
season. • 




FIG. 5. — METHOD OF PLACING CONCRETE JACKET AROUND STEEL PIPES. 

Two quite different methods were at first tried by the two 
contractors for lining pipes. Under Contract 62, the contractor 
began at once by grouting with forms; under Contract 68, an 
attempt was made to use the cement gun, and the Hunter's Brook 
siphon, 1493 ft. long, 9 ft. 9 in. diameter of pipe shell, was lined 
in this way. The cement gun* was fully described in Engineer- 
ing Recordf July 1, 1911, and other descriptions have been pub- 
lished, so that it need not be described again in this paper. Several 

* See Proceeding*, Vol. VII. p. 504.— Eo. 



Digitized by 



Google 



Flinn on Protection op Steel Pipe with Cement. 433 

machines are shown m Fig. 7. Charges of dry sand and cement 
mixed in prescribed proportions were placed in the charging 
chamber of the machine and then dropped into the pressure 
chamber. From the latter the mixture was rapidly discharged 
under an air pressure of 50 to 60 lb. per sq. in. through a rubber 
hose, while through a parallel hose joining the former in a special 
nozzle, water under pressure was discharged so that the sprays 
of water and of sand and cement were commingled. A pressure 
of about 30 lb. was maintained at the nozzle and so the mixture 




FIG. 6. — ONE OF THE CONCRETE MIXERS. 

was thrown with considerable force against the surface being 
coated. This aided in securing an excellent union between the 
mortar and the steel and more especially between the successive 
portions or layers of the lining. The operator held the nozzle 
2 or 3 ft. away from the surface being coated, moving it back 
and forth continuously, meanwhile controlling the discharge from 
the nozzle by lever valves. 

The force with which the mixture of sand, cement and water 
is thrown against the surface being covered causes a measurable 



Digitized by 



Google 



434 Flinn on Protection of Steel Pipe with Cement. 

proportion of the sand grains to rebound from the surface and 
fall into the bottom of the pipe, particularly when begiiming the 
first layer on the bare steel. By analysis this dry material was 
found to contain about 1 part of cement to 3f parts sand; it was 
collected and used for making the invert or bottom part of the 
pipe lining, which was deposited as a mortar and screeded to 
shape in advance of the gun work. Because of this separation 
of sand, an excess of it is put into the dry mixture in order to 
prevent the lining being richer than intended. Another result 




FIG. 7. — FOUR CEMENT GUNS ON HUNTER's BROOK SIPHON. 

is that the atmosphere inside the pipe is commonly very dusty 
and in order to make reasonable working conditions for the men 
artificial ventilation was resorted to. Furthermore, the mortar 
thus applied contained less water than the grout, and more atten- 
tion must be given to keeping the lining moist in order to minimize 
shrinkage cracks. Fine cracks have formed rather numerously 
and are probably due, in considerable measure, to the failure 
to keep the lining wet enough while it was setting and hardening. 
In spite of the difficulties, a lining which up to date seems satis- 
factory was secured. When this siphon was completed, the use 



Digitized by 



Google 



Flinn on Protection of Steel Pipe with Cement. 435 

of the cement gun was discontinued, the contractor giving as a 
reason excessive cost. Lining by the grouting method was then 
adopted under Contract 68. 

It has been common practice to place the lining in the invert 
of the pipe by screeding, the width thus placed varjring from 
about 2 to 6 ft. measured along the arc. Practical difficulties 
prevented the use of a simple complete cylindrical form; among 
th^e difficulties were going around curves, either vertical or 
horizontal, collapsing one set of forms sufficiently to pass it for- 
ward through another set which must remain in place while the 
grout was hardening, and cleaning, lubricating and inspecting the 
outside of the form inside the pipe. Consequently, forms in 
panels about 2 ft. wide and 15 ft. long (7^ ft. at curves) were 
adopted. These are supported on wooden ribs or centers adjusted 
to give the proper thickness of lining and firmly braced, Fig. 8. 
The lining has been commonly poured in 15-ft. sections in one 
operation. 

The grout is poured from outside the pipe through a 2i-in. 
wrought-iron pipe secured into a rivet-passing hole. In the 
Southern Department this pouring pipe is at the downhill end 
of the section and is long enough to give a head of about 4 ft. 
on top of the uphill end. A vent pipe is fastened in the uphill 
rivet-passing hole of the section, the bulkhead forming the end of 
the lining being placed just below it. For poiu'ing grout the con- 
tractor for the northern siphons has always used two mortar 
boxes set on a temporary staging over the upper end of the sec- 
tion to be grouted. The mortar is mixed to the proper consistency 
in alternate boxes and is allowed to flow into the pipe through a 
hole controlled by a sUding wooden gate. All mixing is done by 
hand, and materials are carefully graded. It generally takes 
about two hours to fill a section, after which a man is kept on for 
an hour or two in order to feed in sufficient grout to get the desired 
consistency. There is a noticeable tendency to get a porous or 
thin condition at the upper end of the section, near the grouting 
hole. To avoid this, the riser is removed two or three times dur- 
ing the pouring and the thin material which collects at the top is 
allowed to escape. The riser is then put back and grout added 
until desired results are obtained. In some cases it takes nearly 
two hours after the main operation to get grout of proper con- 



Digitized by 



Google 



436 Flinn on Protection of Steel Pipe with Cebient. 

sistency at the foot of the riser. At the finish of the pouring a 
small pipe is inserted through the large pipe, to permit the escape 
of the last air while grout is poured through the larger pipe and 
churned into the small remaining space, to insure complete fiUing. 
For the first batches, or nearly up to the horizontal diameter of 




FIG. 8. — WOODEN FORMS IN PANELS FOR PLACING CONCRETE LINING BY THE 
GROUTING METHOD. 

the pipe, the grout is mixed 1 part cement to 1 part sand, and the 
remainder about 1 to 2. 

Some fine cracking has occurred in the lining placed by the 
grouting method as well as in that deposited by the cement gun. 
None of this cracking is believed to be serious. As an aid to 
preventing cracks and a safeguard against the remote possibility 
of small pieces of the lining becoming loose and falling out, both 
contracts provide for using wire fabric in the mortar, as a rein- 



Digitized by 



Google 



FiiiNN ON Protection of Steel Pipe with Ceiosnt. 437 

foroement, but up to the end of last year none had been used. 
Permanent absolute adhesion of the concrete or mortar to the 
steel IS not being attained at all parts; this has been proven by 
careful sounding with a hammer, but on cutting into hollow sound- 
ing spaces, the space between the mortar or concrete and the 
metal has been found almost infinitesimal in width. Before 
preparing the contracts, such an occurrence was anticipated and 
some tests were made at the Board's laboratory to help in deter- 
mining how serious a matter this would be. A brief statement 
of these experiments follows: 

One experiment was made as follows: Four circular slabs 
15J in. diameter and 3 in. thick were made of concrete in pro- 
portions 1 cement: 2.7 Jerome Park screenings: 6.3 Jerome Park 
stone (gneiss) by weight, the consistency of the mix being rather 
dry. Carefully imbedded in the center of the slabs were four 
i-in. round, softnsteel rods spaced 3 in.* apart, so that there was 
at least IJ in. of concrete in all directions from each rod. When 
the concrete had set, two of these slabs were immersed in water 
in two tanks 1 ft. 6 in. in diameter and 2 ft. deep. The other two 
slabs had two galvanized-iron cylinders 15^ in. in diameter 
cemented to them, and a head of 20 in. of water maintained on 
them. The water in the tanks and cylinders was kept at a constant 
depth and head by the addition of water whenever it was neces- 
sary. The slabs subjected to percolation leaked rapidly at first, 
but became gradually tighter, and during the last few months 
of the tests there was very little leakage. The tests were con- 
ducted in open air, and were subject to the variations in outdoor 
temperature. The tests commenced July 13, 1907. On March 
23, 1909, one year and eight months later, the slabs were broken 
and rods examined for corrosion. It was found that the pro- 
tection of the rods by the concrete was perfect in all cases, there 
being no sign of corrosion on either the black finish of the metal 
left by the rolls or on the bright ends of the rods exposed in cutting 
to length with the hacksaw. The concrete when broken was 
found to be thoroughly saturated, showing that the water had 
full access to the rods. 

Another experiment was conducted as follows: Six steel 
plates 8 in. by 16 in., of 12 gauge, cleaned by pickling and then by 
rubbing with emery cloth, were placed horizontally in a galvanized 



Digitized by 



Google 



438 Flinn on Protection op Steel Pipe with Cement. 

iron tank 18 in. in diameter and 24 in. deep, separated from the 
bottom by two IJ-in. bars of alberene stone, and from each other 
by i-in. wood dowels. The first pair of plates was put in with- 
out any protective covering. The second pair had their upper 
surfaces protected by a slab of cement mortar 2i in. thick, not in 
contact with the steel, but separated from it by two metal strips 
about .04 in. thick. The third pair of plates was protected by 
cement mortar slabs 2 in. thick, cast directly on the steel, and 
apparently adhering to it firmly. The tank was then filled with 
proton water to a depth of 4 in. above the top of the uppermost 
mortar slab and kept filled the entire duration of the test, the 
water being renewed twice monthly. 

After two years' immersion, the plates were taken out and 
cleaned oflf by washing with a sponge. The first pair of plates 
showed heavy corrosion. In numerous places the entire layer of 
oxide had separated from the steel and formed blisters, leaving 
the bright steel surface underneath. The second pair of plates 
showed a very slight corrosion. Most of this washed off, thus 
indicating a considerable protective infiuence of the mortar slabs 
even when separated from the metal by a space of .04 in., open 
at the edges all around. When the mortar slabs were removed 
from the third pair of plates, it was found that a part of the sur- 
face of the steel had a distinctly different appearance from the 
other part. One part was clean and wet; the other part was 
covered by strongly adhering particles of mortar and was dry, 
thus indicating that there had been actual adhesion of the mortar 
only over the latter part of the surface and that the former had 
been separated by a space large enough for the water to enter. 
There had, however, been no rusting except at some places near 
the edges of the wet part of the surface, where apparently the 
space had been big enough to allow circulation of the water. 
The water in the rest of the space had evidently been so highly 
charged with lime that no corrosion could take place. 



Digitized by 



Google 



Flinn on Protection of Steel Pipe with Cement. 439 

Extracts from Contract 62. 
general sections. 
Order of Work, 
Sec. 19. The steel pipe shall be laid, tested, and made tight against 
hydrostatic pressure; then surrounded by concrete while still under the normal 
hydrostatic pressiu-e, after which the mortar lining shall be placed. Except 
at overhead stream crossings, no stretch of pipe shaU be left not well protected 
from frost between stretches of pipe around which concrete has been placed. 
As soon as practicable after the concrete covering and mortar lining have been 
placed, the pipe shall be covered with earth; and all pipe covered with con- 
crete, whether or not mortar lining had been placed in it, shall be protected 
in freezing weather by at least one foot of earth. 

MORTAR UNINO FOR STEEL PIPES. 

(Item 26.) 
Work Included. 
Sec. 26.1. Under Item 26 the contractor shall build a mortar lining 
inside the steel pipe and cast-iron bell castings as specified, directed or 
approved. 

Description and Proportions, 
Sec. 26.2. The lining shall consist of Portland cement and sand, mixeJ 
in ordered proportions, probably one part of cement to two parts of sand 
The quality of the sand shall be as specified under Items 28 to 30. Reinforce- 
ment, if used, will be paid for under Item 27. The lining shall be of sub- 
stantially uniform thickness throughout the entire circumference except for 
the unavoidable variations due to lap of the plates, butt straps, and rivet 
heads. The thickness over the inner course of plates is to be 2 in., that is, 
the internal diameter of the lining shall be 4 in. less than the nominal diameter 
of the steel shell. 

Forms. 
Sec. 26.3. Forms shall be of steel, or of wood covered with galvanized 
sheet steel, and shall be especially constructed so as to have sufficient strength 
and yet be adjustable so as to give a uniform space between them and the shell 
of the pipe. Great care shall be exercised to secure forms which will leave the 
surface of the lining perfectly smooth. Forms which give unsatisfactory 
results after use shall be satisfactorily repaired or replaced. The length of 
the sections of forms will not be restricted provided satisfactory means are 
adopted for controlling the uniform thickness of the lining and the correct 
spacing of the reinforcement, but sufficient sections shall be provided adapted 
for lining the pipe on curves, where lining shaU be placed in sections about 7 ft. 
long. Each time the forms are used they shaU be thoroughly cleaned and then 
coated with some approved inadhesive substance which wiU prevent the mortar 
from sticking to the forms without injuring the mortar. The lining of manhole 



Digitized by 



Google 



440 Flinn on Protection op Steel Pipe with Cement. 

castings, and of the blow-off elbows to the sockets, shall be monolithic with 
the lining of the steel shell, and forms shall be so constructed as to permit this 
method. 

Method of Placing. 
Sec. 26.4. The lining shall in general be placed by pouring a grout 
around an internal form, through holes cut in the top of the pipe for that pur- 
pose in the manner and at locations specified in Section 19.11. Sections shall 
be so terminated as to bring a hole at the upper end which shall be arranged as 
an air vent. The mortar shall be mixed to a thick creamy consistency and 
allowed to flow into place as uniformly as possible. When the section is filled 
to the top, the pouring of the grout shall be continued until grout runs from 
the air vent; then headers of steel pipe shall be screwed into the inlet and out- 
let holes and filled with grout so as to put a head of at least 4 ft. on the highest 
part of the section, and these headers shall be kept filled with grout until 
the grout has set, when the pipe shall be removed and the hole made water- 
tight by a screw plug. During the pouring of the grout, the form shall be 
tapped to loosen air bubbles, and a careful watch shall be maintained to pre- 
vent leaks. The work shall be so planned that the grout can be poured con- 
tinuously from start to finish of the section. Any interruption greater than 
fifteen minutes, whether due to leaks or any other cause, may be sufficient 
reason for the rejection of the entire section. 

Lining with Plaster. 
Sec. 26.5. Should there be any portion of the interior of the pipe 
which it is impracticable to line by grouting, this portion shall be lined by 
plastering the pipe with mortar, mixed as specified in Section 26.2. Only 
skilled masons or plasterers shall be allowed to do this plastering, and a section 
once started shall be prosecuted imtil finished, with only such pauses as are 
necessary for a sufficient setting of a layer to permit the next layer to be placed. 
Each layer shall be as thick as is feasible to apply, so that as few layers as 
possible may be necessary. The surface of each layer, except the final one, 
shall be brushed to thoroughly remove the laitance and then deeply scratched 
or otherwise satisfactorily treated to give a bond with the succeeding layer. 

Removal of Forma. 

Sec. 26.6. The forms shall be removed within twelve houTB of the time 
set by the engineer, and the section, if accepted, shall receive immediately 
such repairs as required, in the manner directed. It is possible that the lower 
part of the Lining will, in many cases, show a sandy surface, and if so, it shall 
be brushed with enough neat cement wash to fiU the pores and no more, and 
troweled to a smooth finish. Any section not accepted shall be immediately 
removed by the contractor at his own expense and replaced by acceptable 
lining. 

Prevention of Freezing; Bvlkheada. 

Sec. 26.7. Suitable bulkheads shall be erected in the pipe to prevent 
freezing inside the pipe cither during the placing of the Lining or after its 



Digitized by 



Google 



Flinn on Protection of Steel Pipe with Cement. 441 

completion. They shall be removed before the completion of the contract, 
if ordered. Lining shall not be placed in the uncovered pipes over streams 
during, or within a month before, freezing weather, unless the pipe is satis- 
factorily protected, and, after lining these portions of the pipe, water shall not 
be allowed to stand and freeze there. 

Measurement and Payment. 

Sec. 26.8. For placing the lining in the steel pipe and cast-iron bell 
castings, the contractor shall receive the price per linear foot of pipe stipulated, 
the measurement to be made along the axis of the pipe, this price to include 
all labor and materials necessary to complete the lining in a thorough and 
approved manner except only that the cement required will be paid for under 
Item 35 (Portland cement). 



REINFORCEMENT OF MORTAR LINING FOR STEEL PIPES. 

(Item 27.) 

Description. 

Sec. 27.1. Reinforcement may be ordered under Item 27 for any part 
or the whole of the mortar lining. The reinforcing material shall be galvanized 
steel mesh of a style and weight approved, provided, however, that no rein- 
forcement shall be required of which the lowest price obtainable by the con- 
tractor, f .o.b. New York City, exceeds f cent per square foot, for lots of 10,000 
sq. ft. 

Placing. 

Sec. 27.2. The reinforcement shall be placed approximately in the center 
of the mortar lining. The reinforcement may be kept away from the pipe 
by distorting the reinforcement at frequent intervals, so as to make points 
projecting toward the pipe. Unless otherwise permitted, small blocks of 
mortar shall be attached to the reinforcement, for the purpose of keeping it 
away from the form. Metal shall be lapped at least 6 in. at all longitudinal 
joints. 

Measurement and Payment. 

Sec. 27.3. The quantity to be paid for under Item 27 shall be the 
number of square feet of lining, meastu^ as of a mean diameter 9 ft. 4 in., 
in which reinforcement has been ordered and placed. This does not include 
any allowance for lap. The price stipulated shall include the cost of the 
reinforcing metal, royalty if any, cutting, shaping, bending, wires, clips, mor- 
tar, and other devices used for holding the reinforcement in place, or for splic- 
ing the strips; and it shall further include any additional expense of forms, 
tools, appliances, and labor other than the expense that would be required 
for finishing the mortar lining under Item 26 without reinforcement. . 



Digitized by 



Google 



442 Flinn on Protection of Steel Pipe with Cement. 

The gain in smoothness of interior by covering the rivet 
heads and the plate laps has been computed to so increase the 
hydraulic capacity that three pipes equal four without lining. 
The total cost for the siphons with three lined and jacketed 
pipes is estimated at about the same as for four pipes constructed 
and coated in the more usual way. Obvious incidental advantages 
are secured by the more permanent construction. 

The Board of Water Supply of the City of New York is 



Table I. — Contract Prices for Steel Pipes, Concrete Jacket and 

Mortar Lining. 



Description. 



Contract 62. 

9 ft. 6 in. steel pipe, ^/u-in. plate, lap jointed 

ft. 6 in. steel pipe, V't-in. plate, lap jointed 

9 ft. 6 in. steel pipe, Vi-in. plat«, longitudinal 
jointed 

9 ft. 6 in. steel pipe, */ia-in. plate, longitudinal i 
jointed 

9 ft. 6 in. steel pipe. "Ae-in. plate, longitudinal 

jointed 

9 ft. 6 in. steel pipe. */4-in. plate, longitudinal seams butt- 
jointed 

Mortar lining for steel pipes 

Reinforcement of mortar lining for steel pipes 

Concrete masonry around steel pipes 

Portland cement 



butt- 
biitt- 
biitt- 



Contract 
Price. 



Contract Sfi. 

9 ft. 9 in. steel pipe, "'kt-ia. plate, lap jointed 

II ft. 3 in. steel pipe, ^/it-in. plate, lap jointed.. . . 
11 ft. 3 in. steel pipe, Vrin. plate, lap jointed. . . . 
11 ft. 3 in. steel pipe, ^/a-in. plate, longitudinal 

butt-jointed | 

11 ft. 3 in. steel pipe, */i6 in. plate, longitudinal seams 

butt-jointed ' 

Mortar lining for 9 ft. 9 in. steel pii)e ' 

Mortar lining for 11 ft. 3 in. steel pipe. ... 

Reinforcement of mortar lining for steel pipe. . . 

Concrete masonry around steel pipe 

Portland cement 



$31.00 Un. ft. 
36.00 " " 

40.00 '• " 

43.00 •' •• 

47.00 " " 

60.00 " " 
2.50 '• " 
.02 sq. ft. 
6.00 cu. yd. 
1.75 bbL 



t29.00 Un. ft. 
33.00 •• " 
88.00 •' " 

46.00 " •' 

50.00 " " 
3.00 •• '• 
3.50 " " 
.02 sq. ft. 
5.25 cu. yd. 
1.60 bbl. 



Average of 
AUBids. 



(5 bidder:) 

$35.50 Un. ft. 

41.20 " " 

44.80 " " 

48.80 " " 

66.80 " " 

60.60 " " 
4.90 " " 

.03 0Q. ft. 
6.86 cu. yd. 
1.74 bbL 

i8 bidden.) 

$33.26 Un. ft. 

37.37V« " " 

42.76 " •• 

48.76 " " 

63.62Vi " " 
3.56»/« " " 
4.04»/i *' " 
.02«/4 sq. ft. 
6.78 ou. yd. 
1.72t/s bbL 



constructing the Catskill Water. Works. Engineering operations 
are being directed by J. Waldo Smith, chief engineer; Robert 
Ridgway was department engineer, Northern Aqueduct depart- 
ment, until January 14, 1912; Ralph N. Wheeler was appointed 
department engineer of that department, February 1, 1912; 
Frank E. Winsor is department engineer. Southern Aqueduct 
department. In immediate charge of the construction of the 
siphons are division engineers John P. Hogan, Alexander Thom- 
son, Jr,, and George P. Wood (Northern); George G. Honness, 



Digitized by 



Google 



Flinn on Protection of Steel Pipe with Cement. 443 

Emest W. Clarke, and Charles E. Wells (Southern). The draw- 
ings and specifications were prepared and many of the preliminary 
investigations conducted by Senior Designing Engineer Thomas 
H. Wiggin, and Engineer Inspector Ernst F. Jonson has had 
charge of inspection of cement, of steel at the rolling mills and 
of pipe at the shops, all under the immediate supervision of the 
writer. The chief engineer and his personal assistant, Depart- 
ment Engineer Thaddeus Merriman, made many examinations 
of existing steel pipes which furnished the reasons for seeking a 
better protection than the usual coatings. 

All drawings reproduced as illustrations are for the 9-ft. 
6-in. pipes. Corresponding drawings for the other sizes are 
similar; likewise standard dimensions for rivet and joint details. 



Digitized by 



Google 



A FIREPROOF SCHOOL OF CONCRETE. 
By Theodore H. Skinner.* 

Early in the winter of 1910-11 the writer was intrusted by 
the Trustees of Union School District No. 12, Town of Vernon 
and City of Oneida, N. Y., with the task of building a new four- 
classroom school house which should, in addition to complying 
with all the regulations of the New York State Department of 
Education, be as nearly fireproof as possible, keep within an 
appropriation of $17,000, be so arranged as to appear symmetrical 
and complete, while in reality be only one-half of an eight-room 
building ultimately desired. The school house shown by the 
accompanying illustration is the answer to the many problems 
involved in carrying out the task. 

The general plans were gone over with the state inspectors 
for the purpose of securing informal approval of same before 
detailed drawings and specifications were completed. These 
plans were then submitted to several parties with requests for 
estimates and sketches showing how they would build the frame- 
work of floors, roof and enclosing walls. These parties repre- 
sented field-cast reinforced concrete, light steel frame with metal 
lumber and metal lath stuccoed and factory-made reinforced 
concrete. 

The sketches received were carefully studied and general 
drawings made which would be possible to follow should either 
system be selected. Tenders were then invited from a number of 
general contractors. The specifications provided that the bidders 
might use either one of the three systems proposed and asked 
them to name in their bid what would be the difference in cost of 
the building if erected by them under the various systems. The 
bids ranged from $16,298.00 to $24,500.00 for the bare building 
"^dthout plumbing or heating and nothing to spare for moving 
furniture, etc. The local contractor, who proved to be lowest 
bidder, then opened his estimate books for inspection and it was 
found that by combining his own figures for portions of the work 

♦ Architect, Oneida, N. Y. 

(444) 



Digitized by 



Google 



Skinner on Fireproof School of Concrete. 



445 



with the figures named by the party bidding on factory made 
reinforced concrete that it would be possible to carry out the gen- 
eral plans and keep within the appropriation. Accordingly 
detailed specifications and drawings were made up for the frame- 
work of the building and the general contract let. 

The building (Figs. 1 and 2) consists of a rectangular section 
25 X 92 ft. running north and south, containing four classrooms 
each 24 x 32 ft., four coatrooms 7 x 24 ft., stairway 11 ft. wide, 
two playrooms in the basement 24 x 32 ft., separate toilet rooms 




FIG. 1. — KENWOOD SCHOOL, KENWOOD, N. Y. 

each 7 X 24 ft. for boys and girls and a teachers' room 11 x 13 ft. 
with private toilet on the landing over the first- and second-story 
stairway. An extension across the front 13 x 40 ft. contains fur- 
nace room in the basement, and is entirely corridor on the two 
upper floors. There are two minor further extensions forming 
vestibules at front and rear. 

The plans (Fig. 3) provide for two additional classrooms at 
both the north and south ends of the present rectangular section 
to be reached by extensions of the present corridors across the 
west or blank sides of the present classrooms. Additional stair- 



Digitized by 



Google 



446 Skinner on Fireproof School of Concrete. 




{DiAACM^UX ^UHj- 




-Lb li IJ- - r^ Z 

CCCDDDDDD 
inGUDnDDDD 



Tim* r^t^/a^TKm . 



FIG. 2. — GROUND AND FIR8T FLOOR PLANS, KENWOOD SCHOOL. 



Digitized by 



Google 



Skinner on Fireproof School of Concrete. 447 

ways will be provided in the extended corridors and also coat- 
rooms for the new classrooms. When these extensions have been 
made the west or street face of the building which now presents 
large blank panels will be symmetrical on the center line of the 
building and when finished will have large windows indicating 
the lines of the stairways. 

The present classrooms have southeastern exposure with 
unilateral lighting only. The coatrooms are accessible only from 
the classrooms. Each classroom is provided with a built-in 
cabinet of four drawers and two cupboards for the teachers' use 
and has blackboards 4 ft. high entirely around the walls except- 
ing for the window spaces. The basement rooms are 10 ft. in the 
clear, the classrooms 12 ft. The plumbing fixtures are all of 
colonial ware and of the latest pattern, two drinking fountains 
are arranged in the corridors, one in each story. Special anti- 
panic exit bolts are arranged on the outside doors, which open 
out. 

The details provided for a steel frame for the classroom 
section, of 12 columns connected by horizontal steel girders in the 
plane of the walls at several points. The girders carried 2 floors 
or 2 sets of separately molded and cast reinforced concrete joists, 
spaced about 4 ft. on centers which rested on and were anchored 
to them. The joists carried in turn a series of ribbed reinforced 
concrete slabs separately molded and cast in the factory (Fig. 4). 
The roof was constructed in same maimer as the floors, the only 
difference being that the joists were not of uniform section 
throughout being severally graded or warped so as to give the 
roof slabs resting on them the proper pitch to throw the water to 
desired points. The floors and roofs of corridors and entrance 
porches were built also of separately molded members supported 
by bearing walls of masonry. The type of imit members employed 
is shown in Fig. 5. 

A short description of the process of manufacturing separately 
molded reinforced concrete members may be of interest. A 
frame was made up of the steel rods necessary to reinforce the 
concrete, carefully designed to take care of all the tensile stresses 
which might be developed in the member when finished, set in 
place and loaded, and also all shear which the concrete would not 
take care of. Longitudinal tension rods were carefully bent to 



Digitized by 



Google 



448 Skinner on Fibepboof School of Concrete. 



-v{^ 



U 



Ol "^ 






trW 



Si 









^1 



^ 



§ 

O 



O 

QQ 

I 



o 

M 
OQ 






^ 



Digitized by 



Google 



Skinner on Fireproof School of Concrete. 



449 



the desired form and united into rigid frames by means of vertical 
loops and small rods womid around the larger ones which also 
take care of the shear. Some smaller longitudinal rods were 
built into the frames at their tops to take care of any excessive 
compression or any tension which might be caused by negative 
bending moments produced in handling the beams. Special loops 
were attached to the tension rods at their quarter-points extend- 
ing up above the top rods in the frames and above the finished 
concrete at the top by which the beams were lifted, the strain all 
coming on the reinforcement. 




PIO. 4. — REINFORCED CONCRETE SLABS IN YARD OP CONTRACTOR. 

Sand molds were prepared on large casting floors of the 
shape desired for the finished beams and in these molds the unit 
reinforcing frames were suspended in proper place, and then 
liquid concrete was poured into the molds and thoroughly worked 
in and around the reinforcement. The molds were filled to the 
top and struck off with the straight edge. A mixture of 1:2:4 
concrete was used, the largest aggregate passing a 1-in. screen. 

The beams were allowed to remain in the sand 7 days after 
which they were lifted by the loops, carried by a traveling crane 
into an open yard and stored until ready for shipment. 



Digitized by 



Google 



450 



Skinner on Fireproof School of Concrete. 



The manufacture of the floor panels would not have been 
possible without a vibrating machine. A frame of small rods to 
act as reinforcing for the webs was first made, this was then 
covered with wire reinforcement well wired to the rods. The 
frame was then placed in molds on vibrating machines; the con- 
crete poured and vibrated. 

The beams were about 10 in. wide, 18 to 24 in. deep, of Tee 
section with the top edge rebated on each side to receive the 
floor slabs. All joists were approximately 25 ft. long. Floor 
slabs were about 4 ft. square webbed or thickened around the 
edges and once across the middle to 3 in. thick, the centers were 
thin panels only IJ in. thick. Joists and slabs were designed to 




FIG. 5. — GENERAL TYPE OF UNIT CONCRETE MEMBERS. 

carry a live load of 100 lb. per sq. ft. in addition to their own 
weight, with a factor of safety of 4. 

The footings for the 12 columns were isolated, 7 ft. square 
and 16 in. thick. The basement walls between the columns were 
made of 12-in. hollow terra cotta tile carried by independent 
footings. The tiles were laid up in cement mortar, plastered with 
cement mortar on the outside and coated with asphalt up to 
grade. The exterior walls between columns above the basement, 
being curtain walls only, were laid up of 8-in. hollow terra cotta 
tile and were stuccoed on the outside and plastered on the inside. 
The columns and girders were covered with metal lath and con- 
creted in solidly. 



Digitized by 



Google 



Skinner on Fireproof School of Concrete. 451 

The cornices were supported by a cast stone bed molding, 
ornamented with egg dart topped out with a 12-in. tile covered 
with stucco. Some small brick inserts were made just below the 
cornice to relieve the absolute monotony of the color of the stucco 
and shallow lines were drawn in the stucco facia and elsewhere 
while it was still soft to form panels and accents. With these 
exceptions the exterior is a plain gray floated-finish cement 
stucco. It will be given a brush coat of some waterproof cement 
stain or finish before another winter to fill up minor checks and 
prevent water and frost damage. 

All exterior windows were combined frame and sash of metal, 
glazed with clear double-thick glass in the lower panels and with 
rippled glass in the upper. Each sash was provided with 2 ven- 
tilating sections. 

The roof is a flat concrete slab pitching only enough for 
drainage from the level verge to two outlets near the center line. 
This was covered with 5-ply slag and composition roofing and is 
not visible from the ground. 

The interior finish was the simplest possible. The masonry 
walls were plastered with two coats of mortar and left under the 
float, then painted 4 ft. high with oil paint two coats, and above 
with water color two coats. No wood trim was used around 
either doors or windows, but plank jambs of wood were used in 
the doorways to which the doors were hung by pivot hinges at 
top and bottom. No wood door sills were used, in short the only 
wood in the entire building being the door jambs and doors, the 
molding at the top and bottom of blackboards, cleats for coat 
hooks in the wardrobes and panels of wood flooring in the center 
of each classroom. 

These panels, 4 in number, were laid of one thickness maple 
floor, I X 2J in. face over 2 x 4 in. hemlock sleepers, on top of the 
concrete construction previously described, were oiled on the 
under side before being laid and given a coat of oil immediately 
after laying to prevent their absorbing water from the composition 
borders, subsequently laid against them. The balance of the 
classroom floors were finished on top of the cement slabs with 
concrete surfaced with colored material to match the wood center. 
These borders merged into a sanitary or cove base 4 in. high 
everywhere. Corridors, stairways and wardrobes all have a 



Digitized by 



Google 



452 



Skinner on Fireproof School of Concrete. 



granolithic cement coat over the structural concrete slabs, this 
finish also merging into a sanitary base 4 in. high. There was no 
projection to these sanitary bases at the top, the same being 
finished flush with the finish plaster. 

The four classroom ceilings were plastered upon expanded 
metal lath carried by angle-iron furrings which in turn were 




FIO. 6. — CEILING OF PLAYROOM, KENWOOD SCHOOL. 

attached to numerous iron lugs left projecting from the cast con- 
crete floor joists for this purpose. These furred ceilings were 
used for a double purpose, first to make a ceiling free from all 
shadows, and, second, to provide a deadening air space so that the 
noise of walking on the top of the concrete construction should 
not be heard in the room below. The basement, corridor and 



Digitized by 



Google 



Skinner on Fireproof School of Concrete. 453 

stairway ceilings were left unfurred, with the cast joists and panel 
slab construction showing, as may be seen in Fig. 6, all painted 
with water color paint presenting a very attractive appearance. 

The contract, let May 15, 1911, provided for the completion 
of all work by September 15, but delays were experienced from 
the start, first in getting the necessary steel fabricated and later 
in securing competent labor for erecting the steel and laying the 
heavy concrete floor and roof joists. It was not until November 
26, or a few days over six months from the start, that the work 
was completed. 

The joists and slabs were loaded on cars at the factory and 
shipped by local freight 28 miles; were unloaded by means of a 
small traveling crane on to farm wagons and then hauled a little 
better than a mUe to the job. Here they were unloaded and 
swxmg into place by seemingly inadequate apparatus, all at less 
expense than the contractor figured he could erect forms and cast 
the same number of members in situ. The joists and slabs stood 
shipping remarkably Well and reached the job in good condition 
in spite of the rough handling given by the local train, on which 
the cars are shunted back and forth at every station a great num- 
ber of times. Only one joist and four slabs required replacing as 
they were too green when loaded, 30 days* seasoning should be 
allowed before shipping. 

The straightforwardness of the various operations at the 
building appealed both to the general contractor and to the 
architect as it offered opportunity to inspect each member before 
erection and the job was not littered up nor complicated with the 
forest of supports necessary for field concrete construction. No 
unexpected difficulties were encountered and no members failed 
to fit their respective locations accurately, and everyone con- 
nected with the work would be satisfied to repeat the operation 
again, with the exception possibly of the maker of the cast work, 
who might wish to add slightly to the original allowance for 
handling the cast members at the building. 

The estimates of cost for the various types of construction 
not used were perhaps confidential and will not be given in detail ; 
it is sufficient to say that they were higher than those obtained 
for the construction adopted. 



Digitized by 



Google 



454 Skinner on Fireproof School of Concrete. 

The contracts for the building as erected were as follows: 

Steel frame $1,500.00 

Structural concrete floors, roof and stairs 3,000 . 00 

Suspended ceilings, unplastered 225 . 00 

Granolithic in halls and coatrooms 196 . 00 

Colored surfacing and sanitary base in classroom 379.00 

Iron stair rails 200.00 

General contract for balance of work and materials, including ex- 
cavation, grading, carpentry, painting, etc 9,545.00 

$15,045.00 

Plumbing 767.00 

Heating 825.00 

Hardware 225.00 

Total $16,862.00 

In addition to the above a flowing spring was encountered in 
the excavation which gave standing water over part of the base- 
ment and made it desirable to underdrain same and waterproof 
basement walls; it also necessitated a change in the plans of the 
cold-air boxes from beneath the basement floor to overhead. 
These two items cost respectively $200.00 and $227.97. Some 
additional fireproofing of steel to that provided under the con- 
tract was thought desirable and added at a further expense of 
$400.00. Cement platforms or walks outside at the entrance 
cost $38.00, making a total of extras on the work over and above 
the first contracts of $863.97 and bringing the total cost of the 
building as delivered by the contractor to the Trustees, $17,527.97. 

Not coimting the area of vestibules, this building has an 
area of 2820 sq. ft. on the ground, is 40 ft. from bottom of base- 
ment floor to top of roof and contains 112,800 cu. ft. Its cost 
per cu. ft. is 15.5 cts., per classroom $4381.99 and per pupil as 
$97.37. 



Digitized by 



Google 



THE PRESENT STATUS OF UNIT CONCRETE 
CONSTRUCTION. 

By James L. Darnell.* 

To put it briefly it might be said that the present status of 
Unit Construction is one of progression. Every day new things 
are developing about this very logical and rational method of 
constructing and erecting concrete structiures. Within the past 
two years there has been more progress in both the design and 
in the field work, that is, the construction and erection of sep- 
arately molded structures, than in all the time previous. 

In heavier structures, such as bridges and viaducts, the 
engineering forces of the Chicago, Burlington and Quincy Railway 
Company under the direction of Mr. C. H. Cartlidge, Bridge 
Engineer, and of Mr. George E. Tebbetts, now bridge engineer 
for the Kansas City Terminal Railway Company, have done 
much valuable work and they have pioneered the way for others 
less advancced. The Chicago, Milwaukee and St. Paul Railwaj'^ 
Company under Mr. C. F. Loweth, Chief Engineer, has also 
done some work in this line, but not to the same extent. Mr. 
R. E. Gaut, while bridge engineer of the Illinois Central Railway 
Company, also used this system in the design of highway struc- 
tures in connection with track elevation work in Chicago. All 
of this work was successfully done and showed pronounced 
advantages over ordinary methods. 

If this is the "Concrete Age" as most of us fondly believe, 
it will certainly come to pass that the railroads particularly will 
have to adopt unit construction methods for their concrete struc- 
tures because it lends itself with peculiar fitness to railway work 
of all kinds and shows such marked economies in both time and 
money, that ordinary or monolithic construction is out of the 
question. This is shown in the present track work of the Kansas 
City Terminal Railway Company, where Mr. Tebbetts has 
adopted this form of construction for every structure along the 

* Manager. Kansas City Unit Construction Company, Kansas City. Mo. 

(455) 



Digitized by 



Google 



456 Darnell on Unit Concrete Construction. 

line except those where surrounding conditions are such that the 
use of some other type was imperative, as the McGee Street 
Viaduct, where the 124-ft. span required a concrete encased 
steel girder. 

In building construction the progress has been more marked 
perhaps than in bridges and viaducts. The first recorded example 
of a building of any magnitude built of separately molded units 
is that of the two kiln houses for the Edison Portland Cement 
Company at New Village, N. J., early in 1907.* These two 
buildings were simple one story sheds of the plainest possible 
type. No attempt was made to elaborate or to refine the design, 
as nothing more was necessary than a simple assemblance of 
columns, roof beams and roof slabs. In these two sheds however, 
the soundness of the principle of separately molded units was 
demonstrated. 

At about this same time the idea of building construction 
with separately molded units seems to have been taken up inde- 
pendently in other sections of the country. In the east Mr. 
E. L. Ransome was developing a system which he is employing 
successfully up to the present time. Perhaps the most notable 
example of the Ransome type of Unit Construction is the four 
story building 60 ft. x 200 ft. built for the United Shoe Machin- 
ery Company at Beverly, Mass. This building has successfully 
met all its requirements and is in every respect equal to and in 
some respects superior to similar buildings of ordinary mono- 
lithic construction. 

Mr. Charles D. Watson of Syracuse, N. Y., has also done 
considerable work along this line, having successfully constructed 
several separately molded buildings, f 

In St. Louis, at about this same time, Mr. Albert J. Meier 
in conjunction with Mr. John E. Conzelman, at that time engi- 
neer for the Corrugated Bar Company, and Mr. C. D. Morely 
a contractor, were working to perfect a Unit system independently 
and without knowledge of the work that was being done by others. 
From the beginning there were two very apparent difficulties to 
be met and surmounted. The first one was the difficulty in devis- 
ing or designing connections between the separately molded units 



♦ Seo Proceedings, Vol. IV. 1908. p. 48— Ed. 

t See ProceediTHjB, Vol. IV, 1908. p. 97; Vol. VI. 1909, p. 391. 



Digitized by 



Google 



Darnell on Unit Concrete Construction. 



457 



which would be as strong as the units themselves. Second, there 
were well taken objections on the part of many people to sep- 
arately molded buildings of concrete on account of the pure and 
monotonous ugliness of such buildings. 

The first difficulty, that of the connections, has been most 
successfully overcome by the use of interlocking and overlapping 
reinforcement, broad bearings and improved methods of grouting 
the connections. The second or aesthetic objection is being 
gradually overcome both by improved and advanced design and 
by a process of evolution, in which the engineer in designing is 




FIG. 1. INTERIOR OF WAREHOUSE, NATIONAL LEAD COMPANY, KANSAS CITY, MO. 

slowly but surely drawing away from the architectural faults of 
its monolithic predecessor. The progress is best shown in the 
illustrations which follow. 

The first building constructed with separately molded 
units by the Unit Construction Company of St. Louis, Mo., 
was a warehouse for the National Lead Company, erected on 
West Thirteenth Street, in Kansas City, Fig. 1. This build- 
ing was built under great difficulties in a very restricted 
space. It was attempted to mold the units within the building 
lines, which proved to be very expensive by reason of the neces- 



Digitized by 



Google 



458 



Darnell on Unit Concrete Construction. 



sary rehandling of the unite from time to time. Because of the 
unsightly appearance which would have been presented by the 
use of concrete unite in the front of the building, the National 
Lead Company at that time wanted a brick face. The rear of 
the building, however, was built of concrete unite and ite appear- 
ance in a measure justifies the objections of the owners on the 
score of unsightliness. The different pointe between the unite 
are very cleariy indicated and while the wall on the whole does 
not present a very beautiful appearance, it is after the lapse of 
some four years, perfectly serviceable, thoroughly substantial 
and water-tight. We are certainly proud of this, our first build- 




FIG. 2. CASTING YARD FOR CONCRETE UNITS. 

ing, because it is still standing up, apparently more substantial 
than when built and because it has been so thoroughly satis- 
factory to the owners. 

This first building was so successful that the company was 
enabled, in competition with various other contractors, to secure 
a contract for a very much more extensive construction for the 
National Lead Company in St. Louis. Fig. 2 is a general view 
of the casting yard laid out for this construction. Fortunately 
there was in this case plenty of room to spread out in, conse- 
quently it was not necessary to cast the unite on the ground 
within the building lines. An extensive plot of ground was avail- 



Digitized by 



Google 



Darnell on Unit Concrete Construction. 459 




FIG. 3. — SETTING A WALL SLAB. 




FIG. 4. — ERECTION OF CORRODING STACK HOUSE, ST. LOUIS, MO. 



Digitized by 



Google 



460 



Darnell on Unit Concrete Construction. 



able adjacent to the site of the building and thereon was erected 
the construction plant. Concrete materials were elevated in a 
belt and bucket conveyer from cars to bins in the top of the 
tower, seen in the left center of the picture. Immediately below 
these bins was the concrete mixer, which discharged by gravity 
into the cars on the oval elevated track. Movable spouts con- 
ducted the concrete from this track level to the molds for the 
units which were laid horizontally in the casting yard. A travel- 
ing derrick was erected to run between the elevated tracks and 
was used to handle the completed units from the yard onto flat 
cars for conveyance to the building site. 




FIG. 5. — ROOF SHOWING CONCRETE SKYLIGHTS AND LOUVRES. 

Fig. 3 is a view of the building after construction had some- 
what further progressed. This shows many of the columns in 
place and the workmen setting a partition slab. It may be seen 
that the foundations were run in monolithically with slots left 
for the reception of the columns. A key was cast in each slot 
to engage a corresponding slot cast in the columns, both of which 
may be plainly seen in the picture. Fig. 4 shows a view of the 
structtire when about two-thirds complete. This view illustrates 
another advantage of unit construction in that as soon as any 
part of the building is in place it is ready for service. This was 
peculiarly demonstrated in this particular building. To make 
room for this structure it was necessary to dismantle and tear 



Digitized by 



Google 



Darnell on Unit Concrete Construction. 



461 



down the old wooden stack building and necessarily the machine 
capacity was seriously impaired. It proved to be of great 
advantage that after four or five bays were finished, they were 
turned over for service. As it happened the compan}*- was 
making white lead in more than three-fourths of the building by 
the time it was finally completed. Fig. 5 is a view of the roof 
of this same building, showing the sky lights and louvres, and 
illustrates the roof construction which was of the regulation felt, 
tar and gravel type. This building when complete was 750 ft. 
long, 105 ft. wide and about 45 ft. high. 

Fig. 6 is a group of three buildings built for the National 
Lead Company at their St. Louis plant, contracts for which were 




FIG. 6. — lU 1LDINCJ8 OF NATIONAL LEAD COMPANY, ST. LOUIS, MO. 

secured after the stack house had progressed pretty well along 
toward completion. This group consists of an office and welfare 
building on the right, an oil house in the center, and a stable 
and garage on the left of the picture. In this group is seen 
something of the architectural progress which was referred to 
earlier in this paper. They certainly present a not unpleasant 
appearance and we think they appear favorably with some build- 
ings of similar type in any material. 

Figs. 7 and 8 represent the most pretentious structure which 
has engaged the attention of the Unit Construction Company 
up to the present time. This is an oxide mill also for the National 
Lead Company, built in St. Louis, which provided for floor loads 
running from 500 lb. on the first floor to 250 lb. on the top floor. 



Digitized by 



Google 



462 Darnell on Unit Concrete Construction. 





"^ 







FIG. 7. — OXIDE MILL IN COURSE OF ERECTION, FIVE STORIES. 



fc ,- _J^ 







FIG. 8. — OXIDE MILL BUILDING COMPLETED. 



Digitized by 



Google 



Darnell on Unit Concrete Construction. 



463 



A study of these views will give to the observer an excellent 
idea of the construction methods employed as nearly all of the 
details may be observed. 

Fig. 9 shows a corroding stack building erected for the 
National Lead Company at New Kensington, Pa. They are 
interesting in that they show architectural progress as well as 
improvement, both in design and workmanship. 

Last year the first contract was taken for a 50,000 bushel 




FIG. 9. — CORRODING STACK BUILDING, NATIONAL LEAD COMPANY, 
NEW KENSINGTON, PA. 

grain elevator, Fig. 10, for the Highland Milling Company at 
Highland, 111. In this case each slab unit had a column cast on 
the end of it and the whole was held together by big rods running 
entirely through the building with large cast washers on the 
outside. This elevator had nine bins in it; eight were storage 
bins and the ninth carried the elevating machinery. This struc- 
ture has been in service for more than a year and the owners of 
the Highland Milling Company are very enthusiastic advocates 
of Unit concrete construction because during a season in which 



Digitized by 



Google 



464 



Darnell on Unit Concrete Construction. 



it was very hard owing to weather conditions to successfully 
store grain, in this elevator not a bushel was lost. 

A building under construction for the Ohio Cultivator Com- 
pany at Bellevue, Ohio is shown in Fig. 11. This building is 
200 ft. square, three stories high with a high basement. 




FIG. 10. — GRAIN ELEVATOR, HIGHLAND MILLING COMPANY, HIGHLAND, ILL. 

As to the use of Unit construction in railway work, Fig. 12 
shows a reinforced concrete crossing at Sangoman Street in 
Chicago, 111. The work of the Kansas City Terminal Railway 
Company where concrete units are being used for all the via- 
ducts and subways required in the Terminal Company's improve- 



Digitized by 



Google 



Darnell on Unit Concrete Construction. 



465 



ments, which are now going on, includes some forty structures 
in all. Fig. 13 shows a unit bridge; Fig. 14 column molds as 
well as some of the finished columns; Fig. 15 shows an outer 
deck slab which carries the coping and panel fascia. 

The pictures show for themselves the present status of Unit 
Construction both as applied to buildings and to bridges and 
kindred structures. 

A word as to the '*why" of unit construction. At first 




FIG. 11. — BUILDING FOR OHIO CULTIVATOR COMPANY, BELLEVUE, OHIO. 

glance, to an engineer, the system looks novel and interesting, 
but of narrow application and of little practical value or utility. 
Invariably, however, after thorough study the engineer becomes 
more and more convinced of the practicability of the system and 
of its broad and diverse possibilities. In fact those who are con- 
nected with its application cannot find the time to go into the 
many virgin fields of construction effort, but up to the present 
have been compelled to devote all their time and energies to 
the development of the single field of mill and warehouse con- 



Digitized by 



Google 



466 



Darnell on Unit Concrete Construction. 



struction with short excursions into the grain elevator and rail- 
way fields. 

For unit construction are claimed all the advantages of 




FIG. 12. — CONCRETE BRIDGE CROSSING, SANGOMAN STREET, CHICAGO, ILL. 




FIG. 13. — RAILROAD BRIDGE OF UNIT CONCRETE COLUMNS, BEAMS AND SLABS. 

permanency, strength and ultimate economy to which structures 
of reinforced concrete are justly entitled, without any of the 
disadvantages and uncertainties which in the structure built by 



Digitized by 



Google 



Darnell on Unit Concrete Construction. 



467 



ordinary methods are so often attended by unfortunate results — 
accidents which are always serious, sometimes fatal. 

By this method the designer is sure that no expansion joints 




FIG. 14. — UNIT CONCRETE COLUMNS AND MOLDS. 




FIO. 15. — UNIT CONCRETE DECK SLAB WITH COPING. 

will open up in unexpected places because of improper deposition 
of concrete or other carelessness of the workmen. Each unit 
member of the structure is built on the ground under practically 



Digitized by 



Google 



468 Darnell on Unit Concrete Construction. 

factory conditions. Eflfectual inspection is assured in all respects. 
The reinforcement is placed where the engineer designed it to 
be; each unit is a true monolith without seam or flaw; each 
unit is properly cured and is prepared in the building; the pos- 
sibility of getting an imperfect column, beam, girder or slab into 
a structure is entirely eliminated. In short, the manufacturing 
conditions of the units make for a maximum of efficiency in all 
departments with the chance for uncertainties reduced to prac- 
tically a negligible quantity. In erection the same condition 
prevails as the mechanical handling of the units is exactly the 
same as in the handling of steel and with properly designed units 
it is practically impossible to go wrong on the connections. 

Unit Construction today is just beginning to come into its 
own. Its application is wide and the genius of the American 
engineer will soon lead to the general adoption of a system which 
combines in itself the maximum of efficiency, economy and 
permanency and minimizes the uncertainties contingent on the 
vagaries of the American workingman — ^who is principally a 
foreigner. 



Digitized by 



Google 



DISCUSSION. 



Mr. Rudolph P. Miller. — Mr. Chairman, I would like to Mr.Muier. 
inquire if the joints must not be filled up on the work and whether 
any attempt is made to carry the reinforcement from one member 
to an adjoining member so as to effect continuity? 

Mr. James L. Darnell. — ^The joints and connections are Mr. Dameu. 
all grouted in. The reinforcement overlaps and interlocks, i.e. 
goes through the connection which results in practically a mono- 
lith. In fact, extensive laboratory tests show that the connec* 
tions are stronger than the units themselves. It has been demon- 
strated in buildings actually constructed and in service that the 
connections are the strongest part of the structiu'e. 

Mr. John E. Conzelman. — ^The question just asked would Mr. conzciman. 
indicate that there was some doubt in the speaker's mind as 
to the stability or rigidity of the buildings and I want to say 
that the buildings we have constructed seem to be just as rigid 
and stable as similar buildings constructed in the ordinary 
manner. In fact after the joints are poured a imit constructed 
building has the advantages of continuous action and rigidity 
characteristic of buildings made by the ordinary or monolithic 
method. It may be interesting to know that the thin walls used 
on these buildings have proven satisfactory. We have con- 
structed seven or eight buildings with three and four inch walls 
which have passed through two winters with no complaints from 
the owners. 

Mr. E. J. Moore. — The question of shrinkage cracks in Mr. Moore, 
floors seems to be important. The speaker referred to the absence 
of cracks in the floors. Cracks from temperature stresses will 
no doubt occur at the joints where the different sections are put 
together and that would seem to make an added problem of 
waterproofing the joints. 

Mr. Conzelman. — Mr. Darnell stated that cracks often Mr. conzeiman. 
occurred in concrete structures built in the usual way and that 
these cracks did not at all times select their location with due 
regard for the appearance of the structure or the feeling of the 
builder; in fact these cracks often appear at imfortunate places. 

(469) 



Digitized by 



Google 



470 Discussion on Unit Concrete Construction. 

Mr. ConzeinuiA. It is unusual to 866 a long concr6t6 retaining wall or building in 
which there are not some cracks. These cracks (neglecting settle- 
ment or other structural causes) are due to two causes, shrinkage 
of the concrete as it dries out and temperature changes, acting 
separately or in combination. 

A great many of the cracks are due to shrinkage stresses 
but in unit construction these stresses are practically elminated. 
This is due to the fact that each vmit has hardened before it is 
incorporated into the building. The unit itself may have some 
slight internal stresses due to shrinkage, but these are not cumula- 
tive and do not affect the complete structure. Each unit takes 
care of itself. 

Temperature stresses cannot be eliminated, although tem- 
perature changes undoubtedly have less effect on unit construc- 
tion than on monolothic, on the same principle that a brick wall 
(which is constructed of small units) will generally show fewer 
cracks than a concrete wall. We have observed this action care- 
fully and have had very little difficulty, if any, from temperature 
effects and account for it by the fact that during the construction 
of the building and before the connections are grouted, the imits 
attain an average temperature. 

Mr. Moore. Mr. Moore. — I take it therefore that it is necessary to 

waterproof most of this filling. Joints in factory floors are more 
especially referred to as these are washed down and there is 
objection to having the water go through the cracks. 

Mr. conxeinum. Mr. Conzelman. — ^Thc construction shown is absolutely 

waterproof so far as we know. A unit weighing 4 or 5 tons 
when set on. a mortar bed will compress it to such an extent 
that it is practically watertight. One advantage of unit con- 
struction is that we know just where the joints are and as leak- 
age is necessarily confined to the joints proper provision can be 
made. 

Mr. Kinney. Mr. William M. Kinney. — It would secm there would 

certainly be some little unevenness, which is one of the greatest 
troubles of the concrete floor and one of the things we are trying 
to eUminate in floors that have to be trucked over. The trouble 
occurs where the joints are made in a floor and the question 
arises whether in unit construction any {protection is afforded 
over the joints. 



Digitized by 



Google 



Discussion on Unit Concrete Construction. 471 

Mr. Darnell. — ^There is no apparent joint at the edge of Mr. Dameu. 
the floor slabs. The floor slab and beam are cast in one and 
the edges of the slabs are made so that the space between is in 
the form of a wedge, which is filled up level with mortar so that 
practically there is no joint left at all and the floor is perfectly 
smooth. That is perhaps one of the advantages of this form of 
construction. The edge of the floor slab is built to exact dimen- 
sions and if any vmevenness develops it is very easy in setting 
the slab on the girder to level it up smoothly, which is done. 
If the levelling is carefully done and these wedge shaped joints 
filled up with grout, it makes the floor, to all intents and pur- 
poses so far as the surface is concerned, perfectly smooth. There 
has been no trouble with uneven joints in trucking over floors 
in any of the buildings. The National Lead Company's buildings 
in St. Louis, is a group in which trucking is constantly going on 
and there has never been a complaint at any time. 

Mr. Allen Brett. — ^The tests reported by the Committee Mr. Brett, 
on Reinforced Concrete and Building Laws, in the case of the 
beam and slab show that the concrete in the slab acts in com- 
pression practically all the way across. This girder used in unit 
construction is a broad inverted T, practically, and the slabs are 
inverted boxes, the edges of which rest on the flange of the T. 
Is there any compression acting with the girder in the slab? 

Mr. Conzelman. — ^The girders are made with ledges or Mr. conzeinum. 
shelves on each side upon which rest the floor slabs, usually made 
to resemble the cover of a large box. They consist essentially 
of a thin plate or slab carried by beams on each side, these beams 
returning arovmd the ends of the slab; the arrangement gives the 
slabs a continuous or uniform bearing on the girder ledge. The 
thickness of the slab or plate may vary from IJ to 5 or 6 in. 
depending on the load to be carried and the span between beams. 
When the slabs are placed the top of the slab is higher than the 
top of the girder, an amovmt equal to the thickness of the slab, 
and the steel projecting from the slabs on each side of the girder 
interlock in the space so formed; the stirrups from the girder 
also extend into this space. The joints between the girder and 
slabs are then grouted and the space filled with a rich concrete. 
After the grout has hardened these become for all practical pur- 
poses T beam sections. 



Digitized by 



Google 



472 DiscuBsioN ON Unit Concrete Construction. 

Mr. cm&zeiiiuui. This T-beam action has been demonstrated in the laboratory 

on large beams by measuring the deformation in the concrete 
directly over the girder and in the adjoining floor slabs and also 
the deformation of the reinfprcement, by means of an extenso- 
meter similar to that which has been described by the Com- 
mittee on Reinforced Concrete. 

Mr. H. p. Green. Mr. HERBERT P. Green. — What provision is made to take 

up the shear between the columns other than the brackets on the 
columns? 

Mr. Conzeinum. Mr. Conzelman. — The brackets are designed to take the 

entire load from the girder; the brackets are designed for ver- 
tical shear and bending moment. The size of the top of the 
bracket is determined by the bearing area required to properly 
distribute the load from the girder to the bracket. 



Digitized by 



Google 



REPORT OF COMMITTEE ON SPECIFICATIONS AND 
METHODS OF TESTS FOR CONCRETE MATERIALS. 

In this first report of the Committee no definite recommenda- 
tions are presented for specific tests or methods of making tests 
of aggregate, except that the Committee reconmiends for a practi- 
cal test of sand a determination of the strength of mortar made 
with it. The report, therefore, is in the nature of a tentative 
discussion of various tests and of methods that have been used 
in diflFerent laboratories. This progress report should be followed 
during the coming year by further investigations leading to more 
definite reconmiendations. 

The Committee request that information on methods of mak- 
ing tests of aggregates and results obtained by such methods be 
forwarded to the chairman of the committee. Laboratories that 
are in a position to assist the Committee or imdertake research 
work in the line of concrete aggregates are also asked to corre- 
spond with the Committee. 

Importance of Testing the Aggregates. 

The selection of aggregates for concrete, especially the selec- 
tion of sand or fine aggregate, is of as great importance as the 
selection of the cement. So evident is this to the engineer who has 
had experience both in practical construction and in laboratory 
tests, that it is almost inconceivable that so much important 
work should be undertaken and carried through without testing 
the sand. Frequently every carload of cement is carefully sampled, 
but no test whatever, except by inspection, is made of the equally 
important ingredient, sand. When tests are made, they frequently 
are confined to mechanical analysis or granulometric composition. 
While this is a valuable test for comparing the qualities of different 
fine aggregates that other tests have shown to contain no impuri- 
ties, it does not show up some of the worst defects that occur 
occasionally. It therefore cannot be relied upon alone. 

The impossibility of determining the true quality of a siand 
or other fine aggregates by mere inspection cannot be emphasized 
too strongly. 

(473) 



Digitized by 



Google 



474 Report on Tests for Concrete Materials. 

Sampling Aooreoates. 

Samples must be taken in such a way as to obtain a fair 
average of the material to be tested. 

The size of the sample depends upon the character of the 
aggregate and the nature of the test. One should err on the side 
of getting too large a sample rather than one that is too small. 
For tensile or compressive tests of mortar made with fine aggre- 
gate a sample not less than 20 lbs. in weight should be taken in 
order to have enough material left over for other laboratory tests 
that may be deemed necessary. If practical tests of proportions 
with coarse aggregates are to be made, the sample of fine aggregate 
should be several times larger than this. 

The coarse aggregate sample should be larger than that of 
the fine aggregate in order to get fair average of the material, 
because the grains are larger and there is more variation in them. 
Whatever tests are made must be on a larger scale. For tests 
involving both mechanical analysis and volumetric tests of con- 
crete mixtures for proportioning the aggregates, at least 200 lbs. 
of each coarse aggregate are needed. 

Samples should be shipped in a strong box or a bag. It is 
advisable also that the natural moisture be retained as far as po"^- 
sible, so that the laboratory will receive the material in its natural 
condition. 

For sub-dividing the sample to obtain the required amount 
for each test, different methods are employed in different labora- 
tories. One of the common methods is that of quartering. 

Quartering. — To quarter a sample of aggregate, it is spread 
out on a thoroughly clean floor or table, or else upon a large sheet 
of manilla paper. Care is used in spreading to see that particles 
of different size are distributed through the mass. The pile is 
preferably in the shape of a circular disc. The material in this 
shape is divided into four quarters. Two opposite quarters are 
removed, taking care to remove all dust. The remaining quarters 
are then mixed together. After mixing, the material is spread 
out again, as before, and quartered again. This process is followed 
until the quantity remaining is of the size required for the ex- 
periment. 



Digitized by 



Google 



Report on Tests for Concrete Materials. 475 

Tensile or Compressive Test of Mortar. 

The Committee recommends the test of the strength of mortar 
as that best indicating the quality of the fine aggregate. 

To eliminate variations in result, due to the character of the 
cement, the difference in laboratory conditions, and the .personal 
equation of the operator, the test always must be a comparative 
one. For comparison, standard Ottawa sand which is used for 
cement tests, and which every well-equipped laboratory should 
be provided with, is recommended. 

The Joint Committee on Concrete and Reinforced Concrete 
in their 1908 report, makes the following recommendation for this 
test of strength : 

Mortars composed of one part Portland cement and three parts fine aggre- 
gate by weight when made into briquets should show a tensile strength of 
at least seventy (70) per cent, of the strength of 1 : 3 mortar of the same con- 
sistency made with the same cement and standard Ottawa sand. 

While this requirement is far in advance of usual practice, 
where no laboratory tests are required, it is not so severe as 
should now be demanded in the present state of the art of rein- 
forced concrete construction. For the present, the Committee 
recommends that sand used in reinforced concrete be accepted 
only after tensile or compressive tests of 1 : 3 mortar made with 
the sand in question, in comparison with similar tests of mortar 
of standard sand made up at the same time under the same 
conditions. 

The Committee is not yet prepared to recommend a fixed 
value for the ratio of strength for acceptance. It is suggested 
for the present that the ratio be set to suit individual requirements. 

To avoid the removal of any coating on the grains which 
may affect the strength, bank sand should not be dried before 
being made into mortar, but should contain natural moisture. 
The percentage of moisture to use for correcting the weights in 
measuring the proportions may be determined upon a separate 
sample. From 10 to 40 per cent, more water may be required 
in mixing bank sands or artificial aggregates than for standard 
Ottawa sand to produce the same consistency. 

In the mortar tests, enough test pieces should be made to 
test at 72 hours, 7 days and 28 days, the first 24 hours all being 
stored in moist air, maintained at a temperature of 70 deg. Fahr., 



Digitized by 



Google 



476 Report on Tests for Concrete Materials. 

and the remainder of the period in water at the same tempera- 
ture. The 72-hour test is the mast severe, and sand failing 
to attain this requirement frequently reaches it at 7 days or at 
28 days, and can then be accepte^d. If, however, the 72-hour 
briquets break in the clips of the machine, or if the test pieces 
at this age show very low strength, say 25 per cent, or less, of 
the strength of standard sand mortar, the sand should be considered 
dangerous to use on any important work of construction. 

Mechanical Analysis. 

For proportioning fine and coarse aggregate, the tests of 
mechanical analysis are important. Curves of mixtures in dif- 
ferent proportions, based on the combined analyses of the cement, 
fine aggregate, and coarse aggregate, may be drawn and studied 
to obtain the proportions corresponding most nearly to the ideal 
requirements. The proportions thus found may then be used 
in tests of volume, as referred to below. 

Mechanical analysis of fine aggregate is valuable as furnish- 
ing an indication of its quality. It is recommended that fineness 
requirements be introduced into concrete specifications. This 
test cannot be relied upon fully, however, since there may be 
impurities in the sand that will make it unfit for use even when 
the analysis is satisfactory. The chemical and also the min- 
eralogical composition of the sand also may afifect its strength. 
Leaving out of consideration, however, the question of impurities, 
for which specific tests probably will soon be evolved, the test 
of mechanical analysis, or granulometric composition, as it is 
sometimes termed, is worthy of much further development. 
Laws which govern the effect of the sizes of the particles of the 
aggregate upon the resulting mortar or cement are not yet clearly 
formulated. 

Further studies are necessary for the selection of standard 
sieves for use in mechanical analysis. From the report of the 
Committee on Reinforced Concrete of this Association,* the fol- 
lowing paragraphs are quoted: 

19. The relative strength of mortars from different sands is largely 
affected by the size of the grains. A coarse sand gives a stronger mortar 
than a fine one, and generally a gradation of grains from fine to coarse is 

♦ See ProeeedingM, Vol. V, p. 457. — Ed. 



Digitized by 



Google 



Report on Tests for Concrete Materials, 



477 



advantageous. If a sand is so fine that more than 10 per cent, of the total 
dry weight passes a No. 100 sieve, that is, a sieve having 100 meshes to the 
linear inch, or if more than 35 per cent, of the total dry weight passes a sieve 
having 50 meshes per linear inch, it should be rejected or used with a large 
excess of cement. 

20. For the purpose of comparing the quality of different sands a test 
of the mechanical analysis or granulometric composition is recommended, 
although this should not be substituted for the strength test. The percentages 
of the total weight passing each sieve should be recorded. For this test the 
following sieves are recommended:* 

0.250 inch diameter holes, t 

No. 8 mesh, holes 0.0955 inch width No. 23 wire. 
No. 20 " " 0.0335 " " No. 28 " 

No. 50 " " 0.0110 " " No. 35 " 

No. 100 " " 0.0055 " " No. 40 " 

21. The effect of mechanical analysis or granulometric composition 
upon the strength of mortar is illustrated in Appendix. By this table (which 
follows) the relative strength of different sands may be approximately estimated. 



TssTs BY New York Board of Water Supply of 1:3 Mortar 
Made with Sands of Different Mechanical Analysis. 



Peroentaces Paadns Sioves. 


Tenmle Testa. 


CompreasioD Test. 


No. 4. 1 No. 8. 


No. 50. 


No. 100. 


7 days. 


00 days. 


7 days. 


90 days. 


100 ' 70 


12 


5 


213 


613 


2690 


5640 


100 


86 


21 


6 


263 


412 


1915 


4660 


100 


99 


26 


2 


177 


325 


905 


2170 


100 


97 


28 


6 


178 


282 


1070 


1500 


100 


94 


44 


12 


139 


228 


905 


1130 


100 


100 


52 


14 


122 


170 


275 


810 


100 


100 


94 


48 


80 


149 


330 


490 



Void Tests. 

Void tests of coarse aggregates frequently are used to de- 
termine proportions for concrete. They do not give entirely 
correct results, however, unless the tests are made with mixtures 



* Sheet brass periorated with round holes passes the material more quickly than square 
Round holes corresponding to sieves No. 8, 20 and 50 respectively are approximately 
0.13&, 0.060, 0.020 inch diameter. 

t A No. 4 sieve, having 4 meshes per linear inch, passes approximately the same sise 
gnUns as a sieve with 0.25 in. diameter holes. 



Digitized by 



Google 



478 Report on Tests for Concrete Materials. 

of all the ingredients. Usually some of the grains of the fine 
aggregate are so coarse as to force apart the grains of the coarse 
aggregate. If the sand is fine in proportion to the stone, less 
mortar will fill the voids of the stone than if the sand is coarser 
and therefore more nearly the size of the stone particles. 

Void tests of fine aggregate are also affected by the fact 
that the cement forces the grains of sand apart. The voids in 
fine aggregate also are affected to a large degree by the percentage 
of moisture contained in it. If perfectly dry, a fine aggregate 
with grains of uniform size may have nearly the same percentage 
of voids as a coarse aggregate of uniform size grains, although 
the former will produce a very weak mortar and the latter a 
strong one. If the voids were based on the volume of the moist 
aggregate, the results would be more normal, but a slight varia- 
tion in the percentage of moisture produces such a marked effect 
that it is impossible to make true comparisons in this way. 

A common method of determining the voids in an aggregate 
is to place it in a measure, either loose or compacted as desired, 
and measure the quantity of water which can be poured in. The 
percentage of voids, either by weight or volume, is thus found 
directly. This method with a clean coarse aggregate is fairly 
satisfactory. With a fine aggregate, however, air is entrained 
and it is almost impossible to obtain correct results. 

Another plan sometimes followed is to measure the material ; 
place a definite quantity of water in a graduated vessel ; pour the 
aggregate into the water; and determine from the graduated 
scale the difference in volume of the water before and after add- 
ing the aggregate. This difference is the amount of water which 
the aggregate displaces. This is substantially a specific gravity 
method. 

A still simpler plan, if the specific gravity of the aggregate 
is known or can be readily determined, is to weigh a given bulk 
of the aggregate, loose or compacted as required, correct for 
moisture, and compute the voids directly. 

Weight. 

Weight tests have the same limitation as void tests, since 
the weight is affected by the percentage of moisture contained 
in the aggregate. The weight also varies directly with the specific 



Digitized by 



Google 



Report on Tests for Concrete Materials. 479 

gravity. If the specific gravity of the material is known and the 
percentage of moisture is determined, the voids can be computed, 
as indicated above. 

Volumetric Tests op Mortar of Fine Aggregates. 

If there are no organic, or other similar impurities, that ab- 
normally reduce the strength, the aggregate producing, with the 
cement and mortar, the smallest volume of mortar or concrete 
is apt to give the greatest strength. 

A volumetric test is better than an ordinary void or a weight 
test because the aggregate is mixed with cement and water as 
in practice. 

The general method employed in making a test of volume 
is to mix up the aggregates in the given proportions by weight, 
add enough water to produce a consistency slightly softer than 
used in tensile tests, and determine the bulk of mortar or con- 
crete made with this mixture. Knowing the specific gravity of 
all the materials used, the absolute volumes and density can be 
computed. The method of making this test is described more 
fully in a paper on Concrete Aggregates presented in 1906.* 

Volumetric Tests of Concrete Ingredients for 
Determining Proportions. 

One of the most valuable fimctions for volumetric tests 
lies in determining the proportions of concrete. The value of 
the test is based on the principle that, with the same proportion 
of cement, the mixture which gives the smallest volume, and ia 
therefore the densest, usually produces the strongest concrete. 
This rule is not strictly true for permeability because the size 
of the voids as well as the density influence the permeability. 

For determining the density of the concrete, the specific 
gravity of each aggregate must be known. The specific gravity 
of cement may be assumed as 3.10. An average specific gravity 
for bank sand is 2.65. 

The process of making volumetric tests of mixtures of coarse 
and fine aggregates with cement is similar in principle to the 
volumetric test of mortar described. Larger volumes must be 



♦ See Proceedirtifa, Vol. II, p. 27.— Ed. 



Digitized by 



Google 



480 Report on Tests for Concrete Materials. 

used and it is sometimes easier to fill a measure of given size 
and determine the amount left over than to determine the bulk 
of the total material. A blank form for use in this test is referred 
to below. 

Microscopical Examination. 

The mineralogical composition of a fine aggregate may fre- 
quently be determined approximately by an examination under a 
microscope or magnifying glass of high power. Quartz can be 
recognized and if dirt surrounds and adheres to the grains it can 
be seen. 

Chemical Analysis. 

The value of chemical analysis is not clearly defined. A 
quartz sand in general is better than a natural sand of other 
composition, chiefly because it is cleaner. It is claimed by some 
that the amount of clay material in the aggregate affects the sand 
in other ways besides increasing the amount of fine material. 

The effect of colloids and colloidal action remains to be 
studied. 

The ignition test referred to below is really a chemical test. 

Test for Organic Matter. 

Experience with defective concrete indicates that the quality 
of a sand may be very poor through a minute quantity of organic 
or vegetable matter contained in it. The best method of testing 
for this and the limitations which must be placed upon the quan- 
tity are not yet clearly defined. 

A method of test was suggested several years ago by the 
Chairman of the Committee and has been used in practice by 
him and also, more recently with some modifications, by another 
member of the Committee, Mr. Chapman. The methods are 
substantially as follows: 

Two hundred grams of the damp sand just as it is received 
at the laboratory are weighed out and put into a jar or a graduate. 
If desired, the quantity of this may be measured and a given 
bulk, such as 100 c. c, may be used instead of the fixed weight. 
If thus measured by volume the weight is also determined. Water 
is added and the mixture is shaken violently and stirred for 2 
minutes. Then the dirty water is poured off into a separate 



Digitized by 



Google 



Report on Tests for Concrete Materials. 481 

vessel. More water is added and the operation is repeated until 
the water is practically clear. The water poured off is evapo- 
rated, taking care not to raise the temperature much above the 
boiling point, or else by another method, it is poured through 
a filter paper (previously weighed) and this residue, together 
with the paper, dried at a temperature of 212 deg. Fahr. The 
filter paper and residue are then weighed together, the weight of 
the filter paper deducted, and the remainder is considered as 
silt. The percentage is recorded which the weight of this silt, 
or, by the other method, the weight of the silt left from the 
evaporated water, bears to the weight of the original sand. 

The evaporated residue, or else the filter paper with its 
residue, are ignited in a crucible at a red heat, and the loss of 
weight by this process (after allowing for the weight of the filter 
paper) is taken to indicate the amount of organic or vegetable 
matter. The percentage of this is expressed both in terms of 
the silt and of the total sand. 

Blank Forms for Reports. 

Forms are appended to this report that have been used in 
laboratories of members of your Committee for special tests of 
aggregates, see Figs. 1-4. 

Respectfully submitted, 

Sanford E. Thompson, Chairman, 
Cloyd M. Chapman, 
William B. Fuller, 
Russell S. Greenman, 
Arthur N. Talbot. 



Digitized by 



Google 



482 



Repobt on Tests fob Concrete Materiaia. 



SANO TEST REPORT 



grJ^»Ji.M 









rijEAi^wjf 




FIG. 1. — BAND TEST REPORT USED BY WBSTINGHOUSE, CHURCH, KERR 

AND COMPANY. CLOYD M. CHAPMAN, ENGINEER IN CHARGE, 

NEW YORK, N. Y. 



Digitized by 



Google 



Report on Tests por Concrete Materials. 



483 



Contract No. 



TESTS OF SAND fpom bank at ^ , N. Y. 

Proposed for use on contract No , Ree. No Canal, Division..... 

Contract sample No taken , received at Laboratory , made up 

Sand is maitUy quartz and feldspar with some hornbUnd and magnetite. 
Percentage of Voids, 36^; Loam. S.2; Organic matter, trace. 
Parts of sand to cement, by bulk: — S sand to 1 cement. 
Per cent water used -\-13. 

Cement used in tests . 

For test of cement see Vol. , Page 
Temperature, Fah. when mixed, Air , Water 

Briquettes kept in air 24 hours and then inmiersed. 



TENSILE STRENGTH (in pounda per square inch) 



Natural Sako 



I 



Washed Sakd 



SIZE OF SAND 
Paisxno Suyb 



Briquette' _ . 
N©. 7 days 



178 
186 
192 
192 
170 



Total I 918 
Average, 184 



28 daysl 

252 I' 

274 1 

2S0 \\ 
268 

260 ,| 



Briauette 



1294 
269 



Total 
A verage 



7 days l28 days, 
182 ! 286 



196 
190 
198 
196 



962 
192 



276 
264 
262 
274 



I No. 

I 

' 6(1/ 
]10 

130 



Per H 
cent. 

100.0 \ 
lOO.Tl 
99.2 \ 
97.6 \ 
87.0 \\ 
68.0 \\ 



No. 

40 

60 

74 

100 

140 

200 



I Per 

I cent. 

I SS.2 

I 16.0 



7.8 
3.2 
2.0 
14 



, Reported 19 . 

I Examined and Approved 

1362 1 1 Resident Engineer. 

270 11 Accepted . Rejected. 



Tests for strength made by 
Tests of sand made by 



. Recorded by 



FiQ. 2. — Report Card of Sand Test New York State Testtno Labor- 
ATORT, Albany, N. Y., Russell S. Greenman, Resident Engineer 
IN Charge of Tests. 



Digitized by 



Google 



484 Report on Tests for Concrete Materials. 



SANFORD E. THOMPSON. FUb, 8WBS. 

CoNBUiA'iNa Enoinbbr, Date, Feb. 18, 1911. 

Newton Highlands, Mass. Volumetrie or Density Teet. Experimentor, W. O. L 



( 1) Test Number ? V-e39 

( 2) Date S/W/ll 

By volume 1:2:4 

( 3) Nominal mix By weight 1:1.8:S.8 

( 4) Brand of cement 

( 5) Weight of cement 2.000 

( 6) Weight of aggregate parsing a No. 100 sieve 0.0007 

(7) '^ " " coarstt' than a No. 100 sieve 11.180 

( 8) " " vessel and water (before using) 1 .600 

(9) " " " '' " (after using) 0.740 

(10) " •* water used 0.760 

(11) Total weight mixed (5) + (6) + (7) +(10) 1S.947 

(12) Weight of trav+tools (aft«r mixing) 2.760 

<13) " " '' + " (before mixing) 2.626 

(14) Weight of mixed adhering 0.126 

(16) Weight of waste+water 0.126 

(16) Weight of waste 0.094 

(17) Weight of free water O.OSl 

(18) Net mix of set = (11) -(14) -(17) 18.791 

(19) Water left on tray -^5J!^-(|y^^ior ^ ^^ 

(20) Net water set = (10) -(17) -(19) 0.696 

(5)X(14) 

(21) Net cement (5)" (5)^, (6)4- (10) ^ -^^^ 

(22) Net aggregate passing a No. 100 Sieve - (6) — (K\A.{ k\ XTim 0.0O67 

(23) Depth of concrete in cylinder 0.44^ 

(24) Volume of concrete in cylinder 0.0869 

(25) Net water per cu. ft. as mixed (10) h- (24) 8.760 

(26) " " " " set (20) -^ (24) 8.020 

(27) Net cement per cu. ft. as set (21) ^ (24) 22.000 

(28) Net aggregate per cu. ft. as set (22) -f [(7) -f- (24)] 128.60 

(29) Abs. vol. of water per cu. ft. as set (26) -^62.3 0.1286 

(30) Abs. vol. of cement (27) ^62.3 0.11S9 

(31) Abs. vol. of aggregate (28) -^^^.5 0.760 

(32) Abs. vol. total* (29) + (30) + (31) / .0024 

(33) Weight of form +concrete 19.000 

(34) " " " 6.166 

(35) " " concrete 13.844 

(36) Temperature of water 70'*F-. 

(37) Time of mixing 10,00 a.m. 

(38) Remarks on consistency JeUy like 



Pig. 3. — Mechanical Analysis Report, Sanford E. Thompson, Con- 
sulting Engineer, Newton Highlands, Mass. 



Digitized by 



Google 



Report on Tests for Concrete Materials. 



485 



8ANFORD E. THOMPSON, FUe, 8WBE. 

CoNBULTiNa Enoxnesb, Date, Jan. £4$ 1911. 

Newton Highlanda, Mam. Mechanical Analysts. Experimenter, W. E. S. 

Test made for SWBE, 

Description Gravel from Neponsel River. 

Samples Taken (date) Jan. 23 ^ 1911, Shipped to laboratory in bag. 

Samples Received (date) Jan. 24, 1911, 

Weight Total Sample 20 lbs. 

Client's Mark 

Laboratory Mark A, S., 186 SWBE. 

Analysis No A. S. 186. 

Per cent. Moisture 1.6% 



Sise of Sieve 


Total 


Totol 


Per 
cent. 


Totol 


Totol 


Per 
cent. 


Total 


Total 


Inches 


No. 


Weight 
PasSng 


Per cent. 
Passing 


Finer 
than H" 










2.50 


H 
1 
J 

J 

6 
12 
20 
40 
50 

100 

200 










1 






2.00 
















1.60 


998.0 

677.2 

669.0 

388.0 

366.0 

160.6 

64.2 

17.0 

11.8 

6.8 

3.6 


100.0% 
67.8 
57.1 
38.9 
36.6 
16.1 

6.4 
1.7 
1.2 
0.6 
0.36 












1.00 












0.50 






.... j ... . 




0.25 


100.0% 
94.3 
38.8 
16.6 

4.4 

3.1 
1.6 
0.9 




1 






0.16 




1 






0.0583 










0.0335 




. . . 






0.0148 








.... 


0.0110 










0.0055 




::::i:::: 






0.0030 














1 







Remarks 



Washing No. W226 
Tensile No. T926 
Report made Fe&. 1, 1911. 



Approved, 
Noted by S.E.T., 



Date Initials 

Jan. 24, 1911. W. O. L. 
Jan. 28, 1911. S. E. T. 



Fig. 4. — ^Volumbtrig or Density Report. 



Digitized by 



Google 



AGGREGATES FOR CONCRETE. 
By William M. Kinney.* 

Within the past few years has been seen the almost universal 
adoption of the Standard Specifications for Portland Cement f of 
the American Society for Testing Materials. Practically all 
cement made in this country is guaranteed to meet these require- 
ments and it is only occasionally that a shipment fails to pass the 
specifications by a safe margin. Despite the precautions taken in 
the making and testing of cement, however, there are occasionally 
complete or partial failures of concrete work. Fortimately these 
failures usually occur during the process of construction so that 
the loss of life and property is relatively small, but that there 
should be any loss whatever in this type of construction should 
lead such bodies as this Association to strive for greater efficiency 
by a careful and thorough study of the materials entering into 
and the workmanship required for concrete. 

It is seldom possible to determine positively the cause for 
such failures; in fact, the number of reasons given for a particular 
failure usually varies with the number of engineers employed on 
' the investigation. It is essential, therefore, that in the construc- 
tion of any concrete or reinforced concrete structure, that each 
step be made with absolute surety in order to attain success. 

Deficient strength in concrete is usually due to one or more 
of the following causes: poor workmanship, unsatisfactory 
aggregate and unfavorable weather conditions, assuming of course 
that the design is right and that the cement had been carefully 
tested to the standard specifications. Poor workmanship can be 
eliminated from this discussion, because with the widespread 
distribution of literature on the mixing and placing of concrete, 
it should be possible for any contractor or user to handle concrete 
in such a manner that success will be assured. However, the 
question of aggregate and temperature conditions, with usually 
a combination of the two, is a matter deserving of very careful 
consideration. Of these two, the question of aggregates is the 

* Assistant Inspecting Engineer, Universal Portland Cement Company. Pittsburgh, Pa. 
t Standard No. 1. National Assooiation of Cement Users. — Ed. 

(486) 



Digitized by 



Google 



Kinney on Aggregates for Concrete. 487 

paramount issue, as it is almost a certainty that if good, clean 
aggregates were used, at least three-fourths of the frozen concrete 
would be eliminated. The reason for making this statement is 
based upon the fact that when concrete is subjected to actual 
freezing weather during the early stages of hardening, it is usually 
well protected and poor work seldom results; but when it is 
deposited at a temperature above freezing it is not protected, and 
owing to delay in hardening due to unsatisfactory aggregates, 
freezing may occur the following night or even a number of 
days later. It is certain that in a great many cases had clean 
aggregates been used the concrete would have hardened suffi- 
ciently during the favorable weather to have withstood the sub- 
sequent freezing without resultant injury to the concrete. 

The aggregates, therefore, play an important, if not the most 
important part in concrete work, and are without doubt subject 
to the greatest variation, yet have received up until the present 
time the least study of all the adjuncts of good concrete. To be 
sure, there are the studies of Feret, Candlot and other European 
experts. Likewise, the Structural Material Division of the 
United States Geological Survey (now under the Bureau of 
Standards) and individual investigators, such as Thompson, 
Spackman and Greenman have given valuable information 
obtained from their study of sand, gravel and other aggregates. 
This work, however, has of necessity been limited somewhat to 
the study of deposits local to the various laboratories, and the 
results are chiefly valuable in giving an idea of the most advan- 
tageous tests for detecting the properties of a particular aggregate, 
which render it good or poor for concrete. It remains to apply 
these tests in the study of the aggregates being used or proposed 
for use in individual structures or in particular localities. 

The report of the Committee on Specifications and Methods 
of Tests for Concrete Materials has been presented. It would 
seem highly proper for this committee at the earliest possible 
date to confer the committees on Concrete and Reinforced Con- 
crete of the American Society for Testing Materials, American 
Society of Civil Engineers, and Association of American Portland 
Cement Manufacturers, looking toward the establishment of a 
standard set of tests for concrete materials. Having outlined such 
methods, it should then be within the power of this committee to 



Digitized by 



Google 



488 Kinney on Aggregates for Concrete. 

influence the various Government, State and commercial bodies 
interested in this subject to inaugurate at once an extensive study 
of these materials by the methods outlined. 

The magnitude of such an investigation may lead to the 
thought that it is not feasible, but there is no reason why, if 
properly organized under an advisory board composed of rep- 
resentatives of the various large engineering societies, the work 
cannot be systematically done without duplication of tests by the 
Bureau of Standards and the various State Universities and 
Experiment Stations together with whatever assistance may be 
rendered by other laboratories of a public or private nature. 

In 1905 there was established at St. Louis a laboratory under 
the Structural Materials Division of the United States Geological 
Survey, which published in 1908 Bulletin 331 on Portland Cement 
Mortars and Their Constituent Materials, Neither this laboratory 
nor its successor in this line of work, the Bureau of Standards, 
has given any further data on this very important subject. It 
is to be hoped that the Bureau of Standards will continue this 
very necessary work, as it would be unfortunate to lose the prac- 
tical information gained by the investigators in conducting the 
j&rst series of tests. Good work has also been done in several of 
the State Universities and Experiment Stations. It is pleasing 
to note that the University of Wisconsin is to start at once under 
Mr. M. 0. Withey, a comprehensive study of the aggregates of 
Wisconsin. If Wisconsin can do this, why cannot other States 
be interested in a similar investigation? 

Such an investigation will not by any means solve the prob- 
lem and insure good work in the future, but it will give an idea 
of the relative merits of the various aggregates available for con- 
crete in any particular locality. This in itself will be a most 
important step forward, as it is safe to say that less than half of 
the architects and engineers in this country know the crushing 
strength that can be obtained from the mixtures they specify 
with the aggregates that are being used on jobs under their super- 
vision every day. 

This statement was recently borne out by an investigation of 
the aggregates available for concrete in one of our larger cities 
prior to the formation of a Building Code for concrete and rein- 
forced concrete buildings. In this code it was originally planned 



Digitized by 



Google 



Kinney on Aggregates for Concrete. 



489 



to have the requirement for compressive strength of 1:2:4 con- 
crete at 28 days at least 2400 lb. per sq. in. This was later 
reduced to 2000 lb. and the allowable stresses figured accord- 
ingly. An investigation by the writer and others interested in 
the writing of the Code developed the fact that though the supply 
of sand and gravel was limited to practically one source and the 
crushed stone to several quarries in the same vein, very few tests 
on the aggregate were obtainable and these seemed to indicate 
that even 2000 lb. was too high. A series of tests was started 
and the results to date are quite surprising. The specifications 

Table I. — Comparative Compressive Strength in lb. per sq. in. of 
Concrete Made from Various Aggregates. 



Mark. 


7 Days. 


Ordinary gravel.. 
Large gravel 

Stone 


677 
o9o 
603 

626 

898 
677 

874 

816 

819 
858 

962 

879 



28 Days. 


3Mos. 


1578 
1390 
1633 


1512 
1673 
1943 


1533 


1709 


1626 
1640 


1684 
1779 


1580 


1649 


1582 


1704 



1824 
1972 



1880 
1888 



2031 
2206 



2484 
2240 



Proportions. 



1 part typical cement. 

2 parts nver sand passing */u in. screen, 
'cl Vw in. ' "' ' 



4 parts river gravel 



. to V« in. 



1 part tsrpical cement. 

2 parts nver sand — 3 parts through Vi« 

in. screen, 1 part Vi« in. to 1/4 in. 
4 iMtrts river gravel 1/4 in to 1 in. 



1 part typical cement. 

2 parts nver sand — 3 parts throujgh */i« 

in. screen, 1 jmrt '/is in. to 1/4 in. 
4 IMtrts crushed stone 1/4 in. to 1 in. 



Typical cement used was a mixture of five brands. Medium consist- 
ency. Hand mixed. Test pieces 8 in. in diameter and 16 in. long. Aged 
in temperature of about 70^ F. and protected from drying out by cotton bags 
wet twice a day. 

require that the coarse aggregate shall pass a 1-in. ring and be 
retained on a i-in. ring; fine aggregate to be all that passing a 
i-in. ring. The material produced commercially was being 
screened through a |-in. and over a A-in. mesh, so that tests were 
made on the aggregates as conmiercially produced and on samples 
specially prepared to meet the specifications. In the case of the 
sand, this was done by mixing with three parts of sand passing a 
^-in. screen, one part of the fine material passing a J-in. screen 
obtained from the i^-in. to f-in. gravel. The results up to three 
months are shown in Table I and prove conclusively that with 



Digitized by 



Google 



490 



Kinney on Aggregates fob Concrete. 



Tabub II.— Tests on Typical Ceiibnt used in Coiipbessiqn Tests. 

Fineness 100 mesh — 94.8 per cent. Initial Set — 4 Hours. 

Fineness 200 mesh— 77.4 per cent. Final Set— 7 Hours 25 Minutes. 

Soundness — Satisfactory. Normal Consistency — 24 per cent. 

Chemical Analysis. 

Silica 20.72 

Alumina 7.24 

Iron Oxide 2.84 

Caleium (hdde 62.85 

Magnesia 2.47 

Sulphuric Anhydride 1 . 42 

Moisture and undetermined 2 . 46 

Tensile Strength in lb. per sq. in. 





NeM. 






1 : 3 Ottawa Sand. 




24 Houni. 


7 Days. 


28 Days. 


3 Months. 


3 Days. 


7 Days. | 28 Days. 


3 Months. 


310 
815 
330 


600 
660 
660 


740 
780 
750 


765 
775 

740 


180 
190 
200 


340 
300 
300 


420 
390 
400 


440 
480 
460 


318 


640 


757 


760 


190 


313 


403 


460 



I : 


3 Commeroial Sand. 


3 parts throi 


7 Days. 


28 Days. 


3 Months. 


7 Days. 


240 
300 
265 

268 


390 
380 
36.^ 

378 


390 
425 
380 

39S 


290 
340 
290 

307 



1 : 3 Special Sand. 



28 Days. 



460 
420 
400 

427 



3 Months. 



505 
480 
535 

506 



Granulombtric Analysis. 





Commercial 
Sand. 


'^t» 


P«r fwnt retained on No. 4 mcvh . . t . . . . . . t . . . . t - - , - 1 - - - 



10.6 
19.2 
25.8 
11.8 

5.8 
24.4 

0.6 

1.4 





10 ** 


31.5 


20 '• 


14.5 


30 •* 


15 2 


40 " 


10.5 


50 " 


4.2 


80 '* 


18.5 


100 " 


2.3 


Through 100 " 


3.2 







Digitized by 



Google 



Kinney on Aggregates for Concrete. 



491 



such proportions a requirement of 2000 lb. per sq. in. is too hi|^ 
for the local material, and that either the amount of cement will 
have to be increased or the proportions readjusted so as to pro- 
duce higher results. In these tests, which were made on 8-in. 
diameter cylinders 16 in. long, the concrete was accurately pro- 
portioned and mixed by hand and the test pieces were stored in 
an even temperatiu-e approximating 70* F., being protected from 

Table III. — Compressive Strength in lb. per sq. in. of Concrete 
CTLifn)ERS Using Material from Same Source as Table I. 



Brand of Cement. 



10 Days. 

4«6 
498 
600 

521 

439 
539 
409 

462 

620 
527 
515 

520 

395 
521 
536 

484 

430 
396 
434 

420 



30 Days. 


3 Months. 


803 

960 
828 


Ill 


893 


1275 


990 
904 
800 


1182 
1096 
1018 


898 


1098 


820 

999 

1142 


1433 
1172 
1136 


987 


1247 


774 

872 
898 


1138 
1263 
1316 


848 


1235 


706 
834 
1084 


1263 
1219 
1160 


878 


1214 



6 Months. 



looa 

1185 
1131 

1106 

1152 
1063 
1299 

1171 

1333 

949 

1464 

1248 

1084 
1382 
1609 

1358 

1060 
1064 
1046 

1063 



Proportions: 1 part cement (5 brands used individually), 2} parts river 
sand through A in., 5 parts river gravel A »n. to li in. 

Test pieces 8 in. in diameter and 16 in. long, mixed with batch mixer, 
stored in open shed and sprinkled night and morning for first 7 days. 

drying out by cotton bags wet twice a day. The cement used 
was a mixture of five representative brands, the results of tests 
on which are shown in Table II. In the same table are granulo- 
metric and tensile tests on the commercial and special sand. It 
will be noted that despite the fact that the commercial sand 
approximates very closely the tensile strength of Ottawa sand and 
the special sand exceeds in strength that of Ottawa sand, yet 



Digitized by 



Google 



492 



Kinney on Aggregates for Concrete. 



used in concrete with gravel from the same source and a very 
good grade of crushed stone, neither of these materials developed 
a crushing strength of 2000 lb. per sq. in. at 28 days. Other 



Table IV. — Comparative Cold Weather Tests on (8 in. Diameter by 

16 IN. Long) Cylinders Mixed 1 Cement, 2i Sand, 5 Gravel, 

December 4, 1911. Cyunders Protected from Frost by 

Cotton Bags. Hand Mixed, Medium Consistency. 

Crushiiig Strength in lb. per sq. in. 



Age. 



3 days in air and 1 day in laboratory . 



7 days in air. 



14 days in air. 



30 days in air. 



Brand "A" 



116 
122 

119 Average 

227 
265 

246 



4S4 
458 



471 



640 
739 



689 



Brand "B" 



95 
95 

95 Averago 

199 
209 

205 

440 

388 

414 

495 
653 

574 



Temperature — Degrees Fahrenheit. 



Ist day 

2d " 

3d " 

4th " 

5th '• 

6th " 

7th •• 

8th *• 

9th •• 

10th '• 

11th " 

12th " 

13th " 

14th " 

15th " 



Max. 



30 
36 
48 
54 
54 
54 
66 
61 
59 
51 
42 
47 
58 
38 
33 



Min. 



18 
21 
27 
29 
40 
48 
46 
58 
45 
35 
32 
39 
38 
32 
30 



Mean. 



24 
28 
38 
42 
47 
51 
56 
60 
52 
43 
37 
43 
48 
35 
32 



16th day. 
17th ■ 
18th 
19th 
20th 

2l8t 

22d 
23d 
24th 
25th 
26th 
27th 
28th 
29th 
30th 



Max. 


Min. 


Mean. 


34 


28 


31 


41 


24 


32 


49 


36 


42 


52 


44 


48 


47 


33 


40 


39 


33 


36 


45 


34 


40 


56 


40 


48 


67 


23 


40 


23 


14 


18 


32 


18 


25 


45 


32 


38 


64 


28 


41 


29 


26 


28 


34 


23 


28 



results obtained on material from the same source are shown 
Tables III and IV, and the latter table particularly shows^that 
this aggregate is poor for cold weather work. Great care'should 
be exercised in the use of such material in the winter time as 



Digitized by 



Google 



Kinney on Aggregates for Concrete. 493 

unless the materials are heated and the concrete well protected, 
poor results are almost inevitable. 

Having established the value of aggregates obtained from 
any particular soiu-ce, it is then necessary to see that the mate- 
rials received on the job are equal in quality to the samples 
tested. An examination of the bank, pit or quarry may often 
reveal the fact that it would be impossible to obtain a uniform 
product. In such cases, extreme care must be exercised by the 
men obtaining the material and frequent tests should be made 
to see that that received is right. 

The pier shown in Fig. 1 is a good illustration of the effect of 
variation in aggregate. This pier was built in connection with 
two abutments to support a steel girder railroad bridge across a 
small stream. Prior to starting the concrete work the engineer 
secured a sample of the sand and gravel which the contractor 
proposed to use, and the laboratory results obtained indicated 
that the materials as sampled were entirely satisfactory for con- 
crete work. That the material as received was not good, and 
apparently not equal to the sample, is evidenced by examination 
of the condition of the concrete. A photograph of the gravel 
bank (Fig. 2) reveals the seat of the trouble. The dark streaks 
are apparently decayed vegetable matter, while here and there 
through the bank will be noticed strata of very fine uniform size 
sand. Examination of such a bank indicates quite conclusively 
that the aggregate could not rim uniformly, and imdoubtedly the 
delay in hardening which finally resulted in freezing was due to 
the presence of too much fine sand and loam. That this was 
the cause of the trouble is conclusively proven by the fact that 
the poorest results are evidenced at the water level where the 
fine material held the moisture and at the end of a day's work 
about midway of the pier where most of the fine material floated 
to the top of the concrete. 

On another job of a similar nature, but involving the use of 
approximately 50,000 barrels of cement, the results were anal- 
ogous. A large amount of concrete was condemned on account 
of imsatisfactory strength. In the middle of a 1000-yd. founda- 
tion laitance was found over a foot thick which had the appear- 
ance of wet clay and could be readily dug out with a knife even 
though the concrete had been in a month. Run-of-pit gravel was 



Digitized by 



Google 



494 



Kinney on Aggregates for Concrete. 



being used which had been tested for percentage of fine and 
coarse material and found satisfactory, but examination of the 
pit revealed the fact that it would have been impossible to have 
secured a sample which would have fairly represented the deposit. 




Fia. 1. — DISINTEGRATED CONCRETE PIER, DUE TO POOR AGGREGATES. 

In some places there was very little sand, while in others there 
was practically no gravel. As a mixture of one part cement and 
seven parts run-of-pit gravel was being used, the natural result 
of the use of the latter material can be imagined. Further than 



Digitized by 



Google 



KiNNET ON AgGRBGATES FOR CONCRETE. 



495 



this, there were strata of very fine sandy loam throughout the pit 
and the large section of laitance was the result. 

The foregoing are a few of the many examples which can be 
cited where failure to appreciate the value of .having good 
aggregate has led to trouble and such examples emphasize the 
importance of tests of the aggregates. It is quite surprising to 
find architects and engineers basing their designs on a certain 
strength of concrete which cannot be obtained or may be largely 




FIG. 2. — GRAVEL BANK FROM WHICH POOR AGGREGATE WAS OBTAINS^. 

exceeded in actual practice, and surely our work is devoid of the 
fundamentals of good engineering when we use without ^scrim- 
ination aggregates of high and low strength giving values. A 
more thorough study of our sources of supply for concreting 
materials and a more careful inspection of these materials as 
they are received on the work, is of utmost importance to the in- 
dustry and it should be one of the first efforts of this Association 
to study this important subject. 

A partial bibliography of the literature on aggregates for con- 



Digitized by 



Google 



496 Kinney on Aggregates for Concrete. 

Crete is given below with the hope of furthering study on the 
subject. 

Concrete Aggregates, by Sanford E. Thompson, Proceedings National 
Association of Cement Users, Volume II, p. 27. See also Engineering Record, 
January 27, 1906. 

Sands and Their Relation to Mortar and Concrete, by Henry S. Spackman 
and Robert W. Lesley. Proceedings, American Society for Testing Materials, 
Volume VIII, p. 429. 

The Value of Sand in Concrete Construction, by E. S. Larned, Proceedings, 
National Association of Cement Users, Volume IV, 205. 

Concrete — Its Constituent Materials, by Russell S. Greenman, Barge 
Canal Bulletin (New York State), November, 1909, p. 429. 

Practical Tests of Sand and Gravel Proposed for Use in Concrete, by Russell 
S. Greenman, Proceedings American Society for Testing Materials, Vol- 
ume XI, p. 515. See also Engineering Record, Volume LXIV, No. 3, p. 66. 

Economical Selection and Proportion of Aggregates for Portland Cement 
Concrete, by Albert A. Moyer, Engineering-Contracting, Volume XXXIII, 
No. 3, p. 52. 

Concrete Aggregates by Dr. J. S. Owens, Concrete and Corutrudional 
Engineering^ March 1909. 

\Portiand Cement Mortars and their Constituent Materials, by Richard L. 
Hui|S)plhrey and William Jordan, Jr., BuUetin 331, United States Geological 
Survey. 

Good Concrete and How to Get It, by F. M. Okey, Municipal Engineering, 
May, 1909. 

• Notes on Concrete, a discussion printed in Journal of Association of Engi- 
neering Societies. See Engineering Record, Volume LXI, No. 5, p. 125. 

Impurities in Sand for Concrete, a discussion, — ^Transactions American 
Society of Civil Engineers, September, 1909. 

BUist Furnace Slag in Concrete, booklet published by Carnegie Steel Com- 
pany, Pittsburgh, Pa. 

Tests t0 Determine Effect of Mica on Strength of Concrete, by W. N. Willis, 
Engineering News, February 6, 1908. 

Directions and Suggestions for the Inspection of Concrete Materials, by 
Jerome Cochran, Engineering-Contracting, Volume XXXVII, No. 5, p. 116. 

A Study of Sand for Use in Cement Mortar and Concrete, by E. S. Lamed, 
Journal Association of Engineering Societies, Volume XLVIII, No. 4, April, 
1912. 



Digitized by 



Google 



DISCUSSION. 



President Humphrey. — The relation of the sand to the Mr. Humphrey, 
mortar, especially of the strength of mortars made of the com- 
mercial sand to that of mortars of standard Ottawa sand, is of 
interest. There has been a great deal of debate in the Joint 
Committee as to whether they should show the same strength 
or possibly less. Engineers from New England, especially from 
Boston, object to requiring the same strength, because in their 
locality it is impossible to obtain a sand that would show this 
strength, therefore making the requirement a hardship. I believe, 
however, with Mr. Kinney, taking sand the country over, that the 
various sands should show at least the strength of the Ottawa 
sand, and that localities where it is not possible to obtain this 
strength must fix the requirement to suit their locality. It is 
certainly a fact that the well graded sands give higher results 
than the standard Ottawa sand. The standard Ottawa is a one 
size sand with a large percentage of voids and the strength of 
the mortar in which it is used is less than that of sands well 
graded. 

Mr. Kinney states that grading gives density and that the 
increased density gives an increased strength. The retaining 
walls with the disintegrating mortar referred to are as fine an 
example of cause and effect as could possibly be had. It did not 
require a technical man to look at the sand bank to see that the 
material was not suitable for mortar. It was quite evident that 
the fine sand in the mortar retarded the hardening, and left the 
mortar with insufficient strength and thus easily damaged by 
frost action. This is one of the most important subjects that we 
can discuss. There is more information desired and great need 
for intelligent understanding of just what important part sand 
plays in mortars and concrete; it probably has a more important 
bearing on the resulting strength of the structure than the strength 
of the cement. I think we are more prone to ascribe the defects 

(497) 



Digitized by 



Google 



498 



Discussion on Aggregates. 



Mr. Kinney. 



Mr. Wilson. 



Mr. Httmphrey. 



in mortar or concrete directly to the cement rather than the sand 
used. 

Mr. Wm. H. Kinney. — The sand from this particular bank 
had been tested by the engineer and was found to have a strength 
which compared favorably with Ottawa sand. In fact, it passed 
the 70 per cent strength requirement by a very satisfactory 
margin. However, the samples did not fairly represent the 
material the contractor was obtaining from the bank. 

Mr. Percy H. Wilson. — There was a case called to my 
attention 5 or 6 months ago where a railroad engineer endeavored 
to obtain in the laboratory a compressive strength of 2,000 lb. 
per sq. in., the requirement of the Joint Committee. On obtaining 
only about 1,800 lb. the test was tried over and over, varying 
the aggregate, but the strength never w^ent up to 2,000 lb. Now 
is this requirement of the Joint Committee too high? 

President Humphrey. — Many contractors and probably 
a good many engineers and architects never seem to think it is 
necessary to test the sand. They simply pass on it from a visual 
inspection, whereas the committee states that a differentiation 
between a good or a poor sand can only be made by actual test. 

I would say I do not think the standard requirement set by 
the Joint Committee of 2,000-lb. concrete is too high. It is a 
standard we should work to. Unfortunately there is the practice 
in this country of specifying the proportions such as 1 : 2 : 4 or 
1:3:6. Such proportions mean absolutely nothing, as the 
proper portions of materials cannot be determined until the 
qualities of the material itself are known. The voids vary with 
all the material all over the United States. A study of Bulletin 
No. 331* on the subject of sand will show that there is a wide 
range of variation, first, as to the size of the particles, and second, 
as to the hardness and character of the material itself. A traprock 
with a high compressive strength used as an aggregate in a certain 
proportion will give a concrete of much higher strength than an 
aggregate of very soft limestone. If a contrac^tor with a fixed 
proportion cannot obtain 2,000 lb. or with any variation of the 
proportions, then more cement must be used or another aggregate 
obtained. The mere fa(!t that his material in standard propor- 
tions will not give 2,000 lb. in my mind does not mean that he 

♦U. S, Geological Survey. 



Digitized by 



Google 



Discussion on Aggregates. 499 

should not try to get 2,000 lb. There are too many structures Mr. Humphrey, 
being erected in this country assumed to have a strength of 2,000 
lb. and upwards where as a matter of fact the strength is very 
materially less. The sooner more attention is given to the strength 
of the concrete as well as of the cement and the steel, the better 
the results will be. 

The Joint Committee report specifies that the relation of 
the cement to the aggregates shall be 1 to 6. The next report 
will contain a table showing the strength of concrete from aggre- 
gates of different character and proportions, so that for any 
aggregate, say a soft limestone, reference to the table will show 
approximately the proportion of cement and aggregate necessary 
to obtain a 2,000-lb. concrete. 

Mr. L. R. Ash. — In connection with the sand and grading of Mr. Ash. 
aggregates suggested by the disintegrated concrete in the pier, 
the water put into a mixture sometimes makes considerable 
difference and also the depositing of concrete in or through water. 
I want to call attention to a very interesting occurrence here in 
Kansas City a couple of months ago. We have a bridge over 
the Kaw River built some 4 or 5 years ago and lately during some 
very intensely cold weather, the cylinder pier split open and went 
down so badly that false work had to be placed to hold the bridge 
up. In the search for the cause the first suggestion was that the 
contractor had not put in any cement but some line that seemed 
to be accessible in the neighborhood. I was talking to the fore- 
man of the job and he told me the circumstances, which it seems 
to me would explain it very clearly. The concrete was deposited 
through several feet of water with the result that there were layers, 
at intervals, of material more like laitance than anything else, 
which was chalky in its consistency and would absorb great quan- 
tities of water. To me this very thoroughly explained the break- 
ing down of the concrete. I have frequently noticed in depositing 
concrete that small pockets or irregularities in the distribution 
of the water would crdate conditions that would show up badly, 
and entirely outside of any irregularities in the aggregate at all. 
Sometimes I think that is a matter which is overlooked in the 
handling of concrete and depositing of the same in the forms. 

Mr. Kinney. — I would like to ask whether anyone has noticed Mr. Kinney, 
in the fracture of gravel concrete that the pebbles pull out rather 



Digitized by 



Google 



500 Discussion on Aggregates. 

Mr. Kinney. than break? This was particularly noticeable even at six months 
in these building code tests and it occurred to me that it 
might be a characteristic of all gravel concrete in the early stages 
of hardening. 

If anyone contemplates the making of test cylinders on a 
job or in the laboratory, I might add that we have found a very 
satisfactory form made of galvanized iron. This form consists of 
a flat piece of galvanized iron 16 in. wide and long enough to come 
together and be soldered to form an 8-in. cylinder. The bottom 
is circular and slightly larger in diameter than 8 in. and is soldered 
to the cylinder. With the use of a sharp instrument these soldered 
joints can be readily opened and the forms stripped. Such forms 
cost 22 cts. apiece and they are worth the price. 

Mr. Humphrey. PRESIDENT HUMPHREY. — Whether or uot tcsts of Cylinders 

would show a fracture through the gravel itself, depends largely 
on the age and the proportions. A proportion of 1 : 3 : 6 probably 
up to a year might not show such a failure. In many tests made 
in St. Louis the gravel did fail, especially 1:2:4 mixtures. The 
fracture was right through the particle, even the hardest particles. 
The governing factor seems to be the manner in which the materials 
are handled; that is, the character of the concrete and its age. 
A well made gravel concrete of 1 : 2 : 4, or in which there is one 
part of cement to six parts of gravel, at the end of six months 
should certainly show a fracture through the gravel. 



Digitized by 



Google 



FIELD INSPECTION AND TESTING OF CONCRETE. 
By G. H. Bayles.* 

In the spring of 1910 we were commissioned to design and 
superintend the construction of a warehouse 100 x 210 ft., 4 
stories high, in the Borough of Brooklyn, New York, for the New 
York Dock Company. The use of reinforced concrete was 
recommended and adopted, but from experience and knowledge 
of the many criticisms of this material and the considerable num- 
ber of failures in its use, it was decided that very careful inspec- 
tion should be maintauied. Consequently when the work started 
in the latter part of July the engineer who designed the building 
was put in charge, assisted by two inspectors, both experts in the 
making and placing of reinforced concrete. The engineer assumed 
general charge of the work. One inspector superintended the 
proportioning and mixing of the concrete and the other the 
placing in the forms. They collaborated in the inspection of con- 
structing and wrecking forms, placing reinforcement and testing 
materials. 

The results of the tests on the above work were on the whole 
so satisfactory and the information obtained was deemed of such 
importance that they were continued on some other work during 
the summer and fall of 1911. This latter work consisted of the 
reconstruction of a block of six warehouses. The whole block is 
210 X 374 ft. in plan and 4 stories high. The buildings were of 
the old type, brick walls with wood interior. All the wood was 
removed and replaced by reinforced concrete, mak'mg the build- 
ing fireproof throughout. As this work was done with greater 
despatch than the first job it was thought best to have more 
inspectors. Consequently a chief inspector was put in charge of 
the whole work, assisted by one inspector on the proportioning 
and mixing, one on the placing concrete and one solely for making 
tests. This arrangement worked very satisfactorily and allowed 
the engineer to devote more of his time to other work. 

There seemed to be no recognized system of testing concrete 

* AraiBtant to J. W. Galbreath. ConBulting Engineer, New York. N. Y. 

(501) 



Digitized by 



Google 



502 Bayles on Inspection and Testing of Concrete. 

or concrete materials on the job, so a series of tests was arranged 
to meet what were believed to be the necessary requirements. 
These tests were for cement: fineness, constancy of volume, time 
of setting, initial and final, specific gravity and crushing strength 
of 1: 2 cement mortar; for sand: tests for loam, fineness and 
percentage of voids; for stone: tests for percentage of voids; 
for concrete: tests of the crushing strength. No tests of the 
reinforcement were made on the job. 

A temporary laboratory was built near the work and the 
necessary apparatus installed. This laboratory was intended to 
Ix* such as could be built on any job and no arrangements were 
made for heating it except the use of a small oil stove in extreme 
weather. It was 12 x 8 ft., built of |-in. sheeting on 2 x 4 in. 
frame ^vith felt roof; one door front, and windows one side and 
rear. 

cement tests. 

At first 6 samples for testing were taken from every car of 
cement, but later this number was reduced to 3. For fineness 
the usual 100-mesh and 200-mesh sieves were used. The inspector 
shook the sieves in his hands and judged by the eye when no 
appreciable quantity of cement was going through. Scales 
graduated to milligrams were used to determine the proportions. 
A nearby tinsmith made the boiler and wire rack for the con- 
stancy of volume tests and the glass plates were cut on the job 
from scrap window glass. The boiler is 15 in. high and 8 in. in 
diameter with a lid, and the rack was made to hold 12 pats at a 
time. The test pats were boiled for 5 hours or more. The Vicat 
needle was used to determine the time of setting. The variable 
temperature made the remits of these tests vary so greatly as to 
be of little value. Le Chatelier's specific gravity apparatus was 
used with gasoUne for determining the specific gravity. 

The test for crushing strength, not being in such general use 
as the other tests, requires more detailed description. It consisted 
in testing to failure 4-in. cubes of 1 : 2 cement and sand mortar. 
More elaborate apparatus was necessary for making these tests. 
For this purpose were required gang molds for the cubes, a damp 
closet large enough to hold 4 gang molds, pans of water in 
which to submerge the test pieces, and a compression testing 



Digitized by 



Google 



Bayles on Inspection and Testing of Concrete. 503 



machine. The gang molds, Fig. 1, were 'made of 1-in. lumber 
held together by iron clamps. The measm-ements were not exact, 
but the surface areas were approximately correct within the limits 
of accuracy of the tests. While only 2 cubes were used for each 
test the gang molds were made for 4 cubes, as the same molds 
were to be used for the concrete tests. They were made by a 
carpenter on the job and the clamps were made by the com- 
pany's blacksmith. 

The damp closet was built into the comer of the laboratory, 
the sides and floor of the laboratory forming two sides and the 
bottom of the closet; the other sides and top were of J-in. 
tongued and grooved sheeting. Two shelves were put in the 



£ 








1 




a. 






1 r 








■' •- 


■^ '" 










0| 


^ 4' " 


-«- 4^ 




L.4 


§6 


^.--^ 







n 






H 






Epj 










! 


1 1 


1 

1 










o 




o 


. 


o 














' 



FIG. 1. GANG MOLD FOR 4-IN. CUBES. 

closet, each large enough to hold 2 gang molds, and space was 
left below for a large pan of water. When a mold with test 
pieces was placed in the closet a wet coffee sack was thrown 
over it, the ends of the sack dipping into the water below. A 
better arrangement would be to have the closet lined with felt, 
the lower edge of the felt dipping into the water. The simpler 
method was used as being one easily provided on any job. 

The pans for water in which the test pieces were to be 
submerged were made 28 x 28 in. and 5 in. deep, large enough to 
hold 36 cubes each. The same size pan was used in the damp 
closet. Five pans were required and they were made by one of 
the company's sheet metal workers. 



Digitized by 



Google 



504 Bayles on Inspection and Testing of Concrete. 

The selection of a compression testing machine was more 
difficult. Most machines designed to do this kind of work are elab- 
orate and expensive. A simple and comparatively inexpensive 
machine was desired and after careful investigation a hydraulic 
machine was acquired, consisting essentially of two horizontal 
plane surfaces between which the test pieces are crushed. The 
top surface is fixed, being held rigidly in place by two side posts 
extending upward from the base. The lower or movable surface 
is fixed rigidly on top of a 5-in. cylindrical ram with cup leather 
packing moving in a copperlined cylinder. The pressure is 
applied by means of a hand pump and is measured on a gauge 
reading total tons on the 5-in. ram. The gauge is graduated from 
to 60 t. and while the rated capacity of the press is 50 t., 
cubes were crushed which required as much as 60 t. pressure. 
As many of the newer cubes crushed at low pressure and «xact 
readings on the gauge that was furnished on the machine were 
difficult, a second gauge was attached reading up to 10 t. only. 
When cubes requiring a greater pressure were being tested this 
second gauge was shut off by a valve. 

On account of the high values which were obtained at the 
beginning of the work suspicion was directed towards the accuracy 
of the machine, but comparisons of the cubes by testing on an 
accurately gauged machine at the testing laboratory of Columbia 
University showed that the small press used on the work was 
correct within reasonable limits* (see Table IV). 

For each test 2 cubes of 1 : 2 cement and sand mortar were 
made. The sand used was taken from that used on the work 
and was washed clean. To insure uniformity a sufficient quan- 
tity for all the tests was taken at one time and stored in a bin 
until required. For each test a little in excess of 128 cu. in. of 
sand and half as much by volume of cement were used. The 
sand and cement were mixed dry, then sufficient water added to 
make a fairly wet mixture, that is, one that was readily puddled 
but would not pour. The whole was then well mixed and placed 
in the molds and puddled ^vith a small trowel. All the operations 
were done by hand. As soon as the mortar was sufficiently set 
to hold the markings the date and hour were marked on both 

* A dcBcriptiou of a test of a similar machine is given in Engineering Neva, February 11, 
1909, p. 167. 



Digitized by 



Google 



Bayles on Inspection and Testing op Concrete. 505 

cubes and the mold placed in the damp closet. At the end of 
24 hours the mold was taken from the damp closet, the forms 
removed and one cube crushed. The other cube was submerged 
in a pan of water where it remained for 6 days and was then 
crushed. All were tested to failure. 

To prevent little inequalities of the surfaces of the cubes 
from seriously affecting the results three thicknesses of blotting 
paper were used top and bottom of each cube in the press. After 
some time it was discovered that the long sides of the gang molds 
were more nearly parallel than the others and thereafter care was 
taken to use the sides of the cubes formed by them as the crush- 
ing faces. By this means better and more nearly uniform results 
were obtained. 

SAND tests. 

Every scow load of sand was tested for percentage of loam. 
If the sand in different parts of the scow appeared to be of dif- 
ferent quality more than one test was made. This test consisted 
in taking a quantity of sand and drying it thoroughly on a stove. 
A thousand grams of it was then taken in a 12-quart pail and 
washed by turning on a hose, giving the water as it flowed from 
the hose just sufficient velocity to keep the mass stirred up and 
moving, the loam being carried off in the overflow. When the 
water ran clear the sand was again dried and weighed, the dif- 
ference in weight giving the percentage of loam. Sand contain- 
ing more than 2 per cent was rejected. 

Tests were also made to determine the comparative fineness 
of the sand. These were only made when there appeared to the 
eye to be a difference in the grade. The test consisted in taking 
a measured quantity of sand (by weight)' and screening it by 
hand through 20-mesh and 30-mesh sieves. These tests showed 
on an average about 21 per cent retained on the 20-mesh sieve, 
27 per cent on the 30-mesh sieve and the remainder passing 
through the 30. 

Occasional tests for percentage of voids in the sand were 
made as follows: A comparatively large amount of sand was 
measured and weighed to show the average weight per cu. cm. 
Then a quantity by weight corresponding to 50 cu. cm. was put 
into a glass graduate and the volume of water displaced meas- 



Digitized by 



Google 



506 Batles on Inspection and Testing of Concrete. 

ured. The average volume of voids was thus found to be about 
40 per cent. 

stone tests. 

As the grade of the crushed stone varied but slightly only 
occasional tests were made to determine the percentage of voids 
in the stone. To do this a quantity of stone was submerged in 
water for 2 hours to permit it to become thoroughly saturated. 
The water was then poured off and the stone exposed to the air 
half an hour to permit the surface moisture to evaporate. A 
vessel of known capacity (nearly 10,000 cu. cm.) was then filled 
with the stone and weighed. Sufficient water was added to fill all 
the voids and the weight taken again. The difference in weight 
showed the percentage of voids. As there was excess of voids 
around the sides of the vessel this test is only comparative and is 
not considered particularly important as the inspector can best 
judge by the appearance of the batch when the voids are prop- 
erly filled. For the stone used on this work the voids measured 
by the above method averaged nearly 50 per cent. Measure- 
ments of concrete in place and stone on the scows indicated that 
the volume of stone used was about 97.5 per cent that of the con- 
crete produced. 

concrete tests. 

What seemed to be the most important test and the test 
of most practical value was that of the crushing strength of the 
concrete itself as it was placed in the forms for the building. 
For these tests the same apparatus was used as for the mortar 
tests. The preparation of the test pieces was simpler. The 
gang mold for four 4-in. cubes was taken onto the work where 
concrete was being poured. The concrete was taken from the 
buggies as it was being poured into the forms and was imme- 
diately placed in the cube mold and puddled with a small trowel. 
There was no excessive puddling as it was intended to approx- 
imate working conditions as nearly as practicable. One and 
sometimes two tests were made every day concrete was poured. 
From this time on the operation was similar to that for testing 
mortar. After 24 hours in the damp closet the forms were 
removed, 1 cube crushed and the 3 others submerged in a pan of 



Digitized by 



Google 



Bayles on Inspection and Testing of Concrete. 507 

water. Of these 3 one was crushed at the end of 7 days, 1 day 
in the damp closet and 6 in water, one at the end of 28 days and 
the other was removed from the water at the end of 28 days and 
laid aside to be tested later. 

As has already been said, no attempt was made to preserve 
an even temperature in the laboratory, the object being to main- 
tain as nearly as practicable the condition of the concrete in the 
work, so that it sometimes happened during the winter that ice 
formed on the water in the pans where the test cubes were sub- 
merged. The effect of this is clearly shown in the results of the 
tests. A practical use of the tests of concrete was to indicate to 
the inspector as well as the contractor the safe time for wrecking 
forms. This time was fixed at first at 7 days, but was afterwards 
shortened to 4 days during the summer and well into the fall, 
when the decreasing strength of the cubes tested caused the time 
to be again extended to 7 days. 

results of tests. 

There is nothing notable in the results of the tests for fine- 
ness, constancy of volume and specific gravity, except that they 
show what may be expected from the ordinary insi>ector who is 
not specially trained in making such tests and where the tests 
are made under the conditions prevailing on the ordinary job. 
The results of the tests for initial and final set were so varied, 
due, no doubt, to the changing temperature, that they lack 
value as determining the quality of the cement. 

Tests of the cubes of 1: 2 cement and sand mortar were 
intended to show a definite relation of strength of the cement 
being tested to what could be expected from the concrete made 
from the cement. Comparisons of the results of these tests with 
the results of the tests of concrete were so simple and proved so 
satisfactory that it is believed they show with sufficient accuracy 
the practical rate of setting of the cement and that the tests for 
initial and final set could be discontinued for field laboratories. 

Strength of Mortar. — The 173 tests of 1 : 2 cement and sand 
mortar on the first job showed an average compressive strength 
of 448 lb. per sq. in. in 24 hours and 2110 lb. per sq. in. in 7 days. 
On the second job, where a different brand of cement was used, 
198 tests were made and the average compressive strength was 



Digitized by 



Google 



508 Bayles on Inspection and Testing of Concrete. 

585 lb. per sq. in. in 24 hours and 2100 lb. per sq. in. in 7 days. 
Table I* shows in detail the results of the tests on the former and 
Table II those on the latter job. The results are recorded as 
found with the inspector's notes, the dates alone being omitted 
for lack of space. In Table I the first 50 tests were made 
between August 17 and October 3, 1910, the next 50 between 
October 3 and November 21, 1910, the next 50 between Novem- 
ber 21, 1910, and January 25, 1911. The remaining 23 were 
made between January 25 and March 16, 1911. In Table II 
the first 50 tests were made between July 11 and August 24, 
the next 49 between August 24 and September 22, the next 50 
between September 22 and October 21 and the remaining 49 
between October 21 and November 25, all in 1911. 

Strength of Concrete, — The 235 tests on concrete cubes on 
the first job showed an average compressive strength at 28 days 
of 3450 lb. per sq. in. and on the second job, 96 tests, 2321 lb. 
per sq. in. at 28 days. The tests on concrete cubes were carried 
on from August, 1910, to December, 1911, tests being made nearly 
every day except during the month of February, 1911, when no 
work was done on account of inclement weather, and from the 
completion of the first job in May, 1911, until the beginning of 
the second job in July following. During the latter part of 
December, 1910, and throughout January, 1911, the sand and 
water used in making the concrete were heated so that the batch 
had a temperature of 60 to 65 deg. when placed in the forms. 
Table III shows the complete record of the tests on concrete 
cubes on the first job, and Table IV those on the second job, 
both giving the inspector's notes of variations from the rule. 
While there are considerable variations of strength from the 
maximum to the minimum most of the cubes crushed near the 
average and reference to the final results indicates that the varia- 
tions were due more to the method of testing than to the quality 
of the concrete, for while a cube crushed after 24 hours may have 
been below the average, the cube of the same set crushed after 
28 days was about as often above the average as below it and vice 
versa. It seems evident that the actual strength of the concrete 



* The insiK'ctor's notes give all the strength testa in tons per 4-in. cube, but (or convenience 
of compaiison with other testa the results have been converted into lb. per sq. in. in printing 
the tables. — Ed. 



Digitized by 



Google 



Bayles on Inspection and Testing of Concrete. 509 



Table I. — ^Tests of Cement and 1 : 2 Mortar. 



Car 

Number. 



1 
2 
3 

4 
5 

f 1 
2 
3 
4 
6 
6 

f 1 
2 
3 
4 

, 5 

1 
2 
3 
4 
6 
6 

1 
2 
3 
4 
5 
6 

1 
2 
3 

4 
5 



1 
2 
3 
4 
5 
6 

1 
2 
3 

f 1 
2 
3 
4 
5 
6 

1 
2 
3 

4 
5 
6 



Percentage of Cement. 



Retained on Sieve. 



100 Mesh. 200 Mesh 



11.250 23.800 



7.150 
5.870 
7.210 



7.050 

8.260 
8.300 
7.455 
7.400 
7.080 
8.020 

6.980 
6.130 
6.300 
6.070 
7.170 

6.830 
7.050 
6.790 
5.700 
5.450 
6.960 

5.950 
6.600 
6.270 
7.110 
6.500 
7.000 

8.370 
8.300 
8.100 
5.720 
7.200 
6.450 

8.400 
7.450 
8.100 
7.900 



20.075 
25.010 
21.050 



Through 
200Meflh. 



64.860 



72.100 
68.430 
71.160 



20.300 71.850 



20.630 
21.400 
18.560 
23.310 
21.490 
22.520 

22.140 
20.700 
21.520 
23.140 
19.000 

20.100 
25.020 
22.940 
22.750 
21.800 
22.850 

20.720 
21.120 
21.270 
24.250 
22.400 
17.020 

20.120 
19.600 
19.550 
17.750 
19.030 
18.620 

19.750 
19.800 
19.870 
20.880 



70.620 
69.750 
73.730 
71.250 
70.000 
68.770 

70.490 
72.850 
72.000 
70.300 
73.100 

73.050 
67.830 
69.050 
71.330 
72.490 
70.110 

i 73.250 
' 72.050 

72.250 
' 68.550 

71.100 
I 75.700 

t 71.470 
71.800 
72.000 
76.380 
73.600 
74.800 

71.600 
72.650 
72.000 
71.000 



6.950 
6.490 



18.900 
18.300 



73.900 
74.700 



Setting. 



Initial. 
Hre. 



m 



2H 

1'^ 



4 
4 

4H 
4 

3H 

3>5 

4 

2H 

3*4 

4H 
4>^ 



3H 



4 
3 

2H 



4 



2H 
'3'" 



6.500 
6.420 
5.900 
6.070 
5.550 
5.320 

4.650 



20.330 
18.850 
17.500 
18.050 
17.700 
17.170 



72.970 ' 

74.700 I 

76.500 

75.800 

76.650 

77.460 



3 



Final. 
Hrs. 



16.550 78.720 



6.490 
"5 .'850 



19.050 74.490 | 
18.520 '75.430 



2% 
3 ■ 



2H 



314 



4^ 

4 

5H 



7 

6^ 

6 

5H 
5 

7 
7 
3M 

5 

6>i 

6% 



6^ 



7 



Crushing 
Strength 
Specific I of 4-in. cubes. 
Gravity in lb. persq. in. 



3.1 



3.15 
3.13 
3.11 
3.11 
3.15 

3.07 
3.12 
3.12 
3.13 
3.10 
3.12 

3.15 
3.14 
3.12 
3.12 
3.11 

3.15 
3.15 
3.16 
3.17 
3.13 
3.12 

3.12 
3.13 
3.12 
3.14 
3.12 
3.13 

3.12 



6H 3.09 
'3.10' 



5>i 
■5"* 



5'i 



5 



0>4 



3.14 
3.13 
3.14 
3.12 
3.13 
3.13 

3.09 
3.13 
3.14 

3.09 
3.11 
3.12 
3.12 
3.06 



Boiling 
5 hours. 



24 hrs. 7 days. 



3.09 
3.12 



3.10 
3.11 
3.11 



500 
563 
500 

1053 
1250 
1000 
1188 
1125 

750 
1250 
657 
500 
500 
845 

875 
813 
813 
375 
500 

750 
750 
750 
1025 
600 
813 

500 
475 
625 
500 
500 
525 

750 
875 
875 
875 
813 
875 

688 
750 
625 
688 
750 
625 

625 
875 
625 

375 
438 
438 
438 
563 
313 

1125 
1250 
938 
438 
375 
438 



2125 
2315 

2583 
3190 
2350 
3190 
3220 

1870 
2500 
2290 
2150 
2350 
2230 

2780 
2370 
2275 
1940 
2750 

2650 
2400 
1975 
3350 
2230 
2750 

2625 
2625 
2775 
2500 
2310 
2190 

2125 
2500 
2500 
2500 
2500 
2625 

2.500 
2750 
2625 
2250 
2250 
2370 

3000 
3500 
2625 

2065 
2813 
2875 
2250 
3130 
1880 

2870 
2440 
2500 
2630 
2500 
3190 



O.K. 



O.K. 
O. K. 
O. K. 
O. K. 
O.K. 

O.K. 
O. K. 
O.K. 
O. K. 
O. K. 
O.K. 

O.K. 
O.K. 
O. K. 
O.K. 
O.K. 

O. K. 
O. K. 
O. K. 
O.K. 
O. K. 
O.K. 

O.K. 
O. K. 
O. K. 
O. K. 
O. K. 
O. K. 

O.K. 
O. K. 
O. K. 
O. K. 
O.K. 
O.K. 

O.K. 
O. K. 
O. K. 
O. K. 
O. K. 
O.K. 

O.K. 
O. K. 
O. K. 



0. 


K. 


0. 


K. 


0. 


K. 


0. 


K. 



O. K. 

O.K. 
I O. K. 
i O. K. 



Digitized by 



Google 



510 Bayles on Inspection and Testing of Concrete. 



Table I. — Tests of Cement and 1 : 2 Mortar. (Continued ) 



Car 

Number. 




Setting. 

Initial. Final. 
Hr». 1 Hrs. 


Specific 
Gravity 


Retained 
lOOMesh. 


on Sieve. 
200Me8h. 


Through 
200Me«h. 




f 1 
2 
3 
4 
6 

.6 

[ 1 
2 
3 

4 
6 

ie 

f 1 

2 
3 

4 
6 
6 

1 
2 
3 
4 
6 
A 


8.270 


29.520 


72.120 


2H 


* 


3.14 


L 


8.200 


19.220 


72.200 


2H 


4K 


3.12 




7.960 


18.710 


'73.300 


2H 


4H 


3.12 


M 


8.110 
8.000 


19.040 
18.660 


72.760 
73.180 


3 


5 

5 


3.12 
3.12 


















7.000 


19.620 


73.260 Z}i 


6 






7.600 


19.010 


73.260 


4H 


6^ 


3.08 


n 


7.680 


18.620 


73.670 


4>i 


6 


3.13 




7.499 


18.400 


73.880 ; 4H 




3.12 




7.090 


18.960 


73.900 ZH 


55i 


3.13 


o 


7.200 


19.570 


73.160 i 4 


h^i 


3.12 




7.330 


20.300 


72.190 


3 


4H 


3.13 


f 1 

P 2 


7.610 
7.730 
7.860 

6.400 
7.000 
7.720 

7.600 
7.100 
7.260 

6.060 
4.770 
3.860 

4.480 
6.910 
6.710 

6.940 
7.070 
7.320 

7.300 
7.610 
7.560 

6.200 
6.350 
6.530 

6.490 
6.440 
7.030 

6.330 
6.430 
6.920 


20.620 
16.370 
20.410 

20.900 
21.210 
18.050 

20.680 
16.760 
20.760 

20.300 
20.400 
14.060 

17.360 
18.780 
23.730 

20.030 
19.300 
19.500 

17.700 
18.860 
21.650 

18 500 
18.000 
20.940 

19.040 
19.530 
18.910 

18.660 
18.780 
18.790 


71.560 
76.700 


2H ' 




3.14 
3.11 
3.13 

3.13 
3.12 
3.09 

3.08 
3.07 
3.09 

3.15 
3.12 
3.12 

3.12 
3.12 
3.12 

3.11 
3.11 
3.13 

3.12 
3.13 
3.13 

3.11 
3.13 
3.14 

3.12 
3.13 
3.14 

3.14 
3.12 
3.13 


3 


71.690 




f 1 

« ll 

1 

R 2 

3 

f 1 

8 2 

,3 

f 1 
T 2 


72.620 
71.780 
73.650 

71.620 
76.030 
71.860 

73.440 
74.700 
82.030 

78.160 
76.260 


4H 
6 


6H 

I'A 

6 
6 

6 
6 
6K 

6« 


13 

r 1 

U 2 
3 

f 1 

V 2 
3 

f 1 

W 2 

3 

f 1 

X 2 

3 

f 1 

Y 2 
3 


69.600 

72.560 
73.400 
73.170 

75.000 
73.500 
70.550 

75 . 150 
75.470 
72.510 

74.320 
74.020 
73.850 

74.970 
74.740 
74.200 


7K 

hV^ 
6K 
5H 

5Ji 

6^4 

84 

XA 

.•■>}4 

6 


8 
8 

IS 

7 
7 
9 

8 
8 



Crushing 

Strength 

of 4-in. cubes, 

inlb. persq. in. 



24 hrs. I 7 days. 



1188 
1188 
1125 
400 
438 
476 

876 
1126 
1125 
1125 
1250 

938 

538 
563 
563 
563 
626 
438 

375 
500 
438 
438 
438 
600 

438 
438 
438 

600 
513 
676 

625 
638 
600 

188 
188 
250 

250 
188 
188 

250 
250 
250 

188 
188 
188 

313 
313 
250 

313 
250 

288 

275 
275 
313 



2260 
2500 
2630 
2000 
2500 
2625 

2625 
2376 
2376 
2625 
3250 
2500 

2250 
2375 
2600 
2875 
2500 
3125 

2437 
2750 
2500 
2500 
2875 
2626 

2000 
1876 
2065 

2376 
2250 
2275 

2500 
2625 
2188 

2260 
2250 
2260 

2688 
2250 
2313 

2188 
2260 
2250 

1875 
1750 
1875 

2000 
1937 
1750 

1937 
1750 
2375 

2125 
2312 
2500 



Boiling 
5 hours. 



O.K. 
O.K. 
O.K. 
O.K. 
O.K. 
O.K. 

O.K. 
O.K. 
O.K. 
O.K. 
O.K. 
O.K. 

O.K. 
O.K. 
O.K. 
O.K. 
O.K. 
O.K. 

O.K. 
O.K. 
O.K. 
O.K. 
O.K. 
O.K. 

O.K. 
O.K. 
O.K. 

O.K. 
O.K. 
O.K. 

O.K. 
O. K. 
O.K. 

O.K. 
O.K. 
O.K. 

O.K. 
O.K. 
O.K. 

O.K. 
O.K. 
O.K. 

O.K. 
O. K. 
O. K. 

O. K. 
O. K. 
O. K. 

O. K. 
O.K. 
O.K. 

O.K. 
O.K. 
O.K. 



Digitized by 



Google 



Bayles on Inspection and Testing of Concrete. 511 



Table I.- 


—Tests of Cement and 1 


: 2 Mortar. 


{Continued ) 




Car 
Number. 


Perce 
Retained 
lOOMeahy 


Qtage of Cement 


Sett 

Initial. 
Hrs. 


ing. 

Final. 
Hrs. 


Specific 
Gravity 


Crushing 

Strength 

of 4-in. ctibea, 

inlb. pcrsq.in. 


Bofling 
5 hours. 


on Sieve. 
200Mesh. 

18.450 
19.000 
19.280 


Through 
200Meah. 

74.900 
73.960 
73.330 




24 hrs. 

313 

438 
250 


7 days. 




f 1 

Z 2 

3 


6.630 
7.030 
7.240 


5>i 


8H 
6>i 


3.13 


2375 
2063 
2125 


O.K. 
O.K. 
O.K. 


f 1 

AA 2 

3 


7.100 
6.980 
7.350 


21.280 
23.500 
19.430 


71.470 
69.100 
73.140 


6 


6 


3.12 
3.11 
3.12 


138 
200 
200 


1250 
1625 
1688 


O.K. 
O.K. 
O.K. 


1 
BB 2 

l3 


6.160 
7.000 
6.440 


18.110 
19.730 
19.270 


74.670 
73.250 
74.230 


8 
8 


8H 
9ki 


3.14 
3.13 
3.13 


125 
188 
200 


1250 
1500 
1025 


O.K. 
O.K. 
0. K 


1 

CC 2 

3 


6.600 
6.350 
6.380 


21.330 
18.650 
22.140 


71.980 
74.800 
71.450 


9 


10 


3.09 
3.12 
3.12 


125 
125 
125 


1437 
1250 
875 


O.K. 
O.K. 
O.K. 


DD 1 2 

l3 


4.950 
5.110 
5.000 


17.360 
18.550 
16.960 


77.650 
76.220 
77.960 


6 
6 
7^ 


7 
7 


3.11 
3.14 
3.11 


125 
160 
126 


1813 
1188 
1000 


O.K. 
O.K. 
0. K 


1 

EE 2 

3 


5.050 
4.460 
4.660 


18.150 
17.670 
20.350 


76.700 
77.810 
74.980 






3.12 
3.12 
3.07 


126 
126 
138 


1937 
1250 
1400 


O.K. 
O.K. 
O.K. 


FFJI 


4.440 
4.730 
6.000 


17.450 
19.480 
19.050 


78.070 
75.770 
75.920 


h 




3.11 
3.09 
3.11 


150 
138 
138 


1400 
1437 
1313 


O.K. 
O. K. 
O.K. 


f 1 

GG 2 
3 


5.000 
5.250 
5.430 


19.500 
20.650 
22.260 


74.860 
73.700 
72.050 


3M 
3K 




3.11 
3.09 
3.12 


150 
150 
163 


1437 
1625 
1563 


O.K. 
O.K. 
O. K. 


1 

HH 2 

3 


4.960 
4.850 
4.950 


21.750 
19.050 
19.020 


73.030 
76.000 
75.900 


2K 
35i 


3K 
3K 
4>i 


3.14 
3.07 
3.12 


150 
150 
138 


1437 
1500 
1126 


O.K. 
O.K. 
O.K. 


' 1 

•II 2 

^ 3 


5.140 
6.010 
5.800 


19.080 
20.520 
21.310 


75.650 
74.380 
72.830 


¥ 


2H 


3.12 
3.11 
3.14 


125 
88 
126 


1188 
1063 
1375 


O.K. 
O.K. 
O.K. 


f 1 
JJ 2 


7.420 


21.920 


70.420 


5k 

5H 




3.12 


125 
150 
160 


1437 
1538 
1563 


O.K. 
O K 


13 


7.640 


23.660 


68.650 




3.08 


O.K. 


f 1 

KK \ 2 

3 


10.230 
10.290 
9.500 


23.930 
23.500 
24.800 


65.600 
66.000 
65.500 


6 

6H 




3.12 
3.16 
3.16 


188 
160 
138 


1376 
1313 
1275 


O.K. 
O.K. 
O.K. 


f 1 

LL 2 

3 


7.030 
8.200 
6.300 


22.260 
23.900 
20.370 


70.600 
67.900 
73.230 


6 
6 
6 


8 
8 
7H 


3.12 
3.12 
3.11 


188 
150 
138 


1663 
1500 
1625 


O.K. 
O.K. 
O.K. 


MM 1 2 


9.350 
10.320 
10.520 

6.850 
6.850 
7.120 


22.500 
23.930 
23.610 

27.170 
25.900 
23.230 


68.000 
65.580 
65.700 

65.740 
67.000 
69.480 


6 
6 


7H 


3.15 


126 
138 
188 

188 
163 
150 


1313 
1313 
1376 

1126 
1437 
1376 


O.K. 
O. K. 


u 




3.14 

3.062 
3.107 
3.137 


O K 


1 

NN ■ 2 
3 


7 
7 
7 






f 1 

OO 2 

3 


6.750 
7.660 
6.980 


23.450 
24.260 
22.930 


69.050 
67.800 
70.000 


7 

8K> 


;;;;;;; 


3.137 
3.122 
3.137 


126 
100 
126 


1250 
1313 
1250 




f 1 


7.370 
7.020 
7.050 


22.830 
21.450 
21.800 


69.600 { 
71.300 ! 
71.040 


9 
9 
7 







75 
38 
38 


1125 
1260 
1376 




PP 2 






3 













Digitized by 



Google 



512 Bayles on Inspection and Testing of Concrete. 



Table I. — Tests of Cement and 1 : 2 Mortar. (Continued.) 



Car 

Number. 



QQ 



RR 



8S 



li 



Percentage of Cement 



Retained., on Sieve. 



Setting. 



Through initial 

100 Mesh. 200 Mesh. 200 Mesh., Hra. 



6.700 
6.670 
6.860 

6.000 
7.100 
4.840 



Final. 
Hrs. 



Specific 
Gravity 



22.570 70.450 

21.180 I 72.960 

21.900 I 70.900 

17.750 76.800 

24.100 68.680 

18.760 76.280 



7.800 23.320 
7.660 22.460 
7.300 ' 22.160 



68.630 8 
69.730 9 
70.440 ' 9 



Crushing 

Strength 

of 4-in. cubes, 

inlb. persq.in. 



24 hrs. 7 days. 



125 
125 
150 

150 
113 



1626 
1688 
1626 

1750 
1876 



125 I 1876 



3 122 
3.107 



163 
163 
150 
150 



1375 
1375 
2000 
1875 



Boiling 
5 hours. 



and the strength which can be safely counted on in construction 
is rather above than below the average of the tests made. 

requirements for materials and concrete. 

The values given in Tables III and IV are not necessarily 
true of all concrete, and perhaps it would not be amiss to describe 
in some detail how this concrete was made. 

In order that good concrete may be produced the first 
requirement is that the specifications be correct, full, exact and 
clear. This enables the inspector to do his work on a definite 
plan without the annoyance of possible objections or interference 
by the contractor's superintendent, who is sometimes more 
interested in the quantity than the quality of the product. With 
this object in view the specifications were made for this work. 
Following are the items appl>'ing to the materials and manufac- 
ture of concrete: 

Concrete will be composed of one (1) part cement, two (2) part^ sand and 
four (4) parts stone, with sufficient water to make a wet mixture. The 
cement, sand and stone will be introduced into a batch mixer of approved 
design and the mixer given 3 turns before the water is added. After the wat^r 
is added the mixer will be revolved at least 12 times and to the satisfaction 
of the engineer before the batch is dumped. Concrete must be placed immedi- 
ately after mixing and well spaded to insure a dense concrete. 

Cement, where used in these specifications, will mean Portland cement 
of such quality as to meet the standard specifications of the American Society 
for Testing Materials, and stand the tests pre^scribed in the rules and regula- 
tions of the Bureau of Buildings, Borough of Brooklyn, City of New York. 



Digitized by 



Google 



Bayles on Inspection and Testing of Concrete. 513 



Table II. — ^Tests of Cement and 1 : 2 Mortar. 



Q 


ar 
nber. 


Percentage of C 


ement 

Through 
200 Mesh. 


Setting. 

Initial. 1 Final. 
Hrs. 1 Hrs. 


Specific 
Gravity 


Crushing 

Strength 

of 4-in. cubes, 

in lb. per sq. in. 


Boiling 
5 hours 


Nut 


Retained 
100 Mesh. 


on Sieve. 
200Meah. 




24 hrs. 


7 days. 




A 


H 


5.93 

6. 

6.28 


18.97 
19.15 
20.20 


74.90 
74.70 
73.50 


ig 

2M 


3 
3 
2»i 


3.152 
3.122 
3.137 


450 
563 
469 


1938 
2375 
2125 


O.K. 
O.K. 
0. K. 


B 


f 1 
2 
3 


6.36 
5.75 
6. 


19.26 
18.41 
18.71 


74.26 
75.75 
75.22 


v 

lu 


3 

2}4 

2H 


3.152 
3.152 
3.137 


750 
844 
875 


2313 
2188 
2563 


0. K. 
O.K. 
O.K. 


C 


f 1 
2 
3 


6.7 

7.26 

6.75 


19.85 
19.90 
19.90 


73.25 
72.70 
73.22 


2H 

1».4 

2 


2»yi 

2H 

2»4 


3.122 
3.137 
3.137 


688 
513 
594 


2125 
1750 
1813 


O.K. 
0. K. 
O.K. 


D 


f 1 
2 
3 


6.14 
5.04 
6.09 


20.21 
19.82 
15.65 


73.43 
74.01 
77.94 


2 
2 

2 


2?4 

3 

3>i 


3.137 
3.16S 
3.152 


406 
406 
406 


2000 
1938 
2063 


O.K. 
O.K. 
O.K. 


E 


1^ 

3 


6.32 

6.5 

6.17 


20 02 
18.20 
19.21 


73.51 
75.21 
74.48 


2 
2 
2 


2H 


3.152 
3.122 
3.122 


313 
469 
388 


1875 
19.38 
2000 


O.K. 
O.K. 
O.K. 


F 


ll 
13 


5.38 
6.67 
6.41 


20.02 
16.98 
19.36 


74.39 
76.09 
74.06 


3K 


4 


3.152 
3.152 
3.168 


475 

388 
438 


1875 
1688 
1938 


O.K. 
O.K. 
O.K. 


G 


f 1 
2 
3 


5.07 
6.32 
6.39 


19.83 
20.13 
20.38 


74.86 
73.38 
73.08 


3 


4 

3»4 

4 


3.152 
3.152 
3.137 


375 
525 
531 


2125 
2813 
1813 


O.K. 
O.K. 
O.K. 


H 


f 1 
2 
3 


6.61 
6.13 
5.92 


18.84 
19.06 
19.41 


74.29 
75.58 
74.43 


2'<i 
2^ 


S14 

3?4' 

3'^i 


3.137 
3.152 
3.152 


.531 
563 
531 


2188 
2188 
2000 


O.K. 
O.K. 
0. K 


I 


f 1 
2 
3 


6.17 
5.06 
6.18 


18.92 

19.09. 

19.32 


74.68 
75.53 
74.29 


3 


3«4 

^1 


3.168 
3.152 
3.168 


531 
525 
513 


2125 
2125 

2188 


O.K. 
O.K. 
0. K. 


J 


f 1 
2 
3 


6.13 
7.73 
7.20 


18.84 
18.96 
17.64 


74.81 
73.08 
74.93 


2H 


4'2 

4 


3.137 
3.16K 
3.137 


500 
625 
563 


2125 
2250 
2250 


O.K. 
O.K. 
O.K. 


K 


f 1 

{ 2 

3 


6.17 
5.81 
6.13 


17.86 
19.07 
19.20 


75.61 
74.89 
74.38 


3 
2^ 

2H 


4>i 

4 

4 


3.152 
3.152 
3.168 


438 
388 
538 


2125 
2188 
2000 


O.K. 
O.K. 
O.K. 


L 


f 1 
2 
3 


5.86 
5.12 
6.09 


18.42 
18.63 
19.13 


75.41 
75.93 
74.59 


2H 


4 

4^ 

4M 


3.137 
3.168 
3.168 


450 
531 
325 


2125 
2063 
2000 


O.K. 
O.K. 
O.K. 


M 


( 1 

2 

, 3 


5.16 
6.03 
5.82 


19.13 
18.79 
19.03 


75.38 
74.92 
74.97 




4 

4 
4M 


3.168 
3.152 
3.152 


613 

688 
5.'>0 


2313 
2375 
2063 


O.K. 
O.K. 
O.K. 


N 


f 1 
2 
3 


5.20 
5.43 
6.12 


19.41 
19.21 
19.02 


75.13 
75.19 
74.70 


3^ 
3H 

3M 


4j.i 


3.152 
3.137 
3.152 


525 
600 
538 


2125 
2438 
2188 


O.K. 
O.K. 
O.K. 





f 1 
2 
3 


6.12 
5.87 
6.02 


18.80 
19.21 

18.84 


74.91 
74.79 
75.01 


35i 


AH 

434 

4^4 


3.168 
3.168 
3.152 


563 
550 
775 


2125 
1875 
2313 


O.K. 
O.K. 
O.K. 


P 


f 1 
2 
3 


5.82 
6.45 
8.24 


19.63 
19.47 
17.14 


74.38 
73.43 
74.14 


4 


5 
5 
6 


3.168 
3.168 
3.137 


594 
525 
525 


2250 
1938 
1938 


O.K. 
O.K. 
O.K. 


Q 


f 1 

2 

.3 


7.42 
9.76 
7.40 


17.39 
19.31 
18.39 


75.02 
70.80 
74.03 


3H 

4 

3>i 




3.152 
3.152 
3.137 


313 
250 
450 


2125 
1875 
2250 


O.K. 
O.K. 
O.K. 



Digitized by 



Google 



514 Bayles on Inspection and Testing of Concrete. 



1 


rABLE 11 

Percei 
Retained 
100 Mesh. 

6.47 
9.50 
8.96 


.— Tests 


of Cemc 

sment 

Through 
200Me8h. 


nt and 

Sett 

Initial. 
Hrs. 

4 
3^ 

4H 


1 :2\ 

ing. 

Final. 
Hrs. 


lortar. 

Specific 
Gravity 

3.137 
3.168 
3.152 


{Conlinued) 

Crushing 
Strength 

of 4-in. cubes. 

in lb. per sq. in. 




Car 

Number. 


atage of d 
on Sieve. 
200 Mesh. 


Boiling 
5 hours. 




24 hrs. 

300 
300 
219 


7 days. 




1 

R 2 

3 


18.33 
16.24 
16.88 


75.02 
74.08 
74.05 


F 


2250 
1938 
1875 


O.K. 
O.K. 
0. K. 


[ 1 

S 2 

, 3 


8.66 
9.34 
8.12 


17.20 
16.70 
16.91 


74.02 
73.81 
74.83 


3' '2 

4'a 


4H 

4*4 

5' 2 


3.168 
3.168 
3.152 


300 
225 
219 


1625 
1750 
2000 


O.K. 
0. K. 
O.K. 


1 

T 2 

3 


6.79 
7.04 

8.48 


18.62 
17.47 
16.60 


74.41 
75.30 
74.81 


5'4 

6>i 
5H 


6»4 

6'i 
6>i 


3.137 
3.122 
3.168 


125 
100 
113 


1875 
1625 
1875 


O.K. 
O.K. 
O.K. 


1 

U { 2 

3 


8.43 
5.87 
7.70 


17.05 
18.92 
17.84 


74.37 
74.97 
74.29 


6 

5»4 

5^4 


7*4 

8 

74 


3.152 
3.152 
3.168 


113 
106 
88 


1938 
2188 
1688 


O.K. 
0. K. 
O.K. 


V 1 2 

l3 


6.88 
7.13 
6.84 


17.92 
18.22 
18.39 


74.94 
74.47 
74.51 


4j; 

3*4 


6 
5 
5 


3.122 
3.152 
3.152 


475 
438 
438 


1963 
1875 
2000 


O.K. 
O.K. 
O.K. 


W 1 2 
13 


7.31 

8.48 
8.82 


18.06 
14.70 
16.93 


74.48 
76.59 
74.97 


3M 

3 

3 


5 

4H 
4h 


3.122 
3.162 
3.152 


419 
425 
288 


2125 
2313 
2063 


O.K. 
0. K. 
O.K. 


Mi 


9.03 

7.14 
8.23 


16.17 
17.09 
16.18 


74.62 
75.49 
75.44 


4 4 

4H 
4H 


6 


3.137 
3.168 
3.168 


213 
181 
150 


1938 
1875 
1688 


O.K. 
O.K. 
O.K. 


Y 11 

13 


7.13 
8.08 
8.32 


17.95 
16.69 
18.17 


74.79 
75.06 
73.29 


4% 
4h 
4H 


6 


3.152 
3.137 
3.122 


169 1 1750 
325 1 2188 
294 2000 


O.K. 
O.K. 
O.K. 


f 1 

Z 2 

3 


7.93 
8.12 
6.92 


18.31 
18.22 
18.05 


73.49 
73.43 

74.88 


4H 
4».i 
4,4 


61 2 


3.137 
3.152 
3.137 


263 
250 
288 


1625 
1437 
1563 


O.K. 
0. K. 
O.K. 


f 1 

AA 2 

3 


8.72 
7.88 
9.47 


18.39 
18.27 
17.62 


72.74 
73.60 
72.79 


4H 


6K 
64 
6»i 


3.122 
3.122 
3.152 


288 
350 
350 


1563 
1688 
1437 


O.K. 
O.K. 
O.K. 


f 1 

BB 2 

3 


11.40 
9.13 
8.72 


18.70 
18.99 
18.04 


69.78 
71.64 
73.07 


k 


64 


3.152 
3.137 
3.137 


363 
344 
213 


1437 
1500 
1437 


O.K. 
O.K. 
O.K. 


1 

CC 2 

3 


6.93 
8.31 
8.43 


19.09 
19.01 
18.77 


73.84 
72.43 
72.63 


4H 
4'i 
434 




3.137 
3.107 
3.122 


219 1813« 
238 I 2«>3« 
282 1 1875> 


O.K. 
O.K. 
O.K. 


r 1 

DD 2 
3 


8.70 
7.93 
8.41 


17.34 
18.04 
18.57 


73.84 
73.71 
72.83 


4l'l 
4H 
4H 


5H 
54 


3.137 
3.137 
3.122 


338 ' 2063« 
344 20002 
350 1 2375» 


O.K. 
O.K. 
O.K. 


EE I2 
l3 


12.24 
9.17 
8.41 


16.84 
17.32 
18.69 


70.79 
73.24 
72.74 


4H 

4U 
6U' 


6 


3.152 
3.152 
3.137 


275 2063 
344 1 2188 
313 2250 


O.K. 
O.K. 
O.K. 


1 

FF 2 

3 


9.96 
9.22 
8.43 


17.58 
17.93 
18.34 


72.29 
72.61 
73.02 


6 
64 


7M 


3.122 
3.137 
3.137 


319 2000 
244 1875 
250 , ^000 


O.K. 
O.K. 
O.K. 


f 1 

GG 2 

3 


10.92 
13.30 
11.76 


18.63 
18.95 
18.72 


70.12 
67.50 
69.06 


6 

5>4 

5h 


7H 


3.184 
3.122 
3.122 


175 1 1563 
275 1375 
100 1 1563 


O.K. 
O.K. 
O.K. 


HH J2 
3 


4.75 
5.28 
5.15 


19.31 
24.87 
21.62 


75.9 

69.72 

73.12 


2 
214 

2U 


3)i 


3.12 

3.2 

3.152 


938 
1063 
1063 


2500 
2000 
3063 


O.K. 
O.K. 
O.K. 


> Eight days 


. 



















Digitized by 



Google 



Batles on Inspection and Testing of Concrete. 515 



Car 


II 


f 1 
2 
3 


JJ 


li 


KK 


f 1 
2 
3 


LL 


f 1 

2 

,3 


MM 


f 1 
2 
3 


NN 


f 1 
2 
3 


GO 


f 1 
2 
3 


PP 


f 1 
2 
3 


QQ 


f 1 

2 

.3 


RR 


f 1 
2 
3 


SS 


f 1 
2 
3 


TT 


f 1 
2 
3 


UU 


1 
2 
3 


VV 


f 1 
2 
3 



Table II, — ^Tests of Cement and 1 : 2 Mortar. {Continued) 

Setting. 



Percentage of Cement 



Retained on Sieve. 
100 Mesh. 200 Mesh. 



WW 



XX 



YY 



(1 
II 



5.36 
7.82 
6.47 

4.75 
5.46 
5.83 

5.00 
5.13 
5.09 

6.64 
6.23 
6.06 

5.18 
6.05 
6.13 

5.97 
6.13 
5.99 

7.10 
5.94 
6.12 

5.68 
6.32 
6.29 

6.13 
6.07 
5.84 

7.02 
6.87 
6.92 

4.53 
4.60 

4.87 

7.03 
6.61 
6.72 

5.87 
5.69 
6.12 

6.35 
5.86 
5.92 

5.88 
7.44 
6.78 

5.97 
6.31 
6.41 

6.74 
6.37 
7.01 



22.28 
26.06 
24.63 

25.93 
26.97 
24.66 

22.72 
21.87 
22.36 

23.01 
23.18 
21.61 

23.64 
24.16 
24.42 

23.19 
22.94 
21.82 

21.78 
22.31 
22.14 

25.04 
22.76 
24.07 

22.77 
22.69 
21.57 

23.62 
20.93 
22.17 

22.05 
20.66 
21.33 

23.65 

23.115 

22.95 

22.06 
23.12 
22.36 

24.13 
23.70 
23.15 

19.26 
23.01 
23.55 

23.82 
23.77 
21.95 

20.65 
22.06 
20.85 



Through 
200Me8h. 



72.19 
65.96 
68.79 

69.16 
67.44 
69.39 

72.16 
72.89 
72.47 

70.24 
70.47 
72.20 

71.06 
69.63 
69.30 

70.71 
70.85 
72.11 

71.01 
71.61 
71.60 

69.15 
70.79 
69.51 

71.01 
71.20 
72.28 

69.20 
72.08 
70.80 

73.05 
74.45 
73.61 

69.20 
69.80 
70.13 

71.91 
71.03 
71.40 

68.70 
70.21 
70.49 

74.45 
69.53 
69.41 

70.03 
69.61 
71.39 

72.40 
71.29 
71.98 



2 



2 
2 

2 

2 

2 
2 

IH 
IH 

2V^ 
2>i 

2 

2M 

IH 

1 

IM 

2H 

2?4 
2H 

2H 

2>4 

3 

2H 

2% 

2H 
2H 
2H 



Initial.! Final 
Hrs. I Hrs. 



2M 
2H 
2 

24 

2H 

2H 

v^ 

2H. 

2 
2 
2 

2 

IH 



Crushing 
Strength | 
Specific of 4-in. cubes, Boiling 
Gravity! in lb. per sq. in. 5 hours. 




O. K. 
O.K. 
O.K. 

O.K. 
O.K. 
O.K. 

O.K. 
O.K. 
O.K. 



Pigitized by 



Google 



516 Bayles on Inspection and Testing of Concrete. 

Tablk II. — Tests of Cement anil 1 : 2 Mortar. (CorUinued) 



Car 

Number. 


PercentaRp of Comont 
Retained on Sieve. ^^^^^^^ 
lOOMeah. 2(M)MeHh. 2(K)Me.sh. 

5.91 18.22 75.05 
6.(19 15.42 ; 77.75 
6.21 15.31 78.14 


Sett 
Initial. 

h™. 

IH 


Final. 
HrH. 

3 
3 

2»4 


1 Speeific 
Gravity 

3.168 
3. IS.? 
3.122 


Crushing 

Strength 

of 4-in. eubes. 

in lb. persq. in. 


Boiling 
5 hours. 




24 hrs. 

938 

688 
625 


7 day b. 




r 1 

ZZ 2 
.3 


2438 
2250 
2375 


O.K. 

. O. K. 

O.K. 


f 1 
AB < 2 

is 


8.06 
5 52 
7.02 


17 93 
17 83 
20.05 


73.60 
75.33 
72.70 


2 


3 
2V 

2U 


3.152 
3.152 
3.152 

1 


563 
375 
375 


1088 
2000 
1750 


O.K. 
O. K. 
O.K. 


f 1 

AC \ 2 

I 3 


9.20 
7.21 
0.92 


20 

19.03 

17.95 


70.66 
73.67 
75 


2 
2 
2 


F 


1 3.162 
' 3.152 
, 3.137 


1250 
1000 
1250 


2375 
2250 
2875 


O.K. 
O.K. 
O.K. 


f 1 

AD 2 

> 3 


8.35 

7 

7.47 


19.35 
19.80 
21.02 


72.12 
73.21 
71.51 


3 4 
33 a 


4 


! 3.137 
3.137 
3.137 


1125 
1188 
1063 


2125 
2063 
2438 


O.K. 
O.K. 
O. K. 


f 1 
AE 2 

13 


6.84 

7 

6.30 


19.35 

20 

19.27 


73.60 
72.89 
74.35 


3 


VA 


3.137 

3.137 

1 3.137 


813^ 
938 
938 


2375' 
2125 
2250 


O.K. 
O.K. 
O.K. 


AF 1 2 

13 


6.44 
6.55 
6.9 


18.45 
18.5<i 
18.8 


75 

74.83 

74.07 


2h 


3 
3 


3.152 
3.152 
3.137 


813 
600 
438 


1750 
2188 
2313 


O.K. 
O.K. 
O.K. 


f 1 

AG 2 

3 


6.7 

6.12 

7.13 


18.71 
19.00 
19.28 


74.55 
74.61 
73.40 


3'4 

3 
3 


4 
4 
4 


3.137 

3.137 

j 3.152 


688 
625 
781 


1750 
1688 
1750 


O.K. 
O. K. 
O.K. 


f 1 

AH 2 

3 


6.23 
5.47 
7.11 


19.66 
19.80 
18.80 


74.05 
74.58 
73.87 


2»4 
2«4 

3 


4M 
4 

4M 


3.152 

1 3.137 

3.137 


656 
875 

781 


2250 
2250 
2438 


O.K. 
O.K. 
O.K. 


AI 1 2 
3 


7.92 
8.46 
8.17 


20.08 
21.19 
20. .58 


71.80 

70 

70.71 


3 

2H 

3'.. 


4li 

1'^ 


3.137 
3.137 
3.137 


656 
719 
663 


2125 
2063 
2250 


O.K. 
O.K. 
O.K. 


AJ 1 2 

13 


7 

6.77 

5.76 


20.47 
21.21 
19.58 


72.32 
71.80 
74.53 


3 

2h 


1« 


3.137 
3.137 
3.137 


531 
503 
594 


2000 
2063 
2188 


O.K. 
O.K. 
O.K. 


1 

AK 2 

3 


6.09 
6.16 
6.04 


19.66 
19.98 
19.82 


74 

73.74 

74.09 


2H 
2H 

2H 


3 
3 
3 


3.152 

. 3.137 

3.152 


500 
375 
594 


1938 
1625 
2250 


O.K. 
O.K. 
O.K. 


f 1 
AL { 2 


5 
5 
5.25 


19.45 
20.21 
20.19 


75.35 
74.70 
74.24 


2 


3 
3j.i 


: 3.168 

3.137 

, 3.152 


531 
600 
406 


2188 
2313 
168S 


O.K. 
O. K. 
O.K. 


f 1 

AM < 2 

l3 


6 

0.82 

7.37 


19.97 
18.20 
18.42 


73.63 

72 

74.03 


2 


2\i 
2'.i 
2U 


3.122 
3.137 
3.137 


638 
663 
513 


2188 
2188 
1813 


O.K. 
O.K. 
O.K. 


f 1 

AN 2 

3 


6.82 
6.53 
6 41 


20 

19.75 

19.90 


72.80 
73.45 
73.45 


f' 


2'i 

2'2 

3M 


3.152 

3.152 

1 3.137 


388 
531 
375 


1750 
2000 
1813 


O.K. 
O.K. 
O.K. 


f I 

AG 2 

, 3 


5.95 
6.25 
6.42 


19.40 
19.05 
20.11 


74.50 
74.60 
73.35 


1?4 


3 
3 

2?4 


3.122 

3.137 

1 3.137 


475 

469 
481 


2750 
2250 
1813 


O.K. 
O.K. 
O.K. 



» Sand contained 58 per cent loam. 



Digitized by 



Google 



Bayles on Inspection and Testing op Concrete. 517 



Table III. — Crushing Strength of 4-in. Concrete Cubes. 





Strength in 


Str'gth 


Arc 

in 

days. 

336 


1 


Strength in 


Str'gth 


Age 

in 

days. 


Date made. 


lb. 
24 hr. 

375 


per sq. in. 
7dy. ,28dy. 

1250' 2425 


in 
lb. per 
sq. in. 

3688 


Date made. 
_ 
11-17-10 


lb. 
24 hr. 

250 


per sq. m. 

7 dy. !28dy. 

■ — 1 

2125* 35006 


in 
lb. per 
sq. in. 


8-18-10 


4688 


240 


8-18-10 


375 .... 2475 


3566 


336 


11-18-10 


188 


1750 2750^ 


5000 


241 


8-19-10 


500 1375 2500 


4088 


335 


11-19-10 


313 


1250 4250 


6260 


252 


8-19-10 


375 


19001 2188 


3750 


335 


11-21-10 


250 


1500 3375 


4375 


238 


8-20-10 


500 


1125 2875 


4250 


334 


11-29-10 


275 


1875 1 3500 


4750 


230 


8-22-10 


500 


1250 2225 


4500 


331 


1 11-30-10 


188 


1250 3726 


4875 


226 


8-25-10 


625 


2000 , 2875 


4375 


329 


12- 1-10 


200 


1875 ' 3125 


4375 


224 


8-25-10 


500 


2000 , 3938 


5626 


328 


! 12- 1-10 


200 


1875 4938 


6750 


239 


8-26^10 


375 


1350 ' 2875 


4375 


327 


12- 2-10 


200 


1875 : 3188 


4625 


223 


8-26-10 


375 


1750 1 2875 


4250 


322 


12- 5-10 


263 


1250 ; 2750 


4875 


225 


8-30-10 


375 


875 3000 


4376 


323 


12- 8-10 


188 1500 1 3188 


5000 


221 


8-31-10 


750 ' 2275 i 3500 


4625 


321 


12-15-10 


200 1 2250 3438 


4750 


214 


9- 6-10 


875 , 1975 


3500 


5000 


316 


12-19-10 


200 1 2375*' 4438 






&- 7-10 


438 


1688 


2500 


3875 


316 


12-20-10 


350 ; 1900 2938 


4125 


izi 2 ■ 


9- 8-10 


563 


1500 


3125 


4375 


313 


12-27-10 


250 1625 3250 


4875 


205 


9-13-10 


563 1375 3125 


4813 


308 


12-28-10 


313 2375 402.'^« 


6125 


203 


9-14-10 


500 1875 3375 


4626 


307 


12-28-10 


2.W 


1875 1 418^" 


4688 


201 


9-14-10 


438 1625 3250 


5063 


307 


12-29-10 


438 


3026 ! 5563 


6125 


200 


9-15-10 


750 


1875 4375 


6250 


307 


1 12-30-10 


275 


2750 5500 


5500 


200 


9-15-10 


625 


2625 3813 


4813 


306 


1 1- 9-n 


188 


1600 3063 


4625 


189 


9-16-10 


563 


2250 3750 


5563 


306 


1-11-11 


150 1 1400 , 2938 


3125 


190 


9-16-10 


600 2188 3750 


5000 


305 


1 1-12-11 


400 3150 1 4000 


7500 


197 


9-17-10 


375 1500 3125 


5250 


304 


1 1-12-11 


250 


1688 3750 


5000 


197 


9-19-10 


500 1938 2500 


5250 


303 


; 1-14-11 


250 


1625 1 3250 


4875 


195 


9-20-10 


688 , 2250 3500 


4500 


302 


1-19-11 


225 


1438 3125 


4875 


190 


9-20-10 


813 2250 3938 


4500 


301 


1-19-11 


313 


1688 3125 


4500 


190 


9-21-10 


725 1875 3500 


4938 


300 


1 1-21-11 


150 


1313 2625 


3625 


188 


9-21-10 


563 


2125 ! 37 .fin 


4760 


301 


1-23-1 1 


125 


1250 2025 






9-22-10 


563 


2500 


3600 


5875 


294 


: 1-23-11 


150 


1313 i 2500 


4250 


im ' 


9-22-10 


500 


1875 


2938 


4375 


299 


1-24-11 


338 


1625 2625 


3750 


185 


9-23-10 


750 


2250 


2688 


5376 


298 


1-26-11 


262 


1250 2375 


4125 


168 


9-23-10 


438 
500 


2000 
2250 


3125 
3125 






1-26-11 
1 3- 8-11 


438» 
188 


1250 2250 
1625 j 3250 


3500 
4750 


183 


9-24-10 


' 4688 ' 


299 


124 


9-26-10 


1000 


1750 


2500 


4688 


297 


! 3- 8-11 


200 


1750 1 287.-) 


4376 


124 


9-27-10 


600 1875 


3500 


4188 


294 


3- 9-11 


175 


1938 2875 






9-27-10 


875 1875 


3125 


4626 


296 


3- 9-11 


225 


1975 , 3938 


■ '4875 ■ 


124 


9-28-10 


563 


1575 


2750 


3750 


294 


3-10-11 


188 ■ 1876 


3750 


5813 


132 


9-29-10 


500 


2063 


3000 


4260 


292 


3-10-11 


263 2375 


5625 


5625 


121 


9-30-10 


563 


1875 


2875 


4125 


293 


1 3-11-11 


250 


2500 


5000 


7188 


123 


10- 3-10 


2.'>0 


1250 


2000 


3125 


288 


1 3-13-11 


313 


2188 


4813 


65(K) 


121 


10- 4-10 


500 


2125 


3125 


4938 


289 


3-14-11 


2.50 2750 


4625 


6750 


128 


lO-U-lO 


500 


2000 


2750 


3500 


279 


3-14-11 


188 2125,4188 


6250 


118 


10-13-10 


375 


1875 


2813 


4313 


277 


3-18-11 


188 


2250 5500 


6125 


116 


10-14-10 


500 


1750 


2750 


4250 


276 


! 3 20-11 


350 


2500 ! 4750 


7125 


114 


10-15-10 


375 


1875 


2250 


3563 


273 


1 3-20-11 


288 2500 4375 


5625 


114 


10-17-10 


500 2250 


2938 


4813 


274 


3-21-11 


375 3(X)0 i 4750 


6625 


113 


10-19-10 


500 


1250 2000 


4375 


268 


3-21-11 


275 2500 5000 


6875 


112 


10-19-10 


688 


1875 2750 


4750 


269 


3-23-11 


313 2375 1 5000 


5750 


110 


10-25-10 


313 


1250 ! 2250 


3625 


261 


3-30-11 


250 


2.500 1 4625 


6750 


103 


10-25-10 


313 


1375 ; 2260 


3126 


262 


3-31-11 


150 


2000 1 4875 


6260 


101 


10-26-10 


375 


2063 2688 


4625 


266 


4- 1-11 


313 


2563 5250 


7600 


102 


10-28-10 


350 


1688 2938 


4876 


264 


4- 3-11 


250 


2500 . 4625 


6126 


100 


1O-31-10 


250 


1563 2000 


4375 


256 


4- 3-11 


250 


2625 5000 


5125 


100 


10-31-10 


263 


1250 2750 


4750 


256 


4- 5-11 


250 


2125 4625 


6000 


98 


10-31-10 


313 
313 


2125 
1750 


3125 
2750 






4- 6-11 
4-13-11 


375 
375 


2.563 
3125 


4500 
4625 


5125 
6625 


96 


11- 1-10 


4938 


i260 


89 


11- 2-10 


500 


1250 


2875 


6875 


251 


4-13-11 


438 


3125 


4750 


6126 


88 


11- 2-10 


725 


2125 


2875 


4003 


2.58 


4-14 11 


375 


2750 4625 


5625 


89 


11- 7-10 


350 


2438 


4600 


6500 


250 


4-14-11 


43810 


3375-^ 5625 


4750 


89 


11- 7-10 


313 2250 4125 


6063 


255 


4-15-11 


563" 


2250 3500 


4375 


34 


11- 9-10 


313 2250 3125 


5500 


246 


4-18-11 


375 


1625 3600 


4625 


85 


11-10-10 


350 


1625 3125 


4750 


247 


4-22-11 


260 


1625 1 3375 


4063 


79 


11-10-10 


375 


1500 


3000 


6000 


246 


4-25-11 


375 


1875 3750 


4750 


78 


11-11-10 


250 


2000 


3125 


4125 


245 


4-27-11 


438 


2375 ' 4375 


6125 


76 


11-11-10 


188 


2000 


3125 


5125 


2.>0 


4-28-11 


750 2500 5000 


6875 


74 


11-12-10 


250 


2125 


27505 


5875 


248 


5 8-11 


500 2250 4125 


5000 


64 


11-14-10 


188 


1375 


2375 


4250 


243 


5-12-11 


626 1875 4375 


5000 


60 


11-15-10 


225 


1750 


3813 


5875 


246 


5-18-11 


625 2250 4.375 


5375 


54 


11-16-10 


188 


1250 


2500 


4188 


241 


5-20-11 


875 1750 3625 


4063 


52 


11-16-10 


313 


2125* 


3125 






1 


: 














1 14 days 


<12 


days. 




» 27 daj 


^8. W 


21 hours 




» 30 hour 


B, •8 c 


lays. 




•29 da] 


fS, » 


53 hours 




> 30 days 


•70 


days 




» 43 hoi 


ITS. 




r~>^ 


















' 


Dig 


tized by 


Gc 



518 Baylbs on Inspection and Testing of Concrete. 



Table IV. — Crushing Strength of 4-in. Concrete Cubes. 



Date made. 



7-15-11 
7-17-11 
7-lS-n 
7-18 11 
7-19-U 
7-24-H 
7-25~n 
7-25-11 
7-afl-ll 
7-2fl-ll 
7-31-n 
8- J-H 
ft- J-il 
8- 2-11 
8- 2-il 
S- 3-11 

8- 3-n 

8- 4-11 

8- 4-11 
8^- 8-11 

8- li-ll 
S'l&-]1 
8-11-11 
S-ll-ll 
8-H-ll 
SI- II 11 
8-lt>-ll 
8-18-11 
8-25-11 
8-28-11 
8-28-11 

9- 1-11 
9- 1-11 
9- 2-11 
9- 5-11 
9- 7-11 
9- 8-11 
9-12-11 
9-13-11 
9-14-11 
9-16-11 
9-18-11 
9-18-11 
9-19-11 
9-20-11 
9-20-1 1 
9 21-11 
9-22-11 



IK p{!r b(}k In, 
!J4hr 7dy. l2Rdy 



lOOOJi; 
50;t 
08^ 

75LJ 

esH 
7at3 

5m 
m:i 

^35 I 
62A I 

essM 

BOtI 
625 

as8 

525 

sea 
Si:i 
5fla 

635 

m>» 
56:* 
fi2V 

4;^9 I 

563 ! 

563 

625 

563 

563 

375 
1250* 

625 

500 

375 

563 

375 

500 

9387 

325 

450 

487 

413 

438 

438 
1500» 



1500 |2lS8 
1750 1^875 









1750 3I8S 

1500 2rm 

I37f» 2375 

11125 2S75 

1375'* 2813 

ViJii ims 

rim '2ms 

12505 2im 
15(XI^ 22r»o 
i:j7/i '.utm^ 
m:i 2i:iji 

125C] 2750 
llStH if500 
150tl 287,'i 

Umii 2125 

iOf>:i iihH 
H7:j ntiK 

li;S5* 2260 
1188 2.563 
1313 2313 
1125 2188 
1000 ,1813 
1375 '3188 
1250 2(525 
1188 2813 
1250 12500 
1125 ,2625 
875 2063 
1063 1 2688 
1063« '2563 
1063 J2438 
1375 2563 
1000 12250 
1188 2250 
1188 2375 
1250* 1938 
1500» 2250 
15(K) 2563 
1688 12875 



in I Age 
lb. per , ,»° 
aq. in. I days. 



I i 

I 116» I 

I 177 I 

, 115» I 

! 176 

I 109» I 

' 170 I 

I 170 

1 107» 

I 169 I 
102>,, 



Date made. 



Strength in 
lb. per aq. in 



,24 hr. 7 dy. '28dy.| aq. in. 



3625 
3950 
4875 
4519 
5000 
4031 
5063 
4000 
4313 
4688 
4269 



I 2938 

3813 

1 3875 

4069 

4519 

5125 

4169 

I 4288 

I 5625 

, 3625 

3637 

! 3875 

3719 

I .3875 

I 3438 

, 4063 

I 3788 

3750 

3138 

4813 

I 4.375 

3938 

4625 

I 3688 

3875 

5000 

4125 

4375 

4.500 

39,38 

I 3813 

I 42.')0 

34.38 

4063 

4250 

4875 



38 1 
1001 
100» 

991 

99» 
*98» 
<98« I 

94» 
155 
154 

91» 
1.53 

881 I 
1.50 

861 I 
146 

771 

1.36 

136 

1.32 

132 

131 

128 

126 

125 

121 

120 

119 

117 

115 I 

115 

114 i 

113 

113 

112 

111 



^ Broken at Columbia University. 

* 6 days, 7 hours. 

* 5 day^, 2 hours. 

* Contained 6 per cent loam, extra 10 per 
oent cement used. 

* 5 days. 

* 3 days. 

' 53 hours. 



9-25-11 
9-25-11 
9-27-11 
9-28-11 
9-28-11 
10- 2-11 
10- 2-11 
lO- 3-11 
10- 3-11 
10- 5-11 

10- 6-11 
10-10-11 
10-10-11 
10-12-11 
10-12-11 
10-16-11 
10-17-11 
10-17-11 
10-19-11 
10-20-11 
10-23-11 
10-24-11 
10-25-11 
10-26-11 
10-30-11 
10-31-11 

11- 1-11 
11- 2-11 
11- 2-11 
11- 3-11 
11- 4-11 

11- 6-11 
11-11-11 
11-1.3-11 
11-13-11 
11-14-11 
11-14-11 
11-17-11 
11-20 11 
11-20-11 
11-21 11 
11-22-11 
11-24-11 
11-28-11 
11 2t)-ll 
11-29-11 

12- 1-11 
12- 1-11 



719 

6.50 

688 

388 

344 

263 

294 

469 

438 

419 

4.38 

438 

388 
i 375 
I 375 

406 

375 

450 

325 

294 I 

300 

.344 

263 1 

263 

263 

300 . 

238 I 

113"> 

lOO" 
I ..", 

250" 

400»i 
I 450"; 
I 94 ' 

169 

2.50 1 

275 I 

125 

225 I 

194 • 

331 

100 

175 

200 
•'188>a 
"156"! 
, 219 

156 




2250 
2313 
24.38 
2000 
2188 
2313 



I 



1375 
1313 
1625* 
1000 
875 
9.38 
875 ;2063 
1188 2563 
1125« 12063 
1000 2250 
1063 |2813 
1063 2188« 
1188«;18757| 
1063 I2OOO 
875 11938 I 
1000 2063 I 
969 12313 
1.500« ,2438 
788 '2125 
825 11875 
788 I2I88 ' 
919 2063 I 
800» 11813 ' 
675 17,50 I 
888 12875 | 



969 12563 
600t 1 2125 
469 12063 I 
400 i2188 I 
6.56»|l625 1 
600 12438 , 
781 ;i625 I 
663 I2OOO 
4,38 11625 
588 19.38 I 



713 

788 
875> 
800 
888 
1125 



1625 

19.38 

1813 I 

1625 

1813 

2000 



I 



813 '2000 
7.50 13063 
1000 1938 
688 ,1688 
875u!l813 
719 11,563 
869 ,1813 



4500 
4000 
4625 
3625 
3625 
3813 
3625 
4375 
3688 
3625 
4250 
4250 
3563 
3688 
3875 
4125 
4313 
4188 
4063 
3375 
3438 
3875 
3688 
3375 
5125 
5063 
3625 
3875 
3875 
3625 
3438 
3438 
2875 
3250 
3000 
3000 
3250 
3125 
2813 
3125 
3125 
.3000 
4438 
3063 
2500 
2188 
2625 
3250 



Age 

in 
dasrs. 



108 
108 
106 
105 
105 
101 
101 
100 
100 
98 
97 
93 
93 
91 
91 
87 



84 
83 
80 
79 
78 
77 
73 
72 
71 
70 
70 
69 
68 
66 
61 
59 
59 
58 
58 
55 
52 
52 
51 
50 
48 
44 
43 
43 
41 
41 



•6 days. 

•29 days. 
" Cube not set. 

11 Cube broke when removed from forms. 
" 48 hours. 
" 9 days. 
»« 4 days. 
» 7 days. 



Digitized by 



Google 



Bayles on Inspection and TESTnjra op Concrete. 519 

All sand must be washed, clean, sharp, silicious and free from injm'ious 
matter. 

Ail stone shall be clean crushed trap rock, free from dust and to contain 
no particle that will not pass through a f-in. ring. 

There was little diflSculty in getting satisfactory materials. 
Standard brands of Portland cement were used. The sand was 
what is known in New York as "Cow Bay," and was satisfactory 
with a few exceptions when, for some reason, it was impossible to 
get washed sand from the usual source and other sands were tried 
temporarily to take its place in order not to delay the work. 
Some of this sand was satisfactory, some was condemned out- 
right and one scow load was accepted on condition that lO per 
cent additional cement be used, — a doubtful expedient. The stone 
was a Hudson Palisades trap, commercial J-in. size and proved 
generally acceptable. 

operation of plant. 

The mixer, a batch mixer, was set in a pit so that the top of 
the receiving hopper was a little above the floor of the wharf. 
The sand and stone were measured approximately in wheel- 
barrows or buggies and the cement by the bag, reckoning a bag of 
cement at 0.95 cu. ft. The aggregates were all put into the 
hopper, the trap opened and the batch pushed into the mixer. 
As soon as all the aggregates were clear of the hopper the water 
was added, the mixer revolved 12 times and dumped, while 
another batch was being prepared. 

At first the water was dipped with a pail from a barrel on the 
platform and splashed through the hopper into the mixer. This 
kept the sides of the hopper always wet, preventing the easy dis- 
charge of the dry aggregates and also delayed the work by using 
the hopper as a funnel while another batch should have been in 
course of preparation. These were matters primarily for the 
contractor's attention, but this method produced concrete of such 
varying degrees of consistency that the plan shown in Fig. 2 was 
suggested to the contractor and immediately installed. By this 
method the water flows constantly or nearly so from a f-in. pipe 
into the upper barrel; from this barrel a l|-in. pipe with quick 
opening lever gate valve leads to the lower barrel. From the 
bottom of the lower barrel a 3-in. pipe, also fitted with quick 



Digitized by 



Google 



520 Bayles on Inspection and Testing of Concrete. 

opening valve, leads into the mixer under the hopper trap. A 
number of small holes were bored in the side of the lower barrel. 
By trial the right amount of water for a batch of concrete was 
determined and all the holes below the line corresponding to that 
amount plugged. Only the inspector at the mixer was allowed to 
vary this amount. After all the dry material had run into the 
mixer, which it did more readily since the hopper was always dry, 
the hopper trap was closed and the 3-in. valve opened. Only a 
few seconds was required to empty the barrel, when the 3-in. 
valve was closed, the IJ-in. valve opened and the water allowed 




FIG. 2. — ARRANGEMENT OF WATER SUPPLY FOR MIXER. 



to run until it had filled the lower barrel to the holes in the side> 
which served both to call the attendant's attention and to drain 
off surplus water. By this method a satisfactory mixture was 
achieved and the time of mixing reduced by at least a third. 

The concrete as thus produced was of such consistency that 
it was wheeled to the work — at most 300 ft. — ^without any 
apparent segregation of material and when the buggy was tipped 
the whole mass contained slipped, rather than poured, into the 
forms, leaving the buggy clean. The diflference between concrete 
that is just wet enough and concrete that is too wet is very slight 



Digitized by 



Google 



Bayles on Inspection and Testing of Concrete. 521 

and only experience can determine what is right and constant 
care maintain it. 

The batch was dumped from the mixer into a bucket and 
immediately hoisted and dumped into a hopper above the floor 
where it was to be used. From this hopper it was drawn off into 
concrete buggies and wheeled over plank runways to the work. 
The buggies had a capacity of 6 cu. ft., but in practice each man 
took about two thirds that much. 

The columns were poured first, preferably 24 hours before the 
slab. The concrete in the columns was puddled with long 
wooden bars about 2 x 3 in., which were also used to hold the 
reinforcement in place as the concrete was being poured. Every 
buggyf ul of concrete was thoroughly puddled before the next was 
added. 

In concreting the floors it was the general practice to fill the 
beams and girders first, allowing the concrete to spread on to the 
adjoining slab rather than from the slab into the beams and 
girders. The floor concrete was puddled with narrow wooden- 
handled iron spades. Care was taken that no concrete stood 
long enough, to form a joint before new concrete was added. 
When construction joints were necessary they were made at the 
middle of the span and were vertical. Before new concrete was 
added the surface of the joint was roughened and washed with 
neat cement grout. The concreting gang was followed closely 
by the finishers who screeded the slab to the required thickness 
and troweled the surface to a smooth, hard finish. 

While the inspection was close and strict the contractors 
co-operated heartily and the work was done without friction. 



Digitized by 



Google 



COMPARATIVE TESTS OF THE STRENGTH OF CON- 
CRETE IN THE LABORATORY AND IN THE 
FIELD. 

By Rudolph J. Wig.* 

It is common practice at the present time to define a desired 
strength of concrete by stating that it shall be made of good 
quality materials in certain given proportions. Assuming the 
use of average concrete material a 1:2:4 proportion mixture is 
considered as developing an ultimate compressive strength of 
2000 to 3000 lb. per sq. in. in 28 days, 3000 to 3500 lb. per sq. 
in. in 13 weeks, etc. These values are based upon the results of 
many thousands of laboratory tests which imdoubtedly represent 
the true values which can and should be obtained with good 
materials if properly used in the field. Much attention is given 
to careful selection and the testing of the materials entering into 
the concrete and usually in case of the failure of the. concrete, the 
cause of failure is attributed to the cement or poor aggregate. 

Several years ago an investigation was made in the Struc- 
tural Materials Testing Laboratories of the United States Geolog- 
ical Survey at St. Ix)uis, Mo., to determine the influence of dif- 
ference in workmanship upon the strength of reinforced concrete 
in flexure. In connection with this investigation compression 
tests also were made on the plain concrete molded into 8 x 16 in. 
cylindrical t«st pieces. The work of three prominent contractors 
of the city of St. Louis was compared and the results of the tests 
are of much interest. 

The materials, cement, sand and aggregate, were carefully 
weighed and proportioned under the supervision of the laboratory 
force which ensured each contractor receiving exactly the same 
amount of material so that the variation in the results of the 
tests of the concrete prepared by the contractors was due entirely 
to a difference in the method of mixing and handling the mate- 
rials. 

In the beam tests for hand-mixed concrete in the proportion 

* Assistant Engineer, Bureau of Standards. Washington, D. C. 

(522) 



Digitized by 



Google 



Wig on Tests of Strength op Concrete. 523 

1 part cement to 3 parts sand to 6 parts stone when 4 weeks old, 
there is a difference of over 200 per cent in the maximum load 
carried by beams made by different contractors with the same 
materials. The total maximum load varied from 4850 to 16,310 
lb. For the machine-mixed concrete the variation is about 
90 per cent ranging from 8160 to 14,500 lb. total maximum load. 
The quality of the concrete can perhaps be better observed from 
the compression tests on the cylinders which were molded from 
the same mixtures as the beams. The placing of the material in 
the cylinder molds, however, was done by the laboratory force 
so that the difference in the results of these tests may be entirely 
attributed to the difference in the method and the thoroughness 
of the mixing and the difference in the consistency of the concrete 
as mixed, as each contractor was instructed to prepare a concrete 
such as he used in practice and to exercise no more care in its 
preparation than he would under working conditions. 

This privilege of varying the quantity of water perhaps had 
the greatest effect upon the strength although it was also mate- 
rially influenced by the method of mixing as is shown in the 
results where two of the contractors used the same percentage of 
water in mixing. 

A summary of the results is as follows: 

1:3:6 Gravel Concrete, Machine Mixed. (Average of 3 test pieces.) 
Compressive Strength in lb. i)er sq. in. 

4 weeks. 13 weeks. 

Max. Min. Ave. Max. Min. Ave. 

Company A 998 650 844 1470 1180 1363 

B 577 513 549 1020 878 944 

C 725 623 679 1243 956 1138 

Laboratory 1074 961 1032 1653 1540 1611 

There is an approximate difference of 100 per cent in individual values 
which range from 513 to 1074 lb. per sq. in. at 4 weeks and from 878 to 1653 
lb. per sq. in. at 13 weeks. 

1:3:6 Gravel Concrete, Hand Mixed. (Average of 6 test pieces.) 
Compressive Strength in lb. per sq. in. 

4 weeks. 13 w^ceks. 

Max. Min. Ave. Max. Min. Ave. 

Company A 988 510 746 1030 561 807 

B 672 409 518 590 522 568 

C 640 321 475 881 640 764 



Digitized by 



Google 



524 Wig on Tests of Strength of Concrete. 

There is an approximate difFerence of 200 per cent in individual values 
ranging from 321 to 988 lb. per sq. in. at 4 weeks and an approximate dif- 
ference of 100 per cent at 13 weeks, the values ranging from 522 to 1030 lb. 
per sq. in. 

1:3:6 Limestone Concrete, Hand Mixed. (Average of 6 test pieces.) 
Compressive Strength in lb. per sq. in. 

4 weeks. 13 weeks. 

Max. Min. Ave. Max. Min. Ave. 

Company A 447 398 422 810 668 745 

B 405 364 390 584 522 566 

C 999 507 693 1380 575 864 

There is an approximate difference of 200 per cent in individual values 
ranging from 364 to 999 lb. per sq. in. at 4 weeks, and an approximate dif- 
ference of 150 per cent at 13 weeks, the values ranging from 522 to 1380 
lb. per sq. in. 

1:2:4 Gravel Concrete, Machine Mixed. (Average of 3 test pieces.) 
Compressive .Strength in lb. per sq. in. 

4 weeks. 13 weeks. 

Max. Min. Ave. Max. Min. Ave. 

Company A 1331 1211 1265 1834 1675 1740 

B 1741 1461 1630 2195 2105 2157 

C 2375 1991 2216 2290 1958 2132 

Laboratory 2589 2540 2572 2698 2634 2672 

(All test pieces of this series were exposed to the weather.) 

There is an approximate difference of 100 per cent in individual values 

ranging from 1211 to 2589 lb. per sq. in. at 4 weeks and of 70 per cent at 

13 weeks values ranging from 1675 to 2698 lb. per sq. in. 

1:2:4 Gravel Concrete, Machine Mixed. (Average of 3 test pieces.) 
Comprej^sivo Strength in lb. per sq. in. 

4 weeks. 13 weeks. 

Company A 1650 1160 1443 1890 

B 1942 1691 1787 2607 2294 2485 

C 1966 1615 1828 2389 2220 2308 

Laboratory 2487 2089 2312 2899 2735 2809 

(All test pieces of this series were cured in the laboratory.) 

There is an approximate difference of 130 per cent in individual values 
which range from 1160 to 24vS7 lb. per sq. in. at 4 weeks and of 50 per cent 
at 13 weeks, values ranging from 1890 to 2899 lb. per sq. in. 

The yield point of the concrete in the last two series varied 
from 400 to 1000 lb. per sq. in.; thus in some cases the yield point 



Digitized by 



Google 



Wig on Tests of Strength of Concrete. 525 

of the concrete at 4 weeks was actually below the usual allowable 
working stress of 1:2:4 concrete which is 500 lb. per sq. in. The 
initial modulus of elasticity varied from 2,880,000 to 5,360,000 
or approximately 100 per cent. 

From these results, which it is believed are substantiated by 
other field tests, it would seem that more attention must be 
given to the mixing and handling of the concrete or it is not safe 
to assume an ultimate compressive strength of 2000 lb. per sq. 
in. for 1 : 2: 4 concrete at the age of 4 weeks.' This value is readily 
obtained in the laboratory and by some contractors and there is 
no reason why it should not be obtained by all. As a general 
thing specifications are not sufficiently definite upon the method 
of mixing and the consistency to be used and too much has been 
left to the discretion of the engineer or foreman in charge of the 
work. 



Digitized by 



Google 



DISCUSSION 



Mr.wason. Mr. L. C. Wason.^ — I would like to ask whether the labora- 

tory or the field tests gave the higher results and whether they were 
uniformly one way. 

Mr. Wig. Mr. R. J. Wig. — The laboratory tests were always the higher 

with one exception. There was one contractor who approached, 
in fact met, the laboratory tests in most cases, but the laboratory 
tests were always the higher and more uniform. There was less 
difference between the government test pieces or beams. 

Mr.wafon. Mr. Wason. — Speaking from memory, during the early 

work of the Boston Transit Conmiission in subway construction, 
tests pieces were taken from the actual work to the laboratory 
and invariably gave higher results than the laboratory test pieces. 

Mr. Humphrey. PRESIDENT HUMPHREY. — ^Thc objcct of the tcst was to deter- 

mine if possible the relation between work done under labora- 
tory and under field conditions. The three companies were 
doing what might reasonably be called first-class concrete work. 
The men, however, tried to do their best as they knew they were 
going to the laboratory to do work under competition, although 
an effort was made to eliminate this feeling. 

The results show that no matter how conscientious a con- 
tractor may be that unless he knows in some definite way what he 
is doing, judgment is of no value in determining the quality of 
concrete. 

There is probably no phase of concrete work meriting so 
much careful consideration at the present time as the very feature 
brought out by the paper of Mr. Wig. It certainly is a subject 
we have been hammering ever since this Association has been 
in existence — better concrete. The assumption of 2000-lb. 
concrete without knowledge whether it is obtainable, without 
any attempt to determine whether it is obtained, cannot be too 
strongly condemned. One of the essentials in concrete work, 
which we cannot too strongly emphasize, is the development 
of field inspection and tests of concrete. 

(526) 



Digitized by 



Google 



Discussion on Strength of Concrete. 527 

Our Committee on Concrete Materials have not reported Mr. Humphrey, 
recommendations this year, but we hope they will, because any 
method is better than none. The best education any contractor 
can have at the present time is to use some simple compression 
machine, no matter how inexpensive as long as it gives approximate 
results as far as the load is concerned, that will tell whether the 
materials are giving a 1000- or 2000-lb. concrete, or as was shown 
on the screen by Mr. Kinney, whether the strength is only a 
little over 300 lb. per sq. in. 

The influence of the percentage of water on the strength of 
concrete is not fully appreciated. The concrete can be too wet as 
well as too dry. It must be thoroughly mixed and water cannot 
take the place of mixing. Use the minimum percentage of water 
that you can to get a good consistency, a sticky consistency, 
and mix a half a minute or a minute, if possible, in the mixer, 
and the result will be a stronger and better concrete. The mixing 
must coat the aggregate and rub it together which cannot be done 
by merely adding water. 

It is hoped that definite recommendations on standard field 
methods will be made by the committee next year, not methods 
that can be used in the laboratory, but methods that can be used 
on the work. This is accomplished in Austria and other parts of 
Europe through the control beam. It is very simple, but a rather 
expensive field test for the average contractor. 

Mr. Wig. — The concrete was all proportioned by volume Mr.wig. 
and then weighed, that is, the volume measurement was trans- 
posed into weight measurement. The sand was taken from a 
bin containing a certain amount of moisture. The moisture was 
determined before admission. The quantity of gravel and sand 
to be used and the weight per cubic foot had been predetermined. 
So these results of 1 part of cement, 2 of sand and 4 of gravel, 
may vary slightly, depending upon the method of determining 
the weight per cubic foot. The weight per cubic foot was the same 
with either, practically, so all variables have been eliminated 
except those to be compared: time, strength and workmanship. 

President Humphrey. — There were two series of tests, Mr. Humphrey, 
one on a 1 : 3 : 6 concrete and another on a 1:2:4. The three 
companies had nothing to do with the preparation or the handling 
of the materials. They simply fixed the consistency, which was 



Digitized by 



Google 



528 Discussion on Strength op Concrete. 

Mr. Htunphrey. Variable, by requesting the quantity of water. Some tests were 
made by hand methods and some were made by machine mixing. 
The average laboratory test on 1 : 3 : 6 concrete at 4 weeks was 
about 1000 lb. per sq. in. The companies claimed this mixture 
was not a practical mixture for reinforced concrete conditions and 
the second series was made on 1 : 2 : 4 proportions under the same 
conditions by the same companies. This concrete gave over 
2000 lb. at 4 weeks. 

Mr.chttbb. Mr. J. H. Chubb. — I would like to ask if there was any 

record kept of the amount of water that the particular contractors 
used, the consistency.* It seems to me that since the variation 
was only in the long time tests, the result of those tests would 
show that nearly all those variations were due to consistency. 
In a long time test the variation was about half what it was in the 
short time test. If they ran 60 days or 6 months it would probably 
be much closer. 

Mr. Humphrey. PRESIDENT HUMPHREY. — The amouut of Water used was 

measured. While the contractor said how much water, we meas- 
ured how much water he actually used, so that the tests when 
they finally appear in the bulletin form of the United States 
Bureau of Standards, will contain the actual amount of water 
used in these various tests. 

It is found that at the end of a year there is less variation than 
at any earlier period. When it is considered, however, that 
concrete structures are put into service in a very short time and 
that the contractor wants to remove the forms some in 2 days and 
most in 4 days, certainly a week, it must be known just how to 
handle the concrete to obtain the strength desirable under such 
conditions. The factor of safety generally increases with age as 
the concrete grows harder, but the important thing to know is 
how to make the concrete so as to get the maximum strength at 
an early period. 

Mr. Wig. Mr. Wig. — I think we have given a little too much attention 

to the ultimate strength of concrete rather than to the yield point 
and elastic properties — ^not so much the elastic properties as the 
yield point. The yield point of concrete in these tests was at a 
strength of 400 lb. in some cases at four weeks. The forms are 
usually removed in one or two weeks. In such cases the yield 
point would probably not be more than 150 or 200 lb. and it would 



Digitized by 



Google 



Discussion on Strength of Concrete. 529 

be a question whether the building would withstand its own load, Mr. wig. 
its own weight. The advantage in this variation in the consist- 
ency is in the variation of the yield point. When the concrete 
yields it will ultimately fail. In putting the test piece in the test- 
ing machine the load was applied so rapidly that the yield point 
would be little affected and carried to failure. When the load is 
left on a little beyond the yield point the concrete will ultimately 
fail. 

Usually the yield point will run around 1500 or 2000 lb.; 
1000 lb. would be very ordinary, and 1000 lb. was obtained in 
these tests. 

Mr. Wm. M. Kinney. — I think that this brings us back again Mr. Kinney 
to the question of aggregates. Some aggregates certainly do 
affect the eariy strength* more than others, and that is some- 
thing that must be studied; that is, these tests must be made 
in order to find out the strength of the concrete in a certain length 
of time. Referring to the paper by Mr. Cummings,* I believe 
the use of steam resulted from the fact that in 14 to 16 days he 
could not get sufficient strength in his concrete made from a 
particular aggregate to drive the pile, which necessitated the use 
of steam in order to cure it more quickly. A job usually can not 
be delayed too long for foundations, and piles are the starting 
of a foundation, of course. 



*See p. 312.~£o. 



Digitized by 



Google 



THE NECESSITY FOR FIELD TESTS OF CONCRETE.* 
By Fritz von EMPERGER.f 

One of the most essential precautions taken in every well 
regulated manufacturing plant is in the control of raw material 
as to its qualities in order to properly guard against poor material. 
Failure to do this results in a reduction of the efficiency of the 
finished work, causing either accident or reconstruction, which 
sometimes has a far-reaching effect, especially in the case of con- 
struction work where it often leads to the financial ruin of the 
contractor even if there is no accident. It would seem natural 
that in concrete construction this precaution should not be omitted. 

At the present time, however, it is considered sufficient to 
test the cement according to standard methods and to judge the 
fine and coarse aggregates by visual inspection. The inadequacy 
of this procedure has been demonstrated several times recently 
through the fact that the concrete in parts of large buildings 
resulted so poorly that they had to be rebuilt and in another 
case the structure was accepted only at the combined risk of the 
contractor and the owner because the only correct manner of 
handling the matter was beyond their financial ability, that is 
rebuilding. In all of these cases the raw materials, although not 
of first class quality, were as good as those used in hundreds of 
other cases where no trouble occurred. It was even proven 
experimentally that the materials, claimed faulty by interested 
parties, when mixed with first-chiss cement gave excellent results, 
but when mixed with the cement used on the particular job, the 
results were below the requirements. Similar results have pointed 
to cement coming from certain mills. However, with first-class 
aggregates this cement produced a concrete of suflicient strength. 
In one case it was found that the cement had not been seasoned 
sufficiently, which was immediately remedied. In this case the 
control beam was used and it is mentioned to illustrate the 
effectiveness of the check on the use of proper materials. 



* Traoslated from the German, by the Secretary, 
t Consulting Engineer, Vienna, Austria. 

(530) 



Digitized by 



Google 



Emperger on Field Tests of Concrete. 531 

The difficulty with concrete control is the long time required 
to obtain results. The speed with which modem concrete struc- 
tures are erected makes each day of great value. Therefore, 
the amoimt of time required to send the test pieces to a laboratory 
should be eliminated because a quick control is possible only on 
the job. The method given in the latter part of this paper makes 
it possible to obtain results within from 8 to 10 days, which will 
clearly indicate all poor concrete. It is hardly necessary to state 
that in conctruction work it is only important to detect consider- 
able variation of the concrete from the average quality. 

In the above mentioned case the conditions were about as 
follows: The aggregates were first class and the mixing and other 
operations were properly done. The first 10-day test showed only 
one half compressive strength and it seemed clear that the trouble 
was most likely with the cement. In the meantime one story 
had been built with such concrete. The test indicated that the 
reduced strength of the concrete would still be sufficient and 
precautions were taken only to secure the original quality of 
concrete for further work. On completion of the structure the 
story in question wa^ subjected to an exhaustive test. This 
indicated clearly much lower results than in the case of other parts 
of the structure; however the results were not low enough to 
necessitate rebuilding so that the contractor came through with 
a bad scare only. If the result of the first test had been much 
lower, it would have been possible to tear down the one story in 
time. This would have been a small expense compared with 
rebuilding the whole structure. 

In several cases of completed buildings the removal of forms 
was delayed time and time again in the hope that the concrete 
would harden sufficiently, which, however, it did not do. Such 
are very exceptional cases and may occur in only one of a thousand 
structures. One case, however, should be sufficient to warn the 
contractor and to demonstrate the necessity of concrete control. 
The extra expense is so little that it need hardly be considered in 
determining the cost of a structure and offers insurance against 
a disaster which may cause the ruin of the contractor's business. 

In order to demonstrate the necessity of concrete control 
reference to such exceptional cases is not required. There is to be 
considered in first line the many small things affecting the manu- 



Digitized by 



Google 



532 Emperger on Field Tests of Concrete. 

facture of concrete which to this time have not received sufficient 
attention. Their importance is much under-estimated and it is 
consequently ignored that the combined action of these conditions 
may have a very bad result. In the same way as it has been 
customary to rely upon cement meeting standard specifications; 
one has also relied on the richness of the mixture, although the 
incorrectness of the assumption that the strength of concrete 
is proportional to the amount of cement can easily be proven. 
As soon as the mixture of sand and aggregate has reached the 
point where its voids are fully filled by the cement no material 
increase in the strength is to be expected. The extremely, rich 
mixtures called for in many specifications are valueless and an 
unnecessary waste of material. It is without doubt the business 
of the contractor to determine the most suitable mixture of the 
materials at his disposal, work which, in large cities, requires 
occasional tests. It is clear, therefore, that the contractor must 
keep control of the concrete if he desires to use the most economical 
proportions, consistency and mixing of the materials available. 
The owner, i. e., his engineer, or the city building inspector must 
also have certain information along these lines so that no require- 
ments will be specified which it will be impossible for the contractor 
to meet. From several cases it can be proven that the required 
compressive strength in official specifications for a certain pro- 
portion of mixture cannot be obtained with the material available 
or the reverse, that is, the exclusion of certain kinds of stone is 
technically entirely unjustified. It is one of the worst assumptions 
that one has intuition as to the strength of the material. This 
jud^ent can only be obtained through the use for many years of 
such tests as are described later. 

Let it be assumed that the proper preparations in a special 
case have been made and that the strength in compression at 
8 and 28 days of the mixture of the selected proportions is known 
under normal conditions. The one strength serves as the control 
for the concrete; the other for the removal of forms. Now the 
question is, which' extraordinary conditions on the work should 
be considered. In the first case the most natural deviation from 
the conditions originally stated would be the amount of water 
added to the mixture and the procedure of mixing itself. The 
compressive strengths determined on the test beams repeat them- 



Digitized by 



Google 



Emperger on Field Tests of Concrete. 



533 



selves uniformly, which is not usually the case with concrete cubes. 
This is then confirmation that no mistake has been made in the 
manufacture of the concrete and that the completion of the work 
represented by the test pieces can be anticipated without worry. 
Even small deviations are a warning for more carefulness and 
checking of the methods of making the concrete. 

An entirely different picture is brought to us when there is 
a question of the influence of abnormally high or low temperatures. 
In the first case it is possible that the hardening of the cement will 
be so rapid that concrete partially set is placed in the forms. In 
the latter case the action of frost can withdraw the heat necessary 
for the setting of the cement and destroy the concrete through the 

■ ^" 




FIG. 1. FIG. 2. — CROSS-SECTION OF CON- 

TROL BEAM. 

formation of ice crystals. These two possibilities cannot be repro- 
duced in the laboratory. The contractor or the inspector must 
be able to determine on the job just what is happening to the 
concrete. The test pieces must be stored near the part of the 
structure, the concrete in which is to be controlled and must 
be of the identical concrete used in that part of the structure 
which it is desired to check up. In order to make this possible 
it is necessary to use a method which can be carried through 
right on the job without much trouble and without a testing 
machine. 

In line with the thoughts given above, a short description 
of the control beam used by the author covering all the points 



Digitized by 



Google 



534 



Emperger on Field Tests of Concrete. 



noted is appended. The dimensions of the test beam are so selected 
that the test piece will not be too heavy and can, therefore, 
easily be handled, and so that no large load will be necessary 
for the test. Further it has been borne in mind that the com- 
pressive strength should be determinable by a simple calculation 
from the breaking strength without complicated formulse. This 
empirical rule changes according to the method of calculation, 
which should naturally be the same as that used in calculating 
the beams in the structure so that the stresses obtained can be 
compared and the factor of safety directly determined. 

In the following the usual method of calculation is used, which 



7'47'- 




FIG. 3. — ARRANGEMENT OF TESTING APPARATUS. 

assumes the rectilinear relation of the stress and deformation and 
the ratio of the coefficient of elasticity of the reinforcement and 
concrete, n = 15, although this is a secondary detail and not directly 
connected with the method of testing. The following equation 
gives the strength of the concrete, the maximum bending moment 
being M, 

M = Cctn = Tim = <rcWc = -^x b m 

Insofar as the notation is not shown in Fig. 1, h represents the 
total height, h the width of the beam, and x the distance between 
the neutral axis and the extreme fibre at the section in question. 
According to Fig. 2 in our tests the values of h and h are each 3 in 



Digitized by 



Google 



Emperger on Field Tests of Concrete. 



535 



Therefore, the effective cross-section /^ = 3x3 = 9 sq. in. and the 
effective section of reinforcement F = J sq. in., and consequently 
the percentage of reinforcement is 5.55. 

The arrangement of the testing apparatus, Fig. 3, gives 



M=:^X 29 = 14.5 P in. -Z6. 

2 



Calculation shows 



x—2.1 in. and m^h =2.3 in, 

3 




FIG. 4. — METHOD OF MAKING CONTROL BEAMS AT VIENNA. 



Therefore the extreme fibre stress 

29P 



2 X 3 X 2.1 X 2.3 



= P 



The value of P, the breaking load, is represented by the weight 
of the carrjdng device and the bricks and f of the dead weight 
of the beam (exactly 10.25/14.5). Having determined the load 
P under which the beam has failed this figure represents the 
value of (Tc in in.-lb. and is the desired measure of quality of the 
concrete tested. 

Further details of the method can be obtained from For- 
scherhefi XIII, "Eine Guteprobe fur Beton" by Ing. G. Neumann, 



Digitized by 



Google 



536 



Emperger on Field Tests of Concrete. 



as well as the author's paper presented in Berlin, 1911, to the higher 
officials of the building departments. Session II, published by 
W. Ernest und Sohn, Berlin. A description of an extensive 
application of the method in the construction of K. K. Kriegs- 
ministerialgebaude, Vienna, by Oberleutnant J. Kromus, appeared 
in Beton und Eisen, 1911, Heft XIX and 1912, Heft I. Figs. 4 and 
5 illustrate the method of making and testing the test beams on 
this work. 

In order to obtain correct results in such a method of test it 
is only necessary to be careful in maintaining constant the values 




FIG. 5. — CONTROL BEAM UNDER TE§T. 

of Fi, the sectional area of reinforcement, in this case \ sq. in., 
and of A, the distance between top of beam and center of reinforce- 
ment, in this case 3 in. Ever^i^hing else can be done so primitively 
as would be the case without especial care on a job and it would 
not affect the correctness of the results. According to experience 
the results obtained are always more correct, reliable and more 
quickly obtained than results from test cubes l)roken in a hydraulic 
press. This method of testing represents an easy way for every 
contractor to become acquainted with his materials and to deter- 
mine whether he is obtaining the best results at the utmost 
economy. 



Digitized by 



Google 



DISCUSSION. 



Mr. Arthur N. Talbot. — It would seem that if a satisfactory Mr. Taibot. 
control test piece can be found, its use would be helpful to the 
constructor; it would serve as a record of the construction. There 
is a diificulty, however, in getting a beam or test piece which is 
fully satisfactory. I have used a plain concrete beam of a little 
larger crossnsection than mentioned here, 6 in. wide, 8 in. deep 
and 36 in. in span. This of course will break in tension. It is 
subject to the variations which we must expect in the tensile 
strength of concrete and requires that more than one test piece 
be made, as a single result might be abnormal and accidental. 
Mr. Emperger has put in reinforcement in such a way as to make 
practically a compression test, the beam being so long that there 
is little danger of shear failure. Whether this form of test piece, 
of the size shown, would be satisfactory in practice can be told 
through experience with it. It would seem to me, looking at the 
dimensions, that a test piece only 3 in. wide and 3 in. down to the 
reinforcement would require such care in its construction that it 
would not be suitable when the aggregate is of other than very 
small size and that there would be objection to the size in practical 
use. 

President Humphrey. — It is a matter of interest to this Mr. Hamphrey. 
Association to have this subject come up, as it is in line with the 
endeavor of one of our committees to have recognized the neces- 
sity of having some inexpensive field method by which the 
quality of concrete can be obtained. I have seen this control 
beam test made in Vienna and I feel that the care required to make 
this test renders it impracticable. The section of the test piece 
itself had a material bearing on the quality of the concrete and 
the size of the aggregate also afi^ected in a large measure the 
results obtained. I did not feel from the number of tests that I 
saw made that there was any great concurrence in consecutive 
results. It is a question whether or not the same degree of accuracy 
cannot be obtained from a small cylinder or test piece made in a 
simple compression machine. Certain it is that contractors 

(537) 



Digitized by 



Google 



538 Discussion on Field Tests of Concrete. 

Mr. Humphrey, should try various methods of testing, with the hope that some 
simple, inexpensive method for determining the quality of the 
concrete may be devised. That to my mind is of the most vital 
interest at the present time in reinforced concrete construction. 

Mr. Brett. Mr. Allen Brett. — While as some claim the control beam 

in the 3 by 4-in. section might not be accurate enough to determine 
correctly the value of the concrete, yet, especially in running 
concrete in cold weather, it seems to me that the control beam 
would be sufficiently accurate to indicate when the concrete was 
hard enough to pull the forms. I think this point should be more 
emphasized. 



Digitized by 



Google 



REPORT OF COMMITTEE ON TREATMENT OF 
CONCRETE SURFACES. 

The Committee has this year devoted itself to perfecting 
specifications for stucco and to methods of testing compounds 
for dampproofing of surfaces. Every member of the Association 
who was known to have knowledge on these subjects was written 
to, something over four hundred persons being addressed. There 
was no real criticism from anyone^ and we therefore feel justified 
in recommending this specification for adoption this year. 

The specification as printed in Volume VII of the Proceed- 
ings, 1911, should be changed as follows: 

Page 587, near bottom, before paragraph "Sheathing Boards," 
insert: 

"Frame. .The frame of building shall be so rigidly constructed 
as to avoid cracking the stucco." 

Page 589, near bottom, change title from "Brick or Tile" to 
"Brick, Tile, or Cement Blocks." 

Page 590, line 7, at end add "and thoroughly wetted." 

Page 590, paragraph 6, change title to "Brick, Tile, or Cement 
Block." 

Page 590, near bottom, after title "Intermediate Coat," add 
a sentence, "Intermediate coat may be omitted on brick, tile, or 
cement blocks." 

Page 591, near middle, change title "Fmal Coat" to "Finish 
Coat." 

Page 591, near bottom, change title "Finish" to "Surface 
Finish." 

Page 593, after paragraph 3 add: 

Machine Stucco. 

Stucco may be applied by a machine provided the results 
obtained are equal to those produced by hand work. 

The last report of. the Committee is amended by the 
following: 

Insert, page 579, near bottom, xmder V. Waterproofing. 
(g) Method of testing dampproof compounds and coatings. 

(539) 



Digitized by 



Google 



540 Report on Treatment of Concrete Surfaces. 

Introduction — ^As it is obviously not the province of this 
Association to stoop to commercialism by publishing the names 
and merits of any article of trade, your Coromittee below gives 
the method by which anyone may make comparative tests of any 
compound for himself. 

In the tests of concrete waterproofers and dampproof con- 
crete coatings, these materials are divided into the following 
classes: 

(1) Colorless Dampproof Coatings. — ^This class includes all 
preparations intended for surface application for the purpose of 
excluding dampness, which the makers claim will not change the 
appearance of the surface to which they are applied. 

(2) Black Waterproof Coatings, — ^This class includes all tar, 
pitch, and asphalt preparations and other black waterproofing 
paints. Such coatings should be able to withstand some pressure 
head of water and are therefore termed waterproofers. They 
would not be used where a decorative effect is required. 

(3) Integral Waterproofers. — This class includes all materials 
for waterproofing, whether pastes, powders or solutions, that are 
incorporated into the mass of the concrete at the time of mixing. 

(4) Colored and White Concrete Coalings. — ^This class includes 
all coatings of a decorative nature intended for exterior use. 
Many of the paints in this class are recommended by the manu- 
facturer for their dampproofing qualities. 

» 
Method op Testing Dampproofing and Watbrprooping 

Coatings. 

This test is intended primarily for classes 1 and 2, although 
many paints in class 4, when especially recommended by the 
manufacturer as a dampproofer, are included. 

The test piece is a 3 to 1 sand-cement mortar block 3J in. 
square by 2 in. deep, with a depression in the top 2 1 in. in diameter 
by about i in. deep made by inverting a telephone bell in the mold. 
The mold is of wood, J in. oak, and is made in two right-angular 
pieces which hook together at opposite corners, so that it is easily 
removed by unhooking and drawing apart. 

To obtain a porous block or one that will absorb 30 c. c. 
of water in more than one and less than two minutes, the sand 



Digitized by 



Google 



Report on Treatment of Concrete Surfaces. 641 

and cement is gauged with from 9 to 10 per cent, of water by 
weight (this per cent, being figured on the total weight of sand 
plus cement) and tamped evenly into the mold. Some difficulty 
may be found in making blocks of the proper porosity; that is, 
just dense enough to absorb the 30 c. c. of water in not less than 
one or more than two minutes; but after some experimenting 
with different percentages of water and different degrees of 
tamping, the proper result is obtained. 

Denser blocks are made in the same manner, but the per 
cent, of gauging water is higher and may run up as high as 20 
per cent. 

All blocks are kept in a damp closet for 24 hours and then 
allowed to dry out in air for at least one week before being 
used. 

After the block has dried out, 30 c. c. of water are placed 
in the depression and time required for the block to absorb this 
amount of water is noted. The blocks, are divided into two 
classes: those which will absorb 30 c. c. of water in more than 
one minute and less then two, (these are called standard blocks), 
and those which take less than one or more than two minutes to 
absorb the 30 c. c. of water. 

Those blocks which take up the water in less than one minute 
are considered too porous for a proper test. 

After this treatment the block is again thoroughly dried out. 
The time required to absorb the 30 c. c. of water is recorded on 
the block which is now ready for use. 

The blocks which take up water in one or two minutes are 
considered the standard block, but materials are tested on both 
porous and dense blocks for comparison. 

Each block is given a number for purpose of identification. 
Porous or standard blocks are numbered consecutively from 1 to 
500 and the denser ones from 500 up. 

The waterproofing coating is applied only to the surface of 
the depression and to the top surface of the block. Two full, 
liberal and thoroughly brushed in coats are applied, the first one 
being given at least two days to dry out before application of the 
second. 

Not less than one week after date of application of the second 
coat, the block is ready for the first test. 



Digitized by 



Google 



542 Report on Treatment of Concrete Surfaces. 

Twenty-Four Hour Absorption Test. 

The block is weighed dry. Thirty (30) grams of water are 
placed in the depression. The block with the water in is allowed 
to stand for 24 hours, with a watch glass over the depression, after 
which time it is weighed again. The loss in weight is the evap- 
oration. The water is then thrown out and the block wiped 
dry and weighed again. The loss from the weight, including water, 
is the absorption. The following is a tabulation of the dif- 
erent weighings taken: 

(1) Weight of block dry plus watch glass. 

(2) Weight of block plus 30 grams of water plus watch 
glass. 

(3) Same as (2), but after standing for 24 hours. 

(4) Weight of block plus watch glass, after water has been 
thrown out and the block wiped dry. 

Weighings: 

(2) Minus (3) gives weight of water lost by evaporation. 

(3) Minus (4) gives weight of water left in block after 24 
hours. 

The watch glass almost completely prevents evaporation 
from the surface of the water. Before any appreciable amount 
of the water in the depression can evaporate it must pass through 
the waterproof coating and be evaporated from the surface of the 
sides or bottom of the block. 

The most eflSciently waterproofed blocks are those which 
show the minimum amount of absorption and evaporation. 

Test to Determine Time Required for Coated Block 
to Absorb Thirty Cubic Centimeters op Water. 

The 30 c. c. of water are placed in the depression in the block 
and covered with a watch glass to prevent evaporation. A paper 
is placed over the whole to exclude air currents. The block is 
examined from time to time until all the water in the depres- 
sion has disappeared. The time required for the block to absorb 
the 30 c. c. of water is recorded. If all the water has not been 
absorbed in 3 weeks time the approximate per cent, absorbed 
in 3 weeks is recorded. 



Digitized by 



Google 



Report on Treatment of Concrete Surfaces. 543 

The length of time taken for the water to be absorbed by the 
block is an indication of the efficiency of the waterproofing film 
or coating. 

The block is then exposed in the open air to find the effect 
of weather exposure on the coating. It is brought in at intervals 
of about 3, 6 and 12 months and the same tests as above described 
repeated. The decrease, if any, in the efficiency of the coating is 
noted after each repeated exposure. 

Test of Integral Waterproofing. 

For this test, the blocks are made up of a 1 cement, 2 sand 
mixture plus the waterproofing compound, which may be paste, 
powder or liquid. The waterproofing compound is incorporated 
in the mix in the proper amount and according to the directions 
as given by the manufacturer. The mixture is gauged, in all cases, 
with the proper amount of water to give approximate maximum 
density. The consistency resulting from this amount of water 
gives a rather wet, quaking mortar, but one from which no free 
water will rise to the surface on tamping. 

This block is identical in size and shape to that described 
above in the " Test of Waterproofing and Dampproofing Coatings." 

It is kept in the damp closet for 24 hours after making and is 
allowed to dry out in air for at least one week before being 
tested. 

The tests on these blocks are the 24-hour absorption and time 
to absorb 30 c. c. of water tests and are identical with those 
used for the waterproofing and dampproofing coating blocks, 
described above in detail. 

Description of Test of Concrete Coatings. 

This class comprises, chiefly, coatings of a decorative nature, 
intended for exterior use. 

The paints iji this class are tested on the outer face of hollow 
cubes, the faces of which are about 9 in. square, with walls from 
1 in. to 1 J in. thick. These hollow cubes are made of a 1 cement, 
2 J sand, 4 gravel mixture; and gauged to a rather wet consistency. 
The block is made by packing this rather wet mixture into a 9 in. by 
9 in. by 9 in. wooden form provided with an iron core about 12 in. 



Digitized by 



Google 



544 Report on Treatment of Concrete Surfaces. 

high and 6 in. square. The bottom of the block is made about 
1} in. thick, the hollow iron core is then placed on this bottom 
layer and centered up and the concrete packed in between the 
core and the wooden form. 

The core used has considerable taper so that it may be easily 
drawn out after the concrete has begun to set up. After the block 
has had 48 hours to harden up the wooden form is removed 
and the faces are finished with a 1 cement, 2 sand mortar. In 
finishing the faces the mortar is mixed rather wet and is merely 
rubbed into the surface with a wooden float to give a smooth 
surface. The block is ready for use not less than one week after 
date of finishing up the sides. 

Before applying any paints, the block is given a number, 
markcki on the inside of one of the faces, and the faces are numbered 
from 1 to 4, marked at the middle of the top edge. The date of 
making the block is recorded. 

A few minutes before applying a paint, the bottom third of 
the face to be painted is thoroughly dampened by applying water 
to it with a brush. Care is taken to wait until no visible water 
is on the surface before the paint is applied. 

In applying the first coat the paint is thoroughly brushed 
in so that it will penetrate all small irregularities of the surface. 
The second coat is applied not less than two days after date of 
application of the first. 

At the time of painting the fact is recorded that the bottom 
third of the side was dampened. The following data are also 
noted at this time: 

Number of block. 
Number of side. 

Name of maker and name of paint (for both 1st and 2d coats). 
Date of painting (for both 1st and 2d coats). 
Per cent, of sediment in the paint can when it was opened 
and consistency of this sediment. 

Ease of application of paint. , 

Ease of mixing paint. 
Hiding power of paint. 

After the second coat has thoroughly dried, the general 
appearance of the painted surface is recorded. 



Digitized by 



Google 



Report on Treatment op Concrete Surfaces. 545 

The points noted in regard to appearance are as follows: — 

Dull or glossy. 

Whether surface is smooth or granular. 

Whether brush marks are visible or not. 

Whether paint fills well or poorly, i, e., whether the irregulari- 
ties of the concrete surface are well fitted up by the paint or 
not. 

After all four sides of a block have been painted (two coats) 
and at least three days after the application of the final coat, the 
painted surfaces are scrubbed lightly with a medium bristle 
scrubbing brush with powdered soap and water. The effect of 
this scrubbing is noted in detail, as to whether the coating is 
softened, or whether the paint scrubs off, and if so, to what 
extent. 

After the scrubbing test the block is allowed to dry out and 
after not less than one day from time of scrubbing it is filled with 
water and kept filled for a period of 5 days. If the water leaks 
through either the sides or bottom of the block to any appreciable 
extent, the inside of the block is given a neat cement wash to 
remedy this condition. 

After the water has been in the block for a period of 5 days 
it is emptied out and the condition of the coatings on the different 
faces carefully examined and recorded. 

Conditions resulting from the 5 days' water test to be noted 
are as follows: 

Softening of coatings. 
Excrescence. 
Discoloration of coating. 
Cracking or blistering of coating. 
Flaking off of coating. 

After the block has thoroughly dried out the coatings of the 
four faces are given a general comparative rating according to 
their respective general appearance and condition. 

The block is now ready for the weather exposure test, and for 
this piupose it is exposed in the open air. The block is placed 
with the bottom side up. Examinations of the blocks are made 
at regular intervals of about 3, 6, and 12 months. 



Digitized by 



Google 



546 Report on Treatment of Concrete Surfaces. 

Conditions resulting from weather exposure to be noted are 
as follows: 

Chalking of coatings. 
Checking and cracking of coating. 
Scaling or flaking oflf of coating. 
Fading or discoloring. 

Page 574. After line 8 insert the following: 

A Inethod of waterproofing has been submitted by one 
member of the Association and is reported to have given satis- 
faction. It is as follows: 

For a flat surface lay a base of concrete while still wet and 
plaster with i in. of neat cement troweled hard, then follow with 
another layer of finishing concrete, the lower layer being at least 
2 in. thick and the top layer 3 in. 

On wall surfaces as soon as the forms are removed, thoroughly 
wet surface, trowel on J in. of neat cement, and follow immediately 
with 1 in. of 1 : 2 mortar before the neat cement has begun to 
dry or appreciably set. 

If the wall or floor treatment in this manner is large it must 
be reinforced to prevent cracking. 

Respectfully submitted, 

L. C. Wason, Chairman, 
Cloyd M. Chapman, 
Alfred Hopkins, 
Emile G. Perrot, 
Henry H. Quimby. 



Digitized by 



Google 



DISCUSSION. 



The President. — There has been request for a discussion of The President, 
the cement gun. * The principle of the machine is to deposit sand 
and cement with compressed air, the manner of the deposition 
being such as to obtain a density and compactness not possible 
by hand methods. The sand and cement are first mixed dry. 
It is one of the essentials that the mixture shall be dry, otherwise 
it will cake in the hopper or so clog up the pipe that it will 
not flow. 

The character of the surface on which it is applied and its 
preparation are of course the things which determine the per- 
manency of the plaster applied. With this brief statement we 
may begin the discussion by requesting Mr. Chapman to say a 
word on the cement gun. 

Mr. Cloyd M. Chapman. — The Westinghouse, Church Kerr Mr. chapman. 
Company conducted some rather exhaustive tests last sum- 
mer on the product of the cement gun, that is the qualities pos- 
sessed by mortar applied by the gun were investigated. The 
tests covered compressive strength, tensile strength, percentage 
of voids, absorption through the surface, and adhesion to other 
materials. In each of the tests the qualities were much superior 
in the gun mixed mortars than in the best hand made mortars 
made under similar conditions and of the same mixtures. The 
method was to apply a 1-in. coating on the wall. The wall was 
constructed with sheet metal, so hand plaster and gun plaster 
could be applied to the wall an inch thick. A briquette which 
would fit the regular tension machine was cut out with a die or 
cutter. For compression tests a 2-in. thick coating of mortar was 
cut into cubes with a die or cutter. The results of 28 different 
tests were from 25 to 100 per cent better in the case of the gun 
mixture than the hand mixture. The same applies to density 
measured by absorption, surface absorption, absorption through 
the surface film as well as total absorption on boiling 4 hours and 
then cooling. It showed a very dense, high quality of mortar 
and the uniformity was much greater than that of hand work. 

• For additional information see Proceedings, Vol. VII, p. 5()4. — ^En. 

(547) 



Digitized by 



Google 



548 Discussion on Cement Gun Mortab. 

Mr. ciupiiuui. The test pieces cheeked up closer and the leanness of the mix 
that could be used was very much greater. Some beach sand 
was obtained near the laboratory at Whitestone Landing, a very 
high percentage of which would pass an 80-mesh screen; the 
bulk of it, I think, would lie between 80 and 100 mesh. Propor- 
tions of 1 : 6, 1 : 8 and 1 : 10 of that very fine sand gave pretty 
solid mortar. No tests were made — and I want to emphasize 
this — of the operation of the gun as a machine. 

Mr. Brett. Mr. Allen Brett. — It is thought that the product of the 

cement gun is automatically exact. It has been observed that 
although placing into the receiving chamber all proportions of 
materials, as 1 : 4 and 1 : 6 or 1 : 10, that the product on the 
wall is approximately always the same. There always has to be 
a certain amount of cement in the product on the wall and the 
rest of the sand drops down. It is a very peculiar proposition. 
The mortar on the wall always seems to have a certain amount of 
cement, approximately 1 : 2^. Another point is the amount of 
material required. The amount of material was estimated for a 
certain job the same as for an ordinary stucco job. The sand 
and cement were sent accordingly, but the material was all gone 
before the job was half done. It takes about twice as much 
material as in hand stucco or plaster. Of course the material is 
there as the mortar is denser. 

Mr. Chapman. Mr. Chapman. — One point was not brought out in regard to 

the spatter and the failure of sand to adhere when lean mixtures 
were used. We did not find that the spatter was a much greater 
per cent of the total amount applied when lean mixtures were 
used as against rich mixtures, but it was greater when coarse 
sands were used. The trouble is caused by the larger grains 
which have a momentum sufficient to rebound when the mixture 
is not rich enough. Lean mixtures of coarse sand would rebouna 
somewhat more than a rich mixture of coarse sand, and the rich 
mixture would reboimd somewhat less than the lean mixture of 
the same sand whether fine or coarse. A rich mortar forms a 
putty-like surface to which the particles will adhere more easily 
than in the case of a lean mixture. 

The President. The PRESIDENT. — The first Stage in the application of the 
mortar is a rebound of the sand until a sufficient coating of neat 
cement is deposited on the surface in which the sand will bed. 



Digitized by 



Google 



Discussion on Cement Gun Mortar. 549 

Mr. Chapman. — ^That explains the strong adhesion of gun Mr. chApnum. 
applied mortar. The process gives a coating that next to the 
surface is quite rich. 

Mr. Willis Whited. — There are a very large number of Mr. whited. 
bridges in the State of Pennsylvania, the masonry of which is 
built of inferior stone, which has suffered from the effects of the 
weather; and other bodies of masonry, some of them quite exten- 
sive, have suffered from other causes, such as bad foundations, 
bad mortar, etc. To repair them with stone would spoil their 
appearance. My idea was that the defective parts could be 
replaced with concrete, and then the whole faced up with the 
cement gun, so that they looked like concrete, and put in good 
repair for very much less money than it would cost to tear the 
whole wall down and rebuild it, thus making the whole wall 
better, both in quality and appearance, than it was when first 
built. Besides, much of this masonry was built in early times 
when lime mortar was used exclusively. My understanding is 
that mortar from a cement gun can be forced into open joints 
quite a distance and do a good deal better work than any repoint- 
ing could, and some of the masonry could be saved in this manner 
that, otherwise, would have to be torn down and rebuilt. I 
would like to find out whether my ideas are correct. 

The President. — There is no doubt that the pressure under The President, 
which the mortar is applied drives it into all the cracks, so as to 
completely fill them. With the stone in proper condition the 
mortar will stick to the surface, and with a first-class plasterer 
who should always accompany the cement gun to finish the sur- 
face as fast as the mortar is applied, a very smooth, pleasing sur- 
face can be obtained. The gun can be applied a second time to 
rough up, because the gun produces a pebbly surface, which pre- 
vents hair cracks and gives a pleasing finish. 

The great difficulty would probably be the removal of all 
organic growth, otherwise the mortar would not have sufficient 
bond and the frost would lift it off. The gun can be used as a 
sand blast, to clean the surface first before applying the mortar. 
I should think as a general proposition that the cost of reparing 
those bridges would, even with the cement gun, be materially 
less than tearing them down and replacing them entirely. 



Digitized by 



Google 



550 Discussion on Cement Gun Mortar. 

Mr. Heidenreich. Mr. E. Lee Heidenreich. — During the early days of the 
cement gun I had occasion to investigate the same for the New 
York Central Railroad at Grand Central Station in New York. 
We borrowed a gun from the manufacturer and experimented to 
ascertain how to cover structural steel with the gun. The great- 
est trouble was in losing the material going past the edges; the 
necessary thickness could not be obtained around the edges. 
There was a great waste of material, although several kinds of 
backing were placed behind the steel members; naturally con- 
siderable waste would occur as the mortar would pass the side 
of the steel member. Of course, at that time the gun was in a 
very incomplete condition and, I understand, it has been improved 
since. I would like to know if this waste of material in covering 
structural steel with mortar for fireproofing purposes has been 
diminished by some method. 

The President. The PRESIDENT. — As far as the Chair knows the loss con- 

tinues to be approximately about the same. It is interesting to 
note that the material collected is clean sand free from cement, 
which can be used again. The actual amount of material that 
rebounds from the surface on which the mortar is being applied 
remains the same. 

Mr. Heidenreich. Mr. Heidenreich. — Taking for instance a wooden trough to 

stop the stream of cement mortar, which passes the steel member 
of course a certain amount of the charge of the gun passes down 
along the trough to the bottom. Suppose I want to get to the 
edge of a column or flange. More than one-half goes entirely by 
the column, and that contains cement; it sets very rapidly, too. 
The waste was fully one-half the total mortar material, including 
what fell down between the flanges, along the edge of the columns 
and material that passed by. This was one reason, at first, that 
the cement gun was not considered economical for the purpose. 
I do not know whether it was used since at the Grand Central 
Station, but this was nearly two years ago. 

The President. The PRESIDENT. — I am not quitc surc what the conclusions 

of the New York Central Railroad oflficials were on that work, 
but certainly the mortar was driven into all the little cracks and 
crevices of the iron work and produced a tightness which could 
not be obtained by any other method. The conditions in the 
New York Central Station were unusual on account of the 



Digitized by 



Google 



Discussion on Cement Gun Mortar. 551 

extreme rapidity of corrosion of the steel. Some of the steel The President, 
work, erected not over •two years, had rust on it to the extent of 
^ to iV of an inch, and the use of some protective coating 
to prevent further corrosion was a matter of great importance. 
The matter of cost was not as important to the engineers as the 
question of the efficiency of the coating. 

Mr. Chapman. — ^As to the sand which rebounds from the Mr. chapmui. 
cement gun, an analysis of two or three samples of the reboimd- 
ing material showed about one part of cement to twenty parts 
sand, although a 1 : 3 mixture was applied. Samples of the 
material gathered up at the foot of the wall analyzed about 1 : 20. 



Digitized by 



Google 



CEMENT COATINGS. 
By F. J. Morse.* 

The subject of cement coatings, or cement paints, as many 
speak and think of them, is one which will receive a great deal more 
of thought. As an example, in an advertisement after describing 
a cement building from a fireproof standpoint, there is added, 
"Painting not Necessary." This remark, as it stands and as it 
is construed by the majority of users of cement, is correct and 
really applies to oil painting as it is used for the protection of 
lumber in building of frame construction. Good cement work 
does not need any protection from rain or climatic conditions such 
as soon destroy unpainted wood. 

The difference between what is called painting or paint, or 
cement paint and what cement coating really is and should be, is 
that paint, t. e., an oil paint, when applied is only a film that lies 
on the surface of the material applied to and has for a vehicle 
linseed oil. Linseed oil is useless as a covering for cement or 
lime because the alkali, acids and lime in cement, stone and brick 
surfaces, and in the air, immediately attack the linseed oil. After 
a year or two, at best, the life is gone and there is only a film 
left, that under a microscope would be found porous, moisture 
going through it. Water and moisture come in contact with the 
lime underneath and the film cracks and peels, so of course paint- 
ing is not necessary on cement, but absolutely useless. But 
something is necessary, absolutely necessary for use on exterior 
cement work. Large concrete structures are veneered with brick 
or tile and lose their identity as cement buildings. Stucco homes 
are built and after a short time the surfaces crack, sometimes large 
patches fall away, they become very discolored; as the surface is 
porous oxides from the air are carried in by the rain, dirt as well; 
water goes through and corrodes the metal lath, and many times 
entirely disintegrates it. On wood lath, moisture swells them, 
nails rust, and the same trouble is had, cracking surfaces. 



* Heath and Milligan Manufacturing Company, Chicago, III. 

(552) 



Digitized by 



Google 



Morse on Cement Coatings. 



553 



The first thing perhaps that comes to mind would be to 
waterproof the dement so as to prevent the moistm-e and rain 
from going in, discoloring the work, corroding the metal lath 
and swelling of wood lath. This is the very thing cement coatings 
are made for and they not only are waterproof but add with it 
the decoration feature, in that most any color can be obtained 
in several shades for selection from natural cement colors to green. 




FIG. 1. — ROUGH CEMENT STUCCO RESIDENCE COATED WHITE. 

Was originally non- waterproof, like adjoining house and now looks like a 
white cement house; does not show painted effect. 

yellow and reds. Many attractive color schemes can be worked 
out, especially on stucco homes. 

From the fact that cement coatings are applied with a brush, 
like unto an oil paint on wood, many of them are called cement 
paints and the majority of people among users and manufacturers 
of cement, call it paint, so it is plain that either cement does need 
painting, or cement coating must not be called paint. A real 
cement coating must be so made and applied so that there is 



Digitized by 



Google 



554 Morse on Cement Coatings. 

penetration into pores of the surface it is applied to and not leave 
a film, otherwise one reverts right back to an oil paint. 

Cement coatings are in a measure going through an experi- 
mental stage, by both manufacturer and user, for some of the 
harder requirements made of them, and too, some cement sur- 
faces will not take a cement coating. For instance, some cement 
floors that are so hard and crystal-like that it is impossible to 
scratch them, while others are soft and porous. Now on the soft, 
porous floor a cement coating will harden and make waterproof 
and prolong the wearing of the surface a very great deal. On 
the hard surface that is not necessary, but the man who has this 
hard floor wants some uniform or decorative color to it, as it 
may be in a public corridor, dining room, etc. Unless the surface 
is treated to make it porous so that cement coating could be made 
to adhere, it would be useless to put it on because it would soon 
wear oflF. This can be done by using a 10 per cent muriatic acid 
solution as a wash to open up the surface so the coating can get 
a chance to adhere and penetrate. 

There have been cement floors so soft that the owners of the 
building would not accept them and wanted the contractor to 
take up an inch or two and lay a new floor (a very expensive job) 
that have been made hard and dust proof by two coats of cement 
coating. There has been one coat applied to an exposed brick 
wall for 7 or 8 years, protecting the brick and mortar joint, still 
in good condition. There have been swimming pools made water- 
tight and enameled with cement coating. A finishing coat of 
enamel in colors to match can be applied where enamel surface 
is desired, and here is a point too of much interest. A cement 
brick wall can be laid up and enameled with two coats of white 
coating and one coat of white cement enamel at a cost of about 
J6 per thousand brick, or $5 per square of 100 sq. ft., adding the 
cost of the brick, say at $10 per thousand, there is a saving of 
about $50 per thousand over the regular enameled brick, and 
at the same time the joints are as well as brick. There have 
been reservoirs holding 600,000 gallons of water, made water- 
proof by two coats of cement coating, on the inside. 

To cite an example of waterproofing done with cement coating, 
one large corporation had built four big cement tanks, measuring 
about 300 ft. around. They were square, 20 ft. deep, to store 



Digitized by 



Google 



Morse on Cement Coatings. 



555 



soft coal in, filling the tanks with wat^r to keep the coal from air 
slacking. The walls and floor after a time cracked and would 
not hold the water. It was impossible to make a superficial 
application on the inside surface, such as the membranous method 
of waterproofing because the coal was taken out by a large grab 
bucket or clam shell, and it would break the surface; it was impos- 
sible to fill the cracks with cement, because to fill perpendicular 




FIG. 2. — ROUGH CEMENT STUCCO OR PEBBLE WASH HOUSE, KENMORE AVENUE, 

CHICAGO, ILL. 

Cracks filled and entire surface covered with olive green cement coating. 
Color is permanent and surface waterproof. 

cracks with cement in a plaster form stiff enough to hold itself 
did not have sufficient water for proper crystalization and bond- 
ing, and would crumble and fall out, and grouted in, it would of 
course run out. These tanks however were waterproofed with 
cement coating, cracks filled with the coating in a heavy con- 
sistency before being reduced and afterwards all surfaces covered 
with the coating applied with a brush. 



Digitized by 



Google 



556 Morse on Cbbibnt Coatings. 

Another feature of cement coating is, that the cracks in a 
stucco house can be filled in the same manner by the use of a 
syringe with a spout about the size of the crack, the entire surface 
afterwards being coated. Cement coatings are all classed and 
described about alike, in that they are made for the waterproofing 
and decoration of cement, stone and brick and stucco surfaces 
and to make cement floors dustproof and stainless. 

The word waterproofing covers a multitude of sins, and about 
the first question is, will a cement coating waterproof a basement 
wall on the inside to stop the water coming through the wall? 
The answer is a very emphatic "no," as it is not made for that 
purpose, no more than Portland cement is for filling teeth, or 
mending chairs or china. Cement coating is to waterproof only 
when the pressure is against it with the wall to back it up, for the 
exterior walls above grade, against rains and snows and moisture 
from the air and for decoration. 

Some two years ago a rough pebble dash house was found 
so discolored and cracked that the owner was unable to sell or 
rent. The surface was so porous that the moisture had corroded 
the metal lath and the iron oxide showed up on the surface. 
This house has all been coated, cracks filled, and made pure white 
and the surface made \yaterproof so it does not show stains or 
discoloration. 

In a recent popular article on concrete homes there is de- 
scribed a house of cement plaster on Ijong Island Sound built by 
one who spared no expense, and the surface is described as cracked 
and apparently moth eaten. Another house near by is one that 
had been erected by a small property owner and is of solid con- 
crete, described as of a not very inviting color or texture. Such 
examples can be seen in any of the suburban towns surrounding 
the large cities, and is a further example of the necessity of more 
educational work among the builders of cement buildings. 

Cement in itself is naturally absorptive and in all building 
work, where there is an exposed cement surface, not only from a 
waterproofing standpoint but from a decorative point as well, 
something is necessary. That something is a waterproof coating 
such as a cement coating. Those who have studied the situation 
carefully, noting the requirements, have been able to obtain 
many excellent results. 



Digitized by 



Google 



Morse on Cement Coatings. 



557 



Cement coating can be applied to cement surfaces, including 
labor and material at about the following costs. On stucco or 
pebble dash surfaces $2.50 per square for the first coat; about 
$2 per square for the second coat. On sanded finished surfaces, 
such as interior cement plastered walls or exterior, this work can 
be done for approximately $2 per square for the first coat and 
$1.50 per square for the second coat. On cement floors one 



Js 


»i^- 


■^^ 1 : ^S^^Tjm^^^^thw 


■■yjyii 




"itI 



PIG. 3. — CEMENT PLASTER RESIDENCE, EVANSTON, ILL. 

All cracks filled and entire surface made waterproof and uniform in color 
with natural cement color coating. 

coat can be applied for about $1.75 per square for the first coat 
and $1.25 for the second coat. Two coats are recommended. 
There are, however, many surfaces where one coat is sufficient, 
as one coat of a real cement coating will waterproof and harden 
the surface, but will naturally flat in differently in places according 
to the porosity of the surface to which it is applied. The second 
coat will bring the surface to a uniform color. 

From the figures given above it can be readily seen that the 



Digitized by 



Google 



558 



Morse on Cement Coatings. 



ordinary stucco house, which is. so universally liked and which 
generally is of a size of approximately 35 to 40 squares, can be 
coated with cement coating for about $200, a very small addi- 
tional expense, as compared with the satisfying results that can 
be obtained. 

In the above article porosity is referred to as a mark of poor 
concrete, the result of either poor cement or of wrong proportions 




FIG. 4, — CEMENT STUCCO RESIDENCE AND CONCRETE SWIMMING POOL. 

House covered with cement coating and swimming pool with white cement 
coating and enamel. 

in mixing. There is no question but that to a big extent this is 
true, not so much as to the statement of poor cement for having 
been identified personally with the cement business, I feel pretty 
sure that all cements that are allowed to be shipped from the 
mill have passed rigid inspection, and are good. Improper mixing 
and lack of sufficient cement is, in a great measure the cause 
of the necessity of waterproofing. A cement manufacturer 
recently said that if a greater proportion of cement were used in 



Digitized by 



Google 



Morse on Cement Coatings. 



559 



all work there would be less need of waterproofing. Another 
thmg necessary is more careful supermtendence of work, as the 
very first principle of waterproofing is density and this can only 
be obtained by putting into the concrete sufficient quantity of 
fine material to fill the voids in the aggregate. The better con- 
crete and cement work is done, the less need there is for cement 
coatings, from a waterproofing standpoint, but cement coatings 




FIG. 5. — CONCRETE FIRE PRESSURE WATER TANK, 600,000 GALLONS CAPACTTY. 

Made watertight with cement coating applied on the inside. 

still add with the waterproofing feature the decorative feature, as 
has been explained. 

A beautiful white cement house can be had by applying pure 
white cement coating with a brush to a cement house that has 
been discolored and also it can be made soft shades of green or 
buff, all with lime proof colors and permanent. 

In applying cement coatings it is important that the surface 
is free from dirt and grease and that it is porous, and that the 



Digitized by 



Google 



560 Morse on Cement Coatings. 

cement coating is used and applied in a thin enough consistency 
to penetrate into the work. Cement coatings make cement work 
more pleasing to look at, and at the same time protect it from 
atmospheric conditions that soon plays havoc with it, especially 
plaster and block work. 



Digitized by 



Google 



I3ISCUSSION. 



Mr. L. C. Wason. — Mr. Morse spoke of waterproofing tanks. Mr. Waaon. 
I would like to know what pressure, how high a head of water it 
withstood. 

Mr. F. J. Morse. — ^This tank is one built for a fire pressure Mr. Morse, 
tank for the International Harvester Company, holding about 
6,000 gallons of water and 28 ft. in depth. They were not able to 
keep a good pressure. That is the only test that we have had on 
cement coatings, except in tanks used for storing coal. 

Mr. C. W. Boynton. — It seems to me that some of the cases Mr. Boynton. 
referred to by Mr. Morse should be explained more fully. I do 
not believe that poor stucco work can be saved by painting over 
the surface with a cement coating. Proper workmanship and 
proper materials must be put into structures and one should be 
careful not to be misled to believe that something beautiful can 
be made out of stucco work that is fundamentally bad. 

The commercial success of any cement coating is probably 
dependent very largely upon poor workmanship in concrete con- 
struction, or failure on the part of the builder to get the satisfactory 
results and pleasing eflFects which can be obtained in many ways 
without the use of such material. I do not want to be understood 
as taking a position against all coatings, waterproofing materials, 
etc., but I do feel that as cement users we should try to make the 
best possible use of the material with which we are working. Lack 
of knowledge as to how to handle concrete and mortar, lack of 
interest in trying to learn, and lack of business energy and fore- 
sight among cement users have resulted in something like sixty- 
five waterproofing and coating compounds for cement work being 
put on the market in the last four years. It takes money to keep 
these businesses going. The cement user and his customers are 
furnishing it. Just as soon as we learn to get anything like as 
much out of the materials with which we are working, as there is 
in them, this waste of money will stop. I believe, especially for 
the coatings, there is a field, but it must be developed along the 
line of imparting to concrete and mortar work some property or 

(561) 



Digitized by 



Google 



562 Discussion on Cement Coatings. 

Mr. Boynton. element of beauty which is not attained by the efficient handling 
of the materials. It is not sunmsing that the coating and water- 
proofing business as a whole is looked upon very largely as a para- 
site, for its development has depended to a great extent upon poor 
work, and so long as poor work continues, the demand for these 
things will continue; also the fact that poor work many times can 
be temporarily covered up by the use of such materials tends to 
encourage their use. This Association stands for quality of work 
and that is what we all should strive for. 

Regarding the treatment of floors, there are many poor 
floors and we have not made much headway in solving the prob- 
lem. In time we will understand the reason for dusting and vnll 
be able to write a specification, which, if followed, will assure a 
dustless floor. I base this statement on the one fact that dustless 
floors are laid today. I am confident that when men who are 
actually laying floors become interested and carefully note the 
conditions which produce dusting, and those which prevail when 
dusting does not result, we will be nearing the time when dusty 
floors are a thing of the past. I am glad to know that until such a 
time there are treatments which will keep down the dust, or 
possibly make the floor dustless. 

Mr. Morse told us that we could not use a grout to repair 
cracks in a concrete wall because it would run through. Certainly 
his coating must be near the consistency of a grout, yet it seems 
to stay. I believe that if mortar had been mixed to the proper 
consistency and applied in the proper manner it would have 
stopped the cracks just as effectively. 

Another thing Mr. Morse referred to was the making of denser 
concrete by the use of fine aggregates. The government is about 
to publish a bulletin (Technologic Paper 3, Bureau of Standards) 
which shows that the densest concrete is obtained with aggregates 
containing a very small amount of fine particles. 

Doubtless Mr. Morse's coating and other coatings have their 
place, which is in decorating a well built structure, and not in 
covering up the defects in a poorly built one. 
Mr. Morse. Mr. Morse. — I think I can answer in just a few words, as 

stated in the paper, that if better concrete was made, or better 
superintendence, there would be very little use for cement coatings. 
Cement coating is not for good cement work; it is for cement 



Digitized by 



Google 



Discussion on Cement Coatings. 563 

work that is not made good. I think I tried to leave that impres- Mr.Mone. 
sion with you. 

Referring to the remarks about a part of a stucco house 
falhng out. It was my understanding that stucco work was pebble 
dash work, as we know, which contracts and expands. It is a 
fact that the moisture beating through a stucco wall that is not 
waterproof, continually keeps up a corrosion of the metal lath, 
and the same entirely disintegrates. It is at such places where the 
wall falls out, and I do claim that a waterproof cement coating 
applied to that wall would prevent expansion and contraction to a 
large extent and the corroding of the metal lath. 

Another question raised was filling the cracks with cement. 
A perpendicular crack with no bottom, a crack in any concrete 
wall, for example, or stucco wall, it is impossible to grout with 
cement. The cement will keep running out. Using a plastic 
cement stiff enough to hold, in setting the two sides of the cement 
will withdraw or dehydrate, taking the water out of this cement 
and not enough water remains for proper crystallization. That 
has been my personal experience. The cement coating is of a very 
heavy consistency, and put into a perpenilicular crack will harden 
there perhaps better than cement. In this particular work cement 
had been tried and failed, and the grouting was a success with 
cement coating. 

The question of taking care of cement floors is one that has 
caused a great deal of trouble. With a hard floor that is very 
crystalline, you refuse to put anything on it, but on a floor that 
dusts up it must be covered. 

The placing a cement floor that does not dust is to my mind — 
we are all looking for information in this work — almost impossible, 
especially for a man to guarantee or assure a floor will not dust. 
There is a psychological moment to trowel a cement floor which 
can only be determined in the laboratory, and that is immediately 
after the initial set has taken place. The action of troweling a 
cement floor forms a suction and that draws the fine particles to 
the surface. Now if the troweling is done after the initial set has 
taken place it breaks up the crystallization being formed and 
dusting of the surface is probable. 

As to the outside cement surface in reinforced concrete build- 
ings, I have seen some work that has been rubbed down with 



Digitized by 



Google 



664 Discussion on Cement Coatings. 

Mr. MorM. carbonindum stone in panel effects that are beautiful, and it would 
be a crime to apply a cement coating to them. 

Mr. Boynton. Mr. C. W. Boynton. — I think Mr. Morse covered the whole 

thing when he said cement coatings are not necessary in good 
work. I most heartily agree with him. This Association, if it 
stands for anything, stands for good work. It is exactly what we 
are aiming at and what we are coming to. However, I am glad 
that Mr. Morse and his company stand ready to throw themselves 
into the breach and help reclaim the poor jobs. 

The Preddant. The PRESIDENT. — There is probably no more important field 
than the subject of a color of some kind that can be applied to a 
concrete surface so as to be permanent. Most colors are more 
or less injured by the cement itself. They leach out so that after 
a very short time the color is gone. The Germans have a way of 
treating the concrete with an acid wash, which neutralizes the 
surface, and on this surface durable color can be placed. The 
dampness that is often found on the interior of concrete walls is 
in large measure overcome, and I have never seen damp walls 
on the interior of foreign buildings. They are generally insu- 
lated in some way with air spaces, so that the dampness with 
which we are troubled in a great many cases is lacking in those 
structures. Certain it is that the treatment of the interior so as 
to produce pleasing color efifects is one of the most important 
steps that must be considered. 

It is perhaps easier to tint the interior of a structure than it 
is to tint the exterior, and some experiments which have been 
made by the Committee on Treatments of Concrete Surfaces 
have shown that the various pigments, when they are put out in 
the weather, soon lose their color and coherence and practically 
become worthless. So that the whole subject of the treatment 
of the material, the conditions to be observed in applying color to 
concrete, is very important and very difficult. 

The tendency to use selected aggregates as a color and to 
secure color efifects through the material itself, I think, is prob- 
ably the most hopeful field. The use of white cement and an 
aggregate which possesses color value will give tints that cannot 
be duplicated with ordinary pigments. Perhaps the best example 
of color effect is to be found in the Connecticut Avenue Bridge in 
Washington, D. C, where the Potomac River gravel concrete of 



Digitized by 



Google 



Discussion on Cement Coatings. 565 

buff color is in contrast with the concrete which is made out of The Preddent. 
the gneiss rock of bluish color found on the site. The spandrel 
walls, etc., in buflf are in contrast to the archrings and piers in a 
light blue color; and the color is distinctly visible for a consider- 
able distance. These colors of course are permanent. The most 
permanent colors, as far as the exterior exposure is concerned, are 
due to an aggregate which is of itself permanent. That is, red 
can be obtained by granites and some forms of hard sandstone, 
which give us a permanent color effect, blue from traps and lime- 
stones, yellow from sandstones and gravels. 

The great difficulty in securing color effects in concrete and 
mortar is that the cement itself has a muddy color, and perhaps 
the sand is also of a dirty color, resulting in a mottled appear- 
ance, because it is difficult to add enough coloring material to 
overcome the muddiness produced by the sand and cement. 
The use of white cement and white or light sand makes it possible 
to add a very small percentage of color or to use an aggregate 
with color and secure the desired color effect. The study of this 
subject and the grading of the material and selecting it with the 
view to securing color effect is one of the developments which 
will characterize our future work. 



Digitized by 



Google 



REVIEW OF THE PRESENT STATUS OF 
IRON ORE CEMENT. 

By p. H. Bates.* 

It is generally understood that there are three essential con- 
stituents in Portland cement, namely: lime, alumina, and silica. 
It has been shown experimentally that the lime can be entirely 
replaced by strontium or barium, the alumina in part by iron or 
chromium oxide, and hydraulic materials of considerable value be 
obtained; but none of these have been suggested as being of 
commercial value with the exception of iron oxide. 

There exist very extensive deposits of low grade ores in which 
the percentages of iron are too low to admit of their economical 
reduction in the blast furnaces. There are also numerous deposits 
of high iron oxide clays in which the high percentage of this con- 
stituent reduces the value of the clay. Both of these have appealed 
to the cement manufacturer but to a slight degree, owing largely 
to abundant supplies of clays of low iron content, from which it is 
comparatively easy to produce cement of normal composition. 
The manufacture of a cement of high iron content presents some 
difficulties in view of the low fusion point of the raw material, 
there being very readily produced the so-called — from the point of 
view of the manufacturer — overbumed cement. 

There had been noticed in several localities abroad, the failure 
of concrete made from cement of normal composition when placed 
in sea water. Whether this was due to faulty cement is an unset- J 

tied question. But various authorities on casting about for an | 

explanation of these failures quite unanimously came to the con- 
clusion that the trouble was due to the alumina in the cement. 
This constituent combines with the sulphuric anhydride of the sea 
water and the lime, freed in the process of the setting, to form a 
compound of indefinite composition. This latter crystallizes with 
a large water content, which pushes apart the cement particles 
and ultimately destroys the structure. 

Necessarily the first remedy which suggested itself was the 

♦Chemist, Bureau of Staodards, Pittsburgh, Pa. 

(566) I 

I 
I 
I 

Digitized by VjOOQIC j 

I 



Bates on Iron Ore Cement. 567 

replacement of the alumina with some material which would not 
have this property. Michealis, reviving the original suggestion of 
Malaguiti and Durocher, called particular attention to the re- 
placement of the alumina with iron oxide. Attracted by Michaelis' 
arguments, the Krupps in 1901 took patents covering the manu- 
facture of such a material. 

There is no doubt but that the alumina present in cement 
will form a compound with the sulphuric anhydride radicle of 
soluble sulphates and with lime hydrate. But the exact nature of 
this compound is still much in doubt; just as doubtful is the ques- 
tion whether the compound can result from the setting of all the 
alumina constituents which may be present in cement. It is very 
generally agreed, however, that the iron compounds which are 
present in a cement cannot form such a decomposition product, but 
whether the iron is combined in the cement in a manner 
similar to the alumina is a very doubtful matter. The latter 
problem is complicated by the impossibility of securing materials 
for investigation which are so sufficiently free of alumina as not 
to be decidedly influenced by the presence of small quantities 
of the latter. Thus the Geophysical Laboratory has shown that 
0.5 per cent, of alumina present in a silica lime mix of such a com- 
position as to give tri-calcic siUcate, will give the maximum yield of 
this compound. It is quite likely that those who claim the iron 
compounds in cement have hydraulic properties, have been mis- 
led by the influence exerted by the very small quantities of alu- 
mina which have been present in these materials. Without doubt 
the iron compounds lack hydrauUc properties entirely, and the pos- 
sibility of making a cement entirely lacking alumina will never be 
attained. The iron ore cements must have alumina present or else 
they are of no value, and in the finished cement the iron oxide is of 
no other value than any other material which might be present 
which would reduce the alumina content. This statement does 
not deny the value of the iron oxide as a flux in the manufacture. 

It is therefore not surprising that the manufacture of this class 
of cement has not grown, even in Germany, as was hoped for. Its 
use, even in sea water, has not kept pace with the growing use of 
the "Eisenportlandzemenf (or "Slag-Portland'^ cement) or the 
use of natural slag (Puzzuolana) in connection with PortUnd 
cemejit. The theory of the use of the latter is based on the fact 



Digitized by 



Google 



568 Bates on Iron Ore Cement. 

that the lime which is set free in the setting of the cement, will 
combine with the silica of the slag or puzzuolana, thereby rendering 
less lime available for combining with the sulphuric anhydride 
radicle of the soluble sulphates, and consequently less possibility 
of the forming of the injurious sulpho-aluminate of lime. 

Notwithstanding the discussion and notices which this class 
of cements has received in the technical journals, there is little 
authentic data available showing its relative superiority over 
cement of normal composition. Thus, Michaelis, Jr., quotes from 
the ZentrcdblaU der Bauverwaltung in the Cement and Engineering 
News of March, 1911, the following compressive strengths of 
an iron oxide and Portland Cement in the form of 16-inch cubes: 



I I Compreaoiva 

Paris of Cement. i Parts of Sand and Cniflhed Stone. ' strength. 

I . (lb. persq.in.). 



Iron ore 2 

" " 2 

it tt 4 



Portland 2 



3 ' 6776 

3 6476 

6 3120 

6 , 2910 



3 4690 

2:3 6146 

4 6 I 2416 

4 ' 6 I 3270 



The results were obtained by the Department of Public Works 
Berlin at Swinemiinde and represent the strength after three years 
immersion in sea water. While Michaelis concluded from these 
results that the iron ore cements are the stronger after such treat- 
ment, these conclusions are hardly justified, since the above may 
represent the relative strength before treatment. However, the 
description of test pieces, both at this point and at other localities 
along the German sea coast, undoubtedly shows the superiority of 
low alumina cements. 

Quite a number of tests of iron ore cements are given in the 
Toninduatrie-Zeitung (1906, page 1968). The comparison of them 
with normal composition cements was carried out entirely by the 
use of small prismatic test pieces which were placed in the sea water 
and their appearance noted. Another series was conducted in 
which different amounts of plaster were added to the cements from 
which the prisms were made. In every case the resistance of the 



Digitized by 



Google 



Bates on Iron Ore Cement. 569 

iron ore cements to the sea water was greater than that of the Port- 
land cements. Unfortunately, comparison by means of tensile or 
compressive strengths was not made. E. Maynard in the same 
joiunal (1910, page 651) conducted somewhat similar tests; he 
however showed the slight change in chemical composition which 
the iron ore cement underwent by such treatment, but does not 
give results for Portland cement. 

In France there has been little investigative work carried on; 
however, Candlot* gives some interesting results obtained with five 
high iron oxide cements manufactured by himself. The partial 
chemical analyses and the strength of the neat and 1 to 3 sand bri- 
quettes at the end of the 7 and 28 day and 5 year periods, after im- 
mersion in water, are: 

No. 1. No. 2. No. 3. No. 4. No. 6. 

SiO, 19.0 21.0 26.0 24.6 26.0 

AljOi 6.8 5.3 5.3 6.3 6.4 

Fe^, 7.2 6.2 8.8 7.2 6.6 

CaO 66.6 66.0 69.0 61.8 62.6 



98.5 97.5 99.0 98.8 98.5 



Tensile strength, pounds per square inch. 

No. 1. No. 2. No. 3. No. 4. No. 6. 

Neat Briquettes. . 7 days 951 845 142 220 426 

28 " .... 1,036 - 913 213 668 724 
6 years.... 653 824 894 769 717 

1:3 Sand 7 days.... 356 511 92 234 356 

28 " .... 398 525 199 398 497 
5 years.... 678 710 646 724 667 (strength at 

end of 3 yrs.) 

The author calls attention to the high lime of number 1 and 
mentions that the pats in steam disintegrated, although the bri- 
quettes showed no such signs; number 2 has a silica and lime 
content corresponding to a cement of normal composition; both of 
them show the retrogression in strength of the neat cement bri- 
quettes, after the 28-day period, so characteristic of cements of 
normal alumina content. The three other cements are really high 
silica cements and show the characteristics of this class very 

* h9 Ciment, May, 1011. page 32. 



Digitized by 



Google 



570 Bates on Iron Ore Cement. 

strikingly in the tensile strength — namely the very low initial 
strength followed by satisfactory gains. 

The effect of increasing the silica content with corresponding 
decrease of the lime content is shown by numbers 1 and 3, and 2 
and 5. Numbers 3, 4 and 5, which show decreasing amounts of iron 
oxide, show increasing strength at the early periods. However, 
these three cements show too much variation of the other constit- 
uents to attribute this gain to decreasing iron oxide. 

In this same paper Candlot gives the partial chemical analyses 
and tensile strength, at early periods, of some silicious cements 
which he manufactured. 

No. 1. No. 2. No. 3. No. 4. 

SiO, 26.0 25.0 27.0 26.0 

AltOa '. 2.6 1.7 3.4 2.0 

CaO 67.0 66.0 65.0 67.0 

95.5 92.7 95.7 95.0 
Undetermined 4.5 7.3 4.3 6.0 

Tensile strength, pounds per square inch. 

No. 1. No. 2. No. 3. No. 4. 

Neat Briquettes.. 7 days 899 894 692 777 

28" 965 909 885 866 

1:3 Sand 7 days 601 398 426 440 

Briquettes 28 " 667 635 586 535 

These strengths hardly correspond to those of such high silica 
low alumina cements. In view of the large percentage of unde- 
termined shown in the analyses, it would appear as if there were 
present some other constituent, not determined, which had caused 
these abnormally high strengths. In the discussion following the 
reading of the paper, Candlot acknowledged the use of a flux — 
"generally iron oxide" — in their preparation. We have here con- 
sequently a group of high silica, relatively high irqn ore cements, 
which are of much interest. 

No information is given in regard to the superiority of these 
several cements of Candlot's over normal Portland cements, when 
immersed in sea water. They do seem to show the high strength 
which it is possible to reach with high silica cements and usually 
the steady gains in strength, Thege high silicious cementa 



Digitized by 



Google 



Bates on Iron Ore Cement. 571 

should be receiving more attention by the investigator, the manu- 
facturer and the consumer. That Portland cement is overlimed 
and has principally but the ease of manufacture and ease of han- 
dling by the consmner to recommend it, is a growing conviction. 
The latter statement is substantiated by the increasing use of 
trass, tufa, etc., which use would be impossible were not normal 
Portland cement overlimed. 

Until within a very recent period there has been no attempt in 
this country to manufacture a high iron Portland cement and place 
it on the market as such. There have been on the market for a 
number of years, however, two cements, manufactured here, which 
are of exceptional high iron oxide content, although this fact has not 
been used by their manufacturers for recommending them for other 
than the ordinary purposes. In the two cements referred to the 
percentage of iron oxide is little less than the alumina. 

There has been just as much inattention in this country 
devoted to the investigating of this class of cement as there has been 
to its manufacture. About a year* ago there was conducted a series 
of tests at the University of Illinois in which it was desired to 
show the superiority of this class over that of normal composition 
cements for sea water use. The high iron cement used was manu- 
factured in the Ceramic Department of the University, and was 
compared with a normal Portland cement secured on the market. 
According to the tests reported, which consisted of making pats 
and briquettes of the two cements and treating with sea water 
under pressure and observing the disintegration and loss of strength, 
the former was superior to the latter. But other burnings of high 
iron oxide cement made here gave the opposite results; the author, 
however, explains this by improper composition of his burnings, 
so that on the whole his results are not very conclusive. He 
however did find that alumina was absolutely essential; a burning 
in which it was entirely replaced by iron oxide did not give a 
material which could be called a hydraulic cement. 

The only other extensive series of tests made in this country 
are those which were carried out at the Atlantic City Laboratory 
of the Technologic Branch of the U. S. Geological Survey, now a 
part of the Bureau of Standards. 

In Table I are given some of the results obtained, using not 

• See p. 597. Ed. 



Digitized by 



Google 



572 



Bates on Iron Ore Cement. 



Table I. — ^Tensile Strength in Pounds per Square Inch op Neat 

AND 1 : 3 Sand Briquettes of Various Cebients When Immersed 

IN Fresh and Sea Water. 



Slag Cbme>jt (E). 



Natural Cement (I). 





Neat Briquettes. 

1 . . 

Fresh , Sea 
Water. | Water. 


4 weeks .... 
13 " . . . . 
26 " . . . . 
62 " . . . . 


..' 549 1 686 
658 730 

. . 1 634 662 
422 432 



Neat Briquettes. 



Fresh 
Water. 



282 
368 
378 
366 



Sea 
Water. 



302 
368 
236 
283 





Fresh 
Water. 


Sea 
Water. 


116 
196 
182 
260 


Ill 

166 

99 

208 



Mixture, Natural and Portland Cements (J). Portland Cement (K). 





1 Neat Briquettes. 


Sand Briquettes. 


Neat Briquettes. 




Fresh Sea 
Water. Water. 1 


Fresh 
Water. 

247 
201 
267 
218 


Sea 
Water. 


Fresh 
Water. 


Sea 
Water. 


4 weeks .... 
13 " . . . . 
26 " . . . . 
62 " . . . . 


221 1 258 
216 311 
204 286 

. . 1 260 244 


190 
256 
318 
276 


866 
910 
830 
463 


1100 
927 
900 
923 



Portland Cement (L). Portland Cement (M). 



4 weeks 
13 " 
26 " 
62 " 



Fresh 
Water. 



595 
557 
550 
407 



quettes. 


Neat Briquettes. 


Sea 
Water. 


Fresh 
Water. 


Sea 

Water. 


623 
607 
673 
233 


748 
630 
610 
607 


837 
799 
670 
277 





Slag Cement (F). 








Neat Briquettes. 


Sand Br 

Fresh 
Water. 


iquettes. 




1 

Fresh 
Water. 


Sea 
Water. 

468 
512 
348 
210 


Sea 
Water. 


4 ^fkeks 


416 1 


210 
306 
246 
336 


180 


13 " 


511 


241 


26 " 


590 ' 


268 


62 " 


. . . . ' 636 1 


280 









Digitized by 



Google 



Bates on Iron Ore Cement. 



573 



Special Low Alumina Portland 
German Iron Ore Cement (A). Cement (B). 





Neat Briquettes 


Sand Briquettes 


Neat Briquettes 

Fresh Sea 
Water. Water. 


Sand Briquettes 




Fresh 
Water. 

535 
714 
726 
764 


Sea 
Water. 

702 
746 
802 
656 


Fresh 
Water. 

120 
221 
234 
256 


Sea 
Water. 

134 
166 
236 
204 


Fresh 
Water. 


Sea 
Water. 


4 weeks 


1 
764 714 
803 554 
800 622 

768 1 748 


242 
311 
418 
370 


252 


13 " 


355 


26 " 


320 


52 " 


310 







Portland (C). 

Neat Briquettes 



Typical Portland No. 47 (G). 



4 weeks 
13 " 
26 " 
62 " 



Fresh 


Sea 


Water. 


Water. 


672 


748 


744 


808 


708 


768 


636 


764 





Sand Briquettes 


Fresh 


1 
Sea Fresh 


Sea 


Fresh 


Sea 


Water. 


Water. 1 Water. 


Water. 


Water. 


Water. 




604 


821 


268 


206 




1 636 


136 


300 


238 




, 658 


#4 


304 


258 




674 


• * 


350 


205 



Typical Portland Cement No. 47 (G). 



Specimens stored in damp ] Specimens stored in damp 
|Clo8et*for 24 hours; then for closet for 24 hours: then for 
2 days in fresh water. 6 days in fresh water. 



Neat Briquettes. 



r 



Neat Briquettes. 





Fresh 
Water. 


Sea 
Water. 


1 Fresh 
1 Water. 


Sea 
Water 


4 weeks 


672 
765 
713 
622 


959 
904 
852 
738 


700 
723 
734 
660 


926 


13 " 


840 


26 " 


940 


52 " 


598 







White Portland (H). 



French Portland 
Cement (D). 




Sand Briquettes. 



Neat Briquettes. 



Sea 
Water. 



Sea 


Fresh 


Water. 


Water. 


164 


539 


200 




162 


702 


204 


776 



511 

656 
706 



* A second lot of briquettes gave substantially the same results. 
** Broke in handling, before placed in testing machines. 



Digitized by 



Google 



574 Bates on Iron Ore Cement. 

only this class of cements but in addition for the sake of compari- 
son, Slag, Natural, White, Slag-Portland, and normal composi- 
tion cements. Among the latter is a well-known French Portland 
cement which while of normal composition is highly recommended 
and rather extensively used for sea water concrete. Its distinctive 
characteristics are the low sulphuric anhydride content and its 
decided coarseness, 21.3 per cent, being retained on the 100 sieve 
and 38.8 per cent, on the 200. Unless otherwise noted, all the 
briquettes were stored for 24 hours in the damp closet before 
immersion in the waters. In Table II is given the chemical 
analyses of all cements, the tensile strengths of which are reported 
in Table I. 

An examination of Table I shows that the neat briquettes 
made of Portland cement of normal composition retrograde 
considerably in strength after the 13 weeks period. In this 
class have not been placed the cements K and C." since both 
of them show rather high iron oxide content, although they 
could hardly be called iron oxide cements. Cements L and M 
were simply chosen at random from standard Portland cements, 
while the typical Portland is a mixture of equal parts of a large 
number of standard Portland cements. The efifect of aging, even 
for a short period in fresh water before subjecting the cement to 
the action of sea water, is very closely shown, and suggests a 
practice which should be observed, when at all possible, in all 
cement used in sea water. The rather remarkable behavior of 
the White Portland, in view of the theory under which high iron 
oxide cements are used, is striking and unexplainable. The data 
on E has been given largely because this cement, if we are rightly 
informed, belongs to that class of cements called by the Germans 
"Eisenportland" cements, which are mixtures of slag with Port- 
land cement, though contrary to the German practice more slag 
has been used in its manufacture than Portland cement. This 
cement, while it shows retrogression, does not show the decrease 
in strength which the true slag cement does. 

The sand briquettes of all the classes compare rather favor- 
ably, and it would be difficult to form an opinion of the relative 
value of any class from the behavior of the mortars alone. With 
the exception of the Cement J, all the sand briquettes show less | 

strength in sea water than in fresh water at the end of a year; ' 

I 
I 
I 
I 



Digitized by 



Google 



Bates on Iron Ore Cement. 



5^5 



5 



I 

s 

OQ 

s 

o 

B 



s 

5 

H 



s 


fe 


CO 05 CO 


s s 


1^ t^ 

CO »o 


»0 lO 




2 


?s 


« «5 g 


C^ rH 






• 


8 


hq 


3 


2 S S 


g s? 


s; § 


S 8 


8 : 


s 




« t- g 


CO i-H 








8 

1-H 


w 


S§ 


R f: ^ 


g 5 


s » 


S? g 


g : : 


s 


1-H 


n to m 


rH rH 








s 




s 


^ 8 g 


S J2 


s s 


?5 !:: 




rH 
rH 




s 


c « gj 


c3 *^ 


c^ 

^ s 


O 

i-H 




8 






S 5? So 


: § 


gg : : 




8 


rH to 1-t 
CO 


1-H 






»-H 


8 


w 


s 


18 S « 


2 S 


5 3 


!§ 3 


S : : 


^H 

o 


g5 


«> g 


»H 1-H 


2 ^ 




rH 


§ 


o 


s 


8 2 S3 


S§ 2 


g : 


S5 


1-4 


M X C^ 

CO 


CO fi 






• • 


8 


E^ 


05 


S i 2 


t 2 


§ S 


to 


§5 : S 


CO 


s 


-" 3 § 


<N rH 






CO • <N 


8 



» 



f: 8 fe 


& 


5 




1-i 


• o 


S 


: S 


rH 


S 2 


^ 


Tji 


r^ 




<N 


M 


• M 


8 

1-H 



o 



PQ 



s 


rH 


00 

1-H 


s 


s 


5 


gs 


3 




1-H 


o 

rH 


§5 


(N 


t>. 


3 


rH 










1-H 


8 
































1-H 


rH 


S 


£2 


l> 


22 


t^ 


r^ 


X 


Ok 


Csl 


1-H 


U^ 


■^ 


X 


Oi 


Tf 


»H 


Tji 


X 


X 


-^ 


S 


rj* 


W3 


1-H 

CO 




rH 








1-H 


s 






















T^ 


CO 




S 


9 


X 


rH 


s 


s; 


o 


8 : 


n 


o 


X 


CO 


rH 




1-H 


1-H 




r-i 


r^ 


8 


(N 






CO 















4 


? 


g 


i 


g 


S 


S 


g 






S 3 


1-H 


s 


I> 


CT 


iH 

cc 


rH 












8 






















»-< 












































o 


























»o 


























O 










6 


■1 


: ^ 


: <2 


:2 


D n 


• 1 


• S 






■!^ 


^ 




w 


1^ 


< 


u 


i^ 


cc 


Iz; 


W 


w 




^ C 


^ 





Digitized by 



Google 



576 Bates on Iron Ore Cement. 

those of cements I, F and White at 1 year do not show a retro- 
gression since the previous period of testing. 

The relative strength in compression of concrete made from 
German iron ore cement and normal Portland cement is shown 
in Table III. All specimens were stored for 3 weeks in a damp 
closet before immersion in the water, and while the concrete 

Table III. — Comparison of Comprbssive Strengths of Concrete 

MADE FROM AN IrON OrE AND A PORTLAND CeMENT. 

German Iron Orb Cement (A). 

Compressive strength in pounds per square inch (1 part cement, 2 parts 
sand, 4 parts trap rock). 

Stored 8 weeks in damp closet before immersion in — 

Fresh Water. Sea Water. 

13 weeks, 3743 3477 

26 " specimen 3987 A B 

52 " B B 

A. Two of three specimens did not break under a load of 4237 pounds, 
capacity of the testing machine. 

B. All specimens did not break under a load of 4237 pounds, capacity 
of testing machine. 

Typical Portland Cement No. 47 (G). 

Compressive strength in pounds per square inch (1 part cement, 2 parts 

sand, 4 parts trap rock). 

Stored 8 weeks in damp closet before immersion in — 

# 

Fresh Water. Sea Water. 

13 weeks, 3190 3877 

26 " 3457 3979 

52 " 3389 4060 

from the former cement is stronger than the latter, yet it is not 
very decidedly marked. At all periods the Portland shows 
greater strength in sea water than in fresh water, whereas, at the 
early period the Iron Ore shows a less and apparently at the later 
periods about equal strengths in the two waters. 

These abstracts from the data of the Atlantic City Laboratory 
show in a general way the relation of the strengths of the Iron Ore 
to various other cements. Again, it cannot be positively stated 



Digitized by 



Google 



Bates on Iron Ore Cement. 577 

that the former show much superiority over the latter classes of 
cements. The information to be obtained in foreign publications 
is usually of a descriptive character and is concerned but little 
with numerical data, which of necessity are the only positive and 
reliable data. In this respect the information and data are much 
different from that obtainable concerning the "Eisenportland" 
cements, and if one is to judge of the use of the latter from the 
numerous investigations which have been and are still being 
carried on in regard to its general use and use in sea water concrete, 
it must be a very important and valuable product, far more so 
than the "Erzzement" or Iron Ore cement. 



Digitized by 



Google 



MARINE OR IRON ORE CEMENTS. 
By Herman E. Brown.* 

As soon as Portland cement became an accepted building 
material, engineers and chemists commenced investigations as 
to its behavior under the varying conditions to which it was 
exposed. The object of these investigations was two-fold — ^to 
assure themselves of the permanency of the structure already 
erected, and to find the outer limits of the field in which it could 
be used. 

As early as 1854 Malaguti and Durocher (Campt Rendvs, 
Vol. 39, p. 183) were investigating the cause of disintegration of 
concrete masses in sea water. As a result of such investigations 
on commercial hydraulic cements, and on synthetic cements 
manufactured by them, they state: "The inertness of the oxide 
of iron in the hydraulic materials appears demonstrated by these 
synthetic tests. We are forced to conclude that the presence of 
this oxide is able to enhance the stability of cement mortars in 
sea water." 

Malaguti and Durocher had been inspired to make these 
investigations on account of the discoveries of Vicat in 1840, 
who had found that a certain hydraulic cement, after six months 
in sea water, had been greatly damaged, an analysis of the cement 
showing that the magnesium had been largely increased and the 
calcium oxide decreased by about the same amoimt. 

In 1882 Landrin makes the following statement: "The 
calcium aluminates in cement are extremely harmful for perma- 
nency of mortars in sea water, because they dissolve easily." 
(Thon IndiLstrie Zeitungy 1882, p. 177.) Other prominent investi- 
gators now began to take a lively interest in the search for the 
exact cause of the occasional sickness of concrete structures in 
sea water. Le Chatelier and his co-workers commenced a series 
of painstaking experiments. In 1890 Le Chatelier writes: "The 
double sulphate of calcium and aluminum plays an important 



* Chief Engineer, American Cement Engineering Company, New York, N.Y. 

(578) 



Digitized by 



Google 



Brown on Marine or Iron Ore Cements. 579 

role in the destruction of concrete in sea water." Candlot in 
1892 discovered the double salt of calcium and alumina. In 
1899 commissions were appointed by the German and French 
cement manufacturers for the purpose of studying the action of 
concrete in sea water. The scientific and literary heirs and assigns 
of the said commissions are still investigating, and sometimes 
reporting. 

In 1892 Michaelis, in a paper read at London to the Society 
of Civil Engineers, made the following statements: "From the 
chemical point of view, cements or hydraulic limes rich in silica 
and as poor as possible in alumina and ferric oxide should be used, 
for aluminate and ferrate of lime are not only decomposed and 
softened rapidly by sea water, but they also give rise to the forma- 
tion of double compounds, which in their turn destroy the co- 
hesion of the mass, by producing fissures and swelling." (A. S, 
C. E., Vol. 107, p. 375.) 

Schuljatschenko, Eger, Rebuff at and Vicat did notable work 
on the complex questions involved in their attempts to solve the 
difficulty. The most convincing work was done by Le Chatelier, 
who, some time previous to 1900, made several synthetic cements 
and subjected these cements to the action of sea water. These 
cements were manufactured in sufficient quantities to carry out 
long-time tests by actual inmiersion of the test pieces. Portland 
cements were manufactured, consisting of silica, alumina and 
calcium, in which the alumina varied from 4.1 per cent, to 16§ per 
cent. Iron oxide cements were made, in which there was no 
alumina present, and consisting only of silica, ferric oxide and 
calcium oxide; in these, the percentages of iron ran as high as 14 
per cent. He also manufactured cements in which the oxides of 
manganese, cobalt and chromium were used, as substitutes for the 
ordinarily occurring alumina. In 1900, 1901 and 1902 he reported 
before various scientific bodies on the action of sea water on these 
special cements. As a result of his investigations, Le Chatelier took 
the position that the most important, if not the only, cause of 
disintegration of cement in sea water is the formation of calcium 
sulpho aluminate, and that the iron oxide cements were much 
more resistant to sea water than the alumina cements. (See 
Reports of Paris Congress of Testing Materials, 1900; Buda 
Pest Congress, 1901; and Than Industrie Zeitung, 1902, No. 11.) 



Digitized by 



Google 



580 



Brown on Marine or Iron Ore Cements. 



From 1902 to the present, publications and controversies 
have been frequent and confusing. 

In the fall of 1909, the American Cement Engineering Com- 
pany erected an experimental Portland cement plant, two views 
of which are herewith presented — Fig. 1, a general view of the 
entire plant, and Fig. 2, a closer view of the kiln. 

The plant consists of a rotary kiln 20 ft. by 2 ft., set at an 
inclination of f in. per ft., revolving at 2 turns per minute. The 




PIG. 1. — INTERIOR OF LABORATORY. 



kiln has a stack, 1.5 ft. in diameter by 18 ft. high; thickness of 
kiln shell, J in.; lining — Harbison- Walker Special Aluminous fire 
blocks, 3 in. thick. By means of counter-shafting and a variety 
of different sized sprockets, several changes on the speed of the 
kiln can be quickly made. The slowest speed revolves the kiln at 
such a rate that the material requires somewhat over an hour to 
pass from feed to discharge end. It was found that a speed which 
would permit of the material passing through the kiln in one-half 



Digitized by 



Google 



Brown on Marine or Iron Ore Cements. 



581 



hour was entirely satisfactory from the standpoint of quality of 
clinker produced, and output. The kiln is provided with a No. 6 
Rockwell fuel oil burner, complete with connections, and uses 
crude oil, or distillate which is sent into the atomizing line under 
a pressure from the pump of about 30 lb. per sq. in. A small 
three-piston Gould power pump was first installed, but as the 
valves in this pump gave considerable trouble because of residues 
in the oil, this pump was replaced by a single-acting piston 
power pump, of a capacity at least four times greater than was 




FIG. 2. — EXPERIMENTAL KILN. 

required. A steady control of feed was secured by valves in 
the feed pipe, the excess oil returning to the reservoir. The 
oil used was distillate. 

The blower, as shown on the left of the pump, is a Rockwell 
No. 1 type A positive pressure blower, provided with a relief 
valve, set at 2 lb. pressure. 

The reduction machinery in this plant consists of: 

1 Sturtevant 8x5 laboratory roll. 
1 Sturtevant 2x6 laboratory crusher. 



Digitized by 



Google 



582 Brown on Marine or Iron Ore Cements. 

1 Sturtevant sample edge grinder. 

1 Abbe No. 5. 3 ft. x 3 ft. 6 in. pebble mill. 

1 Abbe No. 6. 30 in. x 30 in. pebble mill. 

1 Perfecticon screen. 

1 Pan dryer, with sheet iron surface 5 ft. x 12 ft. 

The whole plant is operated by a 12 H.P., 200 R.P.M. Backus 
gasoline engine. 

This plant was designed for the purpose of manufacturing 
cements in sufficient quantity to make any desired practical or 
laboratory test with the various classes of cement or raw ma- 
terials which might be submitted for examination. 

The ferruginous Miocene shell marls at Yorktown, Virginia, 
furnish excellent raw material for the manufacture of a complete 
series of Marine cements ranging in alumina content from 2 to 
7 per cent., and in iron oxide content from 3§ to 9 per cent. 
Sufficient runs were produced so that practically the entire 
range of cements were secured, from a high ferric oxide cement 
to a high alumina cement. Enough was made at each run so that 
the characteristics of the cement in course of manufacture, the 
behavior of the raw materials in process of clinkering in the kiln, 
and the grindability of the raw and finished product and the 
chemical and physical qualities of the cement were ascertained. 

The shell marls furnished from 75 to 100 per cent, of the 
raw materials entering into the various cements. When additions 
were ■ necessary, clays of such analyses were selected that the 
resulting cement would have the desired content of ferric oxide 
and alumina. 

On account of the fact that the low-alumina cements so 
manufactm-ed are especially adapted for concrete miarine struc- 
tures, I would recommend the name "Marine" cements, in pref- 
erence to the term "Iron ore" cements, since the latter signifies 
that iron ores are used in the manufacture of the cements. 

The ferruginous shell marls at Yorktown, Virginia, carry 
sufficient amounts of lean, ferruginous clays to make a suitable 
mix for Marine cements, without the addition of tempering ma- 
terials. In some of the runs made, however, in order to secure 
as high ferric oxide content as desired, pyrites cinder of the fol- 
lowing analysis was used : 



Digitized by 



Google 



Brown on Marine or Iron Ore Cements. 



583 



Ferric oxide 79.19 

Alumina 4.73 

Calcium oxide 1.82 

Magnesium oxide 1.35 

Silica 7.72 

SO3 4.95 

Sulphur as sulfide 1.96 

In the beginning of the ''Iron ore*' cement industry in Ger- 
many, the cements as made for commercial pmposes carried less 
than 1^ per cent, of alumina. These cements were so slow in 
setting, and had physical characteristics so different from the 
standard Portland cements, that it was difficult to convince the 
buyers of their merit. It is probably on this account that the 
manufacturers of ''Iron ore" cement in Germany have gradually 





















Table I 


• 












eJ'^' 


n D«an. 


»MO0. FluQion. 


s 


Cii^letlAiaiyni. 













^ 1 


1 






i 


II 


1 


i 


i 


ij 


i 


i 




1 


i\ii 


iii 


I 


i 


kt 


1272 
1813 


m 


m 


\ 
im ' Mil 

2iia ftsi 


3IS 
370 


§28 


193 


m 






TO 


mtil 1 
tlK,l9M7JS} 


SJIfll.75 


0JS 




1.47 so.n 

1,43 97,14 


im 


m 


tm 


! ( 


155 


m \n.^ 


21 


70 


0,K.t2M^ZM 


1 1 


EJoLSajdO 64 



•Two days. 

increased the content of alumina, since recent importations show 
the alumina content to run slightly in excess of 3 per cent. 

Non-aluminous cements are practically impossible to manu- 
facture, on account of the difficulties of securing raw materials 
free from alumina; and even if they could be commercially manu- 
factured, are less satisfactory than the cements which carry suf- 
ficient amounts of alumina to render the resulting cement active 
enough to meet the demands of practical constructors, who re- 
quire hydraulic cements to obtain their final settmg within a 
reasonable time — certainly sooner than 48 hours. 

Table I shows the chemical and physical characteristics of 
(first) an imported low-alumina "Iron ore** cement; (second) a 
cement from the middle west; and (third) a Marine cement 
made in the laboratory plant above described, the tests being 



Digitized by 



Google 



584 



Brown on Marine or Iron Ore Cements. 



6| 



GO >f 

-I 



-a 



£itS 



« I 



6" US 



^11 






6' 



si 8l i1 s| 11 11 g| g| %1 1| 11 11 1§ 11 

55^5 5?'.J5 5?:r)5 5kJ ^,r| 5^.J Sr,,^ jr..^ JrJ !?.r^ ^. r^ SStg gj8 S^icg 

I2;2 gg : : : : : : : : : : 

•Ml^- ooeo • • ; • ; ; ; ; ; ; 

S§ K^ SS : : 



SS 



CM CO 
o S 



2?5 



5»o t-'M »o» a CO — «ac »>.»H i>-r>. r-« oo» q»o obi-i ecoo e^wj £J95 
•.f c« i-io —I'. OS— ^o Ot- — ^ ^?i «Bh>. ■*i^ — ■^ «p«5 «cao 
z-^ Olid Qc^ o>>o 00 lo o>>75 on to cc^ eb>o oic cb>c x-^ oo^ qo<4< 



52 g'o §S iS 2JS S2 gS J3f2 S« R?2 i? feS §J§ |g 

«oM !»•«< ao'T a5« ooco o*"* OO'^ t--^ o6i?5 «^ oo'^ t«-* abeo *^ 

SS SS §g Sg 5lg SSS 2:?2 {:g ?« gi5 ffig gg gs II 

•*«-i i^c^i tOM »!5?i <oesi i-w r^ec «cco t>.c«3 oecvi ir?c4 «c^ «»^ «e>» 



o * o o 



JrL. 



d 



o 



_9 

Is 



d 






d 



d 



d 



M ^ 



M 1-1 



?g s 



s s 



(M 



o at 



*s s 



oo -« 



S 5 {2 8 



CJ — 



ao oo 



^ CJ 



O 00 »Q t^ 

i>- 00 00 t>. 



^_ 


!>. 


«o 


CM 


{2 


- 


CO 


2 


S 


S 


S5 


IM 



s ^ 



IS 



5 S 



^ 


s 


s 


04 


CSI 


CO 


S5 


g 


o 


s 


s 


s 



-^ CO 



CI ^ 



^ r- 



■^ CI 



^ O -I '.t" 

CO «-< OS «-« 






CO O -f UJ 

CI CM CO CO 



Digitized by 



Google 



Brown on Marine or Iron Ore Cements. 



585 



Table III. — Effect of Accelerators on German Iron Ore Cement 

No. 1272. 



Accelerator Used. 'Initial Set. 



No. 1272. 
Without addition.' 

2 per cent. Dical- 
cium Aluminate. ' 

4 per cent. Dical- 
cium Aluminate i 

1 per cent. Cooking 
Soda. 



2 per cent. Cooking, 
Soda. I 



7°-00' 
S-OC/ 
8°-30' 

25' 



Final Set. 



Steamed 6 hrs. | Color, etc., at 
after 24 hrs. 24 hrs. 



40«-00' I 

I 
Just about final at 



O. K., but no 
strength. 



22 i hrs. 

Hard set at 
22} hrs. I 

Good final set at { 
22} hrs. 



9°-00' 



I O.K. Hard. 



O.K. Hard. 



O.K. 



O.K. 



n K" TTo«i Slight efflorescence 
O.K. Hard. and white spots. 

I Not very hard at 
O. K. Hard. 24 hrs. Bad efflor- 



1 per cent. Caustic 

Soda. I 

2 per cent. Caustic 



2 per cent. 
Socf. Calcium 



Soda. 

SoJ 

Aluminate. 

4 per cent. 
Sod. Calcium ' 
Aluminate. 

2 per cent. 
Hydrated Lime. 

4 per cent. 
Hydrated Lime. 

2 per pent. 
Hydrated Lime. 

1 per rent. 
Cooljing Sodu. i 

2 per cent. 
Hydrated Lime. ^ 

2 per cent. 
CooKing Soda. | 

2 per cent, 
Hydrated Lime. 

Lab. No. 1023. 
Without addition. 

2 per cent, Dical- 
cium Aluminate. 

4 per cent. Dical- 
cium Aluminate. 

2 per cent. Trical- 
cixim Aluminate. 

4 per cent. Trical- 
cium Aluminate. 



2*'-30' 

45' 
3*'-15' 

2<'-10' 
6°-30' 
2«-30' 
3°-30' 

3°-45' 

3<»-3(y 
3°-2(y 
3°-30' 
3°-00' 



Good final set at 
21 hrs., but not > 
hard. I 

Very hard at I 
21 hrs. I 

Good final set at I 
22} hrs.. I 

I but not hard. 

I Final set at 22| i 
I hrs., but not hara. 



O. K Hard. 'B*<ily frosted all 
O.K. Ha«l. ''"'^ XovJ'.'"*^ 



Leaves glass; but ' Slight gray 
blotches. 



hard. 
O.K. 



Final set at 22 

hrs and quite 

hard. 

Barely set at 
21} hrs. 



I Very hard at 



21 hrs. 



Very hard at 
21 hrs. 



18 hrs. 
6<»-30' 
70-20' 
7*-00' 
7*'-00' 



' Leaves glass whole i 
, but no strength. 



O. K., but no 
strength. 



LeavoH ghws, but 
O. K., hard. , 



O.K. Hard. 



O. K. Not very 
strong. 

O. K., after 
2 days. 



Slight gray 
blotches. 



O.K. 



O.K. 



Dull dead 
surface. 



Snow white. 



Snow white. 



Good. 

Good color. 

Good color. 
Firm pat. 



Digitized by 



Google 



586 



Brown on Marine or Iron Ore Cements. 



carried out according to the standards of the American Society 
of Testing Materials. 

The important differences in these cements are the relative 
percentages of ferric oxide, alumina, magnesia and sulphur tri- 
oxide contained in them. From an inspection of this table, it 
will be noted that the ferric oxide content runs as low as 2.64, 
and as high as 8.71 per cent., the alumina running as low as 1.43 
and as high as 7.8 per cent. The magnesia in No. 1313, while 
not determined, is known to run about 2 per cent. 

Table II gives the characteristics, imder standard conditions, 
of a selected series of Marine Portlands* For the sake of com- 




Oays 



FIG. 3. — CURVES SHOWING COMPARATIVE STRENGTHS BETWEEN IMPORTED IRON 

ORB CEMENT AND MARINE CEMENT MADE AT YORKTOWN, VIRGINIA, 

AND A STANDARD PORTLAND IN THE MIDDLE WEST. 

TESTS MADE ACCORDING TO A. S. T. M. STANDARDS. 

parison. No. 1272 is again inserted, and a highly aluminous cement 
from the Lehigh Valley, No. 1314, is included. 

No. 1250 gives the tests as secured by a well-known inde- 
pendent laboratory on cement No. 1242. This same cement 
(No. 1242) was sent to three different testing laboratories, and 
checked fairly well with the results secured in our own laboratory, 
so far as physical tests were concerned, there being a much wider 
variation in the chemical tests than in the physical. 

As a consequence of the chemical differences in the cements 
in Table II, a striking contrast in the physical properties as re- 



Digitized by 



Google 



Brown on Marine or Iron Ore Cements. 



587 



gards setting time and initial strengths of the cements (or what 
is known as the activity of the cements) is apparent. These 
distinctions are more graphically shown in the curves of strengths 
in Fig. 3, 4 and 5. 

Very low-alumina cements behave on setting much like 
Puzzolans, and it is difficult to decide when the cement takes its 
initial set. The final set in "Iron ore" cements carrying under li 
per cent, of alumina is from 40 to 50 hours. With such slow- 
setting cements the accelerated tests for soimdness (such as 
boiling the pats of neat cement) should not be carried out imtil 
1 day after final set has been secured. 



%AU03 




so 
Age in hours 

FIG. 4. — CHEMICAL ACnVKT CURVES ON CHARACTERISTIC CEMENTS. 
FULL LINES NEAT. BROKEN LINES SAND. 



Plaster accelerates the setting time of low-alumina cements, 
but if used in excess of 3 per cent, shows ill eflfects, the expan- 
sion being so great that the pats imder boiling disintegrate com- 
pletely. The probable explanation for this behavior is that the 
crystallizing action of the plaster starts the setting action in the 
cement before the free lime has time to hydrate completely. 

A series of investigations was started upon these low-alumina 
cements, to ascertain the effect accelerators would have upon 
setting time, soimdness and color. Numerous accelerators were 
mixed with imported **Iron ore" cements No. 1272 and No. 1023. 
Standard methods for setting time and accelerated tests for 
soundness were carried out. These results are shown in Table 
III. 



Digitized by 



Google 



588 Brown on Marine or Iron Ore C'ements. 

The aluminates of calcium gave the most satisfactory results 
as accelerators for these slow-setting cements. The aluminates 
were prepared in a small laboratory kiln, according to the fol- 
lowing molecular formulas: 

2 CaO AI2O3 for dicalcium aluminate. 

3 CaO AI2O8 for tricalcium aluminate. 

Na20 CaO AI2OS for sodium calcium aluminate. 

The effect of tricalcium aluminate, while not given in Table 
III, differs from the dicalcium silicate in that with corresponding 
percentages, the final set is secured with tricalcium aluminate 
from 30 to 45 minutes later than with the dicalcium aluminate. 
It would appear from the-results secured on laboratory No. 1023 
sample, as shown in Table III, as though the aluminate exer- 
cised no hastening action on the setting time. But the set- 
ting time recorded in the table without addition is that which 
was secured when the cement was first obtained. The barrel 
was permitted to stand exposed to the air, and after one 
month's time thus exposed, this cement had lost its activity. 
This behavior is characteristic of the very low-alumina cements 
that have their setting time regulated by the addition of plaster. 
After such loss of activity, it is impossible to decide when these 
cements secure their final and initial sets. The addition of alumi- 
nates to cement No. 1023 caused it to become active, and in so 
far as setting time and soundness were concerned, it then had 
the characteristics of a normal Portland cement. 

Table IV shows the effect of aluminates on tensile strengths 
of slow-setting iron ore cements. The beneficial effect of the 
addition of these aluminates is most marked on the mortar bri- 
quettes, which were made according to usual methods, the pro- 
portions being 1 of cement to 3 of sand, and these propor- 
tions hold good for all of the sand briquettes described in the 
tables. 

The tensile strengths reported in all of the tables given are 
averages of from three to five briquettes in every instance. 

In order to carry out some accelerated tests on Marine ce- 
ments, Feret's methods were followed. Neat briquettes were 
made up in accordance with the A. S. T. M.'s standards, and 
after 3 months' curing, sharp-edged prisms were groxmd out 



Digitized by 



Google 



Brown on Marine or Iron Ore Cements. 



589 



of the briquettes. These test pieces were approximately f in. 
square by 2 J in. long, varying from 1.5 to 3 in. owing to the diffi- 
culty of sawing them from the solid neat briquette. 

Galvanized sheet iron strips 18 in. by 3 in. were bent into 
U-shaped supports, and the prisms of neat cement tied to the 
supports. A galvanized sheet iron container with a tight cover 
was used for the solution, which was kept at all times at a height 
such that there was complete immersion of all the test pieces. 

Table IV. — ^Effect of Aluminaies on Tensile Stbenqths of Slow- 
SEPnNa Iron Ore Cements. 

Imported Iron Ore Cbment. 



Analyses. 



I Cement. I Lime. | Bauxite. 



SiHca 23.26 



Alumina 

Ferric Oxide 

Calcium Oxide . . . 
Magnesium Oxide 
Sulphur Trioxide 



2.09 
8.43 
62.54 
0.73 
1.66 



1.50 

60 

94.96 



)»■ 



2.02 
64.55 



Loss by ignition ; 1 . 36 



2.99 I 29.45 



Neats. 



Without additions i 484 

Addition of 4 per cent. Tricalcium Aluminate 506 
Addition of 4 per cent. Dicalcium Aluminate ... I 489 



Sands. Neats. Sands. 



123 ' 684 
233 I 726 
303 , 643 



163 
379 
451 



Nora. — Id manufaotuiing the Aluminates, the proportions taken for combination were 
based only on percentace content of Calcium Oxide and Alumina in the Lime and Bauxite. 

Clear rain water, to which various percentages of magnesium 
sulphate were added from time to time, constituted the solution 
in which these sticks were immersed. 

The first immersion was made on May 19, 1910, in a J 
per cent, magnesium sulphate solution. This solution was 
changed to a f per cent, solution on May 1, 1910. On May 
the 24th, the solution was changed for a fresh one of ^ per cent, 
magnesium sulphate. Every 5 days thereafter for 5 weeks the 
solution was made up anew, the old solution being thrown away. 



Digitized by 



Google 



590 



Brown on Marine or Iron Ore Cements. 



Thereafter, every week the J per cent, solutions were renewed, 
up to November 16, 1910, when the examination was made upon 
which Table V is based. 

The numerical scale in Table V is based upon the physical 

Table V. 



lAb. No. 


Standa 
Midc 

Standa 
Midc 

Standa 

A Leb 
Cemt 


Description. 




Ck>n<lition. 


1254 


rd Portland Cement from 
UeWest 


the 


10. Comers 
Pits ir 
9. Comers 

9. Comers 

2. 

Perfect. 
8. 
7. 
1. 

Perfect. 

Perfect. 

Perfect. 

Perfect. 

Perfect. 


all gone. 
1 faces. 


1247 


rd Portland Cement from 
lie West 


the 


all gone. 


1314 
1269 


rd brand from Lehigh Valley 

dgh Valley standard Portland 
ant. — No. 1 Piece 


Edlgone. 


1259-A 


A T^high Valley .s 
Cement.— No. 2 Pie 
Silica — Lime Cement- 
Silica — I^ime Cement- 
High-iron Portland — ] 
High-iron Portland — I 


tandard Portland 
ce 




1223 
1171 
1191 


-Run No. 2. . 
-Run No. 1 . . 
^o. 1 Piece . . . 






1273 


^o. 2 Piece. . . 






1239 


Marine 
Marine 
Marine 
Marine 
Import 


Portland 








1231 


Portland 








1265 


Portland 








1242 


Portland 








1272 


ed Iron ore cement 




Perfec 


;t. 

ON. 




Chemical Conbtitubntb which Intlue 


NCB 


DXBINTBORATI 




Lab. No. 


Per eent. of 
CaO. 


Per cent, of 
AUOi. 


►er cent, of 
SOi. 


Per cent, of 
MgO. 


1254 
1247 
1314 
1259 
1191 
1273 
1239 
1231 
1265 
1242 
1272 


62.20 
61.75 
60.99 
61.39 
62.70 
62.77 
63.53 
62.44 
61.33 
62.11 
61.75 


7.80 
7.35 
7.81 
6.65 
5.89 
4.65 
3.96 
4.09 
3.26 
3.17 
1.43 


1 

j 


1.39 
1.47 
1.61 
1.31 
1.21 
1.34 

0.97 
0.82 
0.96 
1.66 


n.d. 
n.d. 
3.43 
3.17 
0.98 
1.12 
0.83 
0.98 
0.93 
0.90 
0.73 



appearance of the test pieces after the adhering calcium carbonate 
or other salts which sometimes covered the pieces had been 
washed from them. No. 10 was chosen for the piece showing 
the greatest disintegration, perfect condition being represented by- 
No. 1. 



Digitized by 



Google 



Brown on Marine or Iron Ore Cements. 591 

No. 1223 and No. 1171 were special cements, manufactured 
by grinding pure quartz sand very finely, and then adding to it 
an equal amount of Portland cement, analyzing as follows: 

Silica 23. 12 per cent. 

Alumina 6 . 95 

Ferric oxide 4.65 

Calcium oxide 60.01 

Magnesium oxide 1 . 45 

Sulphur trioxide 1 .56 

After adding the cement to the finely groimd sand, 10 per 
cent, of thoroughly hydrated lime was added to the mixtm^. 
The whole was then ground to a fineness such that 85 per 
cent, would pass through a No. 200 mesh sieve, and 95 per cent, 
through a No. 100 mesh sieve. The cement so manufactured 
gave excellent long-time tests, comparing favorably with standard 
Portlands on neats and mortar strength, from and after the 28 
day period. 

The object of subjecting this special cement to the accele- 
rated tests was to throw some light on the often-advanced theory 
of the benefits arising from waterproofing and silica additions to 
standard Portlands which are intended to be used in concrete 
structures in sea water. It will be noted that these cements showed 
nearly as much disintegration as the standard Portland cements, 
and were far inferior in their resisting qualities to the Marine 
Portlands. 

From an inspection of the chemical constituents which have 
an influence on disintegration, it appears that the cements which 
are low in sulphur trioxide, alumina and magnesium oxide, with- 
stand perfectly for the time immersed the disintegrating effects 
of the solution. If a comparison be made between the Lehigh 
Valley brand which withstood perfectly the disintegrating effects 
of the solution and the Lehigh Valley brand which was classed as 
"Nine" xmder "Condition," it will be seen that the magnesia, 
sulphur trioxide and alumina all run higher in the brand which 
was disintegrated than in the one which showed perfect condition. 

After November 16, 1910, when the above conditions were 
noted, the test pieces were allowed to remain for 3 months 
without changing the solution. During this time most of the 
pieces became coated with calcium carbonate, which retarded the 



Digitized by 



Google 



692 Brown on Marine or Iron Ore Cements. 

disintegrating effects of the solution. On March 1, 1911, the 
coatings of calcium carbonate were removed by washing with 
weak acid, and subsequent cleansing with pure water, and the 
test pieces were again immersed in a 2 per cent, solution of 
magnesium sulphate. 

In order to ascertain what effect sea water would have on 
the various classes o Marine and ordinary cements, comparative 
tests were made for a period of 9 months. The sea water at 
Yorktown, Virginia, was found to be practically the same as that 
at the mouth of the Chesapeake Bay. This sea water was evapo- 
rated so that it would occupy but one-quarter of its original 
volume. Care was taken to prevent evaporation of this solution, 
the old solution being rejected and a new one made every 3 
months. Briquettes were made according to the A. S. T. M.'s 
standards, and after curing imder a damp cloth for 24 
hours, were immersed respectively in fresh water and in sea water 
solutions. Table VI gives the results of these tensile tests, and 
also the chemical analyses of the cements. 

The curves deduced from the tensile strengths show at a 
glance the difference in behavior of these various cements, Fig. 5. 

No. 1263, No. 1265 and No. 1246 are Marine cements, 
manufactured from the Yorktown shell marls. All of the cements 
subjected to this test showed to some extent the deteriorating 
influences of the concentrated sea water. The Marine cements, 
however, offered much greater resistance under the same condi- 
tions, than did cements Nos. 1313, 1314, and 1259. No. 1314 is 
a high-alumina cement from the Lehigh Valley district, as is 
also No. 1259. No. 1313 is a cement from the middle west. 
The 12 months' test on No. 1246 shows 876 lb. for the neat, 
and 324 lb. for the sand. It is evident, therefore, that the sand 
briquette is steadily gaining in strength from the 6 months' 
period, while the neat briquette shows a loss between the 9 
months' and the 12 months' period. 

The alumina runs practically the same in No. 1314 and No. 
1313, and, as already stated, the magnesia in No. 1313, while not 
determined in this particular sample, runs slightly less than 2 
per cent, in this cement. 

No. 1259 shows much better resistance in the accelerated 
tests than No. 1314 (again it will be noted that No. 1259 runs 



Digitized by 



Google 



Brown on Marine or Iron Ore Cements. 



593 



< 

< 

A 
H 



8 

O 



< 



& 



A 

U 

O 

CO 

o 

s 









I 



U 



I 



3 



S ^ 

s s 



s 


s 


5 


§3 5 S 


•^ 


^ 


•^ 


M eo CM 


^ 


S3 


o 


s « « 


-* 


o 


^ 


^ *H ^ 



8 S 






0» 9 0» 

oa ^ CO 

8csi 1^ 



:$ 3 S 

C4 04 e>i 



S5 ?i 



2 §J 






gS2 55S 



2S SS2 



SS 



as2 §g 

353 r*r^ 

S2 Si 

MOM e«9W 

i ii" 



S2 



5^ ON 



PO PS CO "* 









^i ii § 






ll 



Hii 



U h u 
fi n f 



S iJBtCO fx«cS {x<{^ &C4C^ 



S eo 



s s 



.3 




Digitized by 



Google 



594 



Brown on Marine or Iron Ore Cements. 



lower in alumina, magnesia and in sulphur trioxide than No. 
1314). 

As a result of these tests, it again appears that the content 
of alumina, calcium sulphate, and magnesia plays a very important 
role in the resistance which the cements are able to show against 
concentrated sea water. 

In the manufacture of Marine cements, there are certain 
ideas prevalent in relation to the ease of burning, which do not 

tzoo 



ilOO 
1000 
QOO 
600 
700 
GOO 
300 
400 
300 
£00 
if 00 



/ 3 e 9 tz 

Months 

PIG. 5. — CURVES SHOWING COMPARATIVE TENSILE TESTS OF MARINE AND 

STANDARD PORTLAND CEMENTS IN CONCENTRATED SEA WATER. 

RESULTS GIVEN ARE AVERAGES OP 3 TO 5 BRIQUETTES. 

FULL LINE CURVES NEAT STRENGTHS. 

BROKEN LINE CURVES SAND 3 TO 1 STRENGTHS. 

seem justified when these cements are actually brought under 
commercial methods of manufacture. It is frequently stated 
that the oxides of iron are better fluxing materials for Portland 
cement than alumina. Experience shows that in the burning of 
these cements which carry high percentages of ferric oxide, com- 
bination between the basic and acidic elements does not com- 
mence at as low temperatures as with the aluminous cements. 
Also, the temperature interval between the combination tempera- 



/ 


\ ^ 


-~^^_ 






// 


<^. 


"^ 


\^ 




/ 


\X 


_^^^^ 


1313 ^^-^-^^^^^ 




// 




^ 




Ico^ 


/ 




\ 






/ 




\ 






^ 




\ 






...; 




^^■^■^---\ 






.1 


N 


\ 




fZOS 


M 


^N 




o/S"^^--.....^,.^^ 




1 




---^___ 




/OA^ 








- 


/3/4. 



Digitized by 



Google 



Brown on Marine or Iron Ore Cements. 595 

ture and the temperature of fluxing is much narrower in the 
Marine cements than in the aluminous cements. It is easy to 
distinguish in the kiln the increase of heat, due to the formation 
of the calcium silicates and calcium ferrites; and because this 
increase of heat is given off in a much more restricted area than 
occurs in the manufacture of aluminous cements, care is required 
to prevent fluxing. 

When the raw materials are ground to the same degree of 
fineness, it is not possible to carry the lime to the silicates ratio 
as high in Marine cements as in almninous cements, if sound 
cement is to be secured. This behavior indicates that ferric 
oxide is not a complete substitute for the alumina, and cannot 
carry relatively as much of it in combination, molecule for mole- 
cule, as alumina. 

The percentage of allowable coarse particles, i. c, particles 
failing to pass the No. 200 mesh sieve, is less for Marine cement 
raw materials than for aluminous cement raw materials, if sound 
cements are to be secured. No trouble, however, was encountered 
in the manufacture of sound Marine cements when the raw ma- 
terials were ground to such a fineness that 85 per cent, would 
pass a No. 200 mesh sieve, and 95 per cent, a No. 100 mesh 
sieve. It is quite possible to make sound Marine cements, that 
differ only slightly in density from the aluminous cements. The 
low temperature interval between temperature of combination 
and temperature of fluxing is responsible for the density 
ordinarily seciu'ed in Marine cements. When these cements 
are manufactured with the specific gravity running two-tenths 
higher than the ordinary Portlands, it is found that the clinker 
so produced is very difficult to grind. It is advisable, there- 
fore, from the standpoint of economy in production of Marine 
cements, to prevent incipient fluxing conditions, and thereby 
secure a cement which has only a slightly higher specific gravity 
than the aluminous cements. 

The problem of the underlying causes for the failure of 
cement structures exposed to the action of sea water is so com- 
plex that it is not surprising that there are many and opposing 
views held by able investigators, who have spent much time and 
earnest effort in the attempt to solve the question. 

The chemical and physical action which stands out clearly 



Digitized by 



Google 



596 Brown on Marine or Iron Ore Cements. 

in this work, extending over a period of nearly 2 years, can be 
summed up as follows: When disintegrating effects occur in con- 
crete exposed to the action of sea water, and where there are 
variations in level of» the water, due to wind, wave and tide, 
these disintegrating effects may be rightly divided into two main 
causes — mechanical and chemical. Mechanical action is in- 
fluenced (first) by alternate drying and wetting of the surfaces, 
due to wind, wave and tidal action; (second) abrasions from ice 
or floating objects; (third) formation of expansive crj^stals where 
freezing occurs, or where double salts may be formed within the 
concrete body. 

This brings us to the second main cause of disintegration, 
namely, chemical action, since the formation of the double salts 
is first brought about by substitutive chemical reactions. Such 
double salts as calcium aluminum sulphate occupy more space 
than the original compounds. The dissolving, replacing and 
formation of these new crystalline substances act in the same 
manner as ice crystals within the rocks. Such ice crystals have 
been very effective in producing disintegration of the ancient 
rocks covering the earth's surface. 

The hydraulic cements which are best fitted to withstand 
satisfactorily all disintegrating effects of sea water (excepting 
those which are purely mechanical) are those which give the great- 
est density of concrete structure, and which are relatively high 
in silica, low in magnesia and sulphur trioxide, and in which the 
content of alumina does not exceed that of the ferric oxide. 

In carrying out the work herein described, the author has 
had the constant help and valuable assistance of J. H. Payne, 
who was chemist for the American Cement Engineering Com- 
pany, at Yorktown, Virginia, during the prosecution of these 
investigations. 



Digitized by 



Google 



IRON ORE CEMENT.* 

By Arthur E. Williams, f 

Iron ore cement is a product intended to be used in sea water 
work. This material is now manufactured in Europe under the 
name of Erz cement. According to Mr. William Michaelis, Jr.,t 
the process of manufacture is similar to that of Portland cement 
except that limestone and iron ore are used in place of limestone 
and clay. United States Consul Thackara§ gives a description 
of its manufacture as follows: Chalk, flintstone, and finely ground 
ferric oxide are used. The flint and iron are ground together, 
then mixed with the chalk and water and screened through a 
fine sieve. The screened product is clinkered in a rotary kiln 
and then groimd. An average composition of iron ore cement, 
given by Michaelis is: 

CaO 63 . 5 per cent AI2O3 1.5 per cent 

SiO, 20.5 " MgO 1.5 " 

FejO, 11.0 " Alkali 1.0 " 

The effect of sea water is undoubtedly two-fold. In the first 
place chemical reaction may take place between certain con- 
stituents of the cement and the salts in sea water, and, on the 
other hand, the mechanical action of the waves carrying large 
amounts of sand, freezing, thawing, and the varying pressure 
of the water due to tide help to injure the cement submerged in 
sea water. This work, however, will be confined to the chemical 
action of sea water, for the mechanical action is of minor import- 
ance unless the cement is weakened by chemical changes. 

The reactions which take place between Portland cement 
and sea water are said to be of three distinct kinds. First, the 
action of MgCl2 and MgS04 in sea water on the calcium hydrate 
formed during the hardening process of the cement, forming 
Mg(0H)2, CaCl2, and CaS04. Second, the action of gypsum, 

* Under the direction of Mr. R. T. StuU. 

t Urbana, 111. A Thesis for the Bachelor of Science Degree in Ceramics, University of 
Illinois in 1910. 

X Bng. News, Vol. 68, pp. 645-646. 

I United States Consular Reports, June, 1908. 

(597J 



Digitized by 



Google 



598 Williams on Iron Ore Cement. 

CaS04 formed above, upon the calcium aluminates forming 
calcium sulpho aluminate. Thirds the crystallization of the 
gypsum and calcium sulpho aluminate giving an increase in 
volume, thus causing the disintegration of the mortar. 

That free lime is present in set Portland cements is well 
known. • Lamine* found 32 per cent of CaO in cement sub- 
merged in the Black Sea 15 years. Every analysis of a cement 
exposed to sea water shows a high percentage of MgO. Vicatf 
in 1840 showed this fact clearly, a cement, which was submerged 
in sea water for 6 months, was analyzed. A sample, taken from 
the surface exposed to the sea, showed 10.4 per cent MgO and 
19.3 per cent CaO while the interior, which was not impaired, 
showed 1.87 per cent MgO and 31.33 per cent CaO. 

A. Meyert states that cement loses strength in sea water. 
The MgS04 acting with the silicate of lime forms Mg(OH)i and 
calcium sulphate. The CaS04 reacts with the calcium aluminates 
(AljO,, X CaO) of the cement, forming A1(0H)3 + 3 Mg(OH), 
+ CaS04 + CaClj. 

Charles J. Potter§ says that MgS04 is the most active con- 
stituent in sea water on cement. He found that MgCU softens 
cement but causes no expansion. Potter says that it is now 
definitely believed that magnesium salts act on the feebly com- 
bined lime and alumina compounds which on taking up water 
of crystallization cause bursting of the concrete. ^He mixed 
calcined red brick clay with Portland cement clinker in propor- 
tions of 6 to 10. From this mixture briquettes were made and 
placed, together with Portland cement briquettes, in fresh water, 
sea water, and sea water to which 10 per cent MgS04 was added. 
Both of these cements gained strength in fresh water. In salt 
water, the Portland cement briquettes began to fail after 5 weeks 
and were disintegrated after 5 years. These cements showed 
blistering after one year, which was followed by expansion and 
bursting. The red cement improved continually but took 8 
weeks to obtain the maximum strength that the Portland cement 
had obtained in 5 weeks. In the 10 per cent solution of MgS04, 
the Portland cement tested 500 lb. in a month and then went 

*LeCiment, 1901, pp. 111-691-81. 

t Iron Ore Cement — The P. C. Co. of Hemmoor, Hamburg, Germany. 

X Chemischea Central BUUt, Vol. 73. p. 1368. 

i Jour. Soe. Chem. Ind., Vol. 28. 



Digitized by 



Google 



Williams on Iron Ore Cement. 599 

back to zero in 1 year. The red cement began at 250 lb. and 
increased continually to 1015 lb. in 8 years. Mr. Potter says 
that the chemical combination of CaO, Si02, and AI2O3 and 
water is feeble and that probably accomits for the ability of 
magnesium in sea water to be so active. 

The experiments of Dr. Michaelis* and Le Chatelierf lead 
them to the conclusion that Portland cement suffers in solutions 
containing sulphuric acid salts, which applies to sea water. A 
double salt is formed composed of gypsum and calcium aluminate. 
This sulpho-aluminate, AUOs, CaO + 3CaS04, is said to crystal- 
Uze with 30 molecules of water, which process must be accom- 
panied by considerable expansion. Le Chatelier says that "the 
main cause if not the sole cause, of the injuries which cements 
suffer under the action of sea water is the formation of calcium 
sulpho-aluminate . ' ' 

Rebuffatt says on the contrary that sulpho-aluminates 
cannot exist in cements in sea water but agrees with Michaelis 
and Le Chatelier that calcium aluminates are the parts of cement 
most easily acted upon by salts in sea water. 

It has been shown that calcium ferrates are formed similarly 
to the calcium aluminates and that alumina could be replaced 
by ferric oxide in Portland cement. Dr. Michaelis puts this 
knowledge into use with the idea of overcoming the disintegra- 
tion in sea water. The result of this application is the Iron Ore 
cement of today. 

Dr. Michaelis and the Royal Experiment Station of Charlot- 
tenburg have tested these cements in comparison with Portland 
cements in a very thorough manner. Mr. William Michaelis § 
says in a paper read in the United States that tests of Erz cement 
and Portland cement were made with both neat' and 3 to 1 mix- 
tures which were placed in fresh water, sea water, and water 
containing five times more salt that sea water. In sea water, 
the Erz cement developed a much greater strength than the 
Portland. In the strong salt water, the strength of the Portland 
cement decreased rapidly while the Erz cement showed a steady 
gain. Briquettes were made of Iron Ore and Portland cement 

* Thon Indmtne, 1896. p. 838. 

t Le CimerU, 1901, p. 31-32. 

X Thon Indtutrie Zeitung, 1901, p. 272. 

i Eng. Newt, Vol. 58. pp. 645-646. 



Digitized by 



Google 



600 Williams on Iron Ore Cement. 

which were placed in a salt solution of five times the normal 
strength of sea water under pressure of 15 atmospheres for a 
few days. This condition destroyed the Portland cement bri- 
quettes entirely, while the Iron Ore cement increased in strength. 

The Royal Experiment Station conducted similar tests to 
the above but much more elaborate. Two Iron Ore and three 
Portland cements were made into prisms, using a 3 to 1 mixture 
of standard sand and cement. These prisms were placed in 
sea water and water containing five times the percentage of salts 
in ordinary sea water. In addition to this, these three solutions 
were allowed to act upon test pieces made of cement mixed with 
varied amounts of gypsum. All the Portland cement mortars 
disintegrated in the three- and five-fold salt solutions; all the 
Iron Ore cement mortars remained intact and sound. 

United States Consul A. W. Thackara* investigated this 
cement for use on the Panama Canal. The result of his investi- 
gations was the adoption of this cement for concrete work exposed 
to sea water. Another point in favor of this cement is the property 
of slower setting. The cement is weaker than Portland for the 
first week, but then gradually gains strength and exceeds that of 
Portland. 

Publications of previous experiments do not show definitely 
the best composition for cements giving the greatest protection 
against sea water. With this idea in view, the following investi- 
gations were undertaken: 

The outline of procedure in these experiments is as follows: 
Newberry's cement formula, x (3CaO, Si02) + y(2CaO, AI2O3), 
was used as a basis. Assuming, according to Newberry, that 
FeaOs could replace AI2O3 and form 2CaQ, FcaOs, a triaxial dia- 
gram was plotted (Fig. 1), the three members stationed at the 
three comers being 3CaO, SiOa, 2CaO, AljOj and 2CaO, FejOj. 
By blending these three members, cements could be obtained 
containing various amounts of the calcium aluminate and the 
calcium ferrate. 

The batch weights of these three members were calculated 
and about 15 kg. of each were weighed up, using practically 
chemically pure materials. Whiting, flint, aluminium hydrate, 
and red oxide of iron were the only ingredients. These batches 

* UnUed StaUa Contular and Trade ReporU, June, 1908. 



Digitized by 



Google 



Williams on Iron Ore Cement. 601 

were ground in a ball mill, then passed through a 20(>-mesh sieve; 
thus getting thorough mixing and a finely ground batch. The 
formulfiB for the cements made are given in Table I. 

The following cements, No. 19, 20, 21, 22, 23, 24, 25, 36, 
37, 38, 39, 40, 42, 48, 49, 50, 51, 52, 53, 54, 58, 59, 60, 61, 62, 
and 65 on triaxial diagram were then weighed up, blunged thor- 
oughly, and partially dried by pouring the slip into plaster molds. 



nCoOj/AAOjJ 




FIG. 1. — TRIAXIAL DIAGRAM. 

The cements were then rolled into small balls about the size of a 
marble, dried, dehydrated in a down draft kiln to about 800** 
C. and placed in fruit jars ready for burning. 

These cements were burnt in a magnesite test kiln, designed 
by Mr. StuU of the Ceramic Department, especially for burning 
experimental cements. The construction of this kiln is shown 
in Fig. 2. The success of this kiln is a noteworthy fact as test 



Digitized by 



Google 



602 



Williams on Iron Ore Cement. 



kilns suitable for this purpose, heretofore, have not been very 
satisfactory owing to lack of control, unevenness of temperature 
in the clinkering chamber. Kerosene oil was used for fuel with 
an air pressure of about 50 lb. 

The temperature at the time the clinker was drawn from 
the kiln was determined first by means of a Wanner pyrometer. 
This was given up, however, as the rapid rate of burning required 
a higher temperature than the true temperature of clinker forma- 
tion. 



No. 



10 
20 
21 
22 
23 
24 
25 
36 
37 
38 
39 
40 
42 
48 
49 
50 
51 
52 
53 
54 
58 
59 
60 
61 
62 
65 



Table I. — Formulae of Cements Made. 



Formuls. 



..■l+. 



■■;r ,r i.siM,HK, 



n: '.|( t.Sii ►■■I } 

J ':,( t<ii 1.1 H- 

:ii ' ,1 I - ]< i,,i -,- 

'.* I '" f ?■ H I : J -+ 



Molecular Ratio 
SiOt:A10+FejO« 



2(aCii(J, AitO.) +.7(2Ca0.Fe»0«) . 
l(2CiiO,AM)i) +.8(2CaO.FejOa) 

.um;aO,Fr»rO.) 

K(2Ciin.F.-jOi) 

HaCiK n AM)a) +.7(2Ca0.Fp,0,) . 
?i >r]iO, AbOj) +.6(2Ca().Ft^a) 
:m :*( ill K\ S .' h) +.5(2CaO.Fet( )«) . 
2rJi ;,« > \h()a)+.5(2Ca().FeiC3). 
[ 2(',N \hi)j)+.6(2CaO.Fci()j). 

7 i'rr,n.|,:Oi) 

i-^<'rinJ--.0,) 

U^CiiU, AM)j) +.5(2CaO.FejOa) . 
a(2CaO.AM)a) +.3(2CaO,Fej()a) . 
3(2rM<).Al.Oa) +.2(2CaO.Fe?Os) . 
2f :?ra<>, A t Oa) +.3(2CaO,Fo2()a) 
1 lTl.« *.Sy )a) +.4(2CaO.FeiOa) . 
.■.ij<\-,nj.-,.j)a) 

4.-< riM,|\;Oa) 

I ' Jf M< i,AJ .« )a) +.3(2CaO.FcjOa) . 
'irjt >,< ). M,< ),) +.2(2CaO.FeiOs). 
J Jf ,1" >, A I ' )a) +.1 (2CaO.Fo7C)a) . 
I . Jt ;if K \:.i )a) +.2(2CaO.Fe»Oa) . 

:£.;i( iiHl i:<h) 

2vJCitiKl'f-.Oi) 

laC'wO.Alii h) +.l(2CaO.FeiOa) . 
l(arnri,F»'sOa) 



0.11 
0.11 
0.11 
0.25 
0.25 
0.25 
0.25 
0.43 
0.43 
0.43 
0.66 
0.66 
0.66 
1.00 
1.00 
1.00 
1.00 
1.50 
1.50 
1.50 
2.33 
2.33 
2.33 
4.00 
4.00 
9.00 



Almost all of these cements w^ere fused till the surface was 
glassy in appearance before the cement seemed well clinkered 
and crystals appeared. Cements No. 54, 58, 62, and 65 appeared 
like a Portland clinker, except darker in color and were not fused 
or slag-like in appearance. 

The clinker was first reduced in a jaw crusher and then 
ground in a disc mill; a screen test showed 24.2 per cent on 150 
mesh screen; 12.3 per cent on 200 mesh screen; and the remainder, 
63.5 per cent passed 200 mesh. These cements show that they 
are approximately of the same degree of fineness as the average 
Portlands. After the samples were ground, pats were made from 



Digitized by 



Google 



Williams on Iron Orb Cement. 



603 




1 


- — 






«*f ■'#'■■ 


1 


"""^t 






; 1 1 



i^iti 



iin 






Nl 



r 



OS 



33 , 




ii 4- 







5 








K; 

f 


□ n a 




M. 


j 


- " 





li 

r 

% 



H 
O 

o 
z 

m 
o 



u 





< 




H 


% 


Cl4 


X 


S 


ll 


e 


?5 


n 
o 




H< 


^ 


1 




1 






Digitized by 



Google 



604 



Williams on Iron Ore Cement. 



them in the usual manner to determine the properties of the 
cement. 

The amount of water used for mortar was determined by the 
Boulonge method (Waterbury's Cement Manual, p. 44). The 
initial and final sets were determined with Gilmore needles. 

Four pats were made of each cement with the idea of using 
one for the time of setting tests and placing the other three imme- 
diately in the moist closet, two of which were to be used for the 
boiling test after 24 hours, the third to be allowed to stand in 



Table II. — Results op Tests on Cements. 



No. 



I Time of 
.Initial Set. 
hours. I 



10 
20 
21 
22 
23 
24 
25 
36 
37 
38 
39 
40 
42 
48 
40 
60 
51 
52 
53 
54 
58 
50 
60 
61 
62 
65 



Time of Water 

Final Set. Used, 
hours. per cent. 



1 



1 
1 
2 

3 
2 

1 

l>i 
1 
1 

1 

IH 



3 
5 

5^ 
4 

'5H 



11 


2H 


5 


8 


3^ 


1 2H 


1 7 


3 

1 ;a 



21.0 
20.0 
21.0 
20.0 
21.0 
22.0 
21.5 
20.0 
20.0 
20.0 
21.0 
20.0 
21.5 
22.0 
21.0 
22.0 



10 


21.0 


9 


20 ' 


4 


21.0 1 


4M 


23.5 


3U 


22.0 1 


4^ 


21.0 


5 


21.0 


6 


22.0 




22.0 




21.0 1 



Remarks 
at Time of 
Final Set. 



Cracked in }4 hour 

(). K- Strong 

No cracks 

Small cracks 

Cracked 

Cracked 

Cracked 

Cracked 

Cracked 

O. K. 

Cracked 

Cracked 

O. K. 

O.K. 

Cracked 

Cracked 

O. K. 

Soft 

Cracked 

Cracked 

No crack.1 

O. K. 

Soft and crumbly 

Warped 

Did not harden 

Cracked 



Conditions 
after 48 Hours 
in Moist Closet. 



Cracked 

Warped and cracked 

No cracks 

No cracks 

No cracks 

No cracks 

No cracks. 

O. K. 

No cracks 

No cracks 

Cracked 

Warped 

No cracks 

No cracks 

Cracked 

No cracks. 

Soft 

Soft 

No cracks. 

No cracks. 

Cracked 

Warped 

Warped and cracked 

o!k! 

Warped 



Soft 



Soft 



O.K. 
O.K. 



water for 28 days. All of these cements went to pieces in cold 
water or in the boiling test. The results are given in. Table 11. 
From these cements, one only, i, e.. No. 62, remained sound 
when placed in water. This cement also stood the boiling test 
(J hr.), the others going to pieces. The molecular ratio of SiOj 
to AI2O3 for this cement is four and since the molecular ratio for 
good cements is between 5.1 and 6.8 and since none of these 
cements lie between these limits, it was decided to construct a 
new group. Cement No. 62 approached these ratios nearer than 
any other. 



Digitized by 



Google 



Williams on Iron Ore Cement. 



605 



A new batch was calculated after Bleininger^s formula 
(2.8CaO,Si02) + (2CaO, AI2O3) having different amounts of 
Fe20s and AI2O3 and also the ratio of Si02 to AI2O3 + Fe203 
varied from just above to just below the limits. The using of 
chemically pure raw materials in place of slag and limestone 
gives less efficient mixtures of lime and Si02. It was, therefore, 
thought that sufficient lime would be obtained by the use of 
Bleininger'^ formula. For formulae see Table III. 



Table III. — Formula for Cements Made. 



No. 




Formulffi. 






^1 


5.[i2M:'t,n. 


<iCh^->r\2Ca .F 


'O^y 






At 


5.S(L>.h(':a,l. 


SiiM+{2rfiUJ>'s(l3:) 






A, 


6.1(1* TSCsit i..Si< f3> +^2riiO,F 


-hUi) 






Ai , 7.ll(2.Nr[i<J, 


■^KHJ^ (^<"iin.K 


-n,> 






Bi 1 5.2't ( a 8C';it K^\ i hi 4^0 3 7 > i 2( 


ii^ KAU< )i) + 825{2CaO,Fp7L?a> 




hi 1 (imriM\0.iiHhi h . 171 . 2V 


^0,AW>j>+.S25f2ni 


^,F*M>.) 




Bj 1 GAii i 2 MCtii J ,.Si< J«} + .2tMl(^ C^U, .VIj^Ji > + . «J(1 1 J r t 


.,K,..o,l. 




B« 7.22i2. 8Cii( \M Uii -h. 1 7.1 i liCaO. kU h ) + .H2;» 1 1: ! , 


I.I ,-.1 ij) 




Ci 1 5AA\'lKVai 


,HKIh| -»- atyH^Cnf >,AM h\ + MO. 2£ 1 


iJ-,-,.i ^]. 




Ci o.S'hJ sfZ-iF 


.Si< \:} \ MV.iriQ 


ri(l.AhOif-*-.ii:M>i'ilii 


(M.-H,Oa) 




Ci \ GAiU-^si -.'ii 


,^,[J Jx, -f ..MHSi^C 


v\^,\U h) \ <i(hiiL*rn< i.Fi-Uj) 




Ci 1 7.<Hji:i.HC:tL 


.>iiMi \.\m\H^ 


^l( r \\A 3.. 1- <.Mni2Cii 








Percentage Composition 










j 




Molecular 


No. 


. CaO 


j AljOa 


1 Fe,0. 


SiOj 


1 Ratio 


1 




1 




1 RzOaiSiO. 


Ai 66.0 


1 0.0 


I 11.6 


22.4 


1 ^ 5^1 " ' 


At 66.7 


, 0.0 


10.4 


22.9 


5.8 


Ai 67.2 


0.0 


9.6 


23.2 


6.4 


Aa 67.5 


1 0.0 


, 8.9 


23.6 


1 7.0 


Bi 66.7 


1.3 


9.4 


22.6 


5.25 


Bt 


67.4 


1.1 


. 8.4 


23.1 


6.00 


Bi 


67.5 


1.3 


7.8 


23.4 


1 6.40 


B* 


68.1 


0.9 


7.2 


23.8 


1 7.22 


Cx 


67.4 


1 2.5 


7.2 


22.9 


5.44 


Ca 


68.0 


1 2.7 


6,0 


23.3 


1 5.80 


Cz 


68.2 


2.5 


5.8 


23.5 


6.40 


Ca ' 68.5 

1 


2.3 


5.4 

i 


23.8 


7.00 



These cements were prepared in the same manner except 
that the temperature of clinkering was determined as near as 
possible by the method used. The kiln was allowed to cool to 
about 1000 deg. C. before a batch of cement was put in and tem- 
perature was then gradually raised till clinker was formed, the 
temperature was then read with a Wanner pyrometer. 

The clinkers obtained appeared exceptionally good, being 
dull black in color and glistening brightly in the sim. These 



Digitized by 



Google 



606 



Williams on Iron Ore Cement. 



clinkers were pulverized the same as has been previously de- 
scribed, then tested. 

The results of these tests, Table IV, show that good cements 
can be obtained with a large amount of alumina using the same 
ratio of SiO^ to R^Os as Portland cements require. One very 
noticeable fact, however, is that when no Al^Oj is present as in 
series -A, -42, -As, and A4 these cements all show expansion, thus 
giving evidence of free lime. Although Ai stood the boiling test, 
the cubes made from this cement bulged out from the mold 
considerably. 

The question arises at this point, is it always necessary for 
AI2O8 to be present or can a good cement be made without it? 





Table IV 


— Results op Test. 








Temperature 
when 


Time to 


Appearance 
of Clinker. 


Initial Set. 


Final Set. 


1 

HK). 


No. 


Clinkered, 
deg. C. 


Clinker, 
hours. 


hours. 


hours. 


; per cent. 


Ai 


1300 


y 




24 


62 


24.8 


At 


1320 


}\ 


■ Ail' ' 


22 


50 


24.0 


Ai 


1320 


\)-^ 


elinkered 


26 


56 


23.2 


Aa 


1330 


1^ 


good. 


28 


60 


1 26.0 


Bx 


1390 


M 


colored black 


4H 


40 


26.3 


Bt 


1320 


^H 


and 


4H 


44 


24.4 


Bt 


1350 


glistening 


11 


36 


28.0 


Ba 


1400 


IH 


with 


6 


-*« . 


25.0 


Cx 


1320 


1/^ 


crystals 


5 


30 • 


1 24.4 


Cr 


1320 


^ 


in a 


12 


40 


24.0 


Ci 


1330 


'^ 


bright 
light 


12 


48 


28.0 


Ca 


1380 


17 


40 


27.2 



This ought to be possible by reducing the lime content, as Ai 
was the best of series A and also had the smallest amount of 
lime silicate. 

The slowness of setting is another factor which must be 
considered. It will be seen by Table IV that all of the cements 
required a long time to harden. This must be carried on in a 
moist atmosphere also or the cement will dry out before it has 
completely hydrated and set. The above factors wil