Full text of "Venango County, Pennsylvania, her pioneers and people, embracing a general history of the county"

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/

i^#^5w

f »

.▼ ^

TRANSACTIONS

OP THE

AMERICAN SOCIETY

MECHANICAL ENGINEERS.

VOL, XXV.

XLVIIIth Meeting, New York, N. Y., 1903.
XLIXth Meeting, Chicago, III., 1904.

NEW YORK CITY:

PUBLISHED BY THE SOCIETY,

From the Library Building,

No. 12 West 31ST Street.

1904.

Copyriirht, 1904,
By thb AMERICAN SOCIETY OF MECHANICAL ENGINEERS.

Press of J. J. Little A Co.
Astor Place, New York.

BiiQinsttfin0 Libnvy
HISTORICAL COLLECTION

OFFICERS

OF THE

AMERICAN SOCIETY OP MECHANICAL
ENGINEERS.

1903-1904,

FORMING THE STATUTORY COUNCIL.

PRESIDENT.
Ambross Swasbt Cleveland, O.

VICE-PRESIDENTS,

F. H. Daniei^ , Worcester, Mass.

James Chbtbtib Philadelphia, Pa.

JoHif R. Freeman Providence, R. I.

Terms expire at Annual Meeting of 1904.

D. 8. Jaoobus Hoboken, N. J.

M. L. HoLMAN St. Louis, Mo.

William J. Keep Detroit, Mich.

Terms expire at Annual Meeting of 1905.

MANAGERS,

R. S. MooBB San Francisco, Cal.

H. A. Goxis Richmond, Va.

Chas. H. Corbett Brookijn, N. T.

Terms expire at Annual Meeting of 1904.

R. C. McKmNEY New York, N. Y.

8. S. Webber Trenton, N. J.

Nbwell Sanders Chattanooga, Tenn.

Terms expire at Annual Meeting of 1905.

aBOROB I. RocKWOOD Worcester, Mass.

John W. Libb. Jr New York, N. Y.

Aba M, Mattice Pittsburg, Pa.

Terms expire at Annual Meeting of 1906.

TREASURER,
Wm. H. Wiley Nos. 43-45 East 19th St., New York, N. T.

SECRETARY,
Pbof. F. R.HuTTOir No. 12 West 81st St., New York, N. Y..

Z^\<^\

HONORARY COUNCILLORS.

PAST PRESIDENTS OF THE SOCIETY.

Thubston, R.H 1880—1882 Died Oct 26. 1908

Lbayitt. £. D 1882—1888 Cambridge, Maas.

SwBBT, JoHH E... 188?— 1884 Syracuse, N. Y.

HOLLOWAY, J. P. 1884—1885 Died Sept. 1. 1896

Sellers, Coleman 1886—1886 Philadelphia, Pa.

Babcock. Gboboe H 1886—1887 Died Dec 16, 1898

Sbk. Hobace 1887—1888

TOWNE, Henbt R 1888—1889 Stamford, Coon.

Smith, Oberlik 1889—1890 Bridgeton, N. J.

HtJOT, Bobebt W 1890—1891 Chicago, DL

Lobino, Chables H 1891—1892.. BrooklyD, N. Y.

CoxB, ECKLET B 1892—1894 .Died May 18, 1896

Davis. E. F. C 1894 Died Aug. 6, 1896

Bnxrnos, Chables E.* 1895 Hartford, Conn.

Fbitz, John 1895—1896 Bethlehem, Pa.

Wabbeb, Wobcesteb R 1896—1897 Cleveland, O.

Hunt, Chables Wallace 1897—1898 New York, N. Y.

Melvillb. Qeobob W 1898—1899 Philadelphia, Pa.

MoBOiN, Chables H 1899—1900 Worcester, Mass.

Wellmah.S.T 1900—1901 '. Cleveland, O.

Reynolds, Edwin 1901—1902 Milwaukee, Wis.

DoDOE, James M 1902—1903 Philadelphia, Pa.

[Note.— According to the Constltation, Article C 27, the five rarviring Past PrMidents who last
held the oflSce shall be members of the Coancil, with all the rights, privileget and daties of the
other members of the coancil.]

* Unexpired term of E. F. C. Davis.

NOTE.

The considerable bnlk of the Yolame of Transactions has induced the Pnbli-
cation Committee to direct the insertion of a sammary of the Society member-
ship in place of the complete list of members which was published in the earlier
Tol nines. The summarj attaching to this issue is that which appears in the
catalogue of the Society issued with corrections to July 1st, 1904. Reference for
the complete list should be made to the ** Oeographical List" for July, 1904.

Africa

Australia

Belgium <

Canada

Central America

China 8

Cuba 2

France 9

Germany 7

Great Britain (England) 44

Great Britain (Scotland) 4

Total foreign membership

FoRBiON Countries.

Membenhip.
.. 20

3
.. 31

Holland

India

Jamaica, W. I

Japan

Mexico

Norway

Russia

South America. . .

Sweden

Switzerland

Trinidad, B. W. I.

Kembenhip.
1

.161

Unitbd States.

Membenhip.

Alabama 6

Aliska 1

Arizona. • 1

Arkansas 2

Galifomia 30

Colorado 23

Connecticut Ill

Delaware 14

District of Columbia 28

Georgia 18

Hawaiian Islands 1

niinoia 164

Indiana 33

Iowa , 3

Kentucky

Louisiana

Maine

Maryland

Massachusetts 249

Michlinn 67

Minnesota 13

Mississippi 1

Missouri 41

Total membership in the

Membenhip.
9

Montana

Nebraska

New Hampshire 15

New Jersey 130

New Mexico
New York . . . . ,
North Carolina
North Dakota.

Ohio

Oklahoma

Oregon

1
802
4
1
188
1
4

Penlosylvania 377

~ ~" 2

48
3
1
4
5
2
11

Porto Rico
Rhode Island .
South Carolina
South Dakota.
TeuDessee ...

Texas

Utah

Vermont

Virginia 25

Washington 5

West \^ginia 8

Wisconsin 62

United States 2,577

VI TERBITORIAL LIST.

GEOGRAPHICAL SUMMARY.

Total foreign memberahip 101

Total memberehip in United States 2,577

* Present address unknown 2

Total membership 2,740

SUMMARY OF MEMBERSHIP BY GRADES.

Honorary members 18

Members 1,856

Associates 221

Junior members 645

Total membership 2,740

f Life members. 108

* These are Lawrence V. Melville and Frank Pettit, both Junior Members, and if any
member knows their present addresses he will confer i favor by so advising the Secretary.

t These Life Members are included in the total membership above, in the class to wliich they
belong.

rPTDEX

TO THE

COIfrSTlTUTIOK", BY-LAWS AND EULES,

AMERICAN SOCIETY MECHANICAL ENGINEERS.

Amendment to the Constitution C 57-58

" •• By-Laws C59

" «« " Rules C60

Annual Meeting, when to be held C 41, B 87, R2, 8

" Report of the Council C 31

Application of candidate C 14, B 1, 8

Applications, disposition and scrutiny of C 14, B 2^

Appropriations by a general meeting C 43, B 21

" Council B21, 23

Associate, qualifications of CIO

" advanced to grade of member C 15

Audit of bUls B21,23

" " books B23

Badges of members .... B 42

" for meetings R 2

Ballot for membership C 16, B 6, 7

" " " second baUot C 16, B 9

" " Officers ....C 35, B 12-17

Bequests aiid gifts C 28

Bond of Treasurer B 39

Book-keeping and account books B 23

Budget, annual : B 21, 23

By-Laws, how amended * C 59

Candidate, qualifications C 5-11

" application C 14, B 1,3

" application (non-resident) B 2

" references B 1-5

" for offices B 30, 31

Certificate of membersliip B 40

Committees, appK)intment ' C 45-51, B 22-31

" duties of B 22-31

" removal of members C 51

" Secretary of Standing .• C 50

Conduct, unprofessional C 25

Constitution, goes into effect C 61

vm INDEX, CONSTITUTION, BY-LAWS AND RULES OF THE

Constitution, how amended C 57, 58

" subscription to C J8, B 11

Cop)rright, not exclusive C 54

Council, annual report of C 31

" composed of C 26

" may order letter-ballot C 23, 44, 57, B 36

Delegate, representative B 32

Directors of the Corporation C 26

Discussions, professional R 4-11

Dues, annual, arrears of C 24, B 19, 20

" when payable C 21, B 18

Election, announcement of the results B 15, 34, 35

" sealed ballot C 16, 35, B 6, 10, 12

" tie in the vote B 16

Emblems of the Society B 42

Endorsers of applicant B 1-5, 9

Entertainments at meetings R 14

Executive Committee of the Council C 30, B 29

" " duties of C30, B29

" " vacancy in C.51

Expenses of Committees C 49

Expulsion of members C 25

Fee, initiation to the several grades C 19, 20, B 18

Finance Committee, appointment C 45, B 23

" " duties B 23

Financial administration B 21

Gifts and bequests C 28

Government, by Constitution, By-Laws and Rules C 3

Grades of membership , C 6

Honorary Vice-Presidents B 32

House Committee, appointment C 45, B 28

" " duties of B 28

Initiatiou fee for (the) several grades C 19, 20, B 18

" " when to be paid C 18. B 18

investments, gifts and bequests- C 27

John Fritz Medal Committee C 46, B 32

Junior dues Rl7

'* member, qualifications of C 11

" " procedure to change grade. . . •. C 15, 20

Letter-ballot on subjects ordered by the Council C 44, B 36

Library Committee, appointment, duties of .C 45, B 27

" Maintenance C 2

" when open R16

Life membership fee C 22

^Managers, members of the Council — C 26

" term of office C 26, 34

Meetings, Annual C 41, B 37, R 2, 3

" Committee, appointment, duties of C 45, B 24

" General C 41, B 37, R 2, 3

•f Special C 42

AMERICAN SOCIETY OF MECHANICAL ENGINEERS IX

Members most sign the Constitution C 18

" qualifications of C 9

Membership Committee, appointment, duties of C 45, B 26

" grades of C 6

Nominating Committee, appointment C 47, 48, B 13, 30, 31

" " duties of B 13, 30, 31 .

Objects of the Society C 2

Officers of the Society C 26, 34, 38

Offices of the Society, location C 4

Papers, professional B 24, R 4-11

Past Presidents, members of Council . . .C 26, 29

President, a member of the Council C 26, 39

" term of office C 34, 36

Press, technical C 54, R 13

Professional Committee .C 49

" papers B 24, R 4-11

Programme of meetings B 24, 37

Proxies C 7. B 41

Publication Conunittee, appointment, duties of C 45, B 25

Quorum for business C 41

Rejection of candidate C 16, B 9

Removal of member of Committee C 51

Report of Professional Conmiittees C 49

" " the Council, annual C 31

Representative delegate B 33

Rights in the Society C 12

Rules, how amended C 60

Secretary, appointment, duties of C 38, B 38

" in the Coimcil C 26

" of Standing Committees C 50

Sections of the Society C 52

Standards shall not be indorsed C 56

Suspension of members C 24

Technical press C 54, R 13

Tellers of Election C 47, B 6-8, 15, 17, 34

Term of office C 34, 3B

Title of the Society CI

Topical subjects B 24

Transactions of the Society. C 53, 54, 55

Treasurer, duties of .C 39, B 39

" term of office.. . . C 34

Vaeancies in office C 28, 33, 40

" " Committees B 22

Vice-Prewdent, member of the Council C 26, 34

" term of office C 34

" Honorary B 33

VotCTB, who are C 6, 18, B 6, 12

." in arrears B 19

AMERICAN SOCIETY OF MECHANICAL

ENGINEERS.

COKSTITUTIOK.

NAME, OBJECT AND GOVEKNMBNT.

C 1. The title of this Society is " The American Society of
Mechanical Engineers.''

C 2. The object of the Society is to promote the Arts and
Sdences connected with Engineering and Mechanical Construc-
tion. The principal means for this purpose shall be the holding
of meetings for the reading and discussion of professional papers,
and for social intercourse; the publication and distribution of its
papers and discussions; and the maintenance of an Engineering
Library.

C 3. The Society shall be governed by this Constitution, and
by By-Laws and Rules in harmony therewith.

C 4. The Society was organized as a Corporation under the
laws of the State of New York, April 7, 1880. Its offices shall
be located in the City of New York.

MEMBEBSHIP.

C 5. Persons connected with the Arts and Sciences relating
to Engineering or Mechanical Construction may be eUgible for
admi^on into the Society.

C 6. The membership of the Society shall consist of Hon-
orary Members, Members, Associates and Juniors. Honorary
Members, Members and Associates are entitled to vote and to
hold office- Juniors shall not be entitled to vote nor to be
oflBcers of the Society, but shall be entitled to the other privileges
of membership.

7. Honorary' Members, Members and Associates are en-

XU CONSTITUTION, BY-LAWS AND RULES OF THE

titled to vote on all questions before any meeting of the Society,
in person or by proxy, given to a voting member. A proxy
shall not be valid for a greater time than six months.

C 8. Honorary Members shall be persons of acknowledged
professional eminence, and their number shall not exceed twenty-
five at any time.

C 9. A Member must have been so connected with Engi-
neering as to be competent, as a designer or as a constructor,
to take responsible charge of work in his branch of Engineering,
or he must have served as a teacher of Engineering for more
than five years. A Member shall be thirty years of age or over.

C 10. An Associate must either have the other qualifications
of a Member or be so connected with Engineering as to be com-
petent to take charge of engineering work, or to co-operate with
Engineers. An Associate shall be twenty-six years of age or
over.

C 11. A Junior must have had such engineering experience
as will enable him to fill a responsible subordinate position in
engineering work, or he must be a graduate of an engineering
school. A Junior shall be twenty-one years of age or over.

C 12. The rights and privileges of every Honorary Member,
Member, Associate and Junior shall be personal to himself, and
shall not be transferable or transmissible by his own act or by
operation of law.

ADMISSION.

C 13. Honorary Members shall be nominated by at least ten
members of the Society. The grounds upon which the nomina-
tion is made, shall be presented to the Council in writing.

C 14. All applications for membership to the grades of
Member, Associate or Junior shall be presented to the Council,
which shall consider and act upon each appUcation, assigning
each approved applicant to the grade of membership to which,
in the judgment of the Council, his qualifications entitle him.
The name of each candidate thus approved by the Council, shall,
unless objection is made by the applicant, be submitted to the
voting membership for election, by means of a letter-ballot.

C 15. Associates or Juniors desiring to change their grade of
membership shall make application to the Council in the sam^
planner as is required in the case of a new applicant.

AMERICAN SOCIETY OF MECHANICAL ENGINEERS xill

C 16. Election to membership shall be by a sealed letter-
ballot as the By-Laws shall provide. Adverse votes to the num-
ber of two per cent, of the votes cast shall be required to defeat
the election of an applicant for any grade of membership. The
Council, may in its discretion, order a second ballot upon a de-
feated applicant, in which case adverse votes to the number of
four per cent, of the votes cast, shall be required to defeat the
election.

C 17. The election of Honorary Members shall be by a vote
of the Council taken by letter-ballot, as provided in the By-Laws.
One dissenting vote shall defeat such election.

C 18. Each person elected, excepting Honorary Members,
shall subscribe to this Constitution, and shall pay the initiation
fee before he can be entitled to the rights and privileges of
membership. If such person does not comply with this require-
ment within six months after notice of his election, he will be
deemed to have declined election. The Council may, thereupon,
declare his election void.

INITIATION FEES AND DUES.

C 19. The initiation fee for membership in each grade shall
be as follows:

For Member Twenty-five Dollars,

For Associate Twenty-five Dollars,

For Junior Fifteen Dollars.

C 20. A Junior, on promotion to any other grade of member-
ship, shall pay an additional fee of Ten Dollars.

C 21. The annual dues for membership in each grade shall
be as follows:

For Member Fifteen Dollars,

For Associate Fifteen Dollars,

For Junior Ten Dollars for the first

six years of his membership and thereafter the

same as for an Associate.

C 22. The Council may in its discretion, permit any Member
or Associate to become a Life Member in the same grade, by the
payment at one time of an amount sufficient to purchase from
the Equitable life Assurance Society of New York, an annuity
on the life of a person of the age of the applicant equal to the

xiv CONSTITUTION, BY-tA\^S AND RULES OP TfiE

annual dues in his grade. Such Life Member shall not be liable
thereafter for annual dues.

C 23. The Council shall have the power, by letter-ballot, to
admit to Life Membership, without the pajinent of a life mem-
bership fee, any person who, for a long term of years, has been
a Member or an Associate when, for special reasons, such pro-
cedure would, in its judgment, promote the best interests of the
Society, provided that notice of such proposed action shall have
been given at a previous meeting of the Council. One dissent-
ing vote shall defeat such admission.

SUSPENSIONS AND EXPULSIONS.

C 24. Any Member, Associate or Junior who shall leave his
annual dues unpaid for one year, shall not receive the volume of
Transactions until such arrears are paid. Any Member, Asso-
ciate or Junior who shall leave his dues unpaid for two years,
shall, in the discretion of the Council, have his name stricken
from the roll of membership, and shall cease to have any further
rights as such.

C 25. The Council may refuse to receive the dues of any
member of any grade, who shall have been adjudged by the
Council to have violated the Constitution or By-Laws of the
Society, or who, in the opinion of the Council by a two-thirds
vote, shall have been guilty of conduct rendering him unfit to
continue in its membership; and the Council may expel such
person and remove his name from the list of members.

THE COUNCIL.

C 26. The afifairs of the Society shall be matiaged by a Board
of Directors chosen from among its Members and Associates,
which shall be styled " The Council." The Council shall consist
of the President of the Society, who shall be the presiding officer,
six Vice-Presidents, nine Managers, the Treasurer and five Past
Presidents. Five members of the Council shall constitute a
quorum for the transaction of business. The Secretary may
take part in the deliberations of the Council, but shall not have
a vote therein. The Chairman of the Finance Committee shall"
attend the meetings of the Council and take part in the discus-
sion of financial questions but shall not have a vote.

AMERICAN SOCiKTY OP MECfiAKlCAL ENGINEERS 5cV

C 27. The five surviving Past Presidents who last held the
office shall be members of the Council with all the rights, priv-
il^es and duties of the other members of the Council.

C 28. The Council thus constituted shall be the legal Trustee
of the Society. All gifts or bequests not designated for a specific
purpose shall be invested by the Council, and only the income
therefrom may be used for current expenses.

C 29. Should a vacancy occur in the Council, or in any elec-
tive office except the presidency, through death, resignation or
other cause, the Council may elect a Member or Associate to fill
the vacancy until the next annual election.

C 30. The Council shall regulate its own proceedings, and
may by resolution delegate specific powers to an Executive
Committee or to any one or more members of the Council. No
act of the Executive Committee or of a delegate shall be binding
until it has been approved by a resolution of the Council.

C 31. The Council shall present at the Annual Meeting of
the Society a report verified by the President or Treasurer or
by a majority of the members of the Council, showing the whole
amount of real and personal property owned by the Society,
where located, and where and how invested, and the amount
and nature of the property acquired during the year immediately
preceding the date of the report, and the manner of the acquisi-
tion; the amount applied, appropriated or expended during the
year immediately preceding such date, and the purposes, objects
or persons to or for which such applications, appropriations or
expenditures have been made; also the names and places of resi-
dence of the persons who have been admitted to membership in
the Society during the last year, which report shall be filed
with the records of the Society, and an abstract thereof shall be
entered in the minutes of the proceedings of the Annual Meeting.

C 32. An act of the Council, which shall have received the
expressed or the implied sanction of the membership at the next
subsequent meeting of the Society, shall be deemed to be the act
of the Society, and shall not afterwards be impeached by any
member.

C 33. The Council may, by a two-thirds vote of the members
present, declare any elective office vacant, on the failure of its
incumbent for one year, from inability or otherwise, to attend
the Council meetings, or to perform the duties of his oflBce, and
shall thereupon appoint a Member or Associate to fill the vacancy

Xvi CONSTITUTION, BY-LAWS AND RULES OF THE

until the next Annual Meeting. The said appointment shall not
render the appointee ineligible to election to any office.

OFFICERS.

C 34. At each Annual Meeting there shall be elected from
among the Members and Associates:

A President to hold office for one year.
Three Vice-Presidents, each to hold office for two years.
Three Managers, each to hold office for three years.
A Treasurer to hold office for one year.

C 35. The election of officers shall be by sealed letter-ballot,
as the By-Laws shall provide.

36. The term of all elective officers shall begin on the ad-
journment of the Annual Meeting of the Society. Officers shall
continue in their respective offices until their successors have been
elected and have accepted their offices.

C 37. A President, Vice-President or Manager shall not be
eligible for immediate re-election to the same office at the expira-
tion of the term for which he was elected.

C 38. The Council, at its first meeting after the Annual
Meeting of the Society, shall appoint a person of the grade of
Member to serve as Secretary of the Society for one year, sub-
ject to removal for cause by an affirmative vote of fifteen mem-
bers of the Council, at any time after one month's written notice
has been given him to show cause why he should not be re-
moved, and he has been heard in his own defense, if he so de-
sires. The Secretary shall receive a salary which shall be fixed
by the Council at the time of his appointment.

C 39. The President, Secretary and Treasurer shall perform
the duties legally or customarily attaching to their respective
offices under the Laws of the State of New York, and such other
duties as may be required of them by the. Council.

C 40. A vacancy in the office of President shall be filled by
the Vice-President, who is senior by age.

MEETINGS.

C 41. The Society shall hold two meetings in each year. The
Annual Meeting shall begin in New York City on the first Tues-
day in December, and a Semi- Annual Meeting shall be held at

AMERICAK SOCIETY OF* MECfiANICAL ENGINEERS Xvii

such time and place a» the Council may appomt. Fifty Memhert
and Associates shall constitute a quorum for the transaction of
business.

C 42. Special meetings of the Society may be called at any
time at the discretion of the Council, or shall be called by the
President upon the written request of fifty members entitled to
vote, the notices for such meetings to state the business for
which such meeting is called, and no other business shall be
entertained or transacted at that meeting.

C 43. Any appropriation recommended by the Society- at a
meeting shall not take effect until it has been approved by the
CounciL

C 44. Every question which shall come before a meeting of
the Society or of the Council or a Committee, shall be decided
by a majority of the votes cast, unless otherwise provided in this
Constitution or the By-Laws, or the Laws of the State of New
York. The Council may order the submission of any question
to the membership for discussion by letter-ballot. Any meeting
of the Society at which a quorum is present, may order the sub-
mission of any question to the membership for discussion by
letter-ballot.

STANDING COMMITTEES.

C 45. The Standing Committees of the Society to be ap-
pointed by the President shall be:

Finance Committee,
Committee on Meetings,
Publication Committee,
Membership Committee,
Library Committee,
House Committee.
C 46. There shall be a John Fritz Medal Committee of three
members appointed as provided in the By-Laws.
C 47. The Annual Committees shall be:

An Executive Committee, appointed by the Council.
A Nominating Committee, appointed by the President.
Tellers as required by the By-Laws, appointed by the
President.
C 48. Special Nominating Committee:

Twenty or more members entitled to vote may constitute

XVm CONSTITUTION, BY-LAWS AND RULES OP tott

themselves a Special Nominating Committee, with the same
powers as the Annual Nominating Committee.

C 49. Pix>fessional Committees:
The Council shall have power to appoint, upon a recom-
mendation of the Society at a general meeting, or upon its own
initiative, such Professional Committees as it may deem desira-
ble, to investigate, consider and report upon subjects of engineer-
ing interest. Eeports of such committees may be accepted by
the Society and printed in the Transactions^ but shall not be
approved or adopted as the action of the Society. Any proposed
expenses of such committees must be authorized by the Council
before they are incurred.

C 50. Each Committee shall perform the duties required of it
in the By-Laws, or assigned to it by the Council. The Secretary
of the Society shall be the Secretary of each of the Standing
Committees.

C 51. The Council may at any time, in its own discretion,
remove any or all members of any Committee, except a Nom-
inating Committee; and the va9ancy, arising from this or from
any other cause, shall be filled by appointment by the President,
except a vacancy in the Executive Committee, which shall be
filled by the Council.

SECTIONS OF THE SOCIETY.

C 52. The Council may, in its discretion, authorize the or-
ganization of sections or groups of any or all grades of member-
ship, for professional or scientific purposes which are in harmony
wi^ the Constitution and By-Laws of this Society. Such sec-
tions or groups may, in the discretion of the Council, be geo-
graphical or professional, and shall have such powers, and act
under such rules and regulations as the Council may from time
to time prescribe.

TBAN8A0TI0N8.

C 53* The official record of technical papers and discussion,
shall be known as the Transactions of the Society, and shall be
published under the direction of the Council. There may be
included therein, the annual report of the Council, reports of
Committees, and business records of the Society.

AMERICAN SOCIETY OF MECHANICAL ENGINEfcR^ XIX

C 54. The Society shall claim no exclusive copyright to any
papers read at its meetings, or any reports or discussions thereon,
except in the matter of their official publication under the So-
ciety's imprint as its Transactions. The policy of the Society
shall be to give the professional and scientific papers read before
it the widest circulation possible, with the view of making the
work of the Society known, encouraging Engineering progress
and extending the professional reputation of its members.

C 55. The Society shall not be responsible for statements or
opinions advanced in papers or in discussions at its meetings.
Matters relating to politics or purely to trade shall not be dis-
cussed at a meeting of the Society, nor be included in the
Transactions.

C 56- The Society shall not approve or adopt any standard
or formula, or approve any engineering or commercial enterprise.
It shall not allow its imprint or name to be used in any commer-
cial work or business.

AMENDMENTS TO THE CONSTITUTION.

C 57. At any semi-annual meeting of the Society any mem-
ber may propose in writing an amendment to this Constitution.
Such proposed amendment shall not be voted on at that meeting,
but shall be open to discussion and to such modification as may
be accepted by the proposer. The proposed amendment shall
be mailed in printed form by the Secretary to each member of
the Society entitled to vote, at least sixty days previous to the
next annual meeting, accompanied by comment by the Council,
if it so elects. At that annual meeting such proposed amend-
ment shall be presented for discussion and final amendment, and
shall subsequently be submitted to all members entitled to vote,
provided that twenty votes are cast in favor of such submission.
The final vote on adoption shall be by sealed letter-ballot, clos-
ing at twelve o'clock noon on the first Monday of March
following.

C 58. The letter-ballot, accompanied by the text of the pro-
posed amendment, shall be mailed by the Secretary to each
member of the Society entitled to vote at least thirty days pre-
vious to the closure of the voting. The ballots shall be voted,
canvassed and announced as provided in the By-Laws. The
adoption of the amendment shall be decided by a majority of the

XX CONSTITUTION, BY-LAWS AND RULES OF THE

Votes cast. An amendment shall take effect on the announce-
ment of its adoption by the Presiding Officer of the semi-annnal
meeting next following the closure of the vote.

AMENDMENTS TO BY-LAWS AND BCJLES.

C 59. For the further ordering of the affairs of the Society,
the Council may, by a two-third vote of its members present,
amend the By-Laws in harmony with this Constitution, pro-
vided that a written notice of such proposed amendment shall
have been given at the previous regular meeting of the Council;
and provided further that the Secretary shall have mailed to
each member of the Council a copy of such proposed amend-
ment, at least thirty days in advance of the meeting of the
Council at which action is to be taken. The amendment shall
take effect immediately on its passage by the Council. The
Secretary shall at once mail a copy of such amendment to the
members of all grades.

C 60. The Council may, by a majority vote of the members
present at any meeting, establish, amend or annul Eules for the
conduct of the business affairs of the Society; for the ordering
and conduct of its professional or business meetings; and for
guidance of its committees in their work and reports; provided
that such Eules are in harmony with the Constitution and By-
Laws of the Society.

CONSTITtJTION GOES INTO EFFECT.

C 61, This Constitution shall supersede all previous Rules of
the Society, and shall go into effect on the announcement by the
Presiding Officer of its adoption.

BY-LAWS.

CANDIDATES FOB MEMBERSHIP.

B 1. A candidate for admission to the Society as a Member
or as an Associate must make application on a form approved by
the Council, upon which he shall write a statement giving a
complete account of his qualifications and engineering experience,
and an agreement that he will, if elected, conform to the Con-

AMEBICAN SOCIETY OP MECHANICAL ENGINEERS XXI

stitution, By-Laws and Eules of the Society. He must refer to
at least five Members or Associates to whom he is personally
known.

B 2. Applications for membership from Engineers who are
Dot resident in the United States or Canada, and who may be
so situated as not to be personally known to five Members of the
Society, as required in the foregoing paragraph, may be recom-
mended for ballot by five members of the Council, after sufficient
evidence has been secured to show that in their opinion the appli-
cant is worthy of admission to the grade which he seeks.

B 3. A candidate for admission to the Society as a Junior
must make application in the same manner as provided for Mem-
bers, except that he must refer to not less than three Members
or Associates to whom he is personally known.

B 4. Keferences shall not be required of candidates for Hon-
orary Membership.

B 5. The references for each candidate for admission to the
Society shall be requested to make a confidential communication
to the Membership Committee, setting forth in detail such in-
formation, personally known to referee, as shall enable the
Council to arrive at a proper estimate of the eligibility of the
candidate for admission to the Society.

ELECTION OF MEMBERS.

B 6. The Secretary shall mail to each member entitled to
vote, at least thirty days in advance of each annual or semi-
annual meeting, a ballot stating the names and the respective
grades of the candidates for membership in the Society which
have been approved by the Council, and the time of the closure
of voting. The voter shall prepare his ballot by crossing out
the names of candidates rejected by him, and shall enclose said bal-
lot in a sealed blank baUot envelope which he shall then enclose in
a second sealed outer envelope on which he shall, for identifica-
tion, write his name in ink. The ballot thus prepared and en-
closed shall be mailed or delivered unopened to the Tellers of
Election. The Secretary shall certify to the competency, and
the signature of all voters. On the closure of voting, the Tellers
of Election shall first open and destroy the outer envelopes, and
shall then canvass the ballots, and certify the result to the meet-
ing of the Society.

Xxii CONSTITUTION, BY-LAWS AND RULES OF THE

B 7. The Tellers shall not receive any ballot after the stated
time of the closure of voting. A ballot without the endorsement
of the voter, written in ink on the outer envelope, is defective,
and shall be rejected by the Tellers of Election.

B 8. The names of those persons elected to membership, with
their respective grades, shall be embodied in a written report,
signed by the Tellers, and presented to the next meeting of the
Society. The President shall then declare them duly elected to
membership in the Society. The Tellers may, through the
Secretary, in advance of any meeting advise each candidate of
the result of the canvass of the votes in his case. The names of
applicants who are not elected shall neither be announced nor
recorded in the Transactions.

B 9. The endorsers of an applicant who has not been elected,
may, with his consent, present to the Council a written request
for a re-submission of his name to ballot. The Council may, in
its discretion, by a three-fourths vote of the members present,
order the name of the applicant placed on the next ballot for
members.

B 10. Election to Honorary Membership shall be by letter-
ballot of the Council. A notice of such proposed election shall
be mailed by the Secretary to each member of the Council at
least sixty days in advance of the date set for the closure of
such election.

B 11. Each person elected to membership, except an Hon-
orary Member, must subscribe to the Constitution, By-Laws and
Kules of the Society, and pay the initiation fee before he can
receive a certificate of membership in the Society,

ELECTION OF OFFICERS.

B 12. The Secretary shall mail to each member entitled to
vote, at least thirty days before the Annual Meeting, the names
of the candidates for office proposed for election by the Nom-
inating Committee.

B 13. The names of the candidates proposed by the Nom-
inating Committee or Committees, and the respective offices for
which they are candidates, shall be printed in separate lists on
the same ballot sheet, each list of candidates to be printed under
the names of the members of the particular committee which
proposed it.
59

AMERICAN SOCIETY OF MECHANICAL ENGINEERS XXlll

B 14. The name of any candidate on the ballot may be
erased, and the name of any person qualified to hold the ofBce
written in its stead. The voter shall make a cross with a pen
or pencil before the name of each candidate for office for whom
he wishes to vote. The ballot thus prepared must be voted and
canvassed in the same manner as for the election of members.

B 15. At the first session of the Annual Meeting, the
Tellers of Election of Officers shall canvass the votes cast for
the officers of the Society in the manner prescribed for the elec-
tion of members, and immediately report the result of the can-
vas to the meeting. The President shall then announce the
candidates having the greatest number of votes for their respec-
tive offices, and declare them elected for the ensuing year.

B 16. In case of a tie in the vote for any officer, the Presi-
dent or, in his absence, the Presiding Officer shall cast the decid-
ing vote.

B 17. A ballot which contains more names marked by a cross
on it than there are officers to be elected, is thereby defective,
and shall be rejected by the Tellers.

FEES AND DUES.

B 18. The initiation fee and annual dues of the first year
shall be due and payable on notice of election to membership,
and upon that payment the member will be entitled to the Trcma-
acti/ms for the year. Thereafter the annual dues shall be due
and payable on the first day of October in each year.

B 19. A member in arrears for one year shall not be entitled
to vote until such arrears have been paid. Should the right to
vote be qu^tioned, the books of the Society shall be conclusive
evidence.

B 20. The Secretary shall present to the Council the name of
any Member, Associate or Junior in arrears for more than one
year, and such member shall not receive the Transaeiions until
such arrears are fully paid. A person dropped from the rolls
for non- payment of dues may, in the discretion of the Council,
be restored to the privileges of membership, upon payment of
all arrears.

FINAKOIAL ADMINISTRATION.

B 91. The Council at its first meeting in each fiscal year,
shall consider the recommendations of the Finance Committee

XXIV CONSTITUTION, BY-LAWS AND RULES OF THE

concerning the expenditure necessary for the work of the Society
during that year. The apportioning of the work of the Society
among the various Standing and other Committees shall be on
a basis approved by the Council and in harmony with the Con-
stitution and By-Laws. The appropriations approved by the
Council, or so much thereof as may be required for the work of
the Society, shall be expended by the various Committees of the
Society, and all bills against the Society for such expenditure
shall be certified by the Committee making the expenditure and
shall then be sent to the Finance Committee for audit. Money
shall not be paid out by any officer or employee of the Society
except upon bills duly audited by the Finance Committee, or by
resolution of the Council.

COMMITTEES.

B 22. The President within one month after the Annual
Meeting shall fill all vacancies in the Standing Committees by
appointment from the membership of the Society.

Each of the Standing and the Annual Committees, shall, at
their first meeting after the Annual Meeting, elect a Chairman
to serve for one year. The President shall appoint the Chair-
man of each Professional Committee. A member of a Standing
Committee whose term of office has expired, shall continue to
serve until his successor shall have been appointed.

FINANCE COMMITTEE.

B 23. The Finance Committee shall consist of five Members
or Associates. The term of office of one member of the Com-
mittee shall expire at the end of each Annual Meeting. This
Committee shall, in the discretion of the Council, have a super-
vision of the financial affairs of the Society, including the
books of account. The Committee may cause the accounts of
the Society to be audited and approved annually by a chartered
or other competent public accountant. The Committee shall
hold monthly meetings for the audit of bills and such other busi-
ness as shall come before it and shall deliver to the Secretary for
presentation to the Council at the end of each fiscal year, a re-
port of the financial condition of the Society for the past year,
find also shall present therewith a detailed estimate of the prob-

AMERICAN SOCIETY OF MECHANICAL ENGINEERS XXV

able income and expenditure of the Society for the following
twelve months. It shall make recommendations to the Council
as to investments, and, when called upon by the Council, advise
upon financial questions.

OOMMITTEB ON MEETINGS.

B 34. The Committee on Meetings shall consist of five per-
sons who may be members of any grade. The term of office of
one member of the Committee shall expire at the end of each
Annual Meeting. It shall be the duty of the Committee to pro-
cure professional papers, to pass upon their suitability for pres- ,
entation, and to suggest topical subjects for discussion at the
meetings. The Committee may refer any paper presented to
the Society to a person or persons, especially qualified by the-
oretical knowledge or practical experience, for their suggestions
or opinions as to the suitability of the paper for presentation.
Papers from non-members shall not be accepted except by unan-
imous vote of the Conmiittee.

The Committee shall arrange the programme of each meeting
of the Society, and shall have general charge of the entertain-
ments to be provided for the members and guests at each meet-
ing. It shall prohibit the distribution or exhibition at the head-
quarters or at the meeting places of the Society of all advertising
circulars, pamphlets or samples of commercial apparatus or
machinery. At the end of each fiscal year, the Committee shall
deliver to the Secretary for presentation to the Council, a de-
tailed report of its work.

PUBLICATION COMMITTEE.

B 25. The Publication Committee shall consist of five Mem-
bers or Associates. The term of office of one member shall ex-
pire at the end of each Annual Meeting. The Committee shall
review aU papers and discussions which have been presented at
the meetings, and shall decide what papers or discussions, or
parts of the same, shall be printed in the Transactions of the
Society. The Committee will be expected to publish all such
data as will be of assistance to engineers or investigators in their
work. At the end of each fiscal year, the Committee shall
deliver to the Secretary for presentation to the Council, a de-
tailed report of its work.

XXVI CONSTITUTION, 'BY-LAWS AND RULES OF THE

MEMBERSHIP COMMITTEE.

B 26. The Membership Committee shall consist of five Mem-
bers OP Associates. The term of oflSce of one member of the
Committee shall expire at the end of each Annual Meeting. It
shall be the duty of this Committee:

To meet monthly to receive and scrutinize aU appli-
cations for membership to the Society.
To send to each voting member the name, qualifica-
tions, engineering experience and references of
each applicant, together with extracts from the
Constitution and By-Laws relating to member-
ship.
To seek further information as to the qualifications
of an applicant, whose evidence of eligibility is
not clear to the Committee.
To report to each session of the Council the names
of all applicants under consideration together
with the action of the Committee on each.
The Committee shall at once destroy all correspondence in
relation to each applicant when his name has been placed on the
ballot by order of the Council, or upon the withdrawal of the
application.

LIBRARY COMMITTEE.

B 27. The Library Committee shall consist of five Members,
Associates or Juniors. The term of oflSce of one member of the
Committee shall expire at the end of each Annual Meeting. It
shall be the duty of the Library Committee to take charge of
the library of the Society, the historical relics, the paintings
and objects of art, and to recommend to the Council suitable
regulations for their care and use. At the end of each fiscal
year, the Committee shall deliver to the Secretary, a detailed
report of its work.

HOUSE COMMITTEE.

B 28. The House Committee shall consist of five Members,
Associates or Juniors. The term of oflSce of one member of the
Committee shall expire at the end of each Annual Meeting. It
shall be the duty of the House Committee to have the care,
management and maintenance of the house of the Society and
its furnishings. They may make rules for the care and the use

AMEKICAN SOCIETY OF MECHANICAL ENGINEERS XXVll

of the Society House, subject to the approval of the Council.
At the end of each fiscal year, the Committee shall deliver to
the Secretary a detailed report of its work.

EXECUTIVE COMMITTEE.

B 29. The- Council shall appoint from its members an Execu-
tive Committee to act for the Council during the interval between
its sessions. The Committee shall make a report of its acts to
each session of the Council for approval. The Secretary may
take part in the deliberations of the Executive Committee, but
shall not have a vote therein.

NOMINATING COMMITTEES.

B 30. A Nominating Committee of five Members, not mem-
bers of the Council, shall be appointed by the President within
three months after he assumes office. It shall be the duty of
this Committee to send to the Secretary on or before October
first the names of consenting nominees for the elective offices
next falling vacant under the Constitution. Upon the request
of any Member or Associate, the Secretary shall furnish to the
applicant the names of such nominees.

B 31. A special Nominating Committee if organized, shall,
on or before October twentieth, present to the Secretary the
names of the candidates nominated by it for the elective offices
next falling vacant under the Constitution, together with the
written consent of each.

JOHN FRITZ MEDAL COMMITTEE.

B 32. The John Fritz Medal Committee shall consist of three
persons of the grade of Member, to be appointed by the Council.
The term of office of one member of this Committee shall expire
at the end of each annual meeting. The duty of this Conmiittee
shall be to represent the Society in the Board of Trustees of the
John Fritz Medal Fund Corporation.

REPRESENTATIVE DELEGATES.

B 33. The Council may in its discretion appoint a member
or members of the Society or other person or persons to repre-

XXViii CONSTITUTION, BY-LAWS AND RULES OF THE

sent it at meetings of Societies of kindred aim or at public
functions. Such delegates shall be designated as "Honorary
Vice-Presidents," and their duties shall terminate with the occa-
sion for which they were appointed.

TELLERS.

B 34. The Presiding Officer shall, at the first session of the
Annual Meeting, appoint three Tellers of Election of officers,
whose duties shall be to canvass the votes cast, and report the
result to the meeting. Their term of office shall terminate
when their report of the canvass is presented to the meeting.

B 35. The President within one month after assuming office
shall appoint three Tellers of Election of members to serve for
one year, whose duties shall be to canvass the votes cast for
members during the year, and to certify the same to the Presi-
dent. They shall notify candidates through the Secretary of
the result of such election.

B 36. The President shall appoint three Tellers to canvass
any letter-baEots which shall be ordered by the Council or by
the Society.

MEETINGS.

B 37. The meetings of the Society shall continue from day
to day as the meeting may decide. The business session of the
Annual Meeting shall be held on Wednesday following the first
Tuesday of December. The professional sessions for the reading
of papers shall be held at such times and places as the meeting
may appoint. Notices of all meetings of the Society shall be
mailed by the Secretary to members of all grades not less than
thirty days before the date of such meeting.

SECRETARY.

B 38. The Secretary of the Society shall be the Secretary to
the Council and also to each of the Standing Committees.

The Secretary shall, under the supervision of the Finance
Committee, have charge of the Books of Account of the Society.

He shall make and collect all bills against members or others.

He shall have charge of all bills against the Society, shall

AMERICAN SOCIETY OF MECHANICAL ENGINEERS Xxlx

keep an account of the same, and shall present •em m proper
form to the Finance Committee for audit.

AIL funds received by any person for the Society, shall be de-
Uvered to the Secretary. He shall immediately enter them in
the Books of Account, and shall immediately deposit such funds
as he receives, to the credit of the Society, in a Bank to be des-
ignated by the Council.

TREASURER.

B 39. The Treasurer shall make payments only on the audit
of the Finance Committee, or upon the direction of the Council,
by resolution of that body. He shall furnish a bond for the
faithful performance of his duties to such amount as the Council
may require, such bond to be procured from an incorporated
Guarantee Company, at the expense of the Society.

TITLES, EMBLEMS, CERTIFOATB.

B 40. Each Member and Associate shall, subject to such rules
as the Council may establish, be entitled on request, to a certifi-
cate of membership, signed by the President and Secretary of
the Society. Every such certificate shall remain the property
of the Society, and shall be returned to it on demand of the
Council-

B 41. Each proxy authorizing a person to vote for an absent
member, shall be signed by such absent member, with an attest-
ing witness, and be submitted to the Secretary for verification
of the member's right to vote at the meeting at which the right
is to be exercised.

B 42. The emblem of each grade of membership approved
by the Council shall be worn by those only who belong to that
grade. The official stationary shall be used only by Officers
and Committees of the Society.

B 43. The abbreviation of the titles of the various grades of
membership approved by the Society are as follows:

For Honorary Members, . . Hon. Mem. Am. Soc. M. E.

For Members, Mem. Am. Soc. M. E.

For Associates, Assoc. Am. Soc. M. E,

For Juniors, Jun. Am. Soc. M. E.

XXX OOitSTITUTiON, BY-LAWS AND RULES OF THE

RULES.

R 1. The Secretary's oflBce shall be open on business days
from 9 A.M. to 5.30 p.m. During the Annual Meeting, the office
shall be open from 9 a.m. to 10 p.m. A register shall be kept
for each regular meeting, to record the attendance of members
and guests.

E 2. The Secretary shall provide a numbered badge or pin
for each member or guest attending the regular meetings, the
number on the badges to correspond with the member's of
guest's number on the register.

E 3. The Secretary shall at each regular meeting of the So-
ciety distribute at the headquarters a printed list of the names
registered at the meeting.

E 4. Copies of papers to be read and discussed at any meet-
ing shall be sent to each member thirty days in ^vance of that
meeting. A paper received too late for such distribution shall
only be accepted for presentation at that meeting by unanimous
consent of the Committee on Meetings. A blank shall accom-
pany the papers by which a member may signify his intention
to discuss any of the papers, and priority in debate shall be given
in the order of the receipt by the Secretary of such notification.

E 5. At professional sessions, each paper shall be read by
abstract only, ten minutes being allowed to the author for the
presentation, unless otherwise ordered by the meeting.

E 6. A member who has given notice of his intention to dis-
cuss a paper, and shall have reduced his discussion to writing,
shall be entitled to ten minutes for its presentation.

E 7. Each speaker shall be limited to five minutes in the oral
discussion of a paper, unless the time should be extended by
unanimous consent. A member who has once had the floor
cannot claim it again until all the others have been heard who
desire to speak on that paper. Authors may have five minutes
to close the discussion on the paper.

E 8. Members unable to attend the meeting may send a dis-
cussion of any paper ill writing, to be presented by the Sec-
retary.

E 9. The Committee on Meetings shall deliver to the Secre-
tary such papers as they reconmiend for presentation to the
professional meetings of the Society.

AMERICAN SOCIETY OF MECHANICAL ENGINEERS XXXi

R 10. The Secretary shall have sole possession of papers and
iUostrations between the time of their approval by the Commit-
tee on Meetings, and their presentation to the professional session
of the Society.

R 11. After the presentation and discussion of a paper, a
copy of both shall be sent to the author, and, so far as possible,
a copy of the reported discussion shall be sent to each member
who presented it, with the request that he correct errors or
omissions, and return the same promptly to the Secretary.

R 12. Members may order reprints of papers at a price suffi-
cient to cover the cost to the Society, provided that said copies
are not for sale.

R 13. The Secretary may furnish to the author twenty copies
of his paper without charge. He may also furnish to the tech-
nical press such papers in advance of the meeting as they may
wish to publish after presentation to the meeting of Society.

R. 14. The entertainments to be provided for the members
and guests at any meeting of this Society in any city shall be in
charge of a Local Committee, subject, however, to the general
approval of the Committee on Meetings.

R 15. A member may invite a non-member to the profes-
sional sessions of the meeting, but the guest shall not take part
in the proceedings without an invitation from the Presiding
Officer. Invitations to guests of members for the entertain-
ments provided for the Society shall be in the discretion of the
Local Committee.

R 16. The Society House shall be open at all hours for access
to members. The Library shall be open on all week days be-
tween the hours of 10 o'clock a.m. and 10 o'clock p.m. It shall
be conducted as a Free Public Reference Library of Engineering
and the Allied Arts and Sciences.

R 17- Juniors who were elected to membership in the Society
six years or more previous to the adoption of this Constitution,
shall pay the same dues as an Associate, beginning with the
fiscal year which opens after such adoption. Juniors, who have
been elected less than six years before that date, shall pay the
dues of an Associate on the expiration of six years after their
election.

CONTENTS OF VOLUME XXV.

New York (48th) Meeting.

TJLQM

No. 1007 Proceedings of the New York

(48th) Meeting 3

No. 1008 Carnegie Gijt to En-
gineering (Second circular, Appendix to Pro-
ceedings) 34

No. 1009 Dodge, James M President's Address, "The Money

Value of Technical Trammg " . . 40 ly^

No. 1010 Barth, Carl G Slide Rules for the Machine Shop as

a part of the Taylor System of
Management 49

No. 1011 Gantt, H. L Modifying Systems of Management. 63 *--^^

No. 1012 Richards, Frank Is Anything the Matter with Piece-
work? 68

No. 1013 ScHEPFLER, F. A Suggestions for Shop Construction... 93

No. 1014 Sweet, John E What are the New Machine Too|s to

Be? 100

Na 1015 WiCKHORflT, Max H. .Air Motors and Air Hanuners —

Apparatus and Methods for Test-
ing 107

No. 1016 Perry, Frank B... . .A Method for Determining Rates

and Prices for Electric Power 120

No. 1017 Bunnell, S. H Improvement in Valve-Motion of

Duplex Air-Compressors 138

No. 1018 Farwell, E. S Tests of a Direct-Connected Eight-
foot Fan and Engine 145

No. 1019 Cioss, W. F. M A Series Dbtilling Apparatus of

High Efficiency 160

No. 1020 Miller, E. F The Pressure-Temperature Curve of

• Sulphurous Anhydride (S Ot) 176

No, 1021 Gregory, W. B The Pitot Tube 184

No. 1022 Allen, Benj. T Construction and Efficiency of a

Fleming Four-valve Engine, Di-
rectly Connected to a 400 K. W.
Generator 212

No. 1023 Morgan, C. H A Compact Gas-Engine : Beam Type 245

No. 1024 Jacjobus, D. S Tests of a Compound Engine Using

Superheated Steam 264

No. 1025 Behtsch, J. C Standard Unit of Refrigeration 292

No. 1026 Report of Conmiittee on Specifica-
tions for Boiler Plate, Rivet Steel,
Steel Castings and Steel Forgings
—Monthly Reunion, Feb., 1904. 321

XZXiy OONTENTS.

Chicago, III. (49th) Mbetino.

PAOI

No. 1027. .* BiRNiE, R Ordnance for the Land Service 355

No. 1028 Proceedings of the Chicago (49th)

Meetmg 421

No. 1029 Carnegie Gijt to En-
gineering (Third circular, Appendix to Pro-
ceedings) 437

Np. 1030 Marks, L. S Use of Superheated Steam and of

Reheaters in Compound Engines

of large size 443

No. 1031 Flint, Wm. P Commercial Gas-Engine Testing and

Proposed Standard of Comparison 509

No. 1032 HrrcHOOCK, E. A Road Tests of Consolidation Freight

Locomotives 550

No. 1033 Testing Locomotives in England —

(Presented by the Institution of
Mechanical Engineers) 589

No. 1034 Campbell, Wm Appendix IV. to Vlth Report of the

Alloys Research Committee,
Effects of Strain and of Anneal-
ing — (Presented by the Institu-
tion of Mechanical Engineers) 599

No. 1035 Nioolson, J. T Experiments with a Lathe-Tool

Dynamometer — (Presented by
the Institution of Mechanical En-
gineers) e37

No. 1036 Wells, J. H Power Plantrof the Tall Office Build-
ing ..... 685

No. 1037 HoDGKiNSON, F Some Theoretical and Practical Con-
siderations in Steam Turbine
Work 716

No. 1038 Rateau, A Different Applications of Steam

Turbines — (Presented by the In-
stitution of Mechanical Engineers) 782

No. 1039 Goss, W. F. M Locomotive Testing Plants 827

No. 1040 Emerson, H A Rational Basis for Wages *868 '

No. 1041 Keep, W. J Cast Iron, Strength, Composition,

Specifications 884

No. 1042 Kerr,C. V Potential Efficiency of Prime Movers 920

No. 1043 Raven, Vincent L Middlesborough Dock Electric and

Hydraulic Power Plant — (Pre-
sented by the Institution of
Mechanical Engineers) 943

No. 1044 Russell, C. N Refuse Destruction by Burning, and

the Utilization of Heat Generated
— (Presented by the Institution
of Mechanical Engineers) 982

No. 1046 Bolton, R. P Power Plant of Tall Office Buildings 1011

DONTSNT& tXX7

PAOI

Ho. 1046 &MMfe^, W. L. R Steam turbine in Modem Egineer^

ing 1041

No. 1047 Lka, E. S., and

Meden E DeLaval Steam Turbine 1066

No. 1048 Watson, Geo Burning of Town Refuse— (Pre-
sented by the Institution of Me-
chanical Engineers) 1074

No. 1049 HiTTTON, F.R Robert Henry Thurston, a Me-
morial 1113

No, 105O Memorial Notices of Members De-
ceased during the Year 1121

;

LIST OF ILLUSTRATIONS.

riQ. PAGE

1. Location New Engineering Building 38

2. Curves iUustrating money value of technical training 42

3. Slide rules for the machine shop 62

4. " rule 54

5. " " Faces 54

6. Qreular time slide rule 69

7. " speed " " 60

8. " spur gear slide rule 61

9. Piece and day work diagram 70

10. Day work diagram 76

11. Piece " " 77

12- Halsey premiimi plan diagram 77

13. Taylor differential piece-work diagram 78

14. Gantt, bonus system diagram 79

15. Emerson parabolic diagram 80

16. Parkhurst combination diagram 80

17. Shop construction diagram 94

18. Buriington route laboratory 108

19. Testing i^paratus 109

20. Arrangement for testing motors Ill

21. Motor record diagram 112

22. Arrangement for testing air-hammers 113

23. Hammer record diagram 116

24. Diagram of No. 6 riveting hammers 118

25. Diagrams of rates and prices for electric power 122

26.
27.
28.
29.
30.
31.

127
127
128
129
130
132

32. " " " " " " 132

33. Arrangement of steam valves — Meyer cut-off and regulating bracket . 139

34. Valve diagram, duplex gas compressor 140

35. Duplex air-compressor 142

36. Baght-footfan 148

37. Pressure, volume, indicated horse-power and efficiency curves 150

38.

151

39. Efficiency curves 153

40. Curves showing relation between vacuum and inlet and cubic feet of

Mr discharged per minute 158

41. Series distilling apparatus 161

XXXVlll LIST OF ILLUSTRATIONS.

wo. FAOl

42. Section of a single elemoit — of series distilling apparatus 162

43. Diagram of a distilling apparatus 163

44. " " " " 165

45. Sevoi chamber series distilling apparatus 172

46. Pressure temperature curve of sulphur dioxide 180

47. Firstformof Pitot tube 185

48. Drawing of tube as described by Pitot 186

49. Form of Pitot tube used by Darcy and Basin 187

50. Pitot tube nozzles 190

51. Curves of velocity and pressure 199

52. Location of piezometers 200

53. Curves of velocity and pressure 201

54. Arrangement of tubes 203

55. A form of Pitot tube — and gauges 205

56. Section through Pitot tube 206

57. Pitot tube gauge 210

58. " " " 210

59. Specimen indicator cards, test at about one-sixth load 213

60. " " " " " five-eighths load 214

61. " " " " " seven-tenths load 215

62. " " " "atfullload 217

63. " " " " at about li^load 218

64. Curve of steam consumption 219

65. Diagram at seven tenths load 220

66. " atfullload 221

67. Friction cards 223

68. Overload " 224

69. Direct connected engine and generator 226

70. Curve showing relation between total steam per hour and total power

developed 229

71. Cross section of valves and cylinders v 236

72. Plan and side elevation of engine 237

73. Inertia governor 238

74. Foundation plan 241

75. Compact gas-engine 246

76. Gas-engines 247

77. " showing partial section through cylinder 248

78. Diagram showing comparison between beam and ordinary type gaa-

engine 249

79. Inlet valve 250

80. Valve gear 252

81 . Connections for beam and direct-connected types of gaa-engines 254

82. Diagram of forces 256

83. Defective forces on crank pin 258

84. Diagrams of test of compound engine using superheated steam 263

Qc u tt it *f tt tt u 2^7

Qrt a tt tt tt tt tt tt 270

QJ tt tt tt tt tt tt tt 271

fifi ** ** ** " " " " . 272

LIST OF ILLUSTRATIONS. XXXix

no. VABM

89. Diagrazns of test of compound engine using superheated steam 273

gQ it tt a it tt it tt 274

91. " " " " " " " 275

92. Standard test specimen >. 323

93. " " " 324

94. Diagram showing test of soft steel 350

95. " " " " " 351

96. " " " " Bessemer machinery stock 352

97. " " " open-hearth, common spring stock 353

98. EHastic resistance of guns 384

99. Diagram of 8-mch B. L. steel rifle 386

100. Diagrams of shrinkage, pressures and strains 387

101. " stresses... '. 388

102. Diagram of initial tension and hollow cylinder. 389

103. Modem field gun 392

104. Limber for field gun 392

105. Breech mechanism 6-inch R. F. gun — model of 1900 398

106. The Before 6-inch R. F. gun 399

107. Sixteen-inch B. L. rifle in proof carriage 401

108. Twelve-inch B. L. motor and carriage 403

109. Fragamentation 12-inch A. P. shell charged with Maxrinate 406

110. " " " " " " explosive D 406

111. Accuracy targets 10 and 12-inch B. L. rifles 409

112. Barbette mount, Taku forts, China, after a battle 411

113. Forty-five-inch shield. Barbette mount, 6-inch gun after proof firing. 413

114. Twelve-inch Buffington-Crozier disappearing carriage, model of 1901. 415

115. Warner and Swasey depression range finder 417

1 16. Engine A 444

117. Sectional elevation of upper half of 60-inch x 56-inch L. P. Cylinder. 446
117a. " plan of 60-inch x 56-inch L. P. Cylinder, showing admission

and exhaust valve 447

117b. Side elevation of 60-inch x 56-inch L. P. Cylinder showing valve gear. 448

118. Engine room of the L. Street Station of the Boston Electric Light

Company showing Engine A 450

119. Test 1 on Engine A, full load with jackets and reheaters 453

120. "2, " u a a u a u u 454

121. " 3, " B, quarter load, with jackets and reheaters 4^7

122. " 4, " " half load, with jackets and reheaters 458

123. " 5, " " " " without jackets and reheaters 459

124. " 6, " " three-quarters load, with jackets and reheaters . . 460

125. " 7. " " " " without jackets and reheaters 461

126. "8, " " full load with jackets and reheaters 462

127. " 9, " " " " without jackets and reheaters 463

128. " 10, " " one and one-quarter load, with jackets and re-

heaters 464

129. Test 11, Engine B, one and one-quarter load, without jackets and re-

heaters 465

130. Test 12, Engine B, full load, with jackets and reheaters 466

131. Results of Tests of Engine B 467

Xl LIST OF ILLUSTRATIONa

FIG. PAOK

132. Engine room at the Atlantic Avenue Station of the Edison Electric

Illuminating Company — showing engines C. D. E. and F 468

133. EngineC 469

134. Test 13, Engine C, full load with jackets and reheatere 472

135. " 14, " " " " without jackets and reheatere 473

136. " 15, " " three-quartere load, with jackets and reheatere 474

137. " 16, " " half load with jackets and reheatere 475

138. "17, " D, full " " " " " 477

139. " 18, " " " " without jackets and reheatere 478

140. " 19, " E, " " with jackets and reheatere 479

141. "20, " "half " " " " " 480

142. " 21, " F, full " " " " " 481

143. " 22, " G, " " without reheatere 482

144. " 23, " " " " with " 483

145. " 24, " H,over " without reheatere 484

146. " 25, " K, full " with " 485

147. " 26, " " half " " reheatere 486

148. " 27, " " quarter load with reheatere 487

149. " 28, " " half load without reheatere 489

150. " 22, " G,full " with " 490

151. "23, " " " " " ' " 492

152. " 24, " H, over " without reheatere 493

153. " 25, " K, full " with " 494

154. "26, " "half " " " 495

155. " 28, " " " " without " 496

156. Temperatvu-e-Entropy Diagram, full load test with reheater. Engine K 497

157. Indicator diagram, high-pressure head end 506

158. " " Engine A, high-pressure crank end 506

159. " " " " low-pressure head end 507

160. Engine A, low-pressure crank end 507

161. 300 B. H. P. double acting tandem gas-engine 510

162. 25 B. H. P. single acting vertical gas-engine 512

163. Gas-engines ready for shop tests 514

164. Diagram, 8 x 10 and 16i x 24 gas-engine, shop tests 516

165. " " " " throttling gas-engine, shop tests 518

166. " 6x7 " 7x10 hit-and-miss gas-engmetests 521

167. Otto suction producer gas-engine 530

168. " " " " 531

169. 60 Horse-power Otto suction 'producer 532

170. Explosion diagram 538

171. " " 538

172. " " 539

173. " " 539

174. Gas-engine cards — ^no load 541

175. " cards— half load 541

176. " caids— full load 542

177. " " 544

178. " " 545

179. Broolcs Lpcomotive No. 230, Run No. 1, cards from left cylinder $51

LIST OF ILLUSTRATIONS.

xli

FIA.

•

TAQM

180. Brooks Locomotive No. 230, Run No. 1, cards from right cylind

552

181.

tt (

1 H It tt

" left

555

Iffi.

it t

t It tt 11^

" right

556

183.

Bddwin '

No. 240, Run No. 3, "

" left

559

184.

II t

t ft ft tt

" right

560

185.

tt 1

t ft tt tt

" left

563

186.

tt t

t ft II ti

" right

564

187.

Brooks

No. 257, Run No. 4, "

'' left

567

188.

tt t

t It It ft

" right

568

189.

tt t

t ft It It

" left

571

190.

ft I

t II It It

" right

572

191.

tt t

" Run No. 5, "

" left

575

192.

tt t

t ft ft It

" right

576

193.

tt t

t ti II ft

" left

579

194.

ft t

t ft It It

" right

580

195.

Diagram showing profile, horse-powers, time,

steam pressures, etc..

etc., for Brooks Locomotive No. 230, Rtm No. 1,

and Baldwin

240, Rmi
Diagram sho

L No. 3

FnoM

580

196.

wing profile, horse-powers, time

steam pressures, etc.,

etc., for Brooks No. 257, Run No. 4, and No. 6

.Faces

580

197.

Diagram showing profile, horse-powers, time.

steam pressures, etc..

etc., for Brooks 257 Run No. 5, and 7. . .

• Faces

580

198.

Diagram sho

wing speeds, horse-powers, etc..

etc., m

road test on

London and Northwestern Railway

.Faces

587

199.

Diagram sho

wing speeds, horse-powers, etc..

etc., in

road test on

London and Northwestern Railway

.Faces

587

200.

Indicator diagrams from road tests on London and Northwestern

Railway

Aluminium cj
ft /

.Faces

587

201.

Etst. X 30 diameters

603

202.

II tt tt

603

203.

It tt tt

603

204.

ft ti tt

603

2C^.

tt tt tt

603

205.

ti tt tt

603

207.

« inirot- X 12 "

603

908

Bismeth

T3 '

Eist and strained x 5 diameters. . . .

603

209

" X 30 diameters

607

210.

" and strained x 30 diameters . .

607

211.

" ingot fractured by 30 diameters.. . .

607

212.

Cadmium cast x 30 diameters

607

213.

" and strained x 30 diameters . .

607

214.

It It It tt 11

607

215.

tt It It II It

607

216.

** rnlloH tLTiA annAJilAH y ^0 diAiTiAtprs

610

217.

" X 30 diameters

610

218.

" and annealed x 30 diameters

610

219.

ti It ft It ti

6M)

220.

It It It tt It

610

221.

II $1 It II II

610

Xlii LIST OF ILLUSTRATIONS.

Fia. TAam

222. Cadmium amiealed and strained x 30 diameters 610

223. " " " " " " 610

224. Copper slowly cooled x 30 diameters 611

225. " electrolytic X 50 diameters 611

226. " impure cast x 30 diameters 611

227. " rolled X 30 diameters 611

228. " foil, unannealed x 50diameters 611

229. " " annealed x 50 diameters 611

230. " slowly cooled x li diameters 611

231. Gold, " " x30 " 611

232. Lead ciystals (reduced) 614

233. Gold slowly cooled x 10 diameters 614

234. Lead Ingots 614

235. " cast, surface x 20 diameters 614

236. " rolled and annealed 614

237. " cast X 35 diameters 615

238. " " x30 " 615

239. " " etched 30diameters 615

240. " " " " 615

241. " " " " 615

242. " " " " 615

243. " " and strained x 30diameterB 615

244. " etched" " " " 615

245. Sheetlead 618

246. Rolled " 618

247. Sheet " (Fig. 245 annealed) 618

248. RoUed " " " 618

249. Platinimi slowly cooled x 30 diameters 619

250. Silver, ingot surface x 30 diameters 619

251. " slowly cooled x 15 diameters 619

252. " cast under salt x 30 diameters 619

253. " " " cover of salt X 30 diameters 619

254. " electrolytic x 8 diameters 619

255. " cast imder cover of salt X 30 diameters 619

256. " electrolytic x 8 diameters 619

257. " cupeUed x30 " 622

258. " " " " * 622

259. Continuation of Fig. 257 622

260. Silver cupelled x 15 diameters 622

261. Continuation of Fig. 259 622

262. Silver cupelled x 30 diameters 622

263. *' " containing copper x 15 diameters 623

264. " " " " " " 623

265. " " " u ' u u 523

266. " " " " " " 623

267. Ingots of tin 623

268. Tin cast on stone x 30 diameters , 623

269. " " " " " 623

;J70. " cast, surface x 30 diameters 624

LIST OF ILLUSTRATIO^& xliii

no. PAoi

271. Tin cast, surface x 75 diameters 624

272. " Dendrites X 30 diameters 624

273. " " deeply etched X 30 diameters 624

274. " cast deeply etched x 30 diameters 624

275. " " etched x 30 diameters 624

276. " " slo^y etched X 30 diameters 624

277. " " " " " " 624

278. " roUed and annealed 625

279. " " 625

280. " cast 625

281. " annealed (Fig. 279) 625

282. " rolled 0.5 mm. thick x 30 diameters 626

283. " " 0.1mm. " " " 626

284. '' annealed 0.9 mm. thick x 15 diameters 626

285. " " " " " " 626

286. " " 0.5 mm. " " " 626

287. " " 0.25 mm. " " " 626

288. " hanmiered 30 diameters 626

289. " " and annealed 35 diameters .626

290. " annealed 35 diameters 630

291. " " 33 " .'. . 630

292. " annealed 10 days x 30 diameters 630

293. " horizontal section of Fig. 292, x 30 diameters 630

294. " heated to melting point, 35 diameters 630

295. " fracture of Fig. 294, 35 diameters 630

296. " strained and etched, 30 diameters 632

297. " " " " " 632

298. " annealed and one end quickly raised to melting point 632

299. " " " " " " " " 632

300. Zinc rolledand annealed 633

301. " " 633

302. " ingotetched 633

303. " " surface 633

304. " cast " x30diameters 635

305. " " base X 30 diameters 635

306. " " strain X 30 diameters 635

307. " " " " " 635

308. " strained etched X 30 diameters 635

309. " " " " " 635

310. Side elevation, lathe-tool dynamometer measuring vertical forces only 641

311. Front view Fig. 310 '. 641

312. Back " " " 641

313. Sectional elevation universal dynamometer 642

314. Plan of Fig. 313 642

315. Front view of Fig. 313 644

316. Side " " 644

317. " " dynamometer in position 646

318. Front " " " 646

319. Plan " '' " 646

Xliv LIST OF ILLUSTRATIONa

ne. WAQt

320. Diagram of test on medium cast iron with dynamometer measuring

vertical force only 654

321. Sameas Fig. 320 654

322. " " 654

323. " " 654

324. Diagram of test on fluid pressed soft steel with dynamometer measur-

ing vertical force only 655

325. Sameas Fig. 324 655

326. " " 655

327. " " 655

328. Diagram showing variation of cutting stress with angle of tool medium

cast iron 659

329. Diagram showing variation of cutting stress with angle of tool, fluid

pressed soft steel 659

330. Failure trials of tools with various cutting angles, medium cast iron. . 659

331. Failure trials of tools with various cutting angles, medium cast iron,

fluid pressed soft steel 659

332. Failure trials of various cutting angles, fluid pressed mediimi steel. . . G31

333. Vertical section through tool and work 6" "

334. Plan view of tool and cut G J '

335. Diagram showing variation of surfacing and traversing forces with

different plan angles 671

336. Diagram showing variation of surfacing and traversing forces with

different cutting angles, fluid pressed soft steel 671

337. Diagram showing variation of percentage of surfacing and traversing

forces with different cuts 671

338. Diagram showing variation of the angle of inclination 671

339. Diagrams showing variation in the angle of inclination 672

340. Diagram showing variation of cutting forces as cut progresses 673

341. " of cutting force on soft steel 674

342. ** showing variation of cutting stress with cutting speed on
fluid pressed soft steel 675

343. Side of Broad Exchange Building (New York City) before commence-

ment of steel work 686

344. Broad Exchange Building (New York City), nearly completed 687

345. Method of supporting columns in side walls (cantilever for two columns) 689

346. Cantilever support for one colunm and side wall 691

347. Same as Fig. 346 showing columns above 693

348. Foundations of New Mutual Life Building (New York City) Faces 694

349. Plan of boiler and pump room, Mutual Life Building, New York

City Faces 694

350. Plan of engine room. Mutual Life Building Faces 694

351. Coal bunker in basement of Mutual Life Building 698

352. Ash hoist from cellar to side-walk, Mutual Life Building 699

353. Safe lifting pumps and bottom of elevator cylinders. Mutual Life

Building 701

354. Plan of boUer room, 60 Wall Street, New York City) Faces 702

355. " basement, 60 Wall Street, showing lay-out of machinexy,
piping, etc Faces 702

LIST OF ILLUSTBATIONa xlv

1«. TAQM

356. Plan of engine and boiler room, Hotel Astor, New York City Faces 702

357. Diagram showing theoretical design of steam turbine diverging

noszle .718

358. Entropy-temperature diagram, showing adabatic expansion of steam 720

359. Entropy-temperature diagram, showing adabatic expansion of steam 720

etc. 721

360. Photographs of jet turbineblaze showing erosion 724

361. CrosB-flection of Zoelly turbine 726

362. SecttODB and connections of buckets of the Zoelly turbine 727

363. " of a 25 stage Rateau turbine 728

364. Typical Westinghouse-Parsons steam turbine 729

365. Stationary and moving blades, Westinghouse-Parsons turbine. 731

366. Indicator cards showing initial pressures Westinghouse-Parsons

steam turbine 733

367. Economy and overload test, 400 K. W., Westinghouse-Parsons Tur-

bine 734

368. 400 K. W. Westinghouse-Parsons Turbine open for inspection 737

369. Plan of Westinghouse-Blachine Company's steam turbine testing

foundations and condensers 743

370. Brake tests of 1250 K. W., Westinghouse-Parsons turbine 745

371. Tests of 1250 K. W., Westinghouse Steam Turbine and generator 747

372. Brake tests of a 400 K. W., Westinghouse-Parsons turbine 748

373. Economy test Westinghouse steam turbine 749

374- Engmeroom plan for four 400 K. W. steam turbines 751

375. " " " " 1000K.W. " " 753

376. " " " " 5500K.W. " " 764

377. 5500K.W. Westinghouse-Parsons steam turbine 757

378. " " " " " 768

379. Westinghouse-Parsons steam turbine at Elyria, Ohio 759

380. Engine room plan at Elyria, Ohio 760

381. Plans showing condenser arrangement at Elyria, Ohio 761

382. Turbine installation of three 1000 K. W. steam turbines at Connells-

ville,Pa. 762

383. Plan of power station at Coimellsville, Pa Faces 762

384. 400K.W. steam turbine installation, Stamford, Conn 763

385. " Westinghouse-Parsons steam turbine at Batavia, N. Y 764

386. Guide vanes and moving vanes of an impulse turbine with speed dia-

gram 784

387. Guide vanes and moving vanes of a reaction turbine with speed dia-

gram 784

388. Reaction drum turbine (Parsons) '. 786

389. Multi-celular turbine (Rateau) 786

390. Rateau turbinedisc with riveted vanes 789

391. " " discs 790

392. Diagram of steam consumption 793

393. Longitudinal section 600 electric horse-power turbo-dynamo (Penar-

roya) 797

394. Plan 500 electric horse-power turbo-dynamo Penarroya. 798

395. Curves of electrical horse-power, steam pressures, etc 799

396. 400 dectncal horse-power turbo-alternator with revolving fidd magnet 802

xlvi List 01^ ILLUSTRATlOKfl.

397. Longitudinal section of horse-power portion of a turbine for Messrs.

Yarrow &Co 805

398. Turbine driven pump at Falkenau (Bohemia) 809.

399. Pump chambers at Bruay; a. Reciprocating steam pump; b. Steam

turbine (Rateau) 810

400. Turbo fan for blast furnace 811

401. Regenerative accumulator (Rateau) 815

402. Water, heat-accumulator (Rateau) 816

403. Low-pressure turbine driving two dynamos 818

404. Comparison of efficiencies 819

405. Method of, locomotive testing employed in experiments of Alexander

Borodin 828

406. Elevation of locomotive and moimtain mechanism first Purdue Loco-

motive Testing Plant Faces 831

407. First Purdue locomotive testing plant, general view 832

408. Elevation of the second Purdue Locomotive Testing Plant 835

409. Plan of the second Purdue Locomotive Testing Plant , 836

410. Elevation showing accessory apparatus, second Purdue Testing Plant 837

411. Floor plan showing accessory apparatus of second Purdue I-iOcomotive

Testing plant 838

412. Exterior of second Purdue Locomotive Testing Plant 840

413. Locomotive Laboratory, Purdue University 842

414. " " " " general view 844

415. Lidicator rigging, second Purdue Testing Plant 848

416. First locomotive testing plant of the Chicago and Northwestern Railway 850

417. General arrangement of engine testing plant, C. & N. W. Ry 851

418. Brakes, Engine Testing Plant, C.&N.W.Ry 853

418a. " " " " " 854

418b. " " " " " 854

419. Removeable track, engine testing plant, C. & N. W. Ry 855

420. Elevation, Cblimibia University Locomotive Laboratory Faces 856

421. Plan, Columbia University Laboratory " 856

422. Arrangement of brakes, Columbia University Locomotive Laboratory 857
422a. " " " " " " 858

423. Alden Absorption Dynamometer, Colimibia University Laboratory.. . . 861
423a. " " " " " " .... 862
424* Side elevation, locomotive testing plant of Penna. R. R. at Louisiana

Purchase Exposition Faces 863

425. 'End elevation, locomotive testing plant, Penna. R.R. at Louisiana

Purchase Exposition Faces 863

426. Plan, locomotive testing plant, Penna. R. R. at Louisiana Purchase

Exposition Faces 863

427. Diagrams for comparison of records of test on test bars 884

428. Strength and chemical composition of test bars 885

429. " " " " " " 885

430. " " " " " " 886

431. " " " " " " 886

432. " " " " " " 887

433. " " " " " " 887

^4^ " " " it tt it OQO

U8T OP ILLUSTRATIONa xlvii

FW. PAOB

435. Strength and chemical composition of test bars. 888

436. " " " " " " 889

437. " " '* " " " 889

438. " " " " " " 889

439. " " " " " " 890

440. " " " " " " 890

441. " " " " " • " 890

442. " " " " " " 891

443. Diagram of average tensile strength per square inch 893

444- " " crushing strength of a half-inch cube 893

445. " of the average transverse strength of section one inch square

by twelve inches 893

446. Diagram showing average tensile strength of various size test bars . . 894

447. Keep's tensile strength chart , 895

448. Diagram showing average transverse strengths, various size test bars 896

448. Continued 897

449. Keep's transverse strength chart 898

450. " shrinkage chart 899

451. Diagram for finding strength of castings 902

452. Sections of various forms of test bars of one square inch area , 907

453. Graphical chart of specifications 908

454. Diagram showing areas of roimd bars 909

455. " " method of taking test bars from a block 9 inches
square x 18 inches long 914

456. Diagram showing tensile strength of bars taken from Fig. 455 914

457. Potential efliciency of water wheels 922

458. Specific heat of superheated steam 925

450. Potential efficiency of Westinghouse Standard engines 928

460. " " the Westinghouse compound engine 929

461. " " Westinghouse three-cylinder compound 930

462. " " Corliss Cross-Compound Condensing and Van-

der Kerchove Tandem Compound Condensing Engines 931

463. Diagram for Van de Kerchove engines , . 932

464. Potential efficiency Westinghouse Steam turbine 200 K. W 934

465. Hiddlesborough Dock, showing electric and hydraulic cranes, ci^>stan8,

etc 944

466. Three-ton electric traveling crane, Middlesborough Docks 948

467. Ten-ton electric traveling crane, Middlesborough Docks 949

468. One-ton electric capstan, Middlesborough and Hartlepool Docks 950

469. Five-ton portable hydraulic crane, Middlesborough Docks 956

470. Ten-ton portable crane, Middlesborough Docks 957

471. Diagram showing work of hydraulic engines 958

472. Electric engine current diagram 958

473. Ground plan of the Shoreditch refuse destructor 985

474. Perspective view of one boiler and two refuse furnaces, Shoreditch . . 986

475. Transverse half-section Shoreditch 987

476. " " " 988

477. Longitudinal section, " 991

478. Section plans, Shoreditch 992

479. Flans showing pipes 996

Xlviii LIST OF ILLUSTBATIONS.

no. pAox

480. Diagram of electrical output of the destructor of the metropolitan

borough of Woolwich 1008

481. Sectional elevation of a typical commercial skyscraper 1012

482. " " of R. G. Dunn's Company's Buildmg (New York) 1015

483. Sample indicator diagrams 1016

484. Typicalfloor plan, 42 Broadway, New YorkCity 1019

485. " " " Park Row Building (New York City) 1020

486. " " " Queen Insurance Building (New Yoricaty) 1022

487. " " " Broadway Chambers (New York City) 1023

488. " " " Bowling Green Offices (New York City) 1026

489. " " " German American Building (New Yoric City) 1027

490. " " " The Hudson BuUding (New York aty) 1030

491. " " " Central Bank BuUding (New York aty) 1031

492. " " " Lords Court (New York CXty) 1034

493. " " " Dunn BuUding (New York aty) 1035

494. Arrangement of buckets and nozzles, Oirtis Steam Turbine 1043

495. Cross-section of first vertical Curtis Turbine 1045

496. Step bearing for Ou^is Vertical Turbme 1047

497. CJonnection of valve mechanism to governor in new 5000 K. W. Curtb

Turbine Faces 1048

498. Controlling valve used with Curtis vertical turbine 1049

499. CJross section showing controlling valve 1050

500. Base for supporting 5000 K.W.Curtis turbine Faces 1051

501. Cross section assembly, 500 K. W. Ou^is vertical turbine 1052

502. " " of detaUs 1053

503. DeLaval wheel and nozzle 1057

504. Sectional plan DeLaval Turbine Dynamo, 30 horse-power 1059

505. " " " " 300 horse-power 1060

506. Section DeLaval tuibme wheel 1061

507. 55 horse-power turbine pump 1066

508. High-pressure centrifugal pimip 1067

509. Diagram of a test of 10 K. W. non-condensing turbine dynamo 1068

510. Tests on a 30 horse-power steam turbine motor 1069

511. " " " " " " " 1070

512. " ona300 " " " " 1070

513. " of a 2-stage high-pressure steam turbine pump 1071

514. " " DeLaval Steam Turbine Pump 1072

515. Single row back-fed destructor furnace 1081

516. Twenty-four ceU destructor 1083

517. Twelve " " 1084

518. Six " " 1085

519. Furnaces for six ceU plant 1086

520. Six ceU destructor 1087

521. Exterior Westminster destructor 1088

522. Cart tipping, Westminster 1089

523. Dust catcher, five ceU destructor. 1098

524. Overhead clinker raUway and bucket 1101

525. Portable destructor 1103

526. Chartof test results, West Hartlepool 1105

527. Portraitof Robert Henry Thurston 1112

PAPERS

OF THE

NEW YORK MEETING

(XLVIIIth)

AMERICAN SOCIETY OF MECHANICAL ENGINEERS.

DECEMBER Ist to 4th, 1903.

BEINO ALSO THE TWKNTY-POURTH ANNUAL MBETINO OF THE SOCIETY.

No. loor.

PROCEEDINGS

OF THE

NEW YORK MEETING

(XLVIIIth)

OF THE

AMERICAN SOCIETY OF MECHANICAL ENGINEERS.

December let to 4tli, 1908.

Opentko Session. Tuesday, December Ist, 1903.

The twenty-fourth annual meeting of the Society, which was
also its forty -eighth convention, was held in New York City, at
the house of the Society and its Library, No. 12 West Thirty-
first Street, during the days December 1st to 4th, 1903.

The opening session was called to order by the President of
the Society, Mr. James M. Dodge of Philadelphia, at nine o'clock
on Tuesday evening.

It became apparent at this first session that the meeting was
to be one of phenomenal size in the matter of members in attend-
ance, and the audience crowded the auditorium to listen to the
address of the President.

After a few words of salutation from the Chair, Messrs. Lane,
Tompkins and Kern Dodge were appointed tellers under the
provisions of the Rules, to count the Officers' Ballot, to be pre-
sented at the next succeeding session, and Messrs. La Forge and
Louer livere appointed tellers to count the letter ballots on the

4 PBOCEEDINGS OF THE

Amendments to the new Rules, which were to constitute the new
Constitution, By-Laws and Rules of the Society.

The President then delivered his address, entitled " The Money
Value of Technical Training," which appears as one of the papers
of this meeting.

After announcements by the Secretary concerning the meet-
ing, a recess was taken to partake of light refreshments served
in the collation room and for an informal reunion of members.

Second Session. Wednesday Morning, December 2nd.

The second or business session of the annual meeting was
called to order at ten o'clock in the Hall of the Mendelssohn
Union, 113 West Fortieth Street. This step was made neces-
sary by the limited accommodation in the auditorium of the So-
ciety, which compelled a choice of a larger meeting room.

The headquarters for registration and other executive business
was retained at the house of the Society, 12 West Thirty-first
Street.

The registration of this session was made noteworthy by the
first eflfort to combine the system in use at previous meetings,
with the wearing of an inconspicuous tag under the lapel button
and number, which carried the name of the member so that it
could be read at short distances. It was believed that by this
procedure the comparative awkwardness of hunting- up a man's
name on the printed register lists would be removed, and the
social approach of members to each other would be stimulated.
The smooth working of the plan in its tenative form justified the
experiment, and until further notice it will be carried out. Up
to the adjournment of the meeting on Friday morning there were
823 names registered, of which 538 were members. This is the
largest enrollment of members in the history of the Society.

The first matter of business of the session was the presentation
to the meeting of the Annual Report of the Council and Standing
Committees of the Society.

These reports had been printed and distributed to all members,
in advance of the meeting, for their information ; they were read
by the Secretary, by title, and are presented herewith for record.

Some minor changes in distribution of accounts, in the Report
of the accountant, were presented verbally, but which did not
alter the totals, merely detail of allotment of expenses to certain
accounts.

NEW YORK MEETING. O

The Reports were as follows:

ANNUAL REPORT OF THE COUNCIL.
Fiacdl year, 19(»-1908.

The Council presents herewith, as required by the Rules of the
Society, a report of business which has passed under its hand
during the year which closes with the annual meeting in Decem-
ber, 1903.

The most important business of the year has been the an-
nouncement of the munificent purpose of Mr. Andrew Carnegie, a
member of the Society, to give the sum necessary to make adequate
provision for the accommodation of the four national societies
of engineers, in an appropriate building, and for the Engineers^
Club in another. Mr. Carnegie expressed his willingness to make
the amount of his gift exceed a minimum of one million dollars,
if that sum should be found inadequate to give the accommoda-
tion required for the present and the future needs of the organiza-
tions which he named. The Council was convened in special
session on the afternoon of Thursday, May 7th, to consider the
simple proposition of Mr. Carnegie's letter, at which eleven mem-
bers of the Statutory Council, and seven past presidents of the
Society were present The Council has made full report to the
membership by circular of the resolutions which were passed at
that and a subsequent meeting, concerning the Carnegie gift,
which have been made matters of official record in the Trans-
tctions of the Forty-seventh Meeting at Saratoga. The Council
has also issued full bulletins to the members, which were also
made part of that record.

The four constituent societies named by Mr. Carnegie have
appointed representatives, and from these representatives an Ex-
ecutive Committee has been formed, which since the adjournment
of the Saratoga Meeting has been formulating the details of the
arrangement of the building, and another sub-committee has been
considering the proper method for the management and control
of such a joint imdertaking.

The Report of these committees will be made public in the. near
future.

The Coimcil has convened for routine business at the necessary
intervals during the year.

It took favorable action at its first meeting concerning the

6 PBOCEEDINGS OF THE

issue of a letter ballot whereby the individual members of the
Society might express their opinion as to the effect of the adoption
of legislation making the metric system compulsory on citizens of
the United States. The result of such ballot, with the expres-
sion of such opinion, was reported at the Saratoga Meeting.

The Council accepted on the first of February, the resignation
of Mr. Arthur L. Kice, who had been acting as Assistant to the
Secretary, and confirmed the engagement of Mr. Louis A. Gillet,
under a different financial arrangement.

The Council has received communications during the winter
from members in different cities, concerning the probable attitude
which it would take with respect to the formation of local chapters
of the Society. The Council has in every case directed that until
the Society had taken official action upon that provision in its pro-
posed Constitution, looking towards the formation of sections of
the Society, that it was premature to open discussion on these
details.

Mr. C. J. H. Woodbury of Boston had been asked to represent
the Society at a conference for the revision of the National Stan-
dard Electrical Bules, and as such representative has furnished a
report of the action of the conference.

An arrangement has been made with the Engineers' Club of
Philadelphia whereby the privileges of the house of that Club,
in Philadelphia, may be enjoyed by members of the American
Society of Mechanical Engineers, and similarly that members of
the Club may have the privilege of use of the house of the
A. S. M. E. in New York City, on the presentation of the re-
spective cards of introduction issued by the two organizations to
their respective members.

The Council has expressed the interest of the Society in cooper-
ating with the American Reception Committee of the Iron and
Steel Institute of Great Britain in their undertaking to provide
for a meeting in 1904, of that organization, in the United States.

The Council has acted favorably upon a request that a pro-
visional committee should be created to consider and report upon
a tonnage basis for expressing the effectiveness of refrigerating
machinery, which could be made generally acceptable as a
standard. This Committee consists of Professor D. S. Jacobus,
Chairman, Messrs. E. F. Miller, A. P. Trautwein, G. T. Voor-
hees, P. De Catesby Ball.

The Council has bad under consideration the invitations which

NEW YORK MEETING. 7

have been received from those interested in the success of the
Louisiana Purchase Exposition in St. Louis, which has urged
that the Society shall select the City of St. Louis as a convention
city, daring the summer of 1904, while the exposition was in
progress.

It was the sense of the Council that it would be more service-
able if the convention of that date should be held in a city within
convenient access by rail to the exposition city rather than in St.
Louis itself.

On communicating this opinion to the representative members
in the City of Chicago, HI., the outcome has been a most em-
phatic urging that Chicago should be fixed upon as the point for
the spring convention in the exposition year. The Council has
acted favorably on this invitation and the City of Chicago has
been selected.

It has been further decided by the Council to avail of this op-
portunity to invite the Institution of Mechanical Engineers of
Great Britain to hold an American convention, and that such
convention be a joint session with the Society of Mechanical En-
gineers, at Chicago, with a view to having such guests of the
Society as might come from Great Britain, within convenient
access of the exposition city, upon the same journey which
brought them to the meeting.

The invitation of the Council has been ftordially accepted by
the Institution of Mechanical Engineers of Great Britain, and
the details of such joint meeting are in progress. The Institu-
tion of Civil Engineers of Great Britain was also invited at the
same time, but an invitation to a similar joint meeting in Septem-
ber, and its acceptance by the Institution, made it impossible that
our invitation to that society should be accepted in any official
way for the month of May.

The Council directed that the practice should prevail for the
present of having the Report of the Society's accountant audited
each year by a firm of public accountants, such as the Audit Com-
pany of New York, or similar competent authority.

The Council has considered a request to undertake the respons-
ibility of organizing the Section of Mechanical Engineering in
the suggested International Congress of Engineering, in connec-
tion with the St. Louis Exposition.

It was the sense of the Council after discussion, that in the ab-
sence of a strong demand from the profession itself, for the hold-

8 PROCEEDINGS OF TUE

ing of such a Congress, that it would not be advisable that this
responsibility should be undertaken. The Council has decided
to maintain a headquarters in the gallery of Machinery Hall of
the exposition buildings for the convenience and use of members
of the Society and its guests and the necessary appropriation has
been made for the employment of a suitable person to maintain
such headquarters and attend to the necessary clerical detail in-
volved. It is proposed that such headquarters should be a centre
for registration of members in attendance, and for the dissemina-
tion of information concerning the exposition to visiting mem-
bers, but that it should not be maintained as an exhibit of the
achievements of the profession.

The Council, on being advised of the sudden death of Professor
Robert H. Thurston, the first President of the Society and serv-
ing for two terms, from 1880-1882, has directed the entry on its
minutes of a Memorial Tribute, and that such tribute be made a
matter of record and presentation at the general meeting of the
Society.

The Council has passed votes of thanks to Miss Louisa Lee
Schuyler and Mr. W. A. Gabriel, and others for gifts to the
Society received during the current fiscal year.

The Council would present for record the list of members who
have died during the current year as follows:

James Spiers, August 13, 1902; W. W. Lindsay, November
12th; Thos. J. Borden, November 22d; George Leach, November
27th; Geo. R. Fulton, December 4th; J. F. Pajeken, December
16th; P. F. Greenwood, December 22d; J. O. Nixon, January 3,
1903; A. Christensen, January 16th; David P. Jones, January 1,
1903; John Hulett, January 31st; Wm. Harkness, February 28th;
Chas. M. Day, February; Victor Mackiewicz, February;. Edward
A. Darling, March 16th; John P. McGuire, April 17th; Irving
M. Scott, April 28th; Elihu Dodds, June 10th; George Shaw,
May 28th; Edward Graftstrom, June; George S. Morison, July
1st; Wm. Garrett, July 15th; E. H. Messer, August 12th; Wm.
H. Stratton, August 13th ; John Humphrey, August 24th ; S. J.
Geoghegan, September 7th; Pulaski Leeds, September 8, 1903;
L. C. Crowell, September 16th; Wm. P. Canning, September 17,
1903; J. Q. Wright, October 16th; Robert H. Thurston, October
25, 1903; Y. Aisawa, October, 1903; Sir Fred'k Bramwell,
December 1, 1903.

Pursuant to a desirable i>olicy inaugurated a year ago, the

NEW YORK MEETING. 9

CouncU would call attention to the report of its Standing Com-
mittees covering the Society's work during the year.

The Beports of the Library and House Conmiittee, and the Re-
port of the Executive Committee, and of the Publication Com-
mittee, will be self-explanatory. With respect to the Eei)ort of
the Finance Committee the Council would call attention to the
following facts and deductions from that Report:

Item* deduced from the accounts of fiscal year 1902-3 showing the expense incurred
per member :

(1) Total members as per July, 1903, catalogue 2,573

Deduct for members who have paid no dues:

Life members 107

Deaths and resignations 6

Lapsed memberships 35

Members who have not paid current year
at September 30, 1903 101— • 249

Paying membership, 1902-3 2,324

(2) Total income exclusive of 1 per cent, from dues

carried to Library Development Fund, 90 per
cent, from initiation fees, entire life member-
ship receipts, carried to Reserve Fund, and en-
tire Sinking and Fellowship Fund, subscriptions
of Mechanical Engineers' Library Association . . $38,662 03

Income earned per paying member (computed) 16 63

Income earned per paying member, dues only

(computed) 14 20

(3) Total expense incurred year October 1, 1902, to

September 30, 1903, less cost operating house
($3,347.54), mortgage interest ($1,402.50),
repairs and renewals ($644.32), depreciations
house and furniture ($481 .45)— $31,773 55:

(4) Total expense incurred for publications, October 1,

1902, to September 30, 1903 14,956 74

(5) Total expense incurred for salaries in Society's

office same period 9,308 55

(6) Total expense incurred for all other accounts except

house 7,508 26—31,773 55

(7) Total expense incurred for house, including interest

on mortgage, repairs and renewals and depre-
dations 5,875 81

Deduct income earned from rent of sleeping rooms

and hall 2,002 25— 3,873 56

Net expense incurred for year $35,647 1 1

(Gross expense, $37,649.36 less rental income
$2,002.25, equals $35,647.11).

10 PK0CEEDING8 OF THE

Expense ineurred per paying member, October 1, 1902, to September 30, 1903 :

(8) For all purposes including house $15 33

(9) For house operation including interest and repairs... 1 66

(10) For all purposes exdusive of house 13 67

(11) For publications, printers' work, engraving,

binding and distribution $6 43

(12) For salaries in Society's office 4 04

(13) For all other expenses except house 3 20-;- 13 67

(14) For house operation exclusive of mortgage interest,

repairs and renewals and depreciations 57

(15) For house operation exclusive of mortgage interest,

but including repairs, renewals and deprecia-
tions 1 06

(16) For operating library 31

(17) For postage, circulars, catalogues, and stationary

and printing in Society's office 1 83

(18) For meetings, and all other expense not otherwise

allotted above 1 26

ComparcUive income earned toUh expense incurred per paying member:
Income earned from dues only, per paying

member, per (2) above 14 20

Expense, incurred all purposes, per paying

member as per (8) above 15 33

(19) Excess income earned from dues only over expense

incurred all purposes per paying member 113

NEW YORK MEETINQ. 11

APPENDIX L
REPORT OF THE PUBLICATION COMMITTEE.

To the CouncU of the American Society of Mechanical Engineers:

Gentlemen: The Publication Committee would present the following report
as work under its direction.

At the doee of the fiscal year, September 30, 1902, the Society was under
obfigation to its printer for work ordered but not completed nor paid for, to an
amount of $597.00. This amount has been paid and in addition the expenses
for binding and distribution of Volume XXIII for 1901-2 have amounted to
$2,757.08.

The net cost of Volume XXIII, was $10,677.54, completed, which makes
Uie cost per copy $4.10. The volume had 878 pages.

The voliune of Traneactione for the current year (Volume XXIV), contains
the Proceedings of the New York and Saratoga Meetings. The selection of the
moiith of June for the meeting has made it imposable to bring in as much of
the expense of this volume into the current fiscal year as can be done when the
meeting falls earlier in the year.

The expense incurred for Volume XXIV to date amounts to $8,656.74 and
it b estimated on prices furnished for completed work that a further sum amount-
ing to $6,300 will be needed to complete this volume, making its estimated total
cost $14,956.74 as reported in sheet B herewith.

Volume XXIV will contain 1,560 pages, which is about a hundred pages more
Uian the largest previous volume in the Society's history. Its cost has been
unusual by reason of the very voluminous contributions to the discussion of
the metric system problem, and the distributions of pamphlet copies to all
voters in the membership. The volume will contain, in addition to the
papers and discussions, the addresses which were given at the ceremonies
connected with tjie unvdling of the Fulton Memorial in December, 1901.
These were omitted from their proper place in the last volume by reason
of the necessity imposed for reducing expense at that time. The items which
make up Uie expenditure of the Publication Committee and the totals under each
item, axe as follows:

Expended for work to dale on volume xxiv :

Advance papers $1,301 07

Revised papers 1,269 25

Stenographer's fees 276 75

Engraving 672 49

Compoation and electrotjrping 4,342 48

Hnding extra copies 259 20

Postage and express 255 50

Storage 280 00

Total $8,656 74— $8,656 74

12 PROCEEDINGS OF THE

Amount brought forward 86,656 74

Estimated amount required to complete volume xxiv :

Revised papers, Saratoga 900 00

Composition and electrotyping 2,400 00

Binding. 2,300 00

Distribution expenses. 700 00

Total reserved to complete Volume

XXIV $6,300 00— $6,300 00

$14,956 74
Respectfully submitted.

Publication Coiimittxb.

APPENDIX n.

REPORT OF THE LIBRARY AND HOUSE COMMITTEE.

To the Council of the American Society of Mechanical Engineers:

Gentlemen: The Library and House Committee presents the following report
of action during the current year. This Conunittee is intrusted with the func-
tions of control both of the House as the headquarters of the Sodety and of the
Library, which is housed within it.

I. llie Library has been open every day between the hours of 10 a.m. and
10 P.M. — excluding Sundays and legal holidays, except during the months of
July and August. By reason of sickness in the Library staff the evening open-
ings were suspended during these months. The additions to the Library in the
form of exchanges which have been received as the equivalent of the annual vol-
ume of the Society's Transactions have amounted this year to $533.00.

The Committee has expended for the purchase of books, $130 .00, and for bind-
ing of periodicals and pamphlets in exchange from other societies the sum of
$113.58. There remains a credit to the Society's Library with Uie house of D.
Van Nostrand & Co., for Transactions furnished, for which books are to be pur-
chased from that firm, amoimting to $328.75. There is a similar credit with
Spon & Chamberlain of $16.50. Since the last report a year ago, visitors to the
Library have numbered 1,800, averaging six persons a day.

The Library has received from Miss Louisa Lee Schuyler a gift of interesting
antiquities from the library of her father, the late George L. Schuyler.

For the conduct of the work in the Library, the Committee has had the ser-
\nces of so much of Mr. Louis A. Gillet's time as could be spared from his duties
as Assistant to the Secretary in other lines, and two-thirds of Uie time, including
evenings, of Miss Thornton, as librarian and cataloguer The manuscript of
Uie card catalogue has been supplemented and extended as far as the book titles
are concerned, and the Committee hope in the near future to issue a Library
catalogue in printed form for distribution.

On account of the expense involved this has not been done up to the presoit
time. The number of volumes in the Library at the date of this report is as
follows:

Books 8.500

Pamphlets 3.000

NEW YORK MEETING. 13

The appraisal reports published on page 19 and 20, of Volume XXIV, made the
▼ahie of the Library, 110,000*; the present book valuation is, $10,979.52. This
is based on the additions of new books, and without making allowance for any
depredation.

n. During the fiscal year the Council appropriated for the needs of the house
the sum of $5322.50. The net expense incurred imder this appropriation has
been as follows:

For operating expenses $3,347 54

For interest on mortgage 1,402 50

For repairs and additions ' 644 32

Total $5,394 36

This Lb about $200.00 in excess of last year, which is mainly to be attributed to
the falling due of long term insurance premiums, and to increased expenditure
for necessary furniture. The cost of fuel also this year is in excess of a year ago.
The figures in the financial report include credits on House Account, which are
not inchided in the above totals.

The receipts on accoimt of room and hall rentals for the year have aggregated
$2,002.25. The total expense of operating the house, exclusive of interest
charges on mortgage^ repairs and renewals and depreciations has been $3,347.54.
Sabtiitcting the recdpts makes the net expense of operating the house,
$1,345.29, and the total expense, including interest on mortgage, repairs and
renewals, and depreciations but excluding an interest on the value of the
equity, $5,875.81. The expense incurred in detail for the house has been as
foDows:

Interest on mortgage $1,402 50

Gas and electric light 428 40

Fuel 302 50

Janitor's supplies 195 60

Laundry. 407 32

Insurance 155 63

Repairs and Renewals, house 298 76

Repairs and Renewals, furniture 345 56

Wages 1 ,740 00

Incidentals 118 09

Total exclusive of depreciation $5,394 36

Depreciations 481 45

$5,875 81

The House Committee emplojrs, for the conduct of the house and hbrary ad-
ministititlon, a janitor and his wife (at $60 per month), and has the services for
part of his time each day, of a man whose other duties attach to the work of the
Secretary's office.

The auditorium has been used during the year for some of the meetings of the
Institute of Electrical Engineers, but that organization has outgrown the limited
capacity of the hall, and expects to make other arrangements tor the future.
Tlie same difficulty has arisen as to the accommodation of the New York Rail-

14 PBOCEEDINGS OF THE

road Club. The societies meeting regularly in our auditorium now are the
Society of Naval Architects and Marine Engin^rs, the American Society of
Heating and Ventilating Engineei^, the Society of Municipal Engineers, and
a few smaller bodies. A session of the Institute of Mining Engineers was held
here in the autumn.

The Committee has considered offers for the House and Lot at No. 12 West
Thirty-first Street, ranging between ninety thousand and ninety-five thousand
dollars, in view of the inconvenience which would be entailed by present removal
from the Society House, and in view of the expected rise in value of the property
during the period which must elapse before the Carnegie building is completed,
the Committee has reported against the acceptance of these propositions. It
has not been thought advisable to raise the appraisal value on the books of
the Society.

The Conunittee believes that, from the location of the house and from the chang-
ing character of the street (which is becoming more and more a business centre),
the value of its holding will increase as the date of the completion of the Penn-
sylvania Terminal at Seventh Avenue draws nearer.

Pursuant to the direction of th^ Council that one per cent, of the total income
from dues should be laid aside and reserved for the extension of the Library, the
Committee calls attention to the fact reported in the Financial Statement, that
the one per cent, for the current year amounts to $319.39. This sum has been
deposited in the Institution for the Savings of Merchants* Clerks in New York
City, to be drawn upon by the Society, and will be drawing interest until such
demand is made.

Respectfully submitted,

LrBBARY AND HOUSK COMMrTTER.

APPENDIX m.

REPORT OF THE EXECUTIVE COMMITTEE.

To the Council of the American Society of Mechanical Engineers:

Gentlemen: The Executive Committee of the Coimcil has special oversight
of those expenditures through the Secretary's office, and other channels which
do not attach themselves directly to the work of any of the stated coounittees.
The items which fall under the headings of such expenditure for the current
year exclusive of salaries, are grouped in the following statement with the amounts
attaching to each.

Expense incurred for :

Certificates and introduction cards $179 11

Badges, distribution and repairs 32 17

Circulars 1,863 02

Meetings 924 92

Catalogues 1,696 86

Office accounts, exclusive of salaries 1,687 00

The account "certificates and introduction cards" covers the expenditure
for printing, engrossing and distribution of the diplomas of membership, and the

NEW YORK MEETING. 16

introduction cards which are given by this Society to roemben when their initia-
tion fees are paid. Tlie badge account is the expenditure connected with dia-
tributioQ only, sinoe the badge itself is billed to the member at the jewelers'
price.

Under the head of circulars the expenditure b grouped into three heads.
What are known as "admission circulars" cover application blanks, the con-
fidential inquiries, announcements of election and the Hke, and have amounted
to $580.73. The circulars in connection with meetings are the notices, pro-
grams, registers of members in attendance and the like, but does not cover any
expense connected wiUi the professional papers. The total this year is $595.23.

Under general circulars are all others which do not fall into either of the other
groups. The total this year is $687.06, and is much larger than usual by reason,
first of the expenses of printing the draft of the Constitution, By-laws and Rules
for the use of the Committee, and for distribution to the members, together with
some extra and unusual printing in connection with the expression of opinion
which was ordered concerning compulsory legislation on the Metric System.

The Employers' Bulletins issued this year have been four in niunber, and are
included under this heading.

Under the heading of "Meetings" fall the es^penses in connection with the
two semi-annual meetings, outside of the printing and circulars. Such expenses
this year have amounted to $613.20. In addition under the Conunittee's care
have been the monthly reunions of members in New York City, directed by the
Council, for which the expenditure has been $311.72. There were four of these
meetings held during the months of January, February, March, and ApriL
The to{NC3 were:

"The Steam Truck for Heavy Duty," "The Pich Process for Brazing Cast-
iron," "Varnishes," and the "Turbine as a Recorder of the Flow of Streams."

The paper by Mr. Allen on the Turbine has been published in the TranMocHon^.
The meeting at which the paper on Steam Trucks was presented was the most
fully attended.

Under catalogues is included the expenses for composition, press work, paper,
and postage, of the two issues of the catalogue. By direction of the Council,
these issues were both made this year in the standard size, and the "vest pocket
editioQ" was discontinued. In the July edition a "geographical finding list"
was incorporated, and will be continued as a feature of the catalogues as issued
in the future. Under the heading of "Office Accounts" are included stationary,
postage, telegraph and telephone, office supplies and incidentals. It will be
apparent that this group of accounts under the Executive Committee will vary
with the size of the Society, and the amounts will increase with the Society's
growth.

The Committee would report certain changes under its direction in the matter
of salaries paid in the Secretary's office. It has made arrangements whereby the
salary paid to the Assistant to the Secretary shall be at the rate of $1,500 for the
first year, with an increase of $100 a year to a limit of $2,000. In recognition of
the services to the Society rendered by the accountant, and their increasing re-
sponsibiCty as the Society increases, and tiie amount of income which must pass
through his hands, the Conunittee have recommended that the salary of his
position be placed at $2,400 a year.

The great increase in the aze of the Society, and the volume of business to be
transacted in its offices, has made it necessary to add to the force of stenographere,
90 that a capable stenographer and typewriter should be available for clerical

16 PROCEEDINGS OF THE

work, in addition to the requirements of the correspondence. These changes
have been made at different times during the year, so that the total of salaries
is different this ytar from what it will be here^ter when the full yearly rate
is to be reported. The expenditures for this year are as follows:

Secretary $3,600 00

Assistant to Treasurer and accountant 2,900 02

Assistant to Secretar3r— 4} months $950 00

Asastant to Secretary— 7i months. 028 53— 1^78 53

Stenographer. 780 00

Assistant Stenographer— 1 month 30 00— 810 00

MaU clerk r 720 00

$9,306 55
Respectfully submitted,

ElxxcuTivi CoiiifrmB.

NEW YORK MEETING.

O
O

(3 : .8g

6 jSsr .

;Jfa

« «

3 S!

.a
a

a
o

■g-S- :

d
o

•5; S'S £^

CW £

-I

PROCEEDINGS OF THE

ssssg

SIS

ii*

a I o

^ =^ 5 & 2ii -

OObO

§ §

§ s

S SI

33,385
810

1 1

5,510
4,059

g§

SIS

1-4

NEW YOKK MEETINO.

t»

»-i

i!^

o S

fi-
ll

cot-

^j be

$ssssss

-ii

M|l -till

I
I

PKOCEEDIXOS OF THE

NEW YOEK MEETING.

8 2 8

sssssss

^ o» «o Oft eo c» <^
e9c0 oo ot

7 a o s
S 2 -

1*0

c
I

bo en.

M-n

» a

|.^-i=^

"I ' ^•«-

PBOCEEDINOS OF TUS

s S s^

i |5

S ^

o &

5 iS

a I

8SSSS

C« OOOOv-iCDiO

!8 S

s ssssssss

•^ Q f- o o5 » ^

SSSg§28

-♦• -^ IC^ Oft 00 t*
•* C<1 » « «

isesgs

SI5

.00 rH

♦

8 88§S8S88SS<S8«
S 82!2§ggg*§SSS

oi eo»-ro»o «

'I 1^

GO ? r:^

►mo

a?ffl

£-2 a

1
I ill

Jcf!

' a

I(S

p p e *» 5
g— o^fi S--

-'""'Hi.
H —

I. I
c -

51.

iilll^^

a
3

® a
o a

l-H ►-« ^

^ fl.t: ctf p , .
.§••§ ig-3 g .

3 -O .2 ►^ > n-i ^ = ^

' 9 • fl ^^^

« ^ ^^'^ SS'S.2 $ *-

s.-

M §£Sx

P r^

NEW YORK MEETING.

4»

ss ggsss

s§

103
298

8,543
819
400

4,059
19

©I

eo«D

©}

00 -^

4»

gSSS s^ss

^g e'

PROCEEDINGS OF THE

fi
H

c •

NEW YORK MEETING. 25

At the conclusion of the presentation of these reports, and some
questions and explanations concerning them, the Chair called on
Mr. Charles Wallace Hunt, representative of the Society on
the Joint Committee intrusted with the consideration of the
proposition from Mr. Andrew Carnegie to present a buflding to
the profession of Engineering, for the joint uses of the societies.

The statements made by Mr. Hunt have been embodied in a
separate report, which is made an appendix to these minutes, and
to which members are referred.

The President then called for the reports of the Tellers, next in
order. There were three groups of these reports.

The Tellers of members presented the following report:

REPORT OF TELLERS.

The undersigned were appointed a Committee of the Council to
to act as tellers, under Artile 11 of the Rules, to scrutinize and
count the ballots cast for and against candidates proposed for
membership in their several grades in the American Society of
Mechanical Engineers, and seeking election before the XLVIIIth
Meeting, New York, 1903.

They have met on the designated day in the office of the Society
and have proceeded to the discharge of their duty. They would
certify for formal insertion in the records of the Society to the
election of the following persons, whose names appear to the ap-
pended list in their several grades.

There are 58 members elected, 15 associates, and 46 juniors,
making a total increase of 119 names.

There were 567 blue ballots cast of which 12 were thrown out
because of informalities. The tellers have considered a ballot as
informal which was not endorsed.

8. S. Webber, } ^'"^''' of Flection.

For Members.

Aiken, Cliis. W. ColweU, J. Van V. Foucard, M. L.

Albert, Otto Conrad, E. B. Gardner, Thos. M.

Albright, H. F. ^ Cooke, Fred W. Gilbreth, Frank B.

Allen, Benj. T. Donnelly, Wm. T. Goddard, A. L.

Black, Edward 8. Duncan, J. D. E. Gray, Niel, Jr.

Bloomberg, J. H. Edgar, Ellis F. Greene, F. 8.

Brown, HngkT. Ellicott, Edw. B. Grossman Albert

Carse, Jno. B. Fleming, H. 8. Harper, Lewis E

Child, E. T. Folger-Osbome, G. P. Harrington, Jno. L.

PROCEEDINGS OF THE

Harrington, Nonnan T.
Hayward, S. F.
Helander, A. H.
Hess, Howard D.
Howe. Albert W.
Hulelt, Geo. H.
Ingersoll, Geo. T.
Johnson, Werner
Knox, Lather L.
Lincoln. Robt. B.

Lindstrom» N. 0.
MacDonald, D. H.
Merry weather, Geo. E.
Mix, Edgar W.
Moore, A. B.
Moore. Wm. E.
Pattison, F. A.
Peirce, Arthur W. K.
Pritchard, W. 8.
Quirk, Wm. M.
Reid, Marcellus

Rickey, Walter J.
Robinson, Frank H.
Shepard, Geo.
Shepard, Ijouis A .
Sprado, Carl G.
Stebbins. Theo.
Street, Edgar L.
Tandy, Harry
Warg, Robert
Waterman, Charles

For Promotion to Full

Allan, Percy
Astrom, J. I.
Berg. Hart O.
Blood, J no. Balch
Bunnell, S. H.
Dollar, Wm. M.
Ducommun, Edward

Ekstrand, Charles
Femald, Robt. H.
Hoffman, James D.
Jaquays, H. M.
Kellemen, H. F.
Kirk, Robt. H.
Malvern, L. K.
Morrow, Percy C.

Memdbrship.

Parker, Charles H.
Powell, Emery H.
Rush more, David B.
Smith, Harry E.
Stanley, A. W.
Stevens, Alfred H.
Trowbridge, Amasa

Aldcorn, Tliomas
Bell, Jno. E.
Caldwell, J. R.
Dornin, Geo. A.
Holz, Fred. H. Jr,

For Associates.
Lauer, C. N.
Loscher, A. P.
Nichols, William W.
Peck, Chas. B.
Pell, David W.

Hansom, Allan
Saldana, E. E.
Saunders. E. W.
Umstead, C. H.
Waddell. Chas E.

Bateman, Edw. L.

For Promotion to Associate.

Finley. A. D. Streeter, L. P.

Anderson, H. B.
At wood, Geo. D.
Bailey, Ervin G.
Barlow, E. S.
Bennett, Geo. G.
Berliner, R. W.
Bough ton, J. H.
Brown, E. H.
Case, Albert H.
Chatard, Wm. M.
Col well, Chas. A.
(^oomba, H. A.
Dreyfus. Theo. F.
Ehrmann, Jno. P.
Harvey, Rich. P.

For Juniors.
Hawley, Wm. P.
Helmes, M. J.
Henes. L. G.
Howlett, Lewis G.
Jones, Jarrard E.
Jordan, Wm. A.
Jump, E. P.
Kasson, R. S.
Katzenstein, M. L.
Klein. A. W.
Kleinhans, Frank B.
I^we, Henry L.
Moran, H. P.
Morison. (^eo. A.
Morrison, Hunter
Pettit, Frank

Pitkin, Jos. L.
Heis, Leslie H.
Richards, W. A.
Rossberg, Chas. A
Schuetz. Fredk. F. -
Springer, Jno. J.
Staples, H. A.
Stevens. Robt. H.
Stone, Thos. W.
Thatcher, R. P.
Westerfield, Geo, 8.
Wettengel, C. Albert
Whitteniore, H. L.
Wilson, Henry D.
Woldenberg, I.

NEW YORK MEETING. 27

The report required no action and was ordered on file.

The tellers to count the letter-ballot concerning the adoption
of the proposed Constitution, By-Laws and Rules in the fonn
. favorably acted on by the Saratoga Meeting, presented their re-
port as follows:

REPORT OF TELLERS APPOINTED TO COUNT THE BALLOTT8 CAST ON
THE PROPOSED AMENDMENTS TO THE RULES.

The Committee of Tellers appointed to count the ballots cast by
members for and against the adoption of the Constitution, By-
Laws and Rules, as reported by the Committee on Revision of the
Rules and Methods, begs to submit the following report;

Total ballots cast , 448

Ballots thrown ont ansigned 4

•* *' signed by rubber stamps 2

Other informal ballots

Total informal ballots ezcladed 6

Total ballots counted by tellers 442

Of the regular ballots so counted by the tellers, they would re-
port the following results:

In fayor of adoption of said Constitution, By-Laws and Rules 487 ballots

Against the adoption of said Constitution, By-Laws and Rules 6 *'

Total '. 442 "

Our count, therefore, shows that the Constitution, By-Laws and
Rules, as submitted at the Saratoga Meeting, 1903, is adopted by
the vote of the membership.

RespectfuUy submitted,

C." LoceC'"'^''' I ^'^^^''' of Election.

On presentation of this report, under the provisions of the
new Constitution, the new Constitution, By-Laws and Rules
went into effect on the announcement of the president of the
formal vote.

The tellers of election of officers presented their report as fol-
lows:

REPORT OF TELLERS.

The Committee of Tellers appointed to count the ballots cast by
the members for officers of the American Society of Mechanical

28 PROCEEDINGS OP THE

Engineers, for the year 1903-1904/ begs to submit the following
report:

Total ballots cast 702

Ballots thrown out ansigned 23

" *' ** signed with rubber stamps

Other informal ballots (scratched) 21

Total informal ballots 28

Total ballots counted by tellers 702

Of the regular ballots counted by the tellers, they would report
the following result:

For President.

Ambrose Swasey, Cleveland 676

Scattering —

For Vice-Presidents.

D. 8. Jacobus, Hoboken, N. J 674

M. L. Holman, St. Louis, Mo 669

Wm. J. Keep, Detroit, Mich 669

For Managers.

George I. Rockwood, Worcester. Mass 672

John W. Lieb, Jr., New York City 678

AsaM. Mattice, Pittsburg, Pa 674

Scattering .... 1

For Treasurer.
Wm . H. Wiley, New York City 677

Our count shows, therefore, election of Mr. Swasey for Presi-
dent; Messrs. Jacobus, Holman and Keep for Vice-Presidents;
Messrs. Rockwood, Lieb, and Mattice for Managers. Mr. Wiley
for Treasurer.

Respectfully submitted,
Kern Dodge, \

H. M. Lane, > Tellers of Election.
S. D. Tompkins, )

At the conclusion of its reading by the Secretary, the President
requested Professor Sweet, appointed a special committee for this
purpose, to escort the President-elect to the platform.

Mr. Ambrose Swasey was then greeted by the President and
in a few words of recognition and greeting spoke of his pleasure
at the honor conferred. He took his seat on the platform at the
side of the presiding oflBcer.

NEW YORK MEETINO. 29

The President then called for the regular order of busuiess of
the session.

Professor H. W. Spangler presented the report of the com-
mittee of which he was chairman, covering the proposed Stand-
ard Specifications for Steel Forgings, Steel Casting and Boiler
Plate.

This report was discussed by Messrs. Henning, Lanza, Carpen-
ter, Sandolph, Kent, Dingee, Flint and Beraent.

At the close of the debate on professional questions, Mr. James
Christie moved that the Keport of the Committee be referred to
Committee No. 1 of the American Association for Testing Ma-
terials.

This motion being duly seconded was passed. The report and
discussion appears as one of the papers of this meeting.

Mr. H. H. Suplee referred to the increasing collection of ma-
terial of value in the Library of the Society, and that in his in-
vestigations as to the very early historj^ of some of the trans-
atlantic societies of engineers it had come to his notice that their
early records were very meagre. By the death of those familiar
with this early history, these early records had become unattain-
able, and that while there was a vary excellent account of the
formation of this Society in the second edition of Volume I of
the TransaciionSy there is also a considerable amount of material
now extant in the form of recollections of interested members
which would in time become unavailable.

Mr. Suplee suggested that the Chair appoint a committee to
examine the records of the Society and prepare such formal mem-
orial covering its property and curiosities, as well as recollections
of its early history, which should appear either in routine form
of the Transactions^ or separately, as may be thought best, with
the particular view of having this material available for the
twenty -fifth anniversary of the Society, which would occur within
two years.

Mr. Charles Wallace Hunt in seconding this motion spoke of
the portrait of John Ericsson, with a very interesting history,
and the table of Robert Fulton's, and that in his opinion such a
memorial record would be a most interesting and useful record,
especially to the younger members of the Society.

The question was put by the President and carried.

The Chair subsequently appointed Messrs. John E. Sweet,
Chas. Wallace Hunt and H. H. Suplee.

30 PROCEEDINGS OF THE

Professional papers were then taken >4ip for the remainder of
the morning:

'' Is Anything the Matter with Piece Work " ? by Mr. Frank
Richards; paper by 11. L. Gantt on ^' Modifying Systems of Man-
agement ", and by Prof. John E. Sweet on '^ What are the New
Machine Tools to be " ? The participants in debate were Messrs.
H. L. Gantt, Oberlin Smith, Bates, Fred W. Taylor, Balkwill,
DuBrul, Emerson, Parker, Riggs, Fairfield, Henshaw, and
Schneble.

The Secretary made some announcements and a recess was
taken until the morning of Thursday, December 3rd, at the Stev-
en's Institute of Technology, Hoboken, at 10:30.

Wednesday Afternoon.

For this afternoon the stations of the Interurban Street Rail-
way Company at Ninety-third Street, the Manhattan Elevated
Railway Company at Seventy -fourth Street, and the Waterside
Station of the New York Edison Company at Thirty -eighth
Street, on the east side of the city, were open to members.

The members were assembled at luncheon in the auditorium of
the Society House, and parties were made up for the visit to these
various points.

No assignment was made for the evening of Wednesday with
a view to leaving the visitors free to avail themselves of the op-
portunities of the city in musical and dramatic lines.

Third Session. Thursday Morning, December 3rd.

By invitation of the President, Trustees and Faculty of the
Stevens Institute of Technology, the session this morning was
held in the auditorium of the Institute in its main building.

The session was called to order at half- past ten, with President
Dodge in the chair. The following professional papers were
taken up:

W. B. Gregory, '' The Pitot Tube " ; F. B. Perry, " Method of
Determining Rates and Prices for Electric Power"; D. S. Jaco-
bus, " Tests of a Coinpound Engine Using Superheated Steam " ;
E. F. Miller, ''The Pressure Temperature Curve of Sulphurous
Anhydride (SOj)"; Benj. T. Allen, ''Construction and Effici-
ency of a Fleming Four- valve Engine"; C. G. Barth, "Slide
Rules for the Machine Shop as a Part of the Taylor System of

NEW YORK MEETING. , 81

Management '' ; and the paper by Mr. F. A. Scheffler on '' Sugges-
tions for Shop Construction." The participants in debate on
these papers were Messrs. Ennis^ Ileisler, Suplee, Carpenter,
Kockwood, Kent, Kerr, Seymour, Wheeler, Moss, Ilice, Child,
Goss, Cluett, and R. S. Hale.

At the close of the meeting the members were divided into two
groups, one of which took luncheon in the Carnegie Laboratory,
and the other was escorted by guides, from the Institute students,
through the laboratories and equipment of the Institute.

On completion of their tour, this group was entertained at
luncheon, and at half-past two the members re-assembled in the
auditorium to listen to a most interesting and striking presenta-
tion of the properties of oxides of aluminum in producing intense
local heat sufficient to melt masses of considerable size, such as
rails, flanges and the like. The most striking example was the
melting of a piece of pipe.

Dr. Goldschmidt gave to his lecture the title of Alumino Ther-
mics.

In the evening the usual reception to the visiting members was
tendered by the New York members of the Society, at Sherry's,
Fourty-fourth Street and Fifth Avenue, New York City.

The reception line included President Dodge and Mrs. Dodge,
President-elect Swasey and Mrs. Swasey. Over 600 persons were
in attendance and dancing was kept up until a late hour.

Fifth Session. Friday Morning, December 4th.

The closing session of the convention was called to order in
the auditorium of the iSociety, 12 West Thirty-first Street, at ten
o'clock. The following professional papers were taken up and
discussed: By C. H. Morgan, entitled '*A Compact Gas Engine —
Beam Type"; ''Standard Unit of Refrigeration", by J. C.
Bertsch: "A Series Distilling Apparatus of High Efficiency",
by Prof. W. F. M. Goss; ''Air Motors and Air Hammers, Ap-
paratus and Methods for Testing", by Max H. Wickhorst;
** Valve Motion of Duplex Air Compressor", by S. H. Bunnell.
The participants in debate on these papers were Messrs. Ilobart,
Saplee, Jacobus, "Voorhees, Bunnell, Shipley, Eeeve, Magruder,
Kent, Thwait, Morse, Uehling and Jones.

At the close of the professional papers the President announced
that new or executive business was in order.

The Secretary by direction of the Council presented the follow-

32 PROCEEDINGS OF THE

ing tribute, prepared by Messrs. John E. Sweet, Robert W.
Hunt, and John Fritz, a special committee of the Council and
which that body had directed should be read in the general
meeting.

A TRIBUTE TO DOCTOR THURSTON.

Sudden death has called from us our first President,

Doctor Kobert H. Thurston,
the one of all best known to us through his work for the Society ;
the one of all best known in technical education, and the one of
all best known to engineers of the world.

His name will last as long as this Society lasts, and we who
knew knew him best will hold his name in loving remembrance
as long as we live. We mourn him, and to his wife and children,
to whom his' loss is so much greater, we extend our sincere sym-
pathy.

God has called to Himself one of his noblemen ; the peer of his
associates and a model for the young men of America.

It is such as he who makes life worth living. Remembering
his greatness, let us try to emulate his virtues.

(Signed) John E. Sweet,

EoBERT W. Hunt,
John Fritz.

The Secretary also reported that during the meeting he had
been advised of the death of Sir Frederick Joseph Bramwell,
D.C.L., L.LD., F.R.S., Honorary Member of this Society and
past President of the Institution of Civil Engineers of Great
Britain.

Sir Frederick was apprenticed to a mechanical engineer in
1834, and rose to the position of Chief Draftsman and later to
that of Manager. He entered the practice of engineering on his
own account in 1853; became an associate of the Institution of
Civil Engineers in 1856, a full member in 1862, and President
in 1884.

He l^ad also been president, of the Institution of Mechanical
Engineers in 1874 and 1875, and of the British Association for
the Advancement of Science in 1888. He will be remembered
by those of the Society who were the guests of the Institution of
Civil Engineers in 1899 as one of its most genial and noticeable
representatives in the entertainment of the visiting American

KEW tOBK MEETING. 33

eBglneers. He was as a practitioner especially emphatic as
to the dignity and responsibility of the engineer, particularly
as owing it to the public that no unworthy schemes reached ma-
turity through lack of exposure of their quality by competent
engineers.

The Secretary was on motion directed to transmit a suitable
minute and tribute^ to the Institution of Civil Engineers of Great
Britain.

Messrs. Rockwood, Miller, Hunt, Suplee, Goss, Halsey and
Engel took part in a somewhat informal discussion as to having
the out-of-town membership of the Society represented as bear-
ing a share in the expenses of the annual meeting, recurring each
year in the City of New York, and the matter was referred, on
motion, to the consideration of the Standing Committee on Meet-
ings created under the new Constitution.

There being no further new or general business to be presented,
the President asked Mr. Ambrose Swasey, President-elect, to
come to the platform that he might turn over the responsibilities
of his office to his successor. Mr. Dodge thanked the Society
for the courteous way in which he had been treated during his
incumbency, and on motion the meeting adjourned.

On the afternoon of Friday a large number of members ac-
cepted an invitation from the De Laval Steam Turbine Company
of Trenton, N. J., to be their guests by special train to the works
of the Company.

Luncheon was served on the train and the excursion was much
enjoyed.

34 CARNEGIE GIFT TO ENGIKEEiUNG.

No. 1008.*

OARNEQIE GIFT TO ENOINEERINQ.
SECOND CIRCULAR.

At the annual meeting of the Society held in New York City,
December 1st — ith, a report was called for from the representa-
tive of the Society on the Joint Committee intrusted with the
details of Mr. Andrew Carnegie's proposed gift of a building
for the needs of the engineering societies.

Mr. Charles Wallace Hunt, as such representative, presented
a verbal report, from which the following information is derived.

It will be recalled that in advance of the Saratoga Meeting an
announcement was made to the Society, which will be found at
page 870, Volume XXIV, of the Transactions. It mentioned
the purpose of Mr. Carnegie's donation of one million dollars,
which should bring the libraries, assembly-hails, offices and
meeting rooms of the societies of engineers into one great build-
ing, which, while ample in size for their individual needs, should
arrange for their convenience as respects business and profes-
sional uses. It was proposed, in addition, that the building
should give adequate accommodation for such other technical,
scientific and engineering bodies as might require the use of a
properly equipped auditorium.

During the intervals between that report and the present
meeting, the Joint Committee of fifteen has been in conference
on the details referred to it, and has appointed an Executive
Committee of five to facilitate the work of the general com-
mittee.

A sub-committee of the Executive 'Committee has been en-
gaged in the preparation of the material which might be neces-
sary to submit the ieeds of these societies to a competition of
architects, as to the structural details of the building. A Com-
mittee on Organization has been appointed, consisting of Messrs.
A. R. Ledoux, Chas. Wallace Hunt and S. S. Wheeler, and their

* Appeodix to Proceedings.

CARNEGIE GIFT TO ENGINEERING. 35

recommendations to the Joint Committee have been adopted and
the Organization Committee instructed to take the necessary pre-
liminary steps to secure the legislative action for which their
report called.
The full report of the Committee on Organization is as follows:
The undersigned were appointed a -special Committee at a
meeting of the Executive Committee of the Union Engineers'
Building Association, on July 9, 1903, and were instructed ' to
propose a plan of organization for the bodies exclusive of the
Engineers' Club, which are to participate in the gift.' In trans-
mitting instructions to this Committee, the Chairman of the Ex-
ecutive Committee summarized our duties, as he understood them,
as follows:

1. To suggest an organization to hold title to the land and
building.

2. A method of superintending and administering the build-
ing after completion.

3. Provision for granting the use of parts of the building to
other organizations whose objects and work may be of suitable
character and which may contribute to the maintenance of the
building.

4. Provision for continffencies, such as the withdrawal of one
of the societies, or the failure on the part of any one properly to
carry out its obligations.

The Committee was authorized to consult legal counsel. It has
held several meetings and has informally consulted members of
the American Society of Civil Engineers, and others interested.

After having informally and tentatively approved suggestions
formulated by one of its members, Mr. Hunt, the same were sub-
mitted to the law firm of Butler, Notman, Joline & Mynderse,
from whom a written opinion was obtained.

Having considered this legal advice and taken note of all sug-
gestions received, the Committee unanimously advise as follows:

1. The total amount oflfered by Mr. Carnegie shall be admin-
istered as two gifts; one to the Engineering Societies and the
other to the Engineers' Club, each U> be held and administered
independent of the other. The allocation of the fund to be made
at once, but the buildings to be designed and erected as one oper-
ation ; thereafter the respective titles and administrations to be
entirely independent.

2. The property represented by land, buildings and equipment
of the engineering societies, shall be held and administered by an

36 CARNEGIE GIFT TO ENGINEERINQ.

executive corporate body, preferably under a special charter, to
be obtained from the State of New York, each of the constituent
societies being entitled to name from its membership three per-
sons to act as incorporators and thereafter as directors.

3. Each society annually to elect or appoint, as their By-Laws
may prescribe, one of their voting members to serve on the Board
of I)irectors of the Executive Corporation for a term of three
years; a vacancy in said Board to be filled by an appointment
made by the society, the retirement of whose representative causes
the vacancy.

4. The land and property being held for the societies by an
executive corporation, tne said corporation may, to pay for the
land acquired, issue certificates of indebtedness or bonds bearing
interest at four per cent., and redeemable on six months' notice,
the buildings bemg a gift from Mr. Carnegie.

5. Each of the constituent societies may purchase and hold an
equal amount in value of the said bonds or certificates, but the
Board of Directors of the executive corporation may authorize
any of the constituent societies to hold an additional amount:
that is in excess of its portion, but such excess shall be sulriect to
recall at its par value at any time that the directors of the Execu-
tive Corporation may so order, to the end that each society shall
have an equal interest in the property of the corporation if it so
desires.

The certificates held bv each society shall be inalienable unless
thev are offered to the Executive Corporation at their par value,
and such tender shall not be acceptea by the Board of Directors
within one year thereafter.

6. The property of the Executive Corporation shall be used
perpetually as a meeting-place and headquarters for the constitu-
ent societies, and for such other scientinc associations as may be
temporarily admitted by the consent of the Board of Directors
of the Executive Corporation. Such associations may pay a pro
rata share in the expenses of the headquarters, but no pront shall
be made from such use.

7. Each of the participating societies shall be entitled to rooms
and space in the property adequate to its need, paying its share
of the running expenses in accordance with the amount of space
occupied; said space to be assigned and a proper assessment
therefor determined by the Board of Directors of the Executive
Corporation.

8. The excess of receipts over expenditures, if any, shall be
used for reducing the subsequent contribution of the several
societies for maintaining the building, and for the advancing of
engineering arts and science, by and through the participating
constituent associations. No dividends shall be declared or prof-
its divided, but a reasonable repair and rebuilding fund may be
established.

9. If the income of the Executive Corporation shall be less than

CARNEGIE GIFT TO ENGINEERING. 37

the expenditnre, the deficiency shall be made ffood bv an assess-
ment on each of the constituent societies, so allocated as to be in
proportion to the number of voting members of each society.

An excess of receipts over expenditures may be allocated to
the societies in like manner to reduce their annual assessment.

10. Shoald any of the constituent societies fail or refuse to
appoint directors, the remaining members of the Board of the
Executive Corporation shall aominister the property with all
the force and effect as though the Board contained its full quota
of members.

11. Finally, your committee, in offering the above suggestions,
has had in mind the setting aside of the money used for a build-
ing for the Engineers' Club, so that on the completion of the said
bmlding, the relations of the Club and of the engineering societies
will terftiinate. Thenceforward, the constituent societies are to
carry through the Executive Corporation the administration of
the building and its accessories, leaving the scientific, profes-
sional, inteUectual and financial activity in each organization
entirely independent of the others, and free to develop to any
extent and along any line that may be determined each for
itself.

The details of the superintendence and administration of the
buildings can best be considered after the organization of the
proposed Executive Corporation through the procuring of a
special charter.

(Signed) Albert R. Ledoux, )

Chas. Wallace Hdnt, 1 (^omrmttee on
Schuyler S. Whebleb, ) Orffamzatum.

It may be mentioned that when the sub-committee on building
plans took up the consideration of the necessary areas it devel-
oped at once that the cost of the building which would be required
for the accommodation of the Engineering societies, on 125 feet
front on Thirty-ninth Street, and the Engineers' Club on a fifty
foot front on Fortieth Street, would exceed the preliminary or
tentative figure mentioned by Mr. Carnegie in his official letter
of gift.

When this conclusion was reached, a special committee was ap-
pointed to visit Mr. Carnegie and talk the matter over with him.
He made it evident that it was his desire and intention to erect
a building which should be a monument to the Engineering pro-
fession, and that it was his purpose to make this building adequate
and the qnestion of cost was in his mind secondary to the attain-
ing of the object dear to his heart, which was the bringing to-

CARNEGIE GIFT TO ENGINEERING.

gether of all classes of engineers into a building in which they
could co-operate for the common purpose of the profes-
sion.

It may be desirable to repeat some of the statements of the pre-
vious announcement, that the location of the proposed building,
as given in the sketch plan herewith, is one of the most desirable
ones in the city of New York.

--^

<3«

rOWTY'rOUWTH

ygiw"T

rOWTY-THIWD

FORTY-riRfT

rOWTV'SCCOND

GHAND '
o\CENrRAL\
ISTATIONj,

BRYANT
PARK

I Building

Englntera
Club

■.MNK MTt GO..N.T.\

\\I]

THIRTY-NINTH

THIWTY-CIGHTH

Pig. 1.

»TRErr

STWrCCT

The New York Public Library building occupies the entire
eastern frontage of the block between 40th and 42nd Streets on
Fifth Avenue, while to the westward is the open square known
as Byrant Park.

It will be conveniently accessible from the terminals of the
New York, New Haven and Hartford Railroad, and the New
York Central and Hudson Eiver Railroad at 42nd Street, at Park
and Madison Avenues, and from the Pennsylvania terminals com-
ing in from the West, at 33rd Street and Seventh Avenue. The
Elevated railway at Sixth Avenue has a station at 42nd Street,
and the Third Avenue Elevated, 42nd Street and Park Avenue.
The junction of the Rapid Transit underground lines at 43rd
Street and Broadway* will make this station an important ex-
press point for trains both from down town and from the North.

Convenient surface lines, both up and down town, intersect at
42nd Street. The hotel and theatre district of the city are cen-
tering more and more around this particular region.

The Committee will bavery glad to receiv^e from members of

CARNEOIE GIFT TO ENGINEERING. 89

the Society suggestions Avhich may be helpful in the matter of
interior arrangement or detailed requirement of the proposed
building, to the end that it may be in every respect thoroughly
adapted for the needs of the societies which are to occupy it.

The committee would report that for the three societies which
have so far signified their purpose to make use of the proposed
building, the requirement of floor space in square feet, outside of
the general rooms for auditoriums and library needs, is as follows :

Purpose. Electrical. Mechanical. Hiaing.

Reception room 600 800 620

Editorial room 400 300 860

SecreUrj or AssistaDt 400 800 860

Counting room 600 600 320

Accountant— Stenographers 400 900 700

Botrd and Committee 1,000 600 760

Stationery and Transactions 1,200 1.200 1.000

Cloeete 400 400 400

Sundries 200 400 600

6.200 5,600 5,120

The Society of Civil Engineers has issued a circular to its memr
bers reporting the condition of affairs at the present time and
their probable need for 9,000 square feet of floor space, but defi-
nite action cannot be taken by that society until a letter-ballot is
had and until after the annual meeting of the society in January.

In the letter-ballot taken on the question of the necessary
changes in their By-Laws by the American Institute of Mining
Engineers, the vote showed over 1,600 in favor of the change,
and only eleven in opposition to it, which is regarded as showing
an overwhelming sentiment in favor of the Engineering Building.

It may be further of advantage to report that the action in the
Joint Committee on all questions which have been submitted to
it has been unanimous, and that throughout, in all decisions, the
Joint Committee has had the benefit of the counsel, advice and
co-operation of the representatives appointed by the American
Society of Civil Engineers, who have sat with the Committee in
all its deliberations.

Respectfully submitted,

James M. Dodge, )

C. W. Hunt, > Committee.

F. R. HUTTON, )

40 THE MONEY VALUE OF TECHNICAL TRAINING.

mo. lOM.*

THE MONEY VALUE OF TECHNICAL TRAINING.

BT JAMBS M. DODOB, PHILADBLPHIA, PA.

PRESIDENT'S ADDRESS, 1903.

1. Technical Training may be self -acquired or obtained through
instruction. The ability to drive a nail properly, or to design and
construct the most complex and wonderful of structures or de-
vices, is the result of Technical Training in but different degree.
Up to a very recent date, and within the memory of most of us,
the Apprentice System and that of Independent Delving repre-
sented the sole methods of acquiring training. Research and in-
vestigation carried on in individual lines, with varying degrees
of success, dependent upon the mental makeup of the individual,
were the means of attaining theoretical technical knowledge.
The blending of these two methods developed the earlier
Mechanical Engineers and will, even in the future, enable those
suiSciently gifted by nature and habit to attain eminence. The
progress of the world, however, calls for a better and more s^pgedy
means of producing trained men than could ever be developed
by the methods of self-instruction. The individual, striving for
manual skill, attains his desire under the old apprentice system.
Individuals sufBciently gifted arise above their fellows, and be-
come the leaders in their calling. The gratification of a mechan-
ical appetite and the desire to earn more money than his fellows
are two moving causes which impel a man towards technical
education. A generation or so ago, the universal belief was that
the sooner a young man entered upon his apprenticeship, or
began practical manual work, the better and more rapid would be
his progress in the Mechanic Arts, and Book Learning was de-
rided as being purely theoretical, and of little practical value.

* Presented at the New York meeting (December, 11)03) of the American So-
ciety of Mechanical Engineers, aud forming part of Volume XXV. of the Traru-
actions.

THE MONEY VALUE OP TECHNICAL TRAINING. 41

This belief is, even at this date, all too prevalent, largely due to
inherited error, and to lack of knowledge and reliable data.

2. Obtaining data from which incontrovertible conclusions can
be drawn is now comparatively easy, but a few years ago was
practically impossible. We are all prone to take extreme cases
of success or failure as the basis of our opinions, and lose sight
ol the fact that it is the average man whose career shows the
true force and direction of the current. For convenience of
comparison, I will outline the actual progress made by four
groups of men working in the Mechanic Arts, — the unskilled
labor group, the shop-trained or apprentice group, the trade
school group, the technical school group, and give the results
attained. Each group I will refer to as an individual :

3. The first is the Laborer, with but primitive and rudimentary
training, working under the immediate and constant supervision
of a boss, and earning, as the line on the chart indicates, $10.20
per week at the age of 22, his line remaining horizontal through
the period of his usefulness. Data are lacking as to his progress
before he reaches the age of 22.

4. The second is the Apprentice or Representative of the Shop-
trained group, of good health and habits, entering a machine
shop at the age of 16, and earning an average wage of $3 per
week for fifty weeks per year. This is about the number actu-
ally worked, making $150, or 5 per cent, on $3,000, which is his
Potential or Invested Value, upon which he draws his interest on
pay days.

5. On the chart accompanying this paper (Fig. 2) you will find,
ruled horizontally, lines representing amounts increasing from the
lower line upward by $1,000 each; starting at $1,000, and
terminating at the top at $50,000, these representing IJpten-
tial Values, upon which 5 per cent, is earned for fifty weeks a
year. The vertical Knes each represent one year in time, begin-
ning at the lower left hand corner at 16, and progressing in
regular order until, at the lower right hand corner, we have 32,
representing in all a lapse of 16 years.

6. To illustrate the progress of the four groups graphically, we
indicate on the line representing 16 years of age, and opposite
the figure $3,000, the young man just entering his apprentice-
ship. We will consider him typical of the Shop-trained Group.
Following the line to the right we see his average progress in
earning capacity through the ensuing years, noting that at the

THE MONET VALUE OP TECHmCAL TBAININO.

SAOOO

CMAAi

JKCOWMNViNS

pi»cs<oci«T*ft AOOMCSS. oeccMBCi* 3.

'903

40^000

4«/>00

AT.OOO
4«.000

*S.eoo

*4/>00

•«•«•

4|l,000

4Ci,oee

-0

z
.<

TkM

SflkOOO

^M«

W.OOO

M^ooe

-«»-

SS,eoo

lAOOO
BJIMO

Bt.000
•1,000

—

:5<:

' liJ ti. -

-1

'mm

•o^ooo

ll.

MCOOO

26COOO

< ~

£7.000

M.OOO

•W^ooo

-:i'

•—

«40OO
<AOOO
tt.OOO
2i 000

/"I

.,---

III

£-•*'

z —
3

lO^OOO
IA.OOO

18.000

17.000

■«aoo

< "

li. ■

f ■

"•/

'li.«»

JIJ

_,^

S^iT

•

C »Mt%

A*c or

rm% a

l>0»9

/^\

•

"S*

IZQOO
• 1.000
lO^OOO

•.000
eiooo
11 000
^^^

aooo
^000
1.0OOI
t.ooo
^ooo

jWr""

"T*

••M

<^l(Xtf

/X^

Jitjo

^CM

wjlfl.

**J

^\^

LA8(

CRO

^5^

'y>

/ ^

fe-

• M

17 18 19 20 21 22 23 24 25 26 27 20 29 30 31
EACH VERTICAL LINE REPRESENTS ONE YEAR.

THE MONEY VALUE OF
TECHNICAL TRAINING

Fig. 2.

THE MONEY VALUE OF TECHNICAL TRAINING. 43

age of 20 he is earning $9 per week, which is 5 per cent, on
$9,000, he having increased his Potential or Invested Value in
four years by $6,000.

7. We now note that his accumulated experience enables him
to make more rapid progress for the next year and a half, and
from the age of 20 to 21^ years we find that his pay has in-
creased to $13.20 per week, and his Potential Value to $13,200.
He is now approaching his goal, and his line of progress does not
continue at the same angle that it followed for the past few years,
but deflects toward the horizontal; and at the age of 24 we find
him' earning $15.80 per week, and his Potential Value $15,800.
In other words, in eight years he has increased his Potential
Value $12,800. Observation shows that 5 per cent, of the
apprentices acquiring the machinist trade rise above the line
made by our average man; 35 per cent, follow the line closely,
and that during the period of training 20 per cent. leaVe of their
own accord, and as near as can be ascertained, go to other shops
and continue in the line originally selected; 40 per cent., how-
ever, are found unworthy or incompetent, and are dismissed,
probably never rising to the $15.80 line.

8. Apprenticeship of to-day in many establishments does not
make the man, broadly speaking, a mechanic — in a majority of
cases he is a specialist or tool hand, and not comparable with the
old mechanic, who was a worker in metals, had some practical
knowledge of steam and prime movers, could chip, file, work on
lathe, planer, drill press, or as an assembler, and was competent
to meet the varied and unusual conditions found in general con-
struction and repair work.

9. The third group of young men are those fortunate enough to
have had the opportunity of entering a trade school, which.they
do at 16 years of age, devoting the next three years of their
Hves, or until they are 19 years of age, acquiring a trade under
competent instruction, and at the same time adding to their store
of rudimentary theoretical education. At the age of 19 a Trades'
School man enters the machine shop and can command $12 per
week, equal to the apprentice at 21 years of age, and very quickly
makes his employment profitable to his employer. The three
years in school have increased his Potential Value from $3,000
to $12,000, a gain of $9,000. Thus he has caught up with the
apprentice entering the shop at 16, and who has been working
for five years. Progress of the Trades' School group now follows

44 THE MONEY VALUE OP TECHNICAL TRAINING.

a line which diverges from that of the regular apprentice, and
by the time $15.80 is earned by the regular apprentice, the
Trades' School graduate is earning $20, with a Potential Value
of $20,000, or $4,200 greater than that of the Shop-Trained man.
The Trades' School line continues at substantially the saipe angle
up to an earning capacity of $22 per week, and a Potential Value
of $22,000. Data are lacking as to the further progress, but the
presumption is that this line will bear oflF more toward the hori-
zontal, eventually paralleling the line of the Shop-trained man,
but much higher on the chart

10. The fourth group we will represent again by a boy of 16
studying at school until his 18th year, and preparing himself for
admission to one of our higher Institutions of Technical learning,
such as the Stevens Institute, the Massachusetts Institute of
Technology, Columbia, Cornell and the like, where, after a four
years' course, or at the age of 22, he is ready to begin practical
work. The statistics upon which this chart is based show the
average starting wage at $13 per week, or the same amount
earned by the regular apprentice at the age of 21^, and by tho
Trades' School graduate at the age of 19^. In other words,
apparently a graduate of our technical schools has lost by his
six years of preparatory study, having been beaten by the regular
apprentice by six months and by the Trades' School graduate by
2^ years. From this time, however, there develops a most inter-
esting and instructive line of progress. The regular apprentice,
who is earning $13.50 a week at the time the technical graduate
is earning $13, is overtaken in six months, and we find both
earning $14 per week, and the technical graduate reaches the
$15.80 line nearly one year before the regular apprentice. In
other words, while it has taken the regular apprentice from his
2]st to his 24th year, or three years, to increase his wages from
$11.60 to $15.80 a week, the technical graduate has done the
same in fifteen months.

11. Progress now continues on substantially the same line, and
we find the technical graduate earning $22 per week, and crossing
the line of the Trades' School group in three^ years' time, a
worthy tribute to the higher education and attainment.

12. The line of the technical graduate now continues divergent
from that of the trades' school graduate, with earning capacity
regularly increasing, and a corresponding augmentation of Po-
tential or Invested Value until, at the age of 32, or ten years

THE MONEY VALUE OF TECHNICAL TRAINING. 45

after entering upon the practical work, we find our technical
graduate earning $43 per week, and his Potential Value at
$43,000. In other words, six years of preparation have enabled
him to distance the Shop-trained man and the Trades' School
graduate overwhelmingly. Bearing in mind that this is an aver-
age line, it is of interest to say that most technical graduates
with a better record than the one in the chart have devoted even
more time to their preparation, either by study or by shop work,
after graduation. Those, on the other hand, who have not come
up to this average line represent, in the main, men more or less
incapable of* original work. The reason that higher education,
other things being equal, carries with it the ability to earn high
wages is that consciously or unconsciously, these men are direct-
ing and making it possible for large numbers of Laborers, Shop-
trained men and Trades' School graduates to perform useful
worL A draftsman at his board may never realize that as a
result of his drawing a hundred men or more may be given em-
ployment His design calling for structural steel, for instance,
could not be built were it not for the labor of many men em-
ployed making and rolling the steel before it reaches the shop.
Then come the shop men, who cut, punch and shear, and then
the erectors, who assemble the structure in accordance with the
original plan. For this ability and knowledge our technical man
is paid.

13. It is quite obvious that aU workers in the Mechanic Arts
cannot be technical graduates. Some must, through natural limita-
tions, or lack of opportunity, #q11ow the apprentice line, and
others the Trade School.

14. If is from graduates of the latter that leading shop men and
foremen are largely selected. These two classes, supplemented
by the technical graduate, constitute the vast army of workers
in the Mechanic Arts.

15. Thus we see clearly that preparation pays, and that it pays
in dollars and cents, and that even a long term of years spent in
proper study and technical training is a good investment from
every point of view.

16. Of course, apprentices have made and will make, in rare in-
stances, a better showing than the average technical man of the
charts and many of our greatest men have, by sheer force of
character, excellence of brain fibre, persistence and self-educa-
tion, risen to preeminent positions, independent of all regular

46 THE HONEY VALUE OF TECHNICAL TRAINING.

systems. To the end of time great examples of this kind will be
found. Among those whose names readily come to mind are
the elder Krupp, Joseph Whitworth, George M. Pullman,
Andrew Carnegie, John Fritz, Prof. John E. Sweet, Edwin
Reynolds, George H. Babcock and Coleman Sellers.

17. The same is true of the Trades' School graduate, but as said
before, we are dealing with the average of each class, taken from
actual statistics, with an eanjest desire to ascertain the facts, and
without any preconceived notion of the outcome.

18. It may be stated as a truism that every man pays for the
amount or percentage of bossing he requires, and conversely,
every man's wages increase in proportion to his ability to act as
the boss or foreman of himself and others. The lower the wage
rate the greater the amount of watching and directing constantly
required. The slaves of ancient Egypt received no wages, but
were treated as horses are to-day. They were fed and sheltered
according to the ideas of their owners. No slave worked volun-
tarily, and the foreman's or leader's excellence was gauged entirely
by his physical strength and efficiency as a driver. This was
certainly the zero of labor conditions.

19. The highest wages are paid to the man through whose abil-
ity the largest number of other men may be most profitably em-
ployed. He does his work with his brain. Thus, on the one
hand, we see manual labor receiving no wages, and on the other
mental labor reaping the highest reward. Between these two
extremes is found every condition of human life.

20. A practical man perform* his work within the radius of his
arm, a technical man within the radius of his brain. This fact is,
even to-day, realized by the few, but it is gratifying to kliow that
the number is increasing.

21. The technical training of an individual makes him valuable
just in proportion as his ability is manifested by good judgment
and perception. Trained common sense receives the highest
compensation and reaps the greatest reward.

22. Mental ability to receive ideas and impart them properly and
wisely, rearranged and grouped, is typical of the most brilliant
mentality ; a dull intellect may be compared to blotting paper, fit
only to absorb and inter a heterogeneous mass of impressions.

23. The most interesting of all graphical charts would be that
properly exploiting the value of technical training to manufac-
turing plants and enterprises. To illustrate this more dearly^

THE MONEY VALUE OF TECHNICAL TRAINING. 47

we may fairly assume that the apprentice of our chart corre-
sponds to the old-fashioned primitive shop, having practically no
overhead expense, the proprietor carrying the business " in his
hat,'' priding himself on his non-receptive sturdiness, contempt
for improvements and personal attention to all details. For his
costs he adds together the value of raw materials and labor, and
then adds a few dollars for profit. The line of this establishment
would parallel the $15.80 line of our Shop-trained group.

24. The Trades' School line on the chart truthfully represents
establishmrnts in which some attention has been paid to the im-
provement of system, with an increased so-called non-productive
force, operating possibly in some particulars with brilliancy, but
with defective features in others; acknowledging the value of
improvement if internally originated; moderately but uncon-
sciously absorbent of ideas from without, but tenacious of dogma
and lacking departmental symmetry. Growth, increased earn-
ings and relative immunity from disastrous failure result.

25. The technical graduate line of our chart represents the man-
ufacturing establishment technically trained and " abreast of the
times " in all particulars, and I predict a time not very far dis-
tant when it will be almost universally recognized %hat establish-
ments should be trained as weU as individuals, and that the mar-
vellous development in scientific shop practice and management
will do for the manufacturer fully as much as technical training
is doing for the individual.

26. A change of mental attitude towards the subject of ad-
vanced Shop Practice and Management is noticeable to a marked
degree. Within a very few years, indifference and antagonism
have changed to a growing interest and appreciation.

27. The greatest musical composition contains no new notes;
each note of the scale can be sounded on a penny whistle. Our
greatest composers have only arranged the notes in harmonious se-
quence. The artists that can render their music truly, well
deserve unstinted praise, even though they lay no claim to the
composition of the masterpiece. Truly a listener at the grand
opera could say, " There is nothing novel in this ; I have heard
every one of these notes before. I have even made similar
sounds myself, and the result was far from satisfactory." So
with Shop Management : it must be as fundamentally harmonious
as a musical composition, and need not of necessity embody wittin
it any one element of extreme originahty. Of it the individual

48 THE MONEY VALUE OP TECHNICAL TRAINING.

may truly say, " Nothing novel has been presented; I tried this
feature or that-feature with no beneficial result," but if he can
play the music of the Art of Management thoroughly well he
need not grieve because he is not its composer.

28. Henry R. Towne, F. A. Halsey, H. L. Gantt and Charles
Day have all ably contributed through our Proceedings to the liter-
ature of this most important subject. Fred. W. Taylor, in his
paper of the current year, while claiming no originality of detail,
has presented to the world the most complete and thoroughly
scientific system of shop management ever promulgated. As an
investigator ^nd student he is sowing seeds in the field of the
Mechanics Arts which will bear a bounteous harvest.

29. It may be truly said that this Society, and others allied in
promoting the Mechanic Arts, complete the system of technical
training by going beyond the province of the technical schools,
their students being the men who constitute the management of
our manufacturing enterprises. It should be gratifying to all of
us that the pioneer literature of advanced shop management and
practice for this post-graduate course of technical training was
presented to the world by the American Society of Mechanical
Engineers. •

SLIDB RULES FOB MACHINE SHOP AND TAYLOR SYSTEM. 49

No. 1010.*

SLIDE RULES FOR THE MACHINE SHOP AS A. PART
OF THE TAYLOR SYSTEM OF MANAGEMENT

BY CABL O. BARTH, SWABTHMOIKB, PA.

(Member of the Society.)

1. In his paper on " Shop Management," read at the Saratoga
meeting of the Society in June last, Mr. Fred W. Taylor referred
to certain slide rules that had been invented and developed under
his supervision and general guidance, by means of which it be-
OHnes a comparatively simple matter to determine that feed and
speed at which a lathe or kindred machine tool must be run in
order to do a certain piece of work in a minimum of time.

2. These slide rules were also mentioned by Mr. H. L. Gantt in
his paper " A Bonus System of Kewarding Labor " (New York
Meeting, December, 1901), as being at that time in successful use
in the large machine shop of the Bethlehem Steel Company, and
reproductions of a number of instruction cards were therein pre-
sented, the dictated feeds and speeds of which had been deter-
mined by means of these slide rules.

3. Mr. Taylor early set about making experiments with a view
to obtaining information in regard to resistances in cutting steel
with edged tools, and also the relations that exist between the
depth of cut and feed taken to the cutting speed and time that a
tool will endure; and he advanced far enough along these lines
in his early position as engineer for the Midvale Steel Company
to make systematic and successful use of the information obtained;
but as this, of course, was confined to tempered carbon tools only,
it was not applicable to the modem high-speed steel, so that the
invention and introduction of this steel called for new experiments
to be made.

4. These were first undertaken under Mr. Taylor's directions at

♦ Presented at the New York meeting (December, 1908) of the American So-
oetj of Mechanical Engineers, and forming part of Volome XXV. of the Trani'

50 SLIDE RULES FOB MACHINE SHOP AND TAYLOB STSTSIC.

Bethlehem, so far as the cutting of steel alone was concerned;
and later on at the works of William Sellers & Co., Inc., of
Philadelphia, at which place the writer spent fifteen months in
going over these experiments again, on both steel and cast iron,
and with tools of a variety of shapes and sizes, and for which
nearly 25 tons of material were required.

5. However, it is not the writer's intention at this time, to give
an account of these experiments, or of the results obtained and
conclusions drawn from them, but merely to give some idea of
the slide rules on which these have been incorporated, and by
means of which a most complex mathematical problem may be
solved in less than a minute.

6. He will also confine his attention to the most generally inter-
esting of these slide rules; that is, the slide rules for lathes, and
he will take for an example an old style belt-driven lathe, with
cone pulley and back gearing.

7. Considering the number of variables that enter into the prob-
lem of determining the most economical way in which to remove
a required amount of stock from a piece of lathe work, they may
be enimaerated as follows :

I. The size and shape of the tools to be used.
* II. The use or not of a cooling agent on the tool.
- HI. The number of tools to be used at the same time.
'IV. The length of time the tools are required to stand up to
the work (Life of Tool).
V. The hardness of the material to be turned (Class Num-
ber).
VI. The diameter of this material or worL
VII. The depth of the cut to be taken.
iTU. The feed to be used.
'IX. The cutting speed.
X. The cutting pressure on the tooL

XI. The speed combination to be used to give at the same
, time the proper cutting speed and the pressure re-
quired to take the cut.
Xn. The stiffness of the work.

8. All of these variables, except the last one, are incorporated
in the slide rule, which, when the work is stiff enough to permit of
any cut being taken that is within both the pulling power of the
lathe and strength of the Jtool, may be manipulated by a person
who has not the slightest practical judgment to bear on the matter;

BLIDB RULES FOB HACHINB SHOP AND TAYLOB SYSTEM. 51

but which as yet, whenever the work is not stiff enough to permit
of this, does require to be handled by a person of a good deal of
practical experience and judgment.

9. However, we expect some day to accumulate enough data in
regard to the relations between the stiffness of the work and the
cuts and speeds that will not produce detrimental chatter, to do
without personal judgment in this matter also, and we will at
present take no notice of the twelfth one of the above variables
but confine ourselves to a consideration of the first eleven only.

10. Of these eleven, all except the third and tenth enter into re-
lations with each other that depend only on the cutting properties
of the tools, while all except the second, fourth and ninth also
enter into another set of relations that depends on the pulling
power of the lathe, andjthe problem primarily solved by the slide
rule is the determination of that speed-combination which will at
the same time most nearly utilize all the pulling power of the
lathe on the one hand, and the full cutting eflSciency of the tools
used on the other hand, when in any particular case under consid-
eration values have been assigned to all the other nine variables.

11. If our lathe were capable of making any number of revolu-
tions per minute between certain limits, and the possible torque
corresponding to this number of revolutions could be algebraically
expressed in terms of such revolutions, then the problem might
possibly be reduced to a solution, by ordinary algebraic methods,
of two simultaneous equations containing two unknown quantities;
but as yet no such driving mechanism has been invented, or is ever
likely to be invented, so that, while the problem is always essen-
tially the solution of two simultaneous equations, or sets of rela-
tions between a number of variables, its solution becomes necessa-
rily a tentative one; or, in other words, one of trial and error, and
involving an endless amount of labor, if attempted by ordinary
mathematical methods; while it is a perfectly direct and remark-
ably simple one when performed on the slide rule.

12. The slide jule method of solution may, however, also be
employed for the solution of niunerous similar problems that are
capable of a direct and perfect algebraic solution ; and it will, in
fact, be best first to exhibit the same in connection with the sim-
plest imaginable problem of this kind.

13. In the first place, the solution of two simultaneous equations
may be graphically effected by representing each of them by
a curve whose coordinates represent possible values of the two

SLIDE BULES FOB HAOHINE SHOP AND TAYLOB SYSTEM.

unknown quantities or variables, for then the coordinates of the
point of intersection of these curves will represent values of the
unknown quantities that satisfy both equations at the same time.
14. Example 1. Thus, if we have y + a? = 12 and y — a? = 3,
these equations are respectively represented by the two straight
lines AB and CD in Fig. 3; and as these intersect at a point (1)

(1) (2)

y—x= 8
2y =15

-§=3

y^64

y=7H y«=y54-7J5

^^.

(?)

\ /

(*+3)a?=18

ar-8

-Ji*.

BmrUt,C.a.

2 8

7 8
Fia. 8.

10 11 12 U 14 15

whose coordinates are a? = 4^ and y = 7i, these values will
satisfy both equations at the same time.

15. Example 2. Suppose again that we have x.y = 18 and - = 3,

and these equations are respectively represented by the equilateral
hyperbola EF and the straight line OH; and the coordinates to
the point of intersection of these (2) being respectively » = 2.45
and y = 7.35, these values will satisfy both equations at the same
time.

SLIDE RULES FOR MACHINE SHOP AND TAYLOR SYSTEM. 53

16. Example 3. Similarly, if we have y — » = 3 and y,x = 18,
these equations are respectively represented by the straight lines
CD and the equilateral hyperbola EF; and the coordinates to the
point of intersection of these (3) being x = 3 and y = 6, these
values will satisfy both equations at the same time.

17. The slide rule method of effecting these solutions — to the
consideration of which we will now pass-— will readily be seen to be
very similar in its essential nature to this graphical method,
though quite different in form.

18. In Fig. 4 is shown a slide rule by means of which may be
solved any problem vnthin the range of the rule of the general
form : " The 8um and difference of two numbers being given, what
are the numbers f ''

19. The rule is set for the solution of the case in which the sum
of the numbers is 12 and their difference 3, so that we may write

y + a? = 12 and y — a? = 3,

which are the same as the equations in Ex. 1 above.

20. In the rule, the upper fixed scale lepresents possible values
of the-sum of the two numbers to be found, for which the example
under consideration gives y + « = 12, opposite which number is
therefore placed the arrow on the upper slide.

21. The scale on this slide represents possible values of the
lesser of the two numbers (designated by x) and the double scale
on the middle fixed portion of the rule represents possible values
of the greater of the two numbers (designated by y); and these
various scales are so laid out relatively to each other, and to the
irrow referred to, that any two coincident numbers on these latter
scales have for their sum the number to which this arrow is set;
in this case accordingly 12.

22. The bottom fixed pcale on the rule represents possible val-
ues of the difference of the two nimibers, in this case 3, opposite
which number is therefore placed the arrow on the bottom slide
of the rule, the scale on which also represents possible values of
the lesser of the two numbers, x; and the double fixed scale in
the middle of the rule representing, as already pointed out, pos-
sible values of y, the whole is so laid out that any two coincident
numbers on these latter scales have for their difference the num-
ber to which ihis arrow is set; in this case accordingly 3.

23. Fixing now our attention on any number on the double y
scale in the middle of the rule, we first note the values coincident

SLIDE BDLE8 FOB IIACHINE SHOP AND TATLOB SYSTEM.

t^ _

o _

J2-

3 -

eo _

et -

o» -
1

_ oc

flO -

_ »•

t^ -

;»>

-s

S. eo_

o -
1

-s

_oo

i --

•o -
1

_t;_

_ ^

-55

< S-

"* -

- S-

«eo

_ o

Ul '^

CO —

-12-

-55

_ «

2 s-

a —

-3-

-?:

1
1

_ eo_

-S

-s

•-0 —

-a-

- o»

-S5

»H

-;2-

L-00

_ *-•

et -

_ o_

o —

eo —

-o»-

— «o

- o

I*—

•* —

-00-

- m

- 00

t--

to -

-t,J

- -^

- r-

«0 —

w -

-«o-

- 00

- «o

>o —

1- -

-«-

- 0*

- lO

"* —

OD -

_^-

-y^

- -*

«-

cs —

-eo-

-o-«

- eo

« —

o _

-01-

~ 1

- «

+ o-

;:: -

- r^ —

-^ H

- o-

-00

— ■^

•^ _

-o»-

- Hi

- o«

ee —
1

-00-

- to

- 00
1

s —

-^ —

-T

- -*

M5 —
1

I- _

- lO—

- oo
1

- lO

S-

— w—

-a»

- o
1

~ i

_ o

-*r

-00-

— J5

o»-

-os-

_ o»

- c»

O _

_ o_

_ eo

_ o

^ _

— V"

" "f

t-* _

_ o*..

t- •«

_ e*

00 _

_«_

- o

- eo

*«• _

_ •»»"-.

_ »-

▼^

»o _

_ 11--

_ X

_ »o

» _

_ O -.

, o

t- _

_J_

_ 1-

00 _

_ QD

TRAH8ACTION8 AMEBICAK SOCIETY OF MBCHAKICAL EnGINBBBB. VoL. XXV.

f l' FOR Power.

I . I

:. Tools.

Diameter for Power.

Speed Combination s^sFsm'^m'^f^-

iiiiT II " •

i" 1- I- i- i-l- 1- 1- Vl-i- J

ftlf4 S SS8 10MifsO128a

onmlkjihgfec

0.4" I 0.25" I 0.1S6"! 0.1" | aojas'i aosD"!

l-A-S 2-A-S 3-A-S 4-A-S 5-A-S
IJSO l.Oa 2.47 3.0S 8^

TUT

3-A-P 3-A-F
4.77 5.93 7.40

Diameter for
\\ Cutting Speed.

60" 60" 40" 80" ao-

I'lllllli I li.illillllllllli I II I I

... . . I I i I I I i I I I I I

Class Number oi28456789ioiii2 18iii

Depth of Cut i" re i" f i" f i" for Cuhino Speed.

Fig

L^.

CABii Q. Babth.

Patent Applied for

Cuss Number 9 ^p » » ^ for Power.

I M<,;:ii ii I i t iMMiii M |i|i|i |i I i I
ma. r »• 40' »• »• w wo" 8" 7" c 5'

Power.

lilll II II I

^ I '
f i « 6 a

r •UBTf &4R95'|

LATHE
No. 43.

- - is 8-B-S 4-lL8 5-B-8 1-B-F S-B-F 8-A-F /' CbUBINATION.

'^A-r 4-A-F S-A-F 1-B-S 3-1

•:,g m 11^ 14^ 17 JS a.2 27.St 84.4 42.9 S8.4 <W.fl

J9* 10" 9' 8* 7" «• 6

^ i \ I I t I \ \ t \ i I

>ER NiN.

LATHE No. 43.

: i M I I I I I I I I i I I I I I I I I I I I I ;
13Il = lli718 19 20S122S8^25262723»a0818Sf8884858087ed3340 | FOR SPEED.

Dry. With Water.

_J L

Life of Tool. 2om.ih. 2houm.

^».ama 4m« ckK^. r.

SLIDE BULBS FOB HACHIKB 8U0P AND TAYLOB SYSTEM. 55

to it in the two x scales on the slides; and this done, we readily
discover in which direction we must move along the first scale in
order to pick out that value of y which has the same value of x
coincident with it in both x scales. For the case under consid-
eration this value of y is 7i,and the coincident value in both scales
is 4^. Evidentiy, therefore, y = 7^ and » = 4^ are the num-
bers sought.

24. In the same manner we may make a slide rule for the solu-
tion of the general problem : " The prodtict and qtcotient of two
numbers being given, what a?*e the number's f *'

Such a rule would differ from the above described rule merely
in having logarithmic scales instead of plain arithmetic scales.

25. By the combined use of both arithmetical and logarithmetic
scales we may even construct rules for a similar solution of the
general problems : '* The sum and product, or the s^im and quo-
tienty or the difference and product, or the difference and quotientf
of itoo numbers being given, what are the numbers f^^ and a
multiplicity of others; and the writer ventures to suggest that
slide rules of this kind, and some even simpler ones, might be
made excellent use of in teaching the first elements of algebra,
as they would offer splendid opportunities for illustrating the rules
for the operations with negative numbers, T^hich are such a
stumbling block to the average young student.

26. We now have sufiicient idea of the mathematical principles
involved, for a complete understanding of the working of the slide
rule whose representation forms the main purpose of this paper.

27. This slide rule, in a somewhat ideal form in so far as it is
made out for neither steel nor cast iron, but for ah ideal metal of
properties between these two, is illustrated in Fig. 5. It will be
seen to have two slides in its upper section and three in its lower
section, and it is in so far identical with the rules made for the

( Bethlehem Steel Company, while in the rules more recently made

it has been found possible and convenient to construct it with only

, two slides in the lower section also.

I 28. It is shown arranged for a beR-driven lathe (No. 43*) with

five cone steps, which are designated respectively by the numbers
1, 2, 3, 4, 5, from the largest to the smallest on the machine.
This lathe has a back gear only, and the back gear in use is desig-

♦ The main frame of the rule is used for a number of lathes, and is arranged
to receive interchangeable specific scales for any lathe wanted, as may be seen in
the illastration.

56 SLIDE BULES FOB MACHINE SHOP AND TATLOB SYSTEM.

nated by the letter il, the back gear out by the letter B. It also
has two counter shaft speeds, designated respectively by 8 and jP,
such that 8 stands for the slower, F for the faster of these speeds.

29. The Speed Combination 3— A — S thus designates — ^to
choose an example — the belt on the middle cone step, the back
gear in, and the slow speed of the countershaft; and similarly, the
combination 1 — B — F designates the belt on the largest cone step
on the machine, the back gear out, and the fast speed of the coun-
tershaft; and so on.

30. The double, fixed scale in the middle of the rule (marked
Feed) is equivalent to the y scale of the rule in Fig. 4, and the
scales nearest to this on the slides on each side of it (marked
Speed Combination fob Poweb, and fob Speed, respectively)
are equivalent to the x scales on the rule in Fig. 4. The rest
of the scales represent the various other variables that enter into
the problem of determining the proper feed and speed combina-
tion to be used, fixed values being either directly given or as-
signed to these other variables, in any particular case under
consideration.

31. The upper section of the rule embodies all the variables that
enter into the question of available cutting pressure at the tool,
while the lower section embodies all the variables that enter into
the question of cutting speed / or, in other words, the upper section
deals with the pulling pov)er of the lathe, the lower section with
the cutting properties of the tool; and our aim is primarily to
utilize, in every case, both of these to the fullest extent possible.

32. The example for which the rule has been set in the illustra-
tion is :

A i inch depth of cut to be taken with each of two tools on a
material of class 14 for hardness, and of 20 inches diameter, and
the tools to last 1 hour and 45 minutes under a good stream of
water.

33. The steps taken in setting the rule were:

1. The first scale in the upper or Poweb section of the rule,
from above, was first set so that 2 in the .scale marked Number
of Tools became coincident with ^ inch in the fixed scale marked
Depth of Cut fob Poweb.

2. The second slide in this section of the rule was so set that
20 inches in the scale marked Diametee of Wobk foe Poweb
became coincident with 14 in the scale marked Class Nuhbeb
FOB Poweb.

8LIDB BULES FOB HAOHIHB SHOP AND TATLOB SYSTEM. 57

3. The first slide from below, in the lower or Speed section
of the rule, was so set that the arrow marked Wrrn Water be-
came coincident with 1 hour 45 minutes in the fixed scale marked
LiFB OF Tool.

4. The arrow on the lower side of the second slide in this
section of the rule was set to coincide with ^ inch in the scale
marked Depth of Cut fob Cuttino Speed.

5. The third and last slide in this section was so set that 20
inches in the scale marked Diameteb of Work fob Cutting Speed
became coincident with 14 in the scale marked Class Number
fob Cuttino Speed.

Let us now separately direct our attention to each of the two
sections of the rule.

34. In the Poweb section we find that all the speed combina-
tions marked B (back gear out) lie entirely beyond the scale of
feeds, which means that the estimated effective puD of the cone
Iteh reduced down to the diameter of the work, does not represent
enough available cutting pressure at each of the tools to enable a
depth of cut of ^ inch to be taken wiUi even the finest feed of the
lathe. Turning, however, to the speed combinations marked A
(back gear in), we find that with the least powerful of them
(5 — A — F) the e feed, which amounts to yfff ^^^^ = 0.039
inch, may be taken; while the / feed, which amounts to {^
inch = 0.05 inch, is a little too much for it, though it is within
the power of the next combination (5 — A — 8), and so on until
we finally find that the most powerful combination (1 — A — 8)
is nearly capable of pulling the t feed, which amounts to -^ inch
= 0.1 inch.

35. In the Speed section of the rule we likewise find that all the
B combinations lie beyond the scale of feeds, while we find that
the combination 5 — A — F (which corresponds to a spindle speed
of 11.47 revolutions per minute), can be used in connection with
the finest feed (a) only, if we are to live up to the require-
ments set for the life of the tool; while the next combination
(4 — A — F) will allow of the e feed being taken, the combination
3 — A — F of the / feed, and so on until we finally find that the
combinations 3 — A — S is but a little too fast for the coarsest (o)
feed, and that both of the slowest combinations (1 — A — 8 and
2 — A — 8) would permit of even coarser feeds being taken, so far
as only the lasting qualities of the tools are concerned.

36. We thus see that there is a vast difference between what the

58 SUDS BULES FOB HAOHINB SHOP AND TAYLOB 8T8TE1C

PowEB section of the rule gives as possible combinations of feeds
and speeds for the utilization of the full pulling power of the
lathe, and what the Spbkd sections of the rule gives for such com-
binations for the utilization of the tools up to the full limit set
However, by again running down the scale of feeds we find that,
in both sections of the rule, the % feed (iV i^^oh = 0.1 inch), is
but a trifle too coarse for the combination 1 — A — F, while the h
feed {^ inch = 0.078 inch) is somewhat too fine in connection
with this speed combination 1 — A — F^ both for the full utilization
of the pulling power of the belt on the one hand, and for the full
utilization of the cutting efficiency of the tools on the other hand.

37. In this case, accordingly, the rule does not leave a shadow
of doubt as to which speed combination should be used, while it
leaves us to choose between two feeds, the finer of which does not
allow us to work up to the full limit of either the belt or the took,^
and the coarser of which will both overload the belt a trifle and
ruin tlie tools a trifle sooner than we first intended to have them
give out.

38. The final choice becomes a question of judgment on the part
of the Slide Rule and Instruction Card Man, and will depend upon
how sure he is of having assigned the correct Class Numbeb to
the material or not ; and this latter consideration opens up a num-
ber of questions in regard to the practical utilization of the rule,
which for the lack of time cannot be taken up in the body of
this paper, but which will be fully answered by the writer in any
discussion on the subject that may arise.

39. Having decided upon the speed and feed to use, the Instruc-
tion Card Man now turns to the Time slide rule illustrated in Fig.
(), and by means of this determines the time it Avill take the tools
to traverse the work to the extent wanted, and making a fair
allowance for the additional time consumed in setting the tools
and calipering the work, he puts this down on the instruction
card as the time the operation sheuld take.

40. For finishing work the pulling power cuts no figure, so that
this resolves itself into a question of feed and speed only; and for
the selection of the speed combination that on any particular
lathe will give the nearest to a desired cutting speed, the Speed
slide rule * illustrated in Fig. 7 is used.

41. It will readily be realized that a great deal of preliminary

* Described in the AtMrxcan Machimd of November 20, 1902.

SUDB BULBS FOB HAOHINB 8U0P AND TAYLOB 8TSTFH.

work has to be done before a lathe or other machine tool can be
successfullj put on a slide rule of the kind described above. The
feeds and speeds and pulling power must be studied and tabulated
for handj reference, and the driving belts must not be allowed

Pio. 6.

to fall below a certain tension, and must, in every way, be kept
in firstK^lass condition.

4:2. In some cases it also becomes necessary to limit the work to
be done, not by the pull that the belt can be counted on to exert,
but by the strength of the gears, and in order to quickly figure
this matter over the writer also designed the Gear slide rule *
illustrated in Fig. 8, which is an incorporation of the formulae
established several years ago by Mr. Wilfred Lewis.

* Described in tlie American Machinist of Jaly 81, 1902.

SLIDE RULES FOR MACHINE SHOP AND TAYLOR SYSTEM.

43. For the pulling power of a belt at different speeds, the
writer has established new f ormnlfie, which take account of the in-
creasing sum of the tensions in the two sides of a belt with increas-
ing effective pull, and which at the same time are based on the
tensions recommended by Mr. Taylor in his paper entitled " Notes

Fig. 7.

on Belting,^' which was presented at the Meeting of the Society
in December, 1893.

44. These formulae have also been incorporated on a slide rule,
but as the writer hopes at some future time to prepare a separate
paper on this subject, he will not go into this matter any further
at the present time.

45. Having thus given an outline of the use of the slide rule sys-
tem of predetermining the feeds and speeds, etc., at which a
machine tool ought to be run to do a piece of work in the shortest

SLIDE RULES FOR MAOUtNE SHOP AND TAYLOR SYSTEM.

possible time, the writer, who has made this matter an ahnost ex-
clusive study during the last four years, and who is at present en-
gaged in introducing the Instruction Card and Functional Fore-
manship System into two well-known Philadelphia machine shops,
which do a great variety of work in both steel and cast iron, will

Fie. 8.

merely add that, in view of the results he has already obtained,
in connection with the results obtained at Bethlehem, the usual
way of running a machine shop appears little less/than absurd.

46. Thus already during the first three weeks of the application
of the slide rules to two lathes, the one a 27 inch, the other a 24
inchy in the larger of these shops, the output of these was increased
to such an extent that they quite unexpectedly ran out of work
on two different occasions, the consequence being that the super-
intendent, who had previously worried a good deal about how to

62 SLIDE BULKS FOB KACHINB SHOP AND TAYLOB ST8TEH.

get the great amount of work on hand for these lathes out of the
way, suddenly found himself confronted with a real diflSeulty in
keeping them supplied with work. But while the truth of this
statement may appear quite incredible to a great many persons, to
the writer himself, familiar and impressed as he has become
with the great intricacy involved in the problem of determining
the most economical way of running a machine tool, the applica-
tion of a rigid mathematical solution to this problem as against
the leaving it to the so-called practical judgment and experience
of the operator, can not otherwise result than in the exposure of
the perfect folly of the latter method.

MODIFYING SYSTEMS OF MANAGEMENT. 63

Mo. 1011.*

MODIFYING SYSTEMS OF MANAGEMENT.

BT H. L. OAMTT, BCHBMKOTADT, N. T.

(Member of the Society.)

1. At the Saratoga meeting the papers on shop management
and the allied subjects covered such a broad field that a thorough
discussion of them in the time available was practically impossible,
and consequently, as the writer has since found in going over the
subject with members, many of the most important points in those
papers were not brought out.

2. Most of the people the writer has talked with regarded the
various things advocated as individual propositions, and approved
or condemned according as they Baw, or did not see how they could,
or could not be adapted to their works and their existing system
of management. Many people apparently would like to adopt
some of the ideas, but do not see how they can do so without
making radical changes. This is due to the fact that few plants
have in practical operation the basis on which the whole system
depends for its ultimate success, namely a complete and accurate
system of getting a record of all work done each day, and the
amount of time spent in doing it. In other words a means of
knowing in the office whether every order has been properly car-
ried out or not, without actually going out into the shops and inves-
tigating. Whether the whole system as advocated is adopted or
not this part is of great value to any manager, superintendent or
foreman, and can usually be gradually introduced without stirring
up any serious opposition. As a matter of fact it has been the
writer's experience that all good men welcome such a system and
do all they can to get it in operation ; and, if a little time is taken,
a radical change may be made by methods that are not at all revo-
lutionary.

♦ Presented at the New York meeting (December, 1903) of the American So-
ciety of Mechanical Engineers, and forming part of Volume XXV. of the Trans-
aeUons.

64 HODIFYINQ SYSTEMS OP KANAQSICSKT.

3. A Complete System : Referring again^ however, to the views
of the members regarding the methods advocated at the Saratoga
meeting, it seems that few have grasped the idea that what was
advocated was not a series of isolated propositions, but a system
of management having a number of parts working in harmony
with each other, designed first to find out in detail what the maxi-
mum output of a plant should be, and then to make it to the in-
terest of all concerned to obtain day after day that maximum
output.

4. Different Views: As an example of how diverse the criticisms
of the papers were, I may quote some of the extreme ones.

5. One man, the manager of a large plant, thought that if the
methods advocated by Mr. Taylor were attempted in his plant
there would be a strike at once.

6. A second man did not see why we did not compel a workman
to give the maximum output without extra compensation if we
knew how it should be done.

7. A third man said to the writer, " Now for the first time I see
how you can get the maximum output from your shop without
having trouble with your men, and if you carry out the methods
described I don't see how you can have trouble."

These opinions, being those of well-known and prominent men,
are worthy of careful consideration.

8. Analysis of Opinions: Regarding the first opinion, that the
introduction of the methods advocated by Mr. Taylor, and it was
the study of unit times that was especially referred to, would
produce a strike in his works, the reply is that s^uch study has
never yet caused a strike. That it is possible to cause a strike in
almost any plant by doing this work in an obnoxious way is un-
doubtedly true ; but it is equally true that the work may be done
without serious opposition in almost any plant if sufficient time
and patience a/re devoted to it. It is realized, however, only by
those who have actually done the work how much time and
patience may be needed, and a man who undertakes this work
without the experience of others to guide him is apt to be dis-
couraged, for he will find that his progress is extremely slow.

9. On the other hand if the workmen are so united in their
determination that no one man can be found who will follow the
wishes of the management and do exactly what is wanted without
objecting to having the details observed and recorded, it would
seem desirable to get such a man as soon as possible. This, how-

ttODIFYlKO SYSTEMS OF MAKAGEHEKT. 65

ever, is a condition the writer has never come across; and, while
it may exist^ is certainly very rare, and probably does not exist at
all in any large plant, although it may take some time and patience
to find the right man. When a start has been made and the good
men begin to realize that what is being done is for their advan-
tage as well as for that of the company, there need be no trouble
provided men are allowed to see the results of one step before
another is taken.

10. The Second Opinion: That if we knew how to get the
mayimnm output of a machine we should compel the workmen to
get it without extrp, compensation is not in keeping with the
spirit of the age. We cannot to-day in this country compel any-
body to do anything. The employer must concede to the work-
man what he demands for himself — that he be allowed to do what
he belifeves it to be lus interest to do. In other words, when the
employer has decided upon what he believes to be his interest, his
only successful method of procedure is to make it to the interest
of the workman to do what is wanted. It may take time and
patience to make the workman realize what his true interest is and
see it in the same light as the employer does, and unions may
oppose it; but if the inducement is a fair one and the employe
is subjected to no real hardship it is only a question of time when
somebody will be found of sufficient independence to work for
his own interest,

11. The Third Opinion: That these methods properly carried
out should give the maximimi output and be practically an insur-
ance against labor trouble the writer leaves for the comment of
others. Suffice it to say that the Avriter is finding an increasing
number of men that are looking at the matter in that light, and
asking for information as to the best method of beginning this
work under the conditions that exist in their shops.

12. Introducing New Methods : It must always be borne in mind
that everybody is suspicious of new methods, and that the only
way to remove this suspicion is to show them that the new
methods are going to help them in their work. If the first thing
that is started is helpful to somebody, it will not 'l)e long before
there is a sentiment in favor of the new system. The time-keeping
department and the foremen usually find the system of daily re-
turns from the men of such assistance to them that it soon has
their support, and when the Graphical Daily Balance begins to
show up weak spots the best foremen realize they have at their

66 MODIFYING SYSTEMS OP MANAGEMENT.

command an instrument which will help them to increase the
eflSciency of their work by enabling them to put their efforts where
they are most needed.

Having thus stimulated an interest in making improvement,
the value of the detail methods as advocated by Mr. Taylor will
soon be realized, after which their adoption is only a question of
time.

13. The Office: The remarks so far have been with reference
to the shop, but tiiey are equally applicable to the office, for to
have a schedule of what should be done in the office each day and
a graphical representation on that schedule of what was done is
of great advantage to the management, and is essential to proper
hannonious relations betwipen the office and the shop. Indeed to
be able to make quickly each day such comparisons as the follow-
ing for the day before is quite as important as to make similar
coinparisons of the shop work :

What drawings should have been completed,

What drawings were completed.

What purchase orders should have been placed,

What purchase orders were placed.

What material should have been received,

What material was received.

14. It also costs but little to make readily available each day a
knowledge of what has been spent in labor and material on any
piece of work up to the close of the day previous.

DISCUSSION.

Mr. Chas, D, Parker, — The discussion has been almost
wholly confined to the management and the workmen. I would
like to make an inquiry of Mr.^Gantt and others who have put
the matter in practice: How about the intermediary agent, the
foreman? What does he get out of it, and how much in-
terested is he in putting it through? Does he get his pay
raised when the men under him increase their earnings? Does
he like it when some of the men under him make much more
than he does, and is there any difficulty arising in putting this
method in practice from that source?

Mr. Ganii,* — I will say that the method adopted at the Beth-

* Aatbor*s Closure under the Rules.

MODIPYIXO SYSTEMS OF MANAGEMENT. 67

lehein Steel Company's Works, when a number of men working
under a foreman each had instructions for doing his work and a
l>oniis for the accomplishment of that work in the time stated was
as follows: the foreman was given a bonus in proportion to the
number of men who earned their bonuses and an extra bonus lJ all
of them earned it. It was thus made to the interest of the fore-
man to give his attention to the poorer workmen, and not to the
best workmen, who needed attention least ; that worked very satis-
factorily at Bethlehem. I also know of another concern that is
putting the same plan into operation now — an entirely different
kind of plant.

6d IS AinrrniNG the matter wtth piece work?

IS ANYTHING THE MATTER WITH PIECE WORKf\

BT FRAHK IUCHABIM« MSW YORK CTTT.

(Member of tbe Society.)

1. The title of this paper, if perhaps slangy in form is not so in
fact, and is, I think, fairly characteristic of what follows. It
may be frankly stated that the purpose is not so much to convey
information as it is to provoke discussion and to accumulate
knowledge upon one of the unsettled questions and one of the
most important which can engage the attention of this Society.
It is also one which can most affect the interests of its members
and of those most interested with them in the safe and successful
conduct of business. A perfunctory "sitting down" upon the
paper is not all that any one could properly desire and cannot
possibly close the case. ^

2. Attention is invited to the accompanying diagram (Fig. 9),
which is easily understood. The purpose of it is to show the actual
earnings of the workman, and of course also the labor-cost to the
employer, for any given amount of work done under either day
work or piece work at different rates, the Rowan premium system
and Mr. Halsey's premium plan. The amount of work done is

* Presented at the New York meetiug (December, 1908) of the American So-
cieij of Mechanical Engineers, and forming part of Volume XXV. of the Trans-
actions,

t For farther dlscossion on this topic consult Transactions as follows :
No. 84t. vol. X., p. 600 : "Gains Sharing " H. R. Towne.
No. 449. vol. xil., p. 765 : " Premium Plan." F. A. Halsej.
No. 647, vol. xvi., p. 860 : ** Piece Rate System." F. W. Taylor.
No. 909, vol. xvii., p. 1040 : ** Drawing Room and Shop System." F. O. Ball.
No. 928. vol. xxiil.. p. 841 : ** Bonus System for Rewarding Labor." H. L. Gantt.
No. 966, vol. xxiv., p. 260 : "Gift Proposition for Paying Workmen." Frank

Richards.
No. 1001, vol. xxlv., p. 1302: ** The Machine Shop Problem.'* Charles Day.
No. 1002. vol. xxiv., p. 1822 : '< Graphical Daily Balance in Manufacture." H.

L. Gantt.
No. 1008", vol. xxiv., p. 1887 : " Shop Management." F. W. Taylor.

18 ANYTHING THE MATTER WITH PIECE WORK? 69

represented by the lengths of the horizontal lines and the wages
paid are represented by the vertical lines.

3. As the Rowan system is not in use in this country all may
not understand its basis of computation. It starts with a fair day's
work, although that may not be the term used to designate it
The unit assumed is the amount or quantity of work which' the
man should ordinarily be expected to do in a day for the ordinary
day*8 wage without any special inducement The premium is
earned only by the work which is done in excess of the regular
day's work, and the premium earned is according to the time
saved in doing the work. If double the work is done in the
given time then one-half the time is saved and the man is paid
one-half in addition to his regular wages. If the man does one
and a half times his day's work then one-third of the time is saved
and he is paid one-third more than his day's wages, and so on.
The basis of computation is thus fixed and cannot be juggled with,
but the inducement constantly decreases with the amount of work
done, so that whatever a man may do he can never by any pos-
sibility double his earnings. Mr. Halse^s premium plan, of
course, requires no explanation here, and it will be designated
hereafter as tlis premium plan.

4. Beferring to the diagram it will be seen that both day work
and piece work, whatever the rate of the latter, are represented
tiuroughout by straight lines. A discouragement curve represents
the Rowan premium system and Mr. Halsey's premium plan has
a bend sinister. It was impossible to include Mr. Oantt's bonus
system in the diagram because a part of it, the part where you do
not quite earn the bonus, must be represented by an invisibh
line.

6. It cannot fail to strike the observer at once that in the pre-
mium plan the work which is done in the earning of the premium
is straight, absolute piece work. The name cannot disguise it. The
line in the diagram for the premium plan at one-half rate is ex-
actly parallel to the half rate piece work line, the wages earned
rise equally in each with equal increments of work done. So the
three-eighth premium rate is parallel to the three-eighth piece
work rate, and so on. If in making the premiutn plan bargain,
the proposition were made to the man to first do his allotted quota
and be credited with his day's wages and that then he should go
to work by the piece for the remainder of the day at one-half
the day rate, that would be the premium plan in every particular.

IS ANYTHING THE MATTER WITH PIECE WORK?

6. The partial piece-work character of the premium plan being
imdeniable, a paper whose topic is piece work must claim the right
to handle it freely and without apology. The premium plan was
invented by its originator nineteen years ago ; it was put in opera-
tion in the shop at Sherbrooke, Canada, thirteen years ago, and
was first brought to the notice, of this Society in a paper twelve

Unit of Work 1

MiehmniM, F.

Awk.B«mk Sm* Ca.^. I*.

years ago. The plan, I know, has been proposed and advocated
in all honesty of purpose; it has been pushed with earnestness
and persistency. As a result the premium plan is in operation in
a few machine shops and nowhere else. I venture the personal
opinion, based on the fullest available information, that perhaps
two per cent, of the machine work in the United States is done
under the premium plan, while ten times as much is done by im-
disguised piece work and much more than half is still done by the
day.

IS ANYTHING THE MATTER WITH PIECE WORK? 7 1

7. It is not at all apparent that there are any peculiar ccmdi-
tions in the machine shop which demand any different plans of
wage adjustment than are prevalent in the other traded. While a
knowledge of the premium plan is now widespread, the plan has
not made itself appear so good a thing that any of the other trades
have taken it up. It would not work with the shoemakers of
Lynn, the hatters of Danbury, the gloveraakers of Gloversville,
or the stitchers and starchers in the collar shops of Troy, for
they all work by the piece, as do most of the manufacturing
trades, and the ultimate possibilities of economical production
are thereby secured as completely as they can ever be claimed
to be under the premium plan.

8. We might by an effort imagine the effect of proposing the
premium plan to one of the trades outside the machine shop. Let
it be tried on a lot of bricklayers. Say that it is first agreed that
the day's wages are earned when five hundred bricks are laid, and
that the premium plan begins right there. The bald proposition is,
first, that if five hundred bricks are laid five hundred bricks will bo
paid for. This is so far meant to be an honest bargain on both
sides. If you don't lay another brick above the five hundred we
will have no cause of complaint. Well, now, having agreed to pay
for the laying of the five hundred bricks, when the five hundred
bricks are laid go on and lay as many more as you can. If you
lay seven hundred and fifty bricks we will pay you for laying six
hundred and twenty-five bricks; if you lay one thousand bricks
we will pay you for laying seven hundred and fifty bricks, and so
on. It will be very plain that under this arrangement the workmen
are clearly the gainers, for if you lay more bricks you ge*t some
more money, and every additional cent you get is, of course,
clear gain to you. The absurdity of this thing, when dealing
with bricklayers, is sufficiently evident; are machinists so vastly
different from bricklayers ?

9. They must be different or else there are some things about
the premium plan upon which I need information, and I take this
way to get it. One of the inherent and inseparable conditions of
the scheme would seem to be the voluntary acceptance of it by
the individual workman. It depends entirely upon himself how
much the man shall do after the allotted amount for the day's
work is done. He may do much or he may do little, and therefore
if he so chooses he may do none at all, but just be content to work
along at his usual rate and just earn his day's wages. The

•ri ^---JUi ^ -*■;

' ^ ,^ ^ dA ^ ^ '^* ..^ , ^:. -*..' -.^ • J

^/ . . ; ", ^ 4- t-'^*' *;irti ;.*^> V5: it luBT lic^ r j» ~&ar

^'.^.. //y,,^ < ^' ♦^ « . /'. O^ t// "^^'/^mr *. X ' r 9BDLT " '^"-^

/ .* // -^ . ^ y ; *^. ^.<^.^ v^f . 1 w ' * .amfiitet IML * -^LSL -xiasm

M *// ^'* '4 , *// '^'- >4 * '^^^.^'vAJf ^-f »:>!» lail MmtfOTTiff SL «ir
^*/./.>^ //^ ,r* /^ *Vr /^/ ** .* M ;/«Vrrf,^>, V> £j: iiSfflCK TOMS. Bill 51

^tft^^h fitU^ h hnh4^*4 th^h nf*i HuffUr/ed and where piece wwk
ffH ^ftfid )tt nil fihfmfUtu hiM, ^9 th^t fM; p^ cent of the prodiie-
I/'/ ^'//^ h\ l^r M»M^ ^/lwMi«l»r^»^ot h 'Jorur hy pi^^ee work, «i«I
)l «/f^«)^ Ki* «n)/1 //f \Smi t^%\M\M%\mm\i that there is no catting of
mHmi IIm-mi, \\\%\,^% IrMly ma I ^Mffp/mr; it in ever said of works where
Mm* I^m HiliMff I'liMi U jn iiM^n All pri(7<)i» when made run for a year.
'lll^V MM* \\\\\i \\\\i\\\\\\\\y Imprmi'd liy the employer or his repre-
*immIm1Iv* rt, hMl' «M» l.l»#» 0Ml,<M»Mu? of fMir ami free and friendly con-
CmmmMi mm(I wImmi nliMMK^'M ^^^ priro are imperative they are
ii(l|iiM|(.i| \\\\\\\\\ Ik llrn fniiiMt wuy. Thn works are prosperous con-
lliiiMilh, mimI IIim O'liHltih of omployors and employes are less
ithMtiMMJ iJiMM llM\y UMio iMidor ollu^r arrangements.

II ll \\\\\A lio <'Ni(lonl (lutt nuiio of tlioHo premium or bonus^
\\\ mIIom oMnoil or l»ohl, or Jofootivo lino schemes, whatever they
oo»v \A\\\\\\ \\\ \\\\\ wwsi K\{ i|ulokouinjc tho pnoo of the worker and
(«h>hut>iM^\ tlio oMlp\ili OMU bo tho uu)!tt ofFootive, for the reason
\\\\\\ lliov ortoi \^ hnhu^ovl iiuvutiv^ at tho priKuse time when the
wvK^A of 0uHnUh\* it lUv^^t ur^vut. It is^ the last piece done whidi
Ns^^u^ t llu^ luulv tt, MUv( U U Ml^iml to offor the man half-price or

IS ANYTHING THE MATTER WITH PIECE WORK? 73

less for doing it. With either of the premium plans doing its
best in the way of increased output and reduced labor cost per
unit, and with piece-work prices adjusted to precisely the same
price per piece, the inducement to the worker to increase his out-
put still further must be greater under the piece work than under
the premium plan. The guarantee that prices shall not be cut
is precisely as applicable to piece work as to the premium
plan. The latter has absolutely no monopoly of honesty, no
assurance '^of price maintenance any more than the other.
With equal temptation to cut, and with the same human
nature in the boss, the chances of cutting will average precisely
equal

12. With no one having the slightest interest in pushing or ad-
vertising piece work, it is advancing on its merits as the most
honest way of paying for repetitive work in the machine trade as
in all others. It is worth while to note its popularity and progress
especially in the extensive line of railroad work. The testimony
at the meetings of the various railroad organizations i& very pro-
nounced in this direction. At the meeting this summer of the
Railroad Master Blacksmiths one man stated that absolutely every
job in his shop was done by the piece. When the price could not
be placed on the work to be done it was placed on the '^ heat''
Perhaps it may not always be possible to do this in the machine
shop, but whenever the opportunity arises to consider the mode of
payment it should always be in order to ask: Is Anything the Mat-
ter with Piece Work?

DISCUSSION.

Mr. Harrington Emerson. — Just two years ago I was fortunate
enough to have Mr. F. W. Taylor explain to me his principles
of shop directions, and the results that he had obtained. It is
Mr. Taylor's great merit that he first applied these principles
to machine shops, but they are as old as humanity, and consist
in setting a definite task, in directing its execution and in appor-
tioning the reward according to the deed. Thirty years ago I
made my first acquaintance with these methods in the strictest
of strict German schools, where we were put under functional
teachers, ten or twelve different men each day, where we were
allotted to the particular classes for which our uneven attain-
ments fitted us : one boy in the highest English and lowest mathe-

74 IS ANITIIIXG THE MATTER WITH PIECE WORK?

matics, his twin brother perhajvs in highest mathematics and
liiedium English, where our methods, books, appliances, hours
were strictly planned for us, our tasks set up to the full limit of
our capacity, and where we were rewarded as individuals, not as
members of a class. In the middle of the term a boy might be
either promoted or debased, yet it was open to each to secure the
same highest marks, and with the marks much-prized rewards
and special privileges.

The results in that school were astonishing. The great
majority of the boys learned more in eighteen months than other
schools could teach in four years, because there was no w^aste
time, no waste process, no waste effort; it was time unit study,
functional foremanship and differential piece work.

At our meeting in Saratoga this year, Mr. Taylor presented
in printed form a full statement of his fundamental principles,
and, in my estimation, it will be many years before anything of
more than detail value can be added to his work. My copy of
his book is worn out \\4th thumbing of the leaves, and marking
of important passages. I discovered that it would have been
easier to mark the few sentences here and there that were not
of prime importance; yet, for our fellow-member, Mr. Frank
Richards, it is as if Mr. Taylor's work had never been made
jmblic.

What is the matter with piece work? Everything is the
matter with it. It is a lazy, haphazard method of shifting
responsibility and direction from employer to employee. It
works for deception in the latter, and gives us a long string of
broken promises from the former; it is hated and opposed by
the unions, and with reason; it brought on the great Union
Pacific strike last year, which is not yet finally settled. Piece
work makes no provision for justice, and any system is wrong
that is not based on justice. In some of the great Burlington
^hops " Xovo " and other modern steels have been introduced,
doubling the output of the workman without extra effort on his
part, yet the superintendent of motive power told me that piece
v.'ork rates would not l>c changed, and in the Union Pacific shops
men came and asked the privilege of j)aying for their own modern
tools if piece rates would be li'ft unaltered.

The fundamental trouble with piece work, in addition to its
lack of justice, is that it makes the workuian sell what is not his
to sell, namely, OUTPUT. When Mr. J. J. Hill formulated

IS ANYTHING THE MATTER WITH PIECE WORK? 75

his famous principle that railroad expenses were by the train mile
and receipts by the ton mile, neither his train crews nor himself
ever dreamed of putting the pay of the men on a tonnage basis.
The engineer who hauls sixty 80,000-pound cars with a hundred-
ton engine gets no more than the engineer who obeys orders,
standing for hours on a side track. The engineer sells his time,
his skill, his intelligence, his obedience, but never output, because
that depends on conditions over which he has no control; and it
lias always been a wonder to me that railroads which manage
their train problems should be so backward in their machine-
shop practices and methods.

What the employe sells, whether in office or shop, is not his
^* output," but primarily his time and his' skill, incidentally his
intelligence and his obedience.

That many shops pay by piece work is no argument in favor
of the plan, since more shops pay by day work; and, as Mr.
Earth in his slide rule paper, presented at this meeting, only too
moderately remarks, the usual way of running a machine shop
appears little less than absurd.

The experiences of Mr. Taylor, Mr. Gantt, Messrs. Dodge and
Day, Mr. Barth, myself and Mr. Parkhurst, who have carefully
studied the output and results in innumerable machine shops,
prove that the wastes going on are more than absurd. As an
example of old practice against new, I hold in my hands the
original figures of the skilled and competent engineer of a large
shop, who estimated the cost of a certain job at $4,575, of which
$3,300 for materials and $1,275 for labor. The work came
under my direction after it was one-third completed, and was
pulled oflF with four men in three months for a total cost of
$3,375.09, of which $622.79 for labor, netting a profit of
$1,824.91, instead of $629, as estimated — nearly three for one,
yet some of the men on that job were paid a bonus of nearly
100 per cent, above their regular wages. I also hold a routing
card of one of my assistants, Mr. Parkhurst, in which a car shop
job, marking and moving 200 pieces of oak, was estimated by
the foreman to require two days, but waS actually completed in
2 hours, 25 minutes on a 50 per cent, bonus basis. This is what
Mr. Taylor's methods will do when applied to an old time shop.

In planning jobs of this kind we pay no attention to what
has or is being done. Former practices have absolutely no inter-
est for us. We figure out the time the job ought to take under

IS ANYTHING THE MATTER WITH PIECE WORK?

existing conditions, and we pay the man a generous bonus, which
must be enough to call out the best that is in him. If the con-
ditions appertaining to the job are changed, either for better or
worse, we again determine the minimum times and pay the man
a bonus for his cooperation. These illustrations show that
astounding results follow the plans Mr. Bichards condemns
without .understanding them, and that there is no argument what-
ever in appealing to present practices.

Mr. Richards's diagrams and his reasoning and conclusions are
erroneous, because he bases them on output, which does not
properly enter into the matter at all, as diagrams based on time
instantly show.

I assume in all cases wages of 25 cents an hour, a usual time
8 hours for a given job, a slow time of 10 hours, a fair time of
6 hours, a piece work time of 5 hours, a Taylor bonus time of
3 hours. Time, days, hours, minutes — ^in this case hours — are
measured on horizontal lines, wages by the week, day, hour or
minute — in this case hour by the hour — are measured vertically.

Diagram 1. Day Work.

(a) Blow day, (b) average, (c) fast day ander good foreman.

-2.00

-1.00

10 Ilours

Normal cost, 8 boars to employer, $2.00 ; wa^es per boar, 25 cents.
Slow •• 10 •* ** •* 2.50; " * *' 25 "

Low •• 6 ** •* " 1.60; *' " ** 25 "

The employer makes all the gain or loss. He is stimulated to
good foremanship and better equipment, but the constant tend-
ency is to deterioration.

IS ANYTHING THE MATTER WITH PIECE WORK?

Diagram 2.— Piece Work.

-2.00

Piece work cost at 6 boura, $1.50 ; wages per hour, .25 cents.

•' ** *' 8 •' 1.50; •• " " .1875 cents. .
•' •* " " 4 •* 1.50; ** ** ** .875 **

This is exactly the reverse of day work The employe makes
all the gain or loss, and is afraid to cut time for fear wages will
be cut.

DiAORAM 8.— Halsbt Prehium.

-2.00

1.00,

Pio. 12.

Cost to emplojer at 10 hours, $2.50
•• •« ** '* 8 •• 2.00
•• •• •• '• 6 " 1.75

" «• •* ** 4 •* 1.50

10 Hours

wages per hour, .25 cents.

It «< (c 25 <<

«< « •« 20 **

** •* ** .87.5 *•

18 ANYTHING THE MATTER WITH PIECE WORK?

The chief merit of this plan is that it obviates the necessity
for change in piece rates. It has worked admirably in certain
shops, steering a half-way course between the injustice of day
work and of piece work, but it is not fitted to cope with the un-
expected. If there are no improvements by the employer there
is no reason why the employe should not get in full the increased
result dvle to his greater diligence and skill, but if improvement
is due to the employer's better equipment there is no justice in
giving the employe any part of it.

Diagram 4.— Tatlor Differential Piece.

2.00

-1.00.

8 10 Hours

Cost to employer at 8 boars, $1.50 ; wages per hour, .50 cents.
.» 4. 4« M2 M 1.60; '• •' *• .75 **

If employe habitually falls below three hours he is not wanted.

Here, for the first time, attention is concentrated on the
reasonable maximum of production and the reward made pro-
portionately great. Xot only is there no attempt made to cut
piece work prices, but the reward is withheld unless the maximum
is done. The great difference between this and ordinary piece
work is that Mr. Taylor demands the payment of a high premium,
often 100 per cent., a figure that would frighten most employers,
in order to effect maximum reduction in cost. If the employer
introduces improvements, times are with justice shortened but
not the premium per hour; if equipment deteriorates times must
be lengthened but the same premium be paid per hour.

IS ANYTHING THE MATTER WITH PIECE WORK?
Diagram 5.— Gantt Bonus.

10 Hours

Fig. 14.

Co6t to employer at 6 boars, $1 .50 ; wages per hour, .25 cents.
.. .* 4. 4. 3 .. J 5Q. .4 .. .. 50 M

" .. M w 2 •• 1.25; " .. •« .625 •*

The <liffereni.e between the Taylor and Gantt plans is that
the fonuer pays by the piece finished in a definite time, while
the latter pays by the definite time for a eompleted job, and pays
the bonus, not for the piece, but for following instructions.

Mr. Gantt does not admit that under his system the work-
man could l>etter the time set and therefore objects to the sup-
jtosition of two hours on a three-hour job; but t extend the
diagram theoretically in order to show the diflFerence between
Taylor and Gantt. Taylor is more severe and more generous.

After careful study of the Taylor and (Jantt diagrams, Mr.
Parkhurst and myself, adhering absolutely to the Taylor and
(iantt theory of time unit study and specific directions of all
operations, have used other diagrams, less severe than Taylor
and (lantt, and permitting us in an old sho]), where tool, machine
and labor conditions are not modern, to keep the ideal always
in view, yet we reward any gain shown by the workman. We
determine with all the skill at our command the time a job
should take, and adopt the Taylor line, based however on time
and not on piece, and then run back to the day line.

IS ANYTHING THE MATTER WITH PIECE WORK?
Diagram 6 — Emerson Parabolic.

-2.00

1.00 2

Cost to emp]

A 6

8 10 Hours

Fig. 15

boars, $2.50

wages per hour, .25 cents.

8 •* 2.00

" •' - .26 "

6 " 1.75

" " .29 "

4 *' 1.50

: " " - .375 ••

3 *• 1.50

M .. 5Q ..

2 •* 1.50

- " " .75 -

Diagram 7--Parkhub8T Combinatiok.

2.00

Fig. 16.
Cost to employer, 8 hoars, $3.00

M M M ^ O ^ 0g

'* ** ** 4 »' 1.56
'* " ** 8 '* 1.50
*' •« .' 2 " 1.50

- 1.00 ,

10 Hours

wages per hour, .25 cents.

.50
.75

IS ANYTHING THE MATTEB WITH PIECE WOEK? 81

The essential difference between these diagrams and the Hal-
sey premium Kne is not that they are curved and it is straight,
but that it begins with an accurate and probably justly deter-
mined rate and drifts mathematically, but not scientifically, into
space.

Mr. Parkhurst and myself begin with the scientific maximum -
of output and reward for endeavor, and, as a mere matter of shop
convenience curve backwards to the day rate line. There is no
special merit in the parabola or in the straight line; other lines
might answer practically as well. The main point is that a
little improvement gets a little taste of reward, and a big im-
provement gets a great big reward.

When all conditions are properly under control, I much prefer
the Taylor diagram based on time. There is something inspirit-
ing in working out a minimum time, in knowing that it can be
made with the regularity that a train makes its fast schedule,
in proving it, in stimulating the workman to it ; but it is equally
discouraging to workman, to expert and to employer to be wrecked
in full flight by hard iron from the foundry, by variable speed
in the engine, by broken belt on main shaft, by any unforeseen
and unforeseeable delay, and in such cases the curve back to
day rate prevents much trouble.

In all these diagrams, except day rate, the employe is bene-
fitted by reduction in time; in all these diagrams, except piece
work, the employer is benefitted by reduction in time, and reward
for reduction in time is apportioned exactly as it should be, only
by the Taylor method and its modifications.

The employer must pay big bonuses or he cannot get results.
He can afford to pay big bonuses, for even if he gives all the
gain in time to the employe he makes on increased efiiciency of
plant and diminished overcharges.

Where, in all these lines and curves, when based on time, is
there any support for Mr. Eichards^s contention that they cannot
be effective because, as he claims, they offer a reduced incentive
at the precise time when the need of incentive is most urgent.
Exactly the contrary is true. They all of them offer, just as
piece work does, ever increasing pay, which, if pushed to the
theoretical limit, would reduce costs to almost nothing, and ^ve
the employe an infinitely large sum per day.

Mr. R. 0, Schnebh, — In answer to the author's question I may
add my experience, and say emphatically: " Nothing."

82 IS ANYTHING THE MATTEK WITH PIECE WORK?

The troubles heretofore attributed to piece work are really
not inherent to the system, but to those who used the system,
they are the results of ignorance, carelessness and cupidity.

Improper rates are the result of ignorance or carelessness.
When an operation or a series of operations are carefully an-
alyzed by a person whose knowledge qualifies him, there is rarely
any trouble after the employees are impressed with the fact that
fair wages may be earned at the rate fijced. The difficulty here-
tofore has been that rates were carelessly made, and such a rate
is nearly always high, and soon leads to restriction of output or
excites the cupidity of his employers, the inevitable in either
case being rate cutting and strikes.

It is much the same qualifications that keeps it out of at least
one-half of the shops of this country. The cupidity of employ-
ers who have risen from small beginnings, who in their begin-
ning were able to hold all the details in their own heads, add'
one foreman after another from the ranks, or from (which is
worse) the family. It is impossible to recruit the necessary abil-
ity from sucli material. They absolutely refuse to pay for this
ability, and all system cr work done that cannot be written down
on a customer's bill is wasted.

I see nothing in the Premium plan nor the Bonus plan, for if
proper care is exercised in setting the time limits, a price might
as well be set at once. 'Tis a sugar coating to fool the work-
man, or at best a stepping-stone toward piece work. In the
Itowan system there is a corrective applied for excessive time
limits (equals piece price). However, in the Rowan works the
time limits are never changed (if I understand rightly), no matter
if methods or machinery are improved.

I most heartily endorse the paper of Mr. Richards. I might
add that in handling odd work in a piece-work shop I found it
convenient to adopt what we termed an " excursion rate " good
for this day and date only, until such time when the work became
regular and it was possible to set a stable price.

Mr. Ganit, — With regard to Mr. Emerson's statement that I
was not serious when 1 suggested that when a workman did
better than his instructions called for he should be made in-
structor, I have to say that I was quite serious, and as a matter
of fact one of the best instructors I know was discovered just
that way.

Mr. Fichards' suggestion with regard to the West Albany

IS ANYTHING THE MATTER WITH PIECE WORK? 8B

shops is quite in order, and the superintendent of those shops
is quite in accord with him; so much so indeed that he has
strongly recommended that this work be begun there as soon as
possible. The knowledge of what has been done in Schenectady
is the reason for his action, which is seconded by some of his
foremen who know the foremen at Schenectady.

Mr. Oberlin Smith. — There is one feature in all this discus-
sion which needs to be gone into further. Everything that has
been said seems to apply to manufacturing shops. Now we
must differentiate between these and the most primitive kind of
a shop, the jobbing shop, which is simply used for making odd
things for people and for doing repairs. Therein we cannot well
have anything much better than " day work." For this we must
have the right kind of a foreman and the right kind of work-
men, specialists in their line. Now, at the other end of the
series, is a manufacturing shop where hundreds and thousands of
things are all made exactly alike; or, at least, there are various
component parts of them made alike. Therein systems of
'* piece " work or " premium " work or " bonus " pay are, of
course, applicable, and they seem to have been thought out and
developed especially with reference to such shops. But there
is an intermediate kind of shop, and probably the great majority
of our machine shops are of this class, whether you call it mongrel
or hermaphrodite, or what not; semi-manufacturing would per-
haps be a good name for it. It has chanced that my experience
has been mostly in a shop of this kind. Our product is presses
and dies for working bar and sheet metals. The dies are almost
all different; they have to be worked on the jobbing shop plan
in a department by themselves, and it would be very difficult to
apply any " piece " or " premium " system to the work upon
them. In our other department we make about five hundred
different kinds and sizes of presses. In consequence of making
so many kinds in a comparatively small shop, employing from
150 to 200 workmen, it has seemed very difficult to put in prac-
tice any of plans of payment in question.

Doubtless some of the gentlemen here have had experience in
shops of this kind and have tested these improved methods of
payment. The larger parts of our machines, such as the frames,
and the component parts of such frames as are of built-up con-
struction must, on the larger sizes, be made only one at a time,
as some certain kind of machine may perhaps be ordered only

84 IS ANYTHING THE MATTER WITH PIECE WORK?

once in a year or two. Of course, very few of the pieces are
made alike and cannot be made up in stock to much extent.
Other more standard kinds are made up in batches of ten. We
must, therefore, consider things which are made in batches of
from one to ten on the larger pieces, and, we wnll say, from one
to fifty on the small pieces. I have tried the premium plan
and the piece work plan a little, but generally we had to fall
back on the old day-work system on account of the great amount
of bookkeeping involved.

When we try to manufacture strictly, we are handicapped by
the small batches. Furthermore, we are much bothered when
customers require modifications and insist that a casting must
be altered; sometimes they merely want changes in the way of
equipping with attachments often \*arying in design. Now, witt
all these complications how far can we apply any of the new
systems ?

In regard to the relative merits of the " piece work " and the
" premium " plan, I do not feel competent to speak. Of course,
we are all liable to meet the diiBculties involved in the labor
problem. The " walking delegate " has fixed upon the term
" piece work " a big black mark. After he found something
was being done to get around him on that, he put another black
mark over the word " premium,'' and then " bonus " came under
the ban. Labor, however, is gradually being educated in eco-
nomics and will, I believe, become more sensible. We must
help it to learn, and must sometimes meet it part way-^remem-
bering that there are two sides to every question.

Mr. E. P. Bates, — I think Mr. Emerson has left an iidpres-
sion that he would not care to have remain with us. As I
understood him, where a price was made for piece work and the
mechanic did the work quicker than was expected, the profit all
went to the mechanic, and where he did it slower the loss went
to the mechanic. My experience is that it may cost from fifty
cents to two dollars a day to furnish this mechanic with tools
and superintendence, including all the items which go to making
the cost of operating a plant, and I think that the operator of
the plant is quite as much interested as is the mechanic in regard
to time, and that the loss or the gain comes to the owner of the
plant fully as much as to the working man.

Mr, Emerson. — Of course, the gentleman imderstands that I
did not read my paper in full; I abridged it very much, but if

18 ANYTHING THE MATTER WITH PIECE WORK? 85

you read it when published in the Transactions of the Society,
you will find this statement in it :

" The employer must pay big bonuses, or he cannot get results.
lie can afford to pay big bonuses, for even if he gives all the gain
m time to the employe he makes on increased efiiciency of plant
and diminished overcharges."

Mr. Taylor, — At the risk of being prosy, and always coming
up with some old remark every time this subject is before the
Society, I want to say again that I think there is no real quarrel
between any of the systems of payment in common use. It is a
curious phenomenon that there are certain men who seem to
wish to attach their names to some one comparatively small and
unimportant element in management. They appear to be unable
to see anything else in the whole line of management except their
one chosen element and attempt to convince themselves and
every one else that this is the whole art of management. Appar-
ently Mr. Richards looks upon piece work as the whole of this
art. I have not heard that he has been able to see any good in
any other element.

Now, I wish again to say what most of you have heard a num-
ber of times, but perhaps some have not yet heard it, that to my
inind there is no quarrel between the various systems of paying
men which are in common use. I think each of these systems
has its proper place; at least four of them can properly be used
in the same shop and at the same time, providing the shop is
large enough. Every large machine shop in the country should
have, I should say, not less than four systems of payment going
on at once in order to do the work in the most economical man-
ner. There cannot be any quarrel, then, between *^ day work "
and " piece work " and Mr. Gantt's ^' bonus plan " and the
*' differential rate system " of piece work since, as I pointed out
in my paper on shop management last spring, each plan has its
own individual field of usefulness ; and I feel convinced that each
one has a field that it is impossible for either the other systems
to entirely fill. And again, outside of these plans I feel that the
Towne-Halsey plan has a large field of usefulness, Mr. Richards
to the contrary notwithstanding.

But back of all systems of paying men and imdemeath them
all, and of vastly more importance than any system of paying
men, lies the true remedy for the fundamental difficulty between
the managers and their workmen. The great difficulty that

86 IS ANYTHING THE MATTER WITH PIECE WORK?

presents itself is that in most cases neither the one nor the other
know accurately how much of any given work a good man can
and ought to do in a day. It is only in rare cases that either the
managers or the men know what really constitutes a day's work.
And what I feel absolutely sure of and wish again to emphasize
is that the only proper solution of the wages question, both as to
the system for paying men to be employed and the compensation
to be paid, lies in a scientific study of how much each man can
do and ought to do; what a really first-class man properly suited
io his work can do if he wishes to and if he has the proper
appliances. It is this study that is so much more important than
Ihe adoption of any one system of paying men, that by com-
parison the differences between the latter sink into insignificance.

After the owner or manager is in possession of the exact
knowledge of how long it ought to take to do the work, even the
day work plan, which in many cases is perhaps the least satis-
factory method, will produce much better results than any of
the other systems without this knowledge. Of course, with the
knowledge there is a choice, and each one of the four systems
lias its proper place in every shop.

Mr. Stephen W. Balkwill. — With regard to Mr. Smith's
remarks about cost, it seems to me that he brought out
some ideas that are worthy of consideration which come more
into actual shop practice, except where the factory manufactures
a specialty and makes nothing but repetition work which does
not require any change of machines. Take,, for instance, an
article of which ten pieces would be required at one time and
a thousand at another time. It is quite obvious that the machine
lo be used takes a definite amount of time to alter, for it is sup-
posed that each of the ten pieces must be as accurately made as
any of the thousand, and the machine must therefore be as
accurately adjusted in either case and therefore makes the cost
of production per piece of the fewer number of pieces a great
deal higher than that of the greater number on account of hav-
ing the same common fixed expenses.

Kegarding Mr. Kichards's remarks about the young man whom
he referred to as boring tires, would say that this is a remarkable
performance presuming it is on a single machine and paid for
by the day which does not seem to leave much room for improve-
ment on the piece work basis. Referring to the statement about
the scarcity of good foremen and the best manner of getting

IS ANYTHING THE MATTEB WITH PIECE WORK? 87

good results in that direction, I agree with the gentleman that
as a general proposition they are not well enough paid for what
is required of them, and if suflicient inducement is held out a
good foreman can be had. I know of a particular case where a
foreman has a sub-foreman under him who has charge of a piece
work department, and although not as intelligent, yet he is able
to earn more money than his foreman, which does not seem
equitable nor liable to contribute to the good feeling of the
foreman.

Mr. E. F. DuBrvl, — I wish to dwell a little on a thought
brought forward by Mr. Oberlin Smith, In this discussion he
has brought up the question of the opposition of the labor unions
to " piece work " and other methods of *^ Shop Management."
Papers on " Shop Management " very frequently state that no
serious diflSculty is encountered from the workmen when such
systems are installed. I should like to know whether this is a
general rule in foundries. I have heard of considerable opposi-
tion on the part of moulders to the introduction of any other
system but straight " day work." I have not heard of many
foundries running anything but straight " day work " or straight
'^ piece work," and under both conditions of " day work " or
" piece work/' I am informed that there is a very widespread
disposition on the part of moulders to limit production and hold
down output.

Coming to the machine shops we find that the last Convention
of the Machinists' Union adopted a resolution declaring for the
abolition of anything excepting straight " go-as-you-please day
work," with no tasks set, no premiums, bonuses, " piece work,"
or anything of the sort. How many here are familiar with that
promulgation ? If this fiat is to go into effect, we must certainly
reckon with the Union as an element in the problem. The
Anthracite Coal Strike Comnussion well says that: Trades
unionism is becoming a matter of business and tha;t employer
who fails to reckon with it as an element of his business makes a
serious mistake and one which he will have to correct sooner or
later.

Perhaps business conditions next year will make inadvisable the
proposed movement to abolish " piece work." Coming events
casting their shadows before appear in the two strikes in New
York City this year, mentioned by Mr. Kichards, against premium
work. I happen to be familiar with the cause of those two

88 18 ANYTHING THE MATTEB WITH PIECE WORK?

strikes and I can assure Mr. Eichards that " piece work," premium
work and all other similar methods look alike to the unions in
question, and that the strikes would have occurred against " piece
work " just as much as premium work. Furthermore, it seems to
me that the employers in question would have had more difficulty
in filling their shops with non-union piece workers than they had
to fill them with non-union premium workers, because even a non-
union man does not like " piece work " over much.

I do not believe that there is so much the matter with " piece
work " as there is with the men who are trying to establish it in
their shops. The greatest difficulty is with the managers who
" know not Jacob." I had occasion to deal with such a case not
over two weeks ago. As you may all know, premium work is very
largely and very firmly established in the shops in Cincinnati
and by general consent of the Associated Manufacturers in that
city the rule is that once a time limit is established it shall not
be cut unless a change has been made in the methods of produc-
tion. A certain shop had a new superintendent, one of the kind
who " knew not Jacob," who had never operated a premium
system before and who thought to make a showing for himself
with his employers by cutting down a time limit on a very efficient
workman, who did a sixty-hour job in eighteen hours. Those of
you who are familiar with the premium or other similar system
will not question the statement that such reductions of time are
common on the part of workmen, and the great difference is very
largely because the men setting the time limits, while they may
have a guide as to how long a job used to take, have absolutely
no way of knowing how long a job ought to take.

The matter coming to my attention, it became my duty to inter-
view that superintendent and show him the error of his way. His
cut of a time limit was in violation of the guarantee that had been
made to all the machinists of Cincinnati by the Associated Manu-
facturers, and his cut would have wrecked the system, not only in
his own shop, but in all the others. If his ideas had been carried
out, the result would have been a rejuvenation of the Walking
Delegate, with wjiom we have not been bothered for some years,
and I am sorry to say that it took quite a while and much vigorous
language, some of which was unfit for publication, to demonstrate
to that superintendent the inadvisability and injustice of his pro-
posed action.

" The matter with piece work " is in my judgment principally

18 ANYTHING THE MATTER WITH PIECE WORK? 89

the matter with the employers or their representatives, who, gen-
erally through ignorance as to how long a job usually takes and
occasionally through downright " hoggishness," have brought
'' piece work '' into a disfavor it does not deserve. Were we all
better informed, and did we all put piece prices, time limits, task
work and other such systems on a scientific and accurate basis, and
once 80 based did we guarantee the men against cuts in prices
unless general reductions in day rates were made, I believe that
all hostile criticism would be forever disarmed.

In the meantime, I believe that we should all keep our eyes
on the union end of the proposition of " Shop Management,"
and not stumble along blindly in that regard any more than we
should in regard to setting rates.

Mr. Smith. — ^I wanted to ask Mr. Taylor, when he spoke of
its always being practicable to find out the cost of a certain job,
whether he thinks it possible to get at such cost where only one
machine is fitted up in a year and, if so, how he does it? Of
course, if it is turning a steel shaft, it can be measured and the
number of pounds to be taken off in chips can be estimated ; but
suppose it is an irregular casting of a new pattern, does it pay to
find out, or can it be found out from the experience of only
having one to make, what the cast will be on the next one ?

Mr. Taylor. — ^I think the best answer to Mr. Smith's question
will be found in Mr. Earth's paper, which will be presented to
the Society at this meeting. Mr. Barth will show how an
ordinary mechanic (with the aid of our slide rules) can deter-
mine accurately and quickly just what combination of cutting
speed and feed should be used in any particular case in order
to do the work in the quickest time on any given lathe planer
or other machine tool, and in finding out the proper cutting
speed and feed to use in cutting forgings or castings, each of the
following elements which affect the answer is given its proper
weight through its own slide on the slide rule. The variable
elements, each of which affects the answer, are :

1. The pulling or driving power of the machine.

2. The strength of the feed mechanism of the machine.

3. The exact coarseness of the feed which can best be used.

4. The diameter of the piece to be turned.

5. The thickness of the layer of metal to be removed or the
proper depth of cut to be taken.

6. The hardness of the metal which is being cut ; i. e., whether

90 IS ANYTHING THE MATTER WITH PIECE WORK?

cast iron, steel or brass, and the exact degree of hardness of the
particular casting or forging.

7. The shape of the cutting tool and its size.

8. The quality of the tool steel from which the tool is made,
and the heat treatment which the tool has received.

9. Whether water, oil, air blast or other cooling medium is
used on the tool.

It is evident that this is a problem which is exceedingly diffi-
cult to solve. And the fact that these difficult problems are now
being daily and most practically and rapidly solved by the ordi-
nary mechanics who are using slide rules makes it evident that
it is a comparatively simple matter to determine the time re-
quired to do the remainder of the work of running a machine,
namely, putting the work into the machine, taking it out and
adjusting the machine, etc. And the study of the time required
to do this hand work is greatly simplified by dividing each job
into its simple elements and then timing each element separately
and systematically mth a stop watch. Such, for instance, as :

Lifting work from floor to machine.

Putting on carrier.

Adjusting work in chuck or on centres.

Calipering.

Setting tool, etc.

If the " hand work '' is studied by single simple operations,
in this way it will be found that the entire hand work of a shop
can be resolved into comparatively few elements which can bo
classified, tabulated and readily used in determining the proper
time for doing even new and complicated work.

Mr. Smith. — ^Does that include the rigging and the unrigging
of the tools?

Mr. Taylor. — Yes; it takes up the work from the time it
leaves the floor to go into the machine until it comes out finished,
and it includes the time required to make all of the changes
and adjustments in the machines. The observations are stop-
watch observations, carefully taken by a man who is trained to
this business. U]) to this time these observations have been
recorded and tabulated on loose sheets or in manuscript books.
But I predict that there will in the future be many books printed
covering in each trade the time required to do each of the
elementary operations which together in various combinations
make up the entire work of the trade.

IS ANYTHING THE MATTER WITH PIECE WORK? 91

Mr. John Colder.* — ^In Mr. Richards' paper too much is
made of the identification of modem management advances with
particular men and methods^ and he appears thereby to miss the
point.

I concur most heartily with Mr. Emerson in his praise of Mr.
F. W. Taylor's paper of June last, which I believe will bear its
full fruit until the appearance of the next volume of the Society's
Proceedings.

To illustrate the point on which issue has been joined with
Mr. Richards' I wish to present a practical problem in which
straight piece work is maintained producing all that Mr. Rich-
ards claims for it, and a great deal more, through expense in-
curred by the management in scientific time study and its neces-
sary accompaniments of betterment.

Where the Taylor system of management and labor reward
has replaced straight piece jobs in manufacturing concerns doing
nothing else at machine and bench but large quantities of light
repetition work, we have an instance in which the issues between
Mr. Richards and his critics can be reconciled. I believe that,
in such circmnstances, if the true time is scientifically deter-
mined after management and equipment conditions of the high-
est eflSciency have been established, it does not matter in the
least, whether (1) a large reward is offered for the daily task,

(2) a piece rate is fixed to give the same maximum daily pay,

(3) or a non-graduated premium or bonus appropriation which
added to a fixed daily time rate will give the same reward.

Under such circumstances Mr. Richards' output and Mr.
Emerson's time abcissse would coincide, in fact, their diagrams
would be identical.

In actual practice, workmen employed under above maximum
output conditions can see no difference between shortening the
time or increasing the task and lowering piece rates; and, as a
matter of fact, there is none.

The one, but very important, improvement upon straight
piece working being that accurate scientific time study has pre-
scribed the maximum and fixed the rate for existing and im-
proved conditions brought about on the initiative of the man-
agement, while ordinary piece rates are more liable to change,
not because they are " piece " rates, but because they are more

* Contribated after adjouniment.

92 IS ANYTHING THE MATTER WITH PIECE WORK?

or less guesses and depend for rectification purely upon what
measure of their true skill the men choose to reveal.

For the class of work assumed, straight piece work based
upon scientifically conducted time study and modernized facil-
ities, will be as easy as any to establish in place of guess piece
rates; and any other name given to the process is a distinction
without a difference. It eliminates the inefficient, and provides
an easy and automatic method of paying and encouraging the
efficient but new men who take a little time to attain the maxi-
mum task — say coming at first within fifteen per cent, of it —
while all others who aspire are tried out on daily time rates
before graduating as "fit."

8U0QB8TI0NS FOB SHOP CONSTRUOTIOK. 93

No. 1013.*

SUGGESTIONS FOR SHOP CONSTRUCTION.

BT F. A. BCBXnxaB.

(Member of the Society.)

1. Some time ago, the writer had occasion to lay out various
buildings for one of the well-known electrical manufacturing firms
in this country, with the view of constructing new shops, covering
a complete equipment for manufacturing on a large scale.

2. At the time various schemes were suggested for the ar-
rangement of the buildings in relation to each other, so that the
resultant buildings could be combined into a scheme of inter-
change between the administration offices, sub-offices, and mate-
rials from one building to another.

3. After discussing the various ideas, above referred to, the
form shown on the accompanying diagram, Fig. 17, suggested itself
to the writer, and as it is entirely novel in its construction, so far
as applied to machine shops and other manufacturing purposes,
he deemed it advisable to place it on record, as there are many
features connected with the layout, which although subject to
modification, would make an ideal manufacturing shop in every
way.

4. This plan was not carried out owing to the fact that the
proposed new buildings were abandoned for several years, and
when the works were eventually built other parties had the
matter in charge, and the proposed plan was not even known to
them.

5. Of course, it would have to be agreed upon in advance that
there would be required for manufacturing purposes a number
of buildings suitable for the various kinds of work to be manu-
factured. The plan herein proposed was primarily designed for
manufacturing electrical apparatus, such as generators, motors,

• Presented at the New Tork meeting (December, 1908) of the American
Sodetj of Mechanical Engineers, and forming part of Volame XXV. of the

SUGOESTIOKS FOB SHOP 00N8TRUCTI0K.

switches, electrical instruments, etc.; but of course, the same
scheme would be applicable to any other kind of manufacturing
where it is desired to have a number of shops, all of which are
easily accessible, both for business purposes and for delivery and
shipping of material.

A brief description of the layout is as follows: —

In the center of the space available for the buildings is located

Mmo».,ir.r.

FlO. 17.

an administration building, constituting the business, accounting,
and sales offices; and on the second story, the draughting room.
This building is octagonal, or hexagonal, which ever may be found
to be most suitable for the purpose. In this case, it has been de-
signed with a view of accommodating seven buildings, which
radiate from each side of the octagon, and has one side reserved
for the main entrance through the building.

6. The end of each shop which is nearest to the administration
building has its individual office for the foremen and shop clerks.
This, it will be seen, is a very harmonious arrangement, as every
shop is then but a short distance from the administration building.

SUOOKSnONS FOR SHOP CONSTRUOTION. 96

SO that intercourse can easily be had between the drawing room,
offices and the offices of each shop.

7. The general arrangement gives each shop plenty of yard
room, which is also very essential; and travelling cranes, either
worked by hand or power, could be located in the yard room be-
tween any two of the shops, for handling raw or finished materiaL

8. A circular track around the administration building con-
nected in front of each shop building by means of suitable turn
tables, worked by hand or power, makes the distribution of mate-
rial between the buildings very easy, and it will be noticed that
the distance the material will have to travel from any one bliilding
to another is comparatively short.

9. At the extreme outer end of each building is another cir-
cular track, primarily to be used for shipping purposes, and the
distributing of such material as may come in or go out over the
connecting railroad lines. This track runs through the end of
each building; and in such buildings where the machinery, cast-
ings, or other goods are to be handled, the heavier travelling
crane in that particular building which should run the length of
the shop, can easily unload or load the cars. This arrangement
makes it possible to go into every shop without having a multi-
plicity of tracks and switches, thus cutting up the available yard
room, as is usually the case in ordinary plants. It is also possible,
if there is sufficient ground available, to extend any one or all of
the buildings, and still retain the best features of the design.

10. A study of the design, which as above stated, is subject
to modification, will be all that is necessary without any further
comment, to make it clear to any one interested in this important
question of the best arrangement of buildings for shop purposes.

11. In connection with this matter, I would add that while
there is as far as the writer's knowledge goes, no manufacturing
plant built on these lines, at the same time there is a plant of
an entirely different character in Pennsylvania, where practically
the same idea is carried out, so far as the location of the buildings,
in relation to the office building is concerned. This is the Eastern
Penitentiary, located in Pennsylvania, a cut of which was pub-
lished in the North American recently, which the writer ran across
by the merest accident, and it really makes a very good perspective
picture of what a manufacturing plant would look like when laid
out as above suggested, eliminating, of course, the walls surround-
ing the grounds.

96 SUGOESTIONS FOR SHOP OONSTfiUCTION.

DISCUSSION.

Mr. TT. D. Ennis. — The arrangement of buildings illustrated
by Mr. Scheffler possesses few advantages over that which is
customary. Those which it does possess are incidental and acci-
dental rather than dependent upon the eccentric distribution of
departments proposed.

For example, the advantages of loading cars by cranes and of
possible extension at low cost are possessed to an equal degree by
this or almost any conceivable grouping of buildings.

The 'scheme is wasteful of land, requiring for the same yard!
and building area very much more ground space than is usual.

The diagram shows a centrally located oflSce. It is ques-
tionable, however, whether it pays to sacrifice other considerations
to those which in plants of any size private telephone systems and
adequate messenger service seem generally able to satisfy.

It also shows one of the worst possible systems of track-
age. There is but one communication with each department,
which must be used both for ingoing and outgoing traflSc. The
trackage provision is not adapted to the requirements of the vari-
ous departments. The same facilities are provided, for example,
for the detail department as for the storage and shipping building.
Should a carload of pig iron come in while a car of coal stood at
the power house, extra switching would be necessary. All the
trackage is in curves, which hampers the movements of cars and
increases the probability of derailment and consequent delay. Ac-
cess from the ofiicc to any of the buildings is impossible without
crossing the industrial railway tracks, a condition which may
counterbalance all the saving in time due to having the ofiice cen-
trally located. The presence of railroad tracks within the build-
ings has been found extremely dangerous to life and limb in grain
elevators, and it would not seem advisable to introduce such an
arrangement in manufacturing plants. Fire ha^zard would be
increased by running locomotives through the departments, espe-
cially such as those housed in the armature, carpenter and storage
buildings.

The power house is not centrally located, is of a wasteful and
awkward shape and cannot be extended in any direction without
heavy expense.

The yard space is accessible only from the rear ; it is distributed
without regard to departmental requirements, the detail depart-

STTGGESTIONS FOR SHOP CONSTRUCTION. 97

ment, for instance, being given somewhat more room than tho
foundry. It is so shaped that it could not be properly covered
by cranes, nor could any compact and systematic grouping of yard
material be practised. As illustrated in the diagram, the space
on the outside of the tracks is too far away from the buildings to
make economical yardage, but unless it is utilized the full benefit
of the track facilities will not be obtained.

It is difficult to see what advantage, excepting in the single
point of accessibility from the office, the suggested arrangement
of buildings has over that in which the departments are housed
under separate roofs, side by side, with the office building on one
side of the group and the railroad on the other side, one or more
switches being run into the buildings where needed. The latter
arrangement would certainly give better results as to amoimt
and distribution of yard room, despatch and economy in shipping
and receiving goods, and convenient access between office and
shops and between the shops themselves. This arrangement
might have to be modified to a moderate extent, but only in order
to obtain direct access between successive departments, as, for
example, the machine and finishing rooms of a paper milL The
feature of Mr. Scheffler's plan, requiring a journey around several
comers to get from one building to another, should be absolutely
prohibited in laying out any plant.

Mr, Chns. L. Heisler* — I offer the following criticism on the
radial plan of arranging shop buildings :

1. There should be a material switch independent of the ship-
ping switch. The plan shows that any car set for loading in the
machine shop, must be reset and shifted each time a load of coke,
coal or sand is delivered to the foundry, and the locomotive must
each time pass through the machine shop. If another switch is
made to parallel the one shown, then when the machine shop is
extended there will be two switches passing through this deparlr
ment.

2. A curved switch is dangerous and seriously inconvenient to
the switching crew, who cannot see two car lengths, so it is very
difiScult to set cars and avoid injuring shop men.

3. It is seriously objectionable to take a train of cars and loco-
motive through several shops for the purpose of reaching another.

4. The circular switch does not come within 150 feet of the

* Sabmitted after adjournment.

98 8U0GESTI0NS FOR SHOP CONSTRUCTION.

foundry cupola, and gives no opportunity for using hopper-bot-
tom cars for cheaply handling foundry sand, coke, etc. The
present arrangement will cause an expenditure of many hundred
dollars^ per year in the extra handling of the raw foundry mate-
rial.

5. The coal for the boilers should be dumped directly in front
of the boilers. In the radial plan shown, each car must be un-
loaded by hand and reset every time any car is^taken from either
the foundry, pattern shop or machine shop.

6. The pattern storerooms should be adjacent to the pattern
department, but sufficiently isolated for fire protection. The
cupola should be farther from the pattern department, and should
be located with respect to prevailing winds, if possible.

7. The arrangement does not comply with the present practice
of arranging buildings as much as possible parallel to each other
to economize in land and in order to utilize their crane columns
and steel framing for supporting traversing yard cranes, which
should cover all the available space adjacent to the several build-
ings, and which can be covered, when necessary, to meet future
growth. The triangular yards evidently will never permit this
without excessive expense and waste.

8. Assume that 30 per cent., more or less, covered floor space
than shown on the plan is required for any one of the depart-
ments. First attempt to lengthen such a department 30 per cent.,
and note that the circular switch then divides the enlarged de-
partment, or, suppose the 30 per cent, enlargement consists of
an unsymmetrical side addition, this cannot then be well fitted
into the very undesirable form of triangular yards. On the other
hand, assume that any department required 30 per cent, less floor
space than shown, in this case the building would not come within
100 feet of the switch, or the central " administration " building.

9. The several receiving and shipping offices must be adjacent
to the switch, and will, therefore, be 350 feet away from the main
office, as shown.

10. The reduction in distances at other points between parallel
buildings, when arranged as usual with an intervening crane yard,
would certainly effect a greater saving than would be lost by the
slightly greater distance between the main and other offices, as
compared with the radial plan.

11. A moment's thought will make it clear that the alleged
saving in time due to the radial plan is lost several times over in

SUGGESTIONS FOB SHOP CONSTRUCTION. 99

the time required in manipulating the many heavy 12 x 14 foot
switch doors. Assume that a switching engine and crew are
making a trip aroimd the circular track, the crew will be required
to make at least 24 distinct and strenuous efforts in opening and
closing the 12 heavy doors, and it will take 48 such operations in
cold weather, if they are all lifting doors, and 96 if they are
double doors. However cold the weather, one round trip, not
considering the resetting of cars, I think would finish even the
most robust switching crew. If they did survive one round, I
fear the shop men would never permit them to make a second.

Mr. Suplee. — In reference to the statement made by the author
of the paper just now about the arrangement of the buildings
being the same in the Eastern Penitentiary at Philadelphia, I may
add that in the centre of the circle formed by those buildings
there is a system of mirrors so arranged that a single watchman
seated in the centre can look down all of the corridors at once.
This ingenious system of mirrors was designed by one of the
prisoners.

100 WHAT ABE THE NEW MAOHINB TOOLS TO BE?

No. 1014.*

WHAT ARE THE NEW MACHINE TOOLS TO BEf

BY JOHN B. eWBBT, 8TBACUBX, N. T.

(Member of the Society.)

1. It is a fact quite apparent to users of machine tools that the
new high-speed tool steel calls for a re-designing of our machines
if we are to get even a fair share of the ultimate possibilities which
the new steel offers.

2. I expect the machine tool builders have already the reply
formulated as follows: " You just keep on building engines and
leave the machine tool business to us.'' But that will not quite do.
If no one but the engine builders had mixed in the engine business,
we would have had no turbine engines, and many of the standard
machine tools were originally devised by those who had use for
them rather than by the man who devised things to selL

3. I think the machine tool builders will admit that the ma-
chines must be re-designed; butto the most of them vnH this mean
anything but just to make the driving elements more powerful and
the machines stronger, which is as much as to say everything has
been all right, and all we need to do is to change the strength and
power. But have they been all right or half right ?

4. It can be shown by figures, I suppose (I know it to be a fact
by a trial with models), that a complete box is thirteen times more
rigid against torsion and four times more rigid against bending
than the same amount of material is in the form of side plates and
thin cross girts. It is probably from four to eight times more
rigid than the cross girt plan in any form, and yet in the case of
lathes, the whole business of whose beds is to resist torsion, only
one or two builders have had the courage to adopt the box form.

6. All planer beds can just as well be box beds with half the
cost in patterns and foundry work, and so too the tablea which are

* Presented at the* New York meeting, December 1908, of the American
Society of Mechanical Engineers, and forming part of Volmne XXV. of the

WHAT ABS THE NSW MAOHINB TOOLS TO BE? 101

spniBg by boltmg down work can just as well be box tables four
times as strong with the same material^ and with a saving of half
the cost in patterns and something in the foundry.

6. The whole tendency of the cut is to slide the work endwise of
the planer bed; but who haA ever tried putting the slots crosswise
in a way to offer the greater resistance and prevent the bending
of the bed by the peening of the upper surface, as now occurs,
which with the springing by bolting down the work are the
primary causes of cut ways.

7. Some planer and boring mill cross rails are of box section in
the centre, but are thinned down at the ends when fastened to
the housings. The most of them are three sides of a box only,
or one-t^itii the strength of a box, where a plain square box
straight through is infinitely better and cheaper. Of course the
boxes are not to be proportioned from what is in use now, but
from what is to be made to meet the new conditions. To select
enough material to meet the new demands and then put the mate-
rial 80 that it will be four times more rigid will be something like
it Housings of box section will be just as rigid fore and aft and
much more rigid against side strain.

8. Milling machines of the planer style are constructed like plan-
ing machines, seemingly without a thought but that the conditions
are identical, while they are not If the bed of a planing machine
and the table were of tJie same length, the weight of the table and
the load over-running the end of the bed would soon wear the top
of the bed crowning and the under side of the table concave to fit,
and it is to counterkct this tendency of gravity to wear them out
of true that the beds are made longer than the tables. With
the milling machine the load is less, more of it in the middle of
the table, because there is less gained by putting on small pieces
end to end, and the down pressure of the big cutter always in the
middle partially, if not wholly, neutralizes the tendency to wear
out of true by gravity. When such a machine has side cutters
or a vertical spindle, ^e pressure is always in the middle, first in
one direction and then the other, exactly the reverse from the
gravity action, and instead of the side guide of the bed being
longer than the table it should be shorter, by just about the same
amount as the bed of a planer needs to be longer.

9. Many times the sliding piece and its guides can be the same
length and keep straight. The things which do not tend to wear
out of true do not wear much, and the things which do wear out

102 WHAT ABB THB NEW MAOHINB TOOIfl TO BE?

of true and have to be refitted are never just right but when new
and when bo refitted. Where a short block slides on a long guide,
if the scraper marks wear out sooner along the middle than at the
ends, the ends of the guide need cutting off, however much over-
run it gives to the sliding block.

10. The draughtsman dare not make a drawing of an engine
cross-head over-running the guide one-third of its length at each
end; the builder would hardly dare to build it if he did, and no
user has the courage to take out the guides and cut them off or cut
away the surface even when he knows it would be money in his
pocket, but it is the thing to do. We find that in the case of a
slipper guide, owing to the effect of inertia and momentum giving
a twisting action to the crosshead, it is necessary to cut away the
guide so that the crosshead will over-run very nearly one-hdf its
length before the scraper marks will show uniform wear. This,
of course, is subject to modification according as the centre of
gravity is higher or lower, or the speed of the engine is greater
or less. We are building engines with the crossheads over-ron-
m'ng that way and people buy them.

11. To get the best out of machines, they not only want to
be rigid and true, but the drive needs to be powerful In this
respect a worm gear is about as perfect as can be, or cutting spur
gear teeth spiral accomplishes about the same result. What
appears as an objection to spiral teeth is end thrust against the
shoulders, which does not amoimt to much, and when the shaft
runs in reverse directions and end play in the journals is permis-
sible, the journals keep in much better condition. The mention
of a worm gear is like the flaunting of a red rag to some people,
but it has its place and a good many more places than it has been
used in. The claimed objection is excessive friction and loss of
power, but the results do not seem to justify the claim.

12. The most perfect worm gear we have (theoretically) is a
screw and nut, and they do waste enormously in friction, and in
proportion to what they do they wear out the most of any piece of
mechanism. The most imperfect worm gear we have (theoret-
ically) is the Seller's planing machine drive, and yet they never
wear out, and hence cannot lose much in friction.

13. In the writer's opinion two of the things which never need
to have been invented are the Hindley worm gear and a machine
for bobbing worm g^. Experience convinces the writer that a
liberal pitch worm skewed round so as to properly mesh with a

WHAT ARE THE NEW MACHINE TOOLS TO BE? 103

plain spur gear, or one with the teeth at such an angle as to skew
the worm a little more will run more easily and last longer than
the other sort. A machine driven with the worm is positive^ and if
there is any chatter it comes from elasticity in the spindle or the
work itself. The value of lathes, particularly those used for face
plate work, is considerably improved by having large and short
main bearings. They should be large to resist torsion and short to
r^ist bending, and the ordinary face plates are ridiculously fraiL
To get the best of a face plate it should be box section and as
large as will swing in the lathe.

14. Owing to the rapid wear of screws the writer is convinced
that a precision screw in any lathe used in manufacturing is of no
special value over a fairly, good one. Wearing the screw in one
place while threading a few himdred pieces destroys the precision
in a way which no future use will ever correct

If the designer will analyze every detail he will find that many
of the old features were not right to meet the old conditions and
not half right for the new.

15. While manufacturing is going to call f oi many more simple
machines — that is, machines to do one thing rapidly and well — ^the
nmchines which will do a variety of work will be still in demand,
for the sparsely settled sections of the country and the colonies
will call for the country machine shop as of old.

DISCUSSION.

Mr, John D, Riggs. — There is no question but that there is
now a demand for a general re-designing of machine tools, but
is this entirely due to the new high-speed steels? Is it not rather
due to the fact that the designs were only half right, and now
with the new steels this half right is being reduced to one-
qiiarter right? If higher speed is all that is required this can
lie had at once in most cases by putting a larger pulley on the
line shaft or at most by adding a high-speed belt between the
line and counter shafts. It may be noted that doubling the
speed in this way doubles the available power as well.

Cone pulleys have gotten into bad repute, largely, I think,
on account of having too many steps and too small diameters.

Established practice may be given credit for some of the weak
IKjints in present machine design. It does not follow that be-
cause our grandfathers sawed off the ends of their wooden lathe

104 WHAT AEE THE NEW MACHINE TOOLS TO BE?

beds square that present lathe builders should do the same with
cast iron ones. Again those builders who borrow designs and
devote their best efforts to getting money do their part in estab-
lishing practice.

In applying the individual electric drive the practice of in-
corporating the motor into the design of the machine seems to
be most commendable. One casting may serve as the frame
for both machine and motor and yet the machine may be fur-
nished with or without the motor drive.

In applying motors to radial drills why not place the motor
on the radial arm very close to the drill spindle and thus dis-
pense with a considerable portion of the mechanism now used
to transmit power from a stationary ^shaft ?

The experience of Professor Sweet with worm gears is cer-
tainly quite different from my own. While the Sellers planer
drive has proved all right, the ordinary worm gear as used in
freight elevators in buildings is not. And the worst feature of
these machines is that you seldom know how near right they
are or when they will go wrong. The lubrication of the two is
essentially different.

Those people who have labored to improve the worm gear for
elevators have my sympathy but none of my orders so long as
the direct hydraulic elevators are in the market.

Mr. H. P, Fairfield, — Any one that has had much to do with
the so-called high-speed steels must have been impressed with
the fact that the ordinary 14" or 16" engine lathe was lacking
in material. About one year ago I put in the hands of some
of our students tools forged from one of the prominent high-
speed steels, my object being to study the uses of the steel. The
students' instructions were to break down the tool if possible,
then to reduce speed and feed until a desirable balance was
reached. On diameters of about two inches the lathes used
were not able to make good, and it was necessary to reduce both
the feed and speed to prevent seriously injuring the machine.
The trouble seemed to be entirely confined to the head stock and
carriage, and after some study of the subject and continued ob-
servation, I came to a conclusion that the most of the trouble
was a lack of material in the he^d stock itself. As the diameters
turned, a high speed of revolution was needed to bring up the
surface speed to the desired point, and to prevent tearing the
head stock to pieces a heavy face plate of a size as large as the

WHAT ARE THB NEW MACHINE TOOLS TO BE? 105

kthe could swing, was put upon the nose of the spindle. This
seemed to correct much of the trouble and I believe is a desir-
able thing to do, although it is not usual to use a fly wheel on an
engine lathe. So far as observed the bed was not affected in any
case.

I would suggest that my conclusions are that the engine lathe
needs more material in its head stock, broad surfaces in its car-
riage, nicely gibbed, a positive drive to its feed works, a less
number of steps on the cone and broader belts^ and a massive
face plate.

Mr. Oberlin Smith. — ^I think the time has now come when
that " anvil principle " that Mr. Porter and I used to talk about
some years ago must come to the fore. It is true that high-
speed steels simply require more horse-power for the higher
speed and not necessarily more torque on the lathe spindle; but
as a matter of fact these new steels are probably stronger and
will take heavier cuts than other steels, without breaking off the
cutting edges. Am I right about that, Mr. Taylor?

Mr. Fred W. Taylor.— No, I think not.

Mr. Smith. — ^Mr. Taylor says I am not right. I thought he
was going to back me up.

Mr. Taylor. — ^I should like to back you up if I could. But
directly the opposite is true. The only advantage which the tool
steels containing tungsten or molybdenum in combination with
chromium and heated to a high heat according to the Taylor-
White process have over ordinary tools is that they will cut from
two to four times as fast and therefore do much more work in
a given time. The presence of tungsten or molybdenum renders
the tools weaker and more brittle in the body of the tool. The
cutting edge of the tool is also more brittle than the edge of the
old carbon steel tool at usual shop temperatures. The Taylor-
White tools, while more brittle at the usual temperatures of the
air from 50^ to 100^ Fahr., have the peculiar property of re-
maining about as hard as they ever were when heated by the
friction of the chip which they are cutting up to the extraor-
dinary temperature of 1000^ to 1200^ ; while tools not so treated
soften and crumble away when heated to 400^ upwards.

iff. Smith. — When I get through I want Mr. Taylor to tell
you what I was trying to tell you, namely, that we need more
strength to our machines as well as more speed. One reason is
l)erhaps that the higher speeds cause more vibration in the thin.

106 WHAT ARE THE NEW 3IACHINB TOOLS TO BE?

fiddle-Kke castings generally used which are attuned only too
well to the new rapidity of motion. At any rate, we all know
that we want very much stronger tools. It may be remembered
that some years ago I told this Society that the proper way to
design a lathe bed was to build it all up in a solid chunk and
then modify it a little by putting a slit through the middle to
let the chips fall through. .1 believe I said also that if lathes
were from three to four times heavier than they are now there
would be a great deal more work turned out in them. If we
would make our lathe heads, too, in great masses of solid ma-
terial it would be all the better. Nobody has yet been brave
enough to make a really heavy lathe, but in my opinion there is
going to be a tremendous revolution in this respect in the next
ten years, in all of our machine-tools. None have had the cour-
age to go at it yet, but it is going to come — ^but gradually, like
all great developments. Another thing we are going to do is to
use milling machines very much more than we do now in the
place of planers. This is because of the great defect in all
planers of moving the heavy weight of the table plus the work.
Thus we must fight inertia in both stopping and starting. An-
other defect of planers as now built is cutting in one direction
only. All this is going to bring us to contrive new forms of
milling machines. Planers will remain, of course, but they will
bo modified, in many cases by making the tool move rather than
the work, thus following the general principle used in shapers
more than we do at present. How all this will develop we cannot
see just now, but it is bound to come. We cannot use a high-speed
tool at high speed on an ordinary planer, and for that reason,
and for other reasons that I might mention, the present machine
is likely to become somewhat obsolete in the near future.

Prof. John E. Sweet,* — ^I am sorry that more time was not
available for discussion. As to the worm gear's not proving
satisfactory for elevators I was not conscious of the fact. If
I were in the business I would not abandon them until I had
skewed around a worm in a spur gear and tried that.

* Author's closoie under tlie rules.

Am ^OlOttS AND AI& HAMllE^. 107

AIR MOTORS AND AIR HAMMERS.
APPARATUS AND METHODS FOR TESTING.

BT MAX H. WICKH0B8T, AUROIIA, IIX.

(Janlor Member of the Society.)

1. The apparatus and methods described below are those used
in some extensive tests of air-drill motors and air hammers made
by the Chicago, Burlington and Quincj Eailroad Co. in ita labora-
tory at Aurora, HL The tests were made for the purpose of deter-
mining the air consumption, horse-power, stalling load of motors,
number and force of hammer blows, etc.

Oeneral Arrangement.

2. The general arrangement of the apparatus is shown in Fig.
18 and in photograph, Fig. 19.

The air used for making the tests was obtained from the shop
supply, which was generally about 60 or 70 pounds. As we de-
sired pressures varying from 60 to 120 pounds, we stepped up
the pressures by pumping the air through a 9-inch Westinghouse
pump which had the air cylinder bushed to a diameter of 7 inches.
This bushing was necessary as the steam pressure at the Labora-
tory is only about 60 pounds. The air was then pumped into
reservoir No. 1, with ordinarily a pressure of 140 or 150^ pounds.
From here the air was allowed to flow into reservoir No. 2, where
it was maintained at any pressure desired by means of a reducing
valve, which was a regular 1-inch Westinghouse air-pump gover-
nor. An oil eup containing a thermometer was screwed into
reservoir No. 2, and was used to determine the temperature of
the air supplied to the tool. Another thermometer was also used
to note room temperature.

♦ Pieeented at tlie New York meeting, December, 1903, of the Americmi
Soeietj of Mechanical Engineers, and forming part of Volume XXV. of the
TrafMaetUm$.

108

Am KOtORd AND AIB fiAMMStUS.

UH^M,

©a=

^inn w»|l»iqUK)

2 i
I I

— 11_^ 4«aiai«q jo

M^OTBOOUailX

•AMA

H ""y'^

AIB MOTORS AKD AIB HAMMERS.

109

3. From the reservoir No. 2 the air passed through a meter.
This is a high-pressure meter made by the Equitable Meter Co.,
Pittsburg, Pa., and in construction is similar to an ordinary gas
meter, the air alternately filling out and exhausting from leather
bellows. From here the air was delivered to the tool to be tested,
through a 2-inch pipe and 1-J-inch hose. At the point where the
air was delivered to the tool we had an expansion consisting of a

Fro. 19.

2-inch Tee, with an internal diameter of about 2^ inches, in which
we determined the pressure by means of a gauge. Care was taken
that between the point where the pressure readings were taken
and the tool to be tested, there was no contraction in the supply
pipe smaller than the openinsr into the tool under test.

4. For the purpose of calibrating the air meter we used the
tank as shown in Fig. 18. This tank had a gauge glass its full
length, and its cubic capacity was determined for each 5 inches on
the glass by weighing the water.

51 The records of time, revolutions of motor, air consumed and
strokes of hammer were obtained autographically, using the record

110 AIR MOTORS AND AIR HAMMERS.

table shown in fig. IS. The records were obtained on glazed
nianilla paper 14 inches wide, moving across the table under
electro-magnetic pens. The driving mechanism of the paper con-
sisted of an air motor and suitable gearing. The electro-magnets
actuated stylographic pens feeding red ink. One of the pens
was actuated by a clock, making a contact every five seconds.
Another was actuated by the air meter, a wiper making contact
with the teeth of one of the gear wheels in the recording mechan-
ism and each contact representing about ^ cubic foot of air.
The third pen was used to record the revolutions of the motors
by arranging a wiper and a simple gearing, so as to make a contact
every 5 revolutions of the socket for holding the drill.

6. In calibrating the meter the method was to have the calibra-
tion tank about full of water, the valve from the meter opened up,
thus allowing full pressure; the outlet valve was opened, the
record paper started going and as the water in the gauge glass
passed the marks 5 inches apart record was made by the observer
pressing a push button. The meter at the same time made its
own record, and thus we were able to figure out the number of
cubic feet per contact or per notch. We also obtained record of
5-second intervals. As the readings of the meter varied some-
what with different rates of flow, calibrations were made at dif-
ferent rates by varying the opening at the outlet valve. A cali-
bration curve was then made by plotting notches per minute as
abscissa? and cubic feet per notch as ordinates. A number of cal-
bration tests were niade during the course of the tests of the tools.

The various gauges used from which pressure readings were
taken were previously checked up and adjusted by means of a
Crosby Dead Weight Tester.

Motor Tests and Calculations.

7. The arrangement used for testing motors is also shown in
Fig. 18 and in the photograph, Fig. 20. The arrangement was to
apply the load by means of a Prony Friction brake, the revolu-
tions being recorded by means of a wiper making contact every
five revolutions. The air consumption and time were recorded as
described above.

8. The procedure was to first regulate the pressure in reservoir
No. 2 at 60 pounds, then put on a light brake load, keeping this
constant during the test with full open throttle. Another test

AIB MOTORS AND AIR HAMMERS.

Ill

was then made with heavier brake load with the same air pressure,
and the increments of load continued in successive tests till the
tool was stalled. Then air pressures of 80, 100 and 120 pounds
Mere used in the same manner. A sample of a motor record
reduced is shown in Fig. 21.

9. Table No. 1 shows a sample data sheet, and the various items
and calculations were obtained as follows:

Fig 20.

No. 1 and No. 2 are pressures as read by an observer on gauges
and are pressures above atmospheric pressure.

No. 3 is temperature in degrees Fahrenheit of the compressed
air in reservoir No. 2.

No. 4 is the meter contacts or notches obtained from the auto-
graphic record included in a strip of record covering one minute
as recorded by the clock.

No. 6 is the cubic feet of compressed air per meter notch as
obtained from the calibration chart.

112 AIB K0T0B6 AND AIB HAMKEBS.

No. 6 ia the cubic feet of compressed air per minute obtained
by iniiltiplyuig Nos. 4 and 5.

EMh tp^ctm S rerolvtioiii of motor

LaO.Mo.»1.84ll46

Burttngton R<mU Laboratory

Aurora, lit.

TEST OF AIR MOTORS

1003.

Fac'9lmtle Motor Rocord

{about i 9ixe)
1 minute.

No. 187 Test No. 72 B

N.P.M. 19.4 R.P.M. US

Aug.l7tta. 1908.

'^"^^Eiirfrnoti-MiTcBbtefc^ ^ ^

^ -H Each 1^-6 Seconds '

Wlckh^t.M.U. *-.•*«. Hvu CO..,

Fio. 21.

No. 7 is the number of cubic feet of free air per minute ob-
tained from No. 6 by the following formula :

where J^ A = free air in cubic feet per minute.

C A = compressed air in cubic feet per minute.
P = pounds gauge pressure at tool.

No. 8 is the revolutions per minute of socket for holding the
drill and is obtained from the autographic record.

No. 9 is the brake load as shown by the weight on the scale at
the end of the lever arm, usually three feet, except with the
smaller motors, where the lever arm was two feet. This brake
load was pre-determined and kept constant by an observer during
each test.

No. 10 is the brake horse-power, and was calculated as per fol-
lowing formula:

AIB MOTORS AND AIB HAKMERS.

BP M x2t X 3.1416 xw

113

BHP^

33000

-where B H P = brake horse-power.

B P M^ revolutions per minute of brake-wheel.
r = radius in feet of brake-lever.
w = weight in pounds on scale.

No. 11 is the load on scale which was just sufficient to stall the
tool

Xo. 12 is the stalling load at one foot radius calculated from
No. 11.

No. 13 is the cubic feet of free air consumed per minute per
horse-power, obtained by dividing No. 7 by No. 10.

10. After obtaining these various data we plotted three curve

Fig. 22.

114

AIR MOTORS AND AIR HAMMERS.

TABLE I.

BURLINGTON ROUTB LABORATORY,
AUBOBA, III.

TEST OP AIR MOTORS,

1908.

Dau Sheet

Lab. No. 681.54 M 47.

Air Motor Test. Testei.

Makbb : Air Tool Co. Name of Tool : Mendota. 8u» : 8. Ttpb : 8 Cyl.

Wkioht : 47 Lbs. Serial Number : 88. Cost :

Remarks:

Brake Lever Arm : 8 ft.

Date : 8/4-03. Temperature Atmosphere : 87 deg.

i ^

C 1 D

1. Press, reservoir

101
100
101

101
100
102

101
100
108

121 121

121

121
120
112

121

2 " at tool

120
106

120
108

120
110

120

3. Tempr. coDip. air

114

4. Notches per min

24.

21.2

19.5

17.6

28.4 24.

20.

20.8

17.5

6. Cu. ft. per notch

.202

.264

.265

.266

.268

.262

.264

.266

6. Cu. ft. C. A. mIn

6.8

5.6

5.2

4.7

6.2

6.8

5.8

6.4

4.6

7. Preeair

52.

42.8

40.

86.

56. 61.8

48.

47.7

41.5

8 Rev. oer mln

138
20

119
25

89
80

72
85

135
25

115
80

91
85

88
40

9. Wt. on scale

10. Brake U.-P

1.67

1.7

1.52

1.45

1.92

l.r

1.82

2.02

11 Stalliiisr weiffht

12 *' 11 "at one ft. radius

111

24.8

144

18. •* Seven " per H.-P

25 26

29 1 81 26.5

1 1

28.5

20.7

(Signed) ..

ObMTver.

AIB MOTOBS AND AIB HAKHEBS. 115

sheets to shoW up the results by the graphic method, representing
in each case the air pressure as abscissse. On one we plotted as
ordinates the stalling load at one foot radius, on another the air
consumption in cubic feet of free air per minute per horse-power
at Tpfl Trimiim horsc-power, and on the other the maximum horse-
power.

Air Hammer Teats and Calculations.

11. The arrangements for testing air hammers is also shown in
Fig. 18 and photograph, Fig. 22. The method in general was to let

— 1 notch" .388 cnbSe feet comp<-aIr—

Lab.No.e21.64 H 11
RIVETINO HAMUEB Burlington Route Laboratory

No. 225 Test No.-525 B. Aurora, III.

V.FJL 28 S.P.1L 906 TEST OF AIR HAMMERS

Aug.l9th. 1908. 1003,

Fac-afmlle Hammer Record
6 aeconde,

lt0ponS*pt6th.19O3,
Paper movM io this direction *-

I o Each "ware" ii one itroke
fj o Vertical height of each wave is the dlstanoe
I -S weiffht U lifted multiplied by 8.

Fig. 28.

the hammer strike upward against a known weight adjusted to the
size of the hammer and to autographically record the distance the
weight was lifted. The weights varied from about 40 to 15C
pounds, and the vertical lift was multiplied 8 times on the record.
The time and air consumption were recorded as above described,
and a sample of one of the records obtained is shown in Fig. 23.

12. The results of test were recorded on a blank, copy of which
is shown in Table 11.

No. 1 and No. 2 are gauge-pressure readings as noted by an
observer.

116

AIB MOTOBS AND AIB HAMMEBS.

Makbb : Air Tool Co.
Wbight : 23 Ibi.
Wt. Pldmgbb : 1 lb.

Bbmabkb :

TABLE II.

Air Hammkb Test.
NambopTool: Valcan.

SXBIAL NUMBBB : 8171.

DiAM. Ctlimdbr : 1^'^

Lab. No. 631.54 Hli
Test 501.

SiZB : 8.
Co§t:
Stroke Inches : &

Datb : 7/22-08.

Temperatcue Atmosphere : 87 deg.

1. PresB. reaervolr , . .

2. " attool

8. Temp. comp. air...
4. Notches per min. . .
6. Cu. ft. per notch . .

6. Cu. ft. C. A. min..

7. Free air

8. Strokes per min. . .

9. Weight

10. Distance raised. ...

1 1. Ft. lbs. per blow . . ,

12. Horse-power

18. ** Seven" per H.-P

.271

.269

7.85

6.45

39.2

40.8

834

892

120

.004'

.0052'

.48

.624

.0123

.01C8

3,200

2,400

100

100
90
25
.27

6.75
51.6
964
120

.006'

.72

.021

2,400

n I

120
120

1 1

' 1 i

.268

1 '

5.63
60.6
1,000

1 i

1 ■ 1

1 1

.0065'

.7818
.0236
2,100

1 1

' , 1

1 1 1

i 1 1

(Signed)

Obset-ver.

BUBLIN6T0N BOUTE LABOBATOBY
Aurora, 111.

TEST OF AIR HAMMERS,

1903.

DaU Sheet.

AlB K0T0B6 AND AllSL HA^M^ilS. 117

Ko, 3 is tlie temperature in degrees Fahrenheit of the com-
pressed air in reservoir No. 2 as noted by an obsenrer.

No. 4 is the meter contacts or notches per minute as obtained
from the autographic record.

No. 5 is the cubic feet of compressed air per notch as obtained
from the calibration chart

No. 6 is the cubic feet of compressed air per minute obtaiaed
by multiplying Nos. 4 and 6.

No. 7 is the cubic feet of free air per minute obtained by the
same formula as given under air motors.

No. 8 is the number of strokes per minute obtained from the
record by counting the strokes made in one minute.

No. 9 is the pounds of the weight placed over the hammer.
No. 10 is the average distance in feet the weight was raised as
obtained from the record.

No. 11 is the foot pounds of effective work 'per blow, obtained
by multiplying the weight in pounds by the distance in feet it was
raised

No. 12 is the horse-power of the hammeri obtained as per fol-
lowing formula:

TT p _ ft lbs. X blows per min.
^ ^ "■ 33000

No. 18 is the cubic feet of free air per minute per horse-power.

13. After obtaining the results of test, three curves were
plotted from them as f oUows, in each case the pounds gauge pres-
sure being plotted as abscissas and the following as ordinates: foot
pounds per blow, cubic feet of free air per minute per horse-power
and horse-power.

14. In conclusion, the author desires to express his special
thanks and acknowledgments to Mr. H. F. Wardwell, who did the
greater part of the work in making the designs and tests.

DISCUSSION.

Mr. Frank 0. Hohart, — The method here described of arrange
ing the hammer to strike against a weight and basing calculations
upon the displacement of this weight would seem useful only for
obtaining comparisons between different hammers.

The facsimile diagram shows a movement of the pencil of about

118

Atft ICOtOHS AND A1& HAMM£!fid.

AIR KOTOBS AND AIR TTAicMiei^fl 119

i inch, corresponding to a movement of the weight of not more
than /f inch. With this small movement, together with the long
lever reductions between the weight and the pencil, the high
si)ee<l and the various questions about impact, weight, friction,
etc., I should not expect the figures based on the diagram could
be accurate.

Fig. 24 shows two diagrams made on one sheet of paper by
1^x6 inch riveting hammers on a machine which we are using
for air-hammer testing. It sho\^s the number of blows per min-
ute, length of stroke of the pistons, velocity of pistons at every
portion of the stroke and the position of the pistons at every in-
stant. Knowing the weight of the piston, its velocity at impact
and the number of blows per minute, the ability of the hanmier
to do work can be calculated. The effect of changes of port areas
or other details of the hammers is very clearly shown by such
diagrams.

The machine consists of a drum driven by a small air motor at
a periphery speed of about 2,000 feet per minute. A centrif-
ugal governor holds this speed very nearly uniform, and a paper
wound on this drum receives the diagram. The hammer is
mounted with its length parallel to the axis of the drimi, and in
such position that a slender steel rod connects the piston to a
pencil slide which marks the position of the piston on the diagram.
All of the moving parts are very delicate and do not interfere, so
far as can be detected, with the action of the piston. The ma-
chine was built for experimental work and for testing hammers
by Fairbanks, Morse & Co. for use at their works at Beloit, Wis.

Mr. M. H. Wickhorst* — Commenting on Mr. Hobart's re-
marks, I would say the tests were made for the purpose of com-
paring the various hanuners on the market, and I wish to correct
his impression concerning the distance the weight was lifted in
the tests. The sample record as printed is only a little over one-
fifth size. The method of studying a hammer which he describes,
I should think, would be decidedly valuable to a designer, and
am very glad that he has presented to the Society an outline of
the apparatus and the sample record.

* Aathor's Closure under the Boles.

120 BATES AND PRICES FOB ELECTRIC FOWEB.

No. 1016.*

A METHOD FOR DETERMINING RATES AND PRICES FOR
ELECTRIC POWER.

BT FRANK B. PERRT, BOBTON, MASS.

(Janior Member of the Society.)

1. That the future will show a rapid growth in the application
of electricity to industrial pursuits, operated from a centrally
located steam-driven electric power station, has led me to present
the following data to our members, many of whom will doubtless
find the methods suggested applicable to their own interests.
The advent of driving textile and other mills from a plant of
this character has opened a question for discussion as to the
establishment of a proper basis for rates to be charged for electric
current supplied in large quantities.

2. The contracts drawn up by electric companies with their cus-
tomers sometimes contain a clause similar to the following, which
is given for the purpose of illustration, and is not intended to
cover any specific case: For a period of years from date the
lessor agrees to furnish, and the lessee to receive and to pay for,
within the times and on the terms set forth, all the power that
may be required to properly operate and light his plant. The
amoimt of power to be determined by meter readings, and to be
billed monthly at the rates recited below, viz. :

1,800 to not more than 2,160 kilowatts at rate of $31.50 per
kilowatt per annum;

More than 2,160 kilowatts and not exceeding 2,520 kilowatts
at rate of $30.00 per kilowatt per annum;

More than 2,520 kilowatts and not exceeding 2,700 kilowatts
at rate of $28.50 per kilowatt per annum;

* Presented at the New York meeting (December, 1908) of tbe American
Society of Meclianical Engineers, and forming part of Volume XXV. of tbe
Tmniaetion$,

RATES AND PRICES FOR ELECTRIC POWER. 121

All in excess of 2,700 kilowatts at rate of $27.60 per kilowatt
per annum.

3. The above rates are based on an annum of 3,000 hours.
From this schedule the following tables are derived :—

TABLE I.

Rate per Kw. SqniT. Ra*e per H. P. Rate per Kw. Rate per H. P.

perAnnooi. per Annum. per Hoar. per Hour.

$81.50 $38,499 $ .0106 $ .007688

80.00 22.88 .0100 C007460

28.50 21.26 .0005 .007087

27.60 20.5896 .0092 .006882

TABLE n.

Kilowatts Utad. Coat per Annom. Coet per Month.

1,800 $56,700.00 $4,725.00

2.160 68,040.00 5.670.00

2.161 64,830.00 5.402.50
2,268 68.040.00 5.670.00

2.520 75.600.00 6.800.00

2.521 71.848.50 5,987.87
2.652.7 75.601.96 6,800.16

2.700 76.950.00 6.412.50

2.701 74,547.60 6.212.80

Table I. is self-explanatory. An examination of Table n.
shows plainly the faults that exist in a schedule such as that
usually followed.

4. By using 2,161 kilowatts, or one additionalkilo\i'attmorethan
2,160 kilowatts, a yearly saving of $3,210 may be made; similarly
an increase of one kilowatt above 2,520 kilowatts effects a reduc-
tion of $3,751.50 per annum, and one extra kilowatt over 2,700
kilowatts lessens the yearly expense by $2,402.40.

5. While these figures show the points in the schedule at which
the greatest saving may be made, many other quantities which
are given in a later table illustrate equally well the irregularity
of prices based on such a list of rates. The inconsistency of this
method is further shown by the fact that 2,268 kilowatts cost the
same as 2,160 kilowatts, or 108 kilowatts may be utilized without
anv increase in expense to the consumer. The price is also prac-
tically the same for 2,652.7 kilowatts as for 2,520 kilowatts,
which indicates that any number of kilowatts between these limits
may be used without additional cost to the person buying power.

6. It would appear from these figures that the electric company
makes more profit at some places in the schedule than at others,

122

RATES AND PRICES FOR ELECTRIC POWER.

and on that account could furnish certain amounts of power
gratis. Of course, this is not true, but tends to convince one of
the absurdity of the system that is customarily used by central
electric stations. It is also evident that the customer, if he
were so disposed, might watch his meter and so adjust his con-
sumption of current as to bring about a substantial decrease in
his bill for the month. I'or example, he might install a few extra
machines that would serve not only to increase his product but

^Horlfonial Unea^

"Ra

lepi

rKn

. pe

' Anffum

bse

eeai

^Ki

ttt.

rtee

Rat

Dta

...

tUi

e«-

Pride pe

1th

asqs

pn.t

RA\

£8

ind

PRH

£8

for

>ow

£R.

6300

aoj

)

9000

flOV>

»j

saon

27J

)

\AR

■He

. /

sooo

•

K!io

vat

4 18^

4700

f*rrp,F.3.

FiQ, 25.

Awu3mikX^0».^.r.

would also be the means of diminishing his bill. As an extreme
case, the use of one additional kilowatt would bring about this
result. It also might prove economical to bum electric lights
during the day in order to increase the current consumption.

7. Both of these examples mentioned serve as illustrations of
the fact that it is within the consumers jurisdiction, by the terms
of his agreement, to use power advantageously with a resultant
financial loss to the electric company. While such a procedure
may be perfectly legitimate, it would be fairer to all concerned
to prevent contingencies of this kind from arising, and at the
same time to provide a system of rates which would be to the
mutual interests of both parties.

8. The horizontal lines on Chart No. 1, Fig. 25, indicate the rates

RATES AND PRICES FOR ELEOTRIO POWER. 123

for power, and the diagonal lines represent the price per month for
electric current covering the entire range of the above mentioned
schedule. The latter lines show at a glance the unfairness of the
system, and prove conclusively that rates cannot be adjusted
properly on the customary basis.

9. The subsequent discussion is intended to give a method for
rearranging a schedule of rates based on the " step-system,'' and
for convenience the aforesaid figures will be used. The principles,
however, may be applied to any other rates made up in a similar
manner.

10. Keferring to Table 11. and applying the well-known formula

(a + /) ^, giving the sum of the terms of an arithmetical pro-

^ression we obtain for each step, assuming a conunon difference
of one kilowatt, the following figures:

(1) 2,161 - 8,520 kw.

(2) 2.521-2.700 " .

Vftlaeof "0."

Value of'/."

Value of "n.

|64,S30.00
71,84S.60

$76,600
76.950

860
180

11. The value of " a,'' or the first term of the series represents
the product of 2,161 kilowatts by its rate per kilowatt per annum.
In a similar way the other terms of the series are obtained, " 1,^'
indicating in each case the last term and " n " the nimiber of
terms.

12. Since the rate is constant from 1,800 to 2,160 kilowatts, we
will compute simply the sum of the series beginning at 2,160
kilowatts.

Sam of (1) = (64,880 + 75.600) 180 = $25,277,400

'• *' (2) =(71,848.5 + 76.950) 90= 18.391.865

2,160 kw. at $81.50 = 68,040

(8) Total sum of two series = $88,737,305

13. With a varying rate of power, diminishing in a fixed ratio
between any two amounts, it is possible, by the substitution of
proper quantities in an equation herein, deduced by the writer,
to compute the value of the sums of the product of each suc-
eeeding kilawratt by its corresponding rate.

14. The horizontal spaces of the charts, or abscissse, represent in
tenns of kilowatts, a progressive increase in each successive
number by the addition of any desired equal amount

124 RATES AND PRICES FOR ELECTRIC POWER.

Let (a) = 1st term. (Z) = last term, {d) = common difference
(n) = number of terms, (s) = sum of terms.
Then 1st term = «,

((

= a + ^,

= a + 2rf,

4th

-a^-Zd,

= a -\- {n —

= (a + 0|.

15. The vertical spaces of charts, or ordinates, represent in terms
of rates per kilowatt annum, or per kilowatt hour, a progressive
decrease in each successive number by the subtraction of any
chosen equal quantity.

Let (J) = 1st term, {k) = last term, {r) = common difference.
(/) = sum of terms, (m) = number of terms.
Then 1st term = ft,

2d " = ft - r,
3d '* = ft - 2r,
4th '' = ft - 3r,

k = b — {m — l)r,

< = (j + i)|.

16. Multiplying the corresponding terms of these two series, we
obtain the following: —

(4) Product of Ist term = aft,

(5) " "2d " ={a + d){h- r),

(6) " " 3d " = (a + 2d) (ft - 2r),

(7) " " 4th " = (a + Sd) (ft - 3r),

(8) " " last " = ja + (n - l)rfj jft - (m - l)r{.

17. Adding these products, and developing the quantities en-
closed in parenthesis, we obtain for the sum of 1st and 2d terms
designated (4) and (5).

{A) Sum of products 1st and 2d terms = 2eift ■{■ hd — or — dr.

„ ,^. , . a u 2 4<ift 4- 2ft</ - 2ar - 2<7r
Multiply mg equation by ^ = -.

Substituting values n = 2 and m = 2 in the abwe, we find
, .. 2aftn + hdn — arm — dnn

U) = o

= *^{2a + (n-l>7}-^^(a+^.

BATES AND PRICES FOB ELECTBIC POWER. 126

18. In a similar manner (S) = sum of (4), (5) and (6),
and {C) = " « (4), (5), (6) and (7),

(5) =1^ j 2« + {n-l)d]-^{2a+8id),
(C) = ^{2a + (n-l>f}-^(3a + 7rf).

19. Comparing equations (4), (B) and (C), we find that they are
alike in all respects excepting the coefficients of " a *' and " d "
of the negative terms.

ao. Substituting these values, viz., for coefficient of " a " (m — 1),

and for coefficient of "rf" | 5 (m — 2) + 1 [ j m — 1 [ , we have

for the sum of any number of products containing "n" and "m"
terms

Total sum = ~ j 2a + (n - l)rf }

21. In order to apply this formula to the case in question, let us
consider the rate as beginning at $31.50 per kilowatt per annum
for 2,160 kilowatts and varying in amount to $27.60 per kilowatt
per annimi for 2,700 kilowatts. By so doing, we preserve the
limiting features of the schedule both for rates and for the quan-
tities at which they are applicable. Since the successive steps
in the first mentioned schedule show an even decrease in rate
for equal increments of power up to 2,520 kilowatts, it is safe to
assume that a list of rates may be made up to vary imiformly
from 2,160 to 2,520 kilowatts and regularly, but at a different
rate, from 2,520 to 2,700 kilowatts.

22. I-et X ~ rate per kilowatt per annum for 2,520 kilowatts.

02 = 2,521
Jj = a?
d2 = l
^2 = 2,700
^ _ (a^-27.6)
^^"" 180
W2 = 180
^=180

«I =

= 2,160

J,=

= 31.50

rf,=

= 1

h--

= 2,.520

T. =

(31.5 -

- 360

«i =

= 361

TOj:

= 361

126 RATES AND PRICES FOR ELECTRIC POWER.

23. Substituting these respective values in equation (D) gives

D, = «^-^ ^ 3^^ (2,160 + 2,520) - (3t-5 - a') x 361 x 360)
2 . 2 X 360

|2,160 + i (722 - 1)1 .

A = ^^ (2,521 + 2,700) -<^- ^^f ^if >- 1^^

12,521 + i (360 - 1)}
From the above,

^ _ 159,655,860 - 81,886,171.5 + 2,599,561a;
^1-^ 6 •

^ _ 2,819,340aj - l,418,038aj + 39,137,848.8
^a- ^ .

24. Then in order to fulfill the conditions of the first named
schedule, J9j + D2 must equal (3) or

Di + 1)2 = 38,737,305.
Substituting values of />i and Dz and solving,

4,000,863a! = 115,516,292.7

X = 28.872.

25. This value of " x " substituted in expressions for rates gives
Ti = $.0073 and ra = $.00706§ as the variations in charges for
each kilowatt per annum from 2,160 to 2,520 kilowatts, and from
2,520 to 2,700 kilowatts respectively.

26. The rates per kilowatt hour become n = $.0000024^, r, =
$.00000235; for 30 kilowatts hours, n = $.000073 and rj =
$.0000706§.

27. Using this system as a basis for rates, the wording of the
agreement for power would be changed to read as follows :

" The charge for electric current furnished under this contract
shall be made as per the chart attached hereto, and made a part
hereof.''

28. The line plotted on Chart No. 2, Fig. 26, indicates the rates
per kilowatt per annum which would, in this instance, replace the
original schedule as given by Chart No. 1, Fig. 25.

RATES Ain> PEIOES FOB ELEOTRIO POWER.

127

Ord

nat^

,^.

fate

per

Kw.

— 1

nifu

Abaci 88a

^=/f

lowatta

Rat

reo

r-Mn

P0\

1
fER

~ f i

flfl i

00 Ktlow

itt«

•«J

TTJl

/>l^

'Mo

Kile

vat\

a Ifl

2TO0

•

Afti

f,r^

^•i.Aa«*M«to>.,/r.r.

Fio. 26.

Una

.0.^

1 1

Line No.2

OUM

Ordin

amu^Kllomatt

Mm,M

PWOf

oura

^700

1ATE

\am

P«(

»r P<[Mr«

6190

640a

tOto

9760

V nil

00 n

M»dtt*

6100

0101

rf^

6060

O100

_j^

6000

soso

sooo

SR80

oom

^/^

5R00

(vm

57S0

OOM

M^:

— *s

5i00

0Q9i

N..^

fiOBO

0099

S600

Mm.

ri».

KmM

m. L.

-L- 1

tr,

Kw.Hn.

t» w

S9U

L-S

to m

J^rrw,r.3,

Fig. 27.

wi«.#«^^iMfC%^»^.r.

128

RATES AND PBICES FOB ELECTRIC POWER.

29. The results obtained by following these lines of rates neces-
sarily gives, on account of the above reasoning, the same average
price per month as that computed by the " step " system. Chart
No. 3, Fig. 27, is an example of the form of chart which would be
embodied in a contract for electric power in order to meet the
rates outlined in the original agreement. It also gives an idea
of the simplicity of interpreting rates and prices for power by
making use of the scheme proposed.

30. Line No. 1 is drawn through points plotted in accordance

Urn

Ho.

5180

Onil
Ab9e

'Urn

#01

*a(i -#.(

writ

>^r.

!lt40

fCfS

for

poJ

!qop

oooo

pto
te4i

990
^600a

Kihmatta
KihmaHt

'r».

9080

9090

40B0

4M0

4900

4800

4890

ARl

4TBp

4740

4W0

•0

Km,

4fl

»-.

Hrr9,F,3,

jlauBamhJhttO».^r'

Fig. 28.

with the rates specified on Chart No. 2, Fig. 26 ; the ordinates being
equal to rates per kilowatt hour and the abscissse representing kilo-
watts. From 1,800 kilowatts to 2,160 kilowatts and above 2,700
kilowatts the rate is constant at $.0105 and $.0092 per kilowatt
hour as respectively indicated by the horizontal portions of the
line. From 2,1C0 to 2,520 kilowatts the rate diminishes in a
fixed ratio for each additional kilowatt used, namely, from
$.0105 to $.010062 per kilowatt hour; also from 2,520 to 2,700
kilowatts the rate diminishes imiformly, but not so rapidly as
between the points previously stated, from $.010062 to $.0092
per kilowatt hour. It will be noted that the ordinates may be

RATES AND FRIGES FOR ELEOTRIC POWER.

129

read easily tojTf'inrof a cent per kilowatt hour and the abscisssB to
3 kilowatts without estimating fractional division of the spaces.

31. line No. 2 is deduced from the foregoing^ and by its means
one may determine at a glance, knowing the wattmeter reading,
the amount of each month's bill for power consumed.

32. The lower line of figures on the chart represent meter read-
ings in kilowatt hours divided by one hundred. These quantities
are computed by multiplying the kilowatt readings by 250, the
average hours per month, or ^ the total for the year as limited by

Ho.

aroo

Of4\

2mm

Mpt

I If J

Km.H

100

fate

=#.0

;m,

^,Hf,

5000

PUK

E8fi

rPO

fTfff

5090

wot

iOto

[2ii

pKU

ioKii

matt

Hrt.

SflBO

5540

aaoo

5400

5I9D

ano

5840

580O

,/^

1AR

T Mi

5900

iQ90

5tW

Km.

^m.

KmM

**--

ir«

Hn,

} ^

Fig. 29.

Am B mt k iht$Ck.Ji:r.

the contract. Two ciphers are omitted from these results to avoid
rei)etition of figures. To obtain the bill per month, find the
number on chart corresponding to that recorded by the watt-
meter, and note the point vertically over same on line No. 2.
In the column at the right-hand side of the chart, horizontally
opposite this intersection, will be found the price of bill for
the amount of power in question. The abscissae may be read
accurately to 750 kilowatt hours and the ordinates to $5 for the
corresponding spaces, and by estimation of tenths these differences
may be proportionately reduced.

33. Charts No. 4 to No. 8 (Figs. 28 to 32) inclusive are similar
to, and may be considered as supplementary to chart No. 3. Each

130

RATES AND PRICES FOR ELECTRIC POWER.

chart includes a section of 180 kilowatts or ^ of the entire range
from 1,800 to 2,700 kilowatts.

34. lines No. 3 and No. 4 indicate prices for power up to and
including 2,160 kilowatts at a constant rate of $31.50 per kilo-

Rati

Un«

No.

I/J^o.

::d

OrdI
AUo

tat-

t$9a*

-Itai

Km.tr.

Ortf/iUtw

W|M IfoA

9B0O

^AUiam Pm

?fsj

>rPlfWEt

5880

Sfc^

i400t

160io23i

OO/f/

omatl

•

ssm

5840

9820

980O

5780

S7S0

5740

.010

srao

iAR

rNo

970O

5080

.010

5000

Km.

5»

Sfl

s«

4oH

.Hn.
100

JUm.SmdtlhUOk^.r.

Fig. 80.

watt per annum, or $.0105 per kilowatt hour. lines No. 5, No. 7
and No. 9 represent rates

varying respectively from $31.50 to $30,186 per kw. per annum.
" " 30.186 " 28.872 " " " "

a u 28.872 " 27.60 " " " "

from

$.0105 to $.010062 per " " hour.
.010062 " .009624 " " " "

.009624 " .0092 " " " "

35. lines No. 6, No. 8 and No. 10 show prices per month for
power, varying in the order given from 2,160 to 2,340 kilowatts,
from 2,340 to 2,520 kilowatts and from 2,520 to 2,700 kilowatts.
Each of the spaces on the charts No. 4 to No. 8 inclusive occupied
by the abscissae represent 1 kilowatt or 250 kilowatt hours as the
case may be, consequently are well within the limit that it is pos-
sible to read a recording wattmeter of 3,000 kilowatts capacity.

RATES AND PRICES FOR ELECTRIC POWER.

131

36. Greater accuracy is also attainable in determination of rates
per kilowatt hour or the total bill for the month. This may be
carried to a still further degree of refinement by an increased sub-
division of the amounts of power and their accompanying rates.
The principles enumerated may be applied equally well to a
pven list of powers with any predetermined rates. The Table
nL i& presented to show the comparison of prices per month that
would be paid for various amounts of power by the original agree-
ment, and also by the lines illustrated on charts No. 3 to No. 8
inclusive (Fig. 27 to 32).

TABLE III.

Kw. Hre.

MoHTHLT Bill-

KWB.

per month
of

Katepor
Kw. Hr.

Dil

•

H.P.

250 Era.

By CUrt.

By Schedule.

2.894.1

2,160

540.000

$.0105

$5,670.00

2.W5.7

2.190

547,500

.010437

5.708.78

5.475.00

—

$288.78

2.J77.2

2.230

555,000

.010854

5,746.47

5.550.00

—

196.47

3,016.1

2,250

562,500

.010281

5,788.06

5,625.00

—

158.06

8.0M.8

2,280

570.000

.010208

5,818.56

5,700.00

—

118.56

8,096.5

2.810

577.500

.010185

5.852.96

5.775.00

77.96

8,186.7

2,840

585,000

.010062

8.886.27

5.850.00

—

86.27

8,176.9

2.870

592.500

.009989

5,918.48

5,925.00

6.52

8,217.2

2,400

600.000

.009916

5,949.60

6,000.00

50.40

8.257.4

2.480

607,500

.009843

5,979.62

6.075.00

-h

95.38

8.897.6

2,460

615.000

.00977

6.008.56

6,150.00

141.45

8.837 8

2.490

622.500

.009697

6,086.88

6,225.00

188.62

8.878.0

2,520

680.000

.009624

6.068.12

6.800.00

286.88

8,418.2

2,550

687.500

.00955831

6.090.25

6.056.25

—

84.00

8.458.4

2,580

645.000

.00948261

6.116.82

6.127.50

11.18

8.498.6

2.610

652.500

.009412

6.141.88

6,198.75

57.42

3,538.8

2.640

660.000

.0098413)

6.165.28

6.270 00

104.72

8,579.1

2,670

667.500

.00927061

6.188.17

6.841.25

158.08

8.619.8

2,700

675,000
Total....

.0092

6,210.00
1118.888.20

6,412.50

-h

202.50

$113,726.25

—

$893.05

37. Referring to the tabulation, one may observe that between
2,190 and 2,370 kilowatts, the chart gives figures higher than
the schedule, while from 2,370 to 2,520 kilowatts the monthly
bills are lower. As previously stated above,' the average is prac-
tically the same. The last line of the table shows that for the
points taken the discrepancy between the chart and the schedule
amounts to about ^^ of 1 per cent. This difference would be
le^ened by choosing smaller sub-divisions and would finally
amount to zero when the points are taken at 1 kilowatt intervals.

182

BATES AND PBIOES FOB ElfCTBIC POWER.

Rati

I/J/Vo.

UninoA

^rht

.010

-Hal
'Klh

"•

On/ZiUte.

II
^6080

Mresam

p«

J£8 for pqWEI

9000

OKii
70 Kt

6D40

.000

,<^

8020

6000

9BB0

.OOR

8960

5040

9030

.000

900O

8880

.000

Km.

2cm

2ft0

2S20

Km.

Km.h
lift

—88

«1

•1

ooo

«»)

..^

Arry.JLB.

.4ai.JtoaJMiCk^r.

Fig. 81.

KaU

i;:^.

I/n«Afo.

-1

Vfo«

Ordtl

a<M

-ir/lc

NMtti

ir«j

OrtffiofMJ

<*
»

f4r£

raiN

/»«*

ffSj

orPiwa

9195

»00^

920(
Oto

o27(i

9180

.oon

9195

9190

9U5

.000

^i^

9120

_M*

9105

9000

.0069

9025

'4^3

9090

XS»

Km.

2fl

Km.

Km.Hr9,

«a

98^

9fl

. «

10^ 6J

Arr»

,r.B.

4lt^

MftJI

mou

JLT,

Fie. ad.

RATES AND PBIOES FOR ELECTRIC POWER. 133

Of course, if the meter should indicate a use — say of 2,520 kilo-
watts per month considerably more would be paid to the central
station by the chart system than fey the schedule. It is equally
true that there are points on the chart lines where, if the meter
should read the proper amount, the bill would be less than by
the schedule.

38. The convenience of this method in determining the value of
bills rendered each month should be appreciated both by the
electric company and by the user of current. There is no con-
flicting of rates or prices with such a system, and bills increase
gradually and consistently, as they ought, in proportion to the
amount of power used. This method, which is equally fair to
producer and to consumer has, to the best of my knowledge,
not heen used up to the present time by any Central Electric
Station for finding costs of current where varying rates per kilo-
watt annum or per kilowatt hour are involved.

39. These ideas are submitted with the belief that their adoption
will bring about more satisfactory results than can ever be realized
by a continuance of the soK^alled " Step '* system of rates.

DISCUSSION.

Mr, R, S. Hale. — Mr. Perry's illustration is nothing more or
less than the usual scheme of discount based upon quantity, al-
though it is stated in figures per kilowatts. That is, he might
set a price of $31 per kilowatt up to $60,000 per year and there
would be no discoimt; above that there would be 5 per cent, dis-
count, and so on. The bad feature of this plan is that it permits
the use of a little more current, in some cases, actually to reduce
the bill, which is taken care of in most electric companies by pro-
viding that the customer need in no case pay more than if he had
actually used the amount necessary to obtain the higher discount.
This, of course, is equivalent to making no charge for some of
the current used. In other cases — ^as, for instance, the Chicago
Edison Company — ^the contract provides, " Intermediate discounts
to be determined by interpolation '' ; and this gives, I understand,
no trouble in practice. It is of course practically just the method
that Mr. Perry suggests. The best way, however, is what is
known as the block system : By selling the first 2,000 kilowatts at
say $41 each; the next 700 kilowatts at $29 each, making, how-

134 RATES AND PRICES FOR ELECTRIC POWER.

ever, no reduction on the first 2,000; the next 200 might be sold
at $28 each, and so on. The proper average rate can be obtained
in this way.

The most important feature, however, in determining rates is
entirely omitted by Mr. Perry, and depends on the difference be-
tween kilowatts and kilowatts hours. Thus 2,000 kilowatts used
for 3,000 hours is six million kilowatts hours, and a price of $30
per kilowatt is one cent per kilowatt hour. If now the amount
should be determined by a kilowatt hour meter, the customer
might use the six million kilowatt hours during only 1,500 hours
of the year, using 4,000 kilowatts during that time; the kilowatt
hour meter would still show six million kilowatt hours as before,
but the central station would have to have twice the investment in
engines, dynamos, boilers and so forth, that it would have to have
if the current was used during 3,000 hours, and the central sta-
tion would lose money. If, on the other hand, the customer used
the six million kilowatt hours during 20 hours per day, or 6,000
hours per year, using day and night service, the central station
would require only 1,000 kilowatts of machinery, only half the
investment and outlay, and would make a much greater profit.

There are a great many systems in use by central stations for
taking care not only of the feature that makes the argument for
Mr. Perry's paper, but other important features that must be con-
sidered when a central station makes its rates. At Niagara Falls
the published rates were $1 per kilowatt, based on the maximum
taken, or the minimum power, that the customer used at any one
time, corresponding to the amount of machinery the company
had to keep ready for that customer, and, in addition, 2 cents per
kilowatt hour for the current used up to 1,000 kilowatt hours, 1^
cent per hour for the next 1,000 kilowatts hours, 1^ of a cent
for the next 1,000 kilowatts hours, and so on. These illustrate
only one of the many methods that have been used in determining
the charges for electric light and power.

Prof. W. W. Crosby* — At Wobum, Mass., a system was de-
vised by Mr. L. R. Wallis, at that time General Manager of the
electric light plant, called by him " The Foresee (4-C) System of
Charging," which is interesting. In this system each customer
paid certain union charges based, first, on capacity demanded and,
second, current used. The name was suggested from the f oUow-

* Sabmitted after adjournrooDt.

RATES AND PBICES FOR ELECTRIC POWER. 135

ing sentence : " A Capacity Charge and a Current Charge." The
capacity charge was figured as follows :

F = Fixed charges per annum on unit cost of plant.
i\r= Number of sixteen-candle-power lamps per unit.
C = Average " capacity charge " per annum per sixteen-can-
dle-power lamp.
F

The following quotation from the paper delivered by Mr.
Wallis before the National Electric Light Association in 1901
explains the system further:

** After obtaining the average capacity charge per sixteen-candle-
power lamp, the minimum and maximiun charge per lamp is easily
determined, and a sliding scale between the minimum and maxi-
mum is readily made that will result in an equitable distribution of
the fixed charges among the various customers. While it is per-
fectly consistent to expect the customer to reimburse the station
for all the fixed charges that it has to meet to furnish him with
the service he requires, the proposition is only to insure the sta-
tion against loss in carrying capacity for him, and to secure this
protection it is not necessary to charge the full amount shown as
being the total fixed charges, as the object is accomplished when
a large proportion of the fixed charges are guaranteed.

" The probable fact that all of the consumers will not demand
their maximum contracted capacity at the same time should be
taken into consideration in establishing the capacity tariff, but
should not be made the basis for individual concessions."

The capacity schedule is made up on the basis of the niunber
of lamps demanded on a yearly rate and a monthly rate. There
is also a schedule of rates fi»om October to June, and then for
June to October, this latter charge being less than the charge for
winter months. I may add that in the practical w^orking of the,
system I found that as a customer there was much to commend it.

Mr. Perry* — Replying to Mr. Hale's remarks, I wish to say
that the contract, which contained rates similar to those chosen
for the illustration in the paper, included no clause of any kind
which would produce a discount or reduction in bills other than
that covered by the schedule itself. The omission from the con-
tract of the usual proviso, which stipulates that " the customer

* Author's Closure under the Hules.

136 RATES AND PRICES FOR ELECTRIC POWER.

shall not be billed a less amount for the use of a larger quantity
of current, alluded to by Mr. Hale, has in this case served to bring
out even more strongly the existing fault in the present " dis-
count system." With the latter it is a fact, as pointed out by Mr.
Hale, that no charge is made for some of the current used. There
is, therefore, something radically wrong with the system on ac-
count of this inconsistency, which could not possibly occur with
the " chart system."

To make the defect more apparent, the following figures are
interesting, because they indicate what ordinarily happens in
applying the " discount system." This data was given me by
a gentleman who formerly operated a 5 horse-power, 500 volt,
direct current motor from a central station circuit • The mini-
mum charge per month in this instance was $3, and the base rate
was 10 cents per horse-power hour. Here are some of the bills :

(1) ApHl Ist to April 24tli, 211 horee-power hours, dincoant 80 p. c $14.77

(2) " 24th " May 27th. 810 " •' " 40 •* 18.60

(8) May 27th " Jane 25th. 102 " ** " 20 *• 15.36

(4) June 25th •* July 28th. 247 •* *• " 80 " 17.29

6) July 28th " Aug. 27th, 194 '* " " 20 " 15.52

Comparing bill No. 1 with bill No. 5, it is noticeable that a
current consumption equivalent to 17 horse-power hours less was
used in the latter case, although the total cost for so doing was
75 cents more than in the first instance. Item No. 2 indicates
that during this period about 25 per cent, more current was con-
sumed than in the time covered by bill No. 4, although the in-
crease in cost was approximately only 8 per cent. The quantities
of current considered in the paper are much larger, consequently
the inconsistencies are more marked; the principle, however, is
the same in both cases.

The title of the paper may be misinterpreted, since it is mis-
leading; the text indicates, however, that it is not intended to
suggest a method for establishing the original basis for the rates.
On the assumption that these have been fixed, the paper illus-
trates, by means of charts, a method for determining rates per
kilowatt hour and prices per month for electric current. No at-
tempt has been made to define what elements should be consid-
ered in finding the cost of central station operation for the pur-
pose of properly adjusting the rates. These vary with local
conditions and must be carefully worked out in every individual
case. Assuming the cost to have been found in any specific case

RATES AND PRICES FOR ELECTRIC POWER. 137

for producing electric current in small quantities, as well as for
successively larger amounts up to the total normal capacity of the
plant, I maintain that rates may be arranged consistently in pro-
portion to the current output without the use of the discount
sheet This result may be brought about by the " chart system,"
which, when once applied to meet the conditions of any plant, is
free from the objectionable features now existing. In most in-
stances the rates would be expressed by curved lines instead of
by straight lines as shown in Figs. 25 to 32 inclusive. In order
to avoid confusion, it would possibly be better to omit the rate
curve from the chart issued to the customer, since that of greater
unportance is the one from which monthly bills may be deter-
mined at a glance.

It is customary in nearly all contracts to insert a clause which
specifically states the minimum monthly charge which the cus-
tomer is obliged to pay to the central station. In view of
this fact, and since all recording wattmeters indicate kilowatt
hours, it is evident that a central station would have to make
a special contract covering the unusual conditions of operation
discussed by Mr. Hale. The subject considered is not due to
any misapprehension of facts or to a lack of knowledge of the
usual forms of contracts made by central stations. The case
discussed was not assumed, but was met with in actual practice,
although, as stated above, the rates and quantities were slightly
altered for obvious reasons. It may be that I have been pre-
sumptuous in calling attention to inconsistencies of a system which
is ahnofit universally followed and with which every one is more
or less familiar. Th€ fact that this system is acknowledged to
be faulty in practical operation is sufficient reason for its displace-
ment if some better and more coherent method may be devised.
The discussion which the paper has merited has borne out my
contention that the " discount system " is an unfair one, conse-
quently is open to improvement Graphical or diagrammatic
representation has long since been recognized as giving solutions
quickly and nearly, if not quite, as accurately as methods involv-
ing mathematical computation, and I am yet to be convinced that
this system is unworthy of consideration as a satisfactory substi-
tute for the " discount '^ or " step system."

Note. — The original charts were made on laboratory cross section paper con-
taining ten more spaces horizontally and yertically than the squares reproduced
in Figs. 25 to 82 inclnslTe. Paragraphs 80, 82 and 85 should be interpreted with
thii understanding in mind ; otherwise it will appear that the text is incorrect.

138 niFBOTEHENT IK TALVE-MOTION OF DUPLEX AIB COHPBE880BS.

No. 1017.»

AN IMPROVEMENT IN VALVE-MOTION OF DUPLEX
AIR COMPRESSORS.^

BT SmtLmO H. BXTHHBLL, LOBAIN, ORIO.

(Janior Member of the Society.)

1. The use of poppet valves held down by springs operates in
pumps handling incompressible fluids merely to increase the work
done by the piston, generally only by a trifling amount. The same
system of valves applied to a compressor working with an elastic
fluid not only involves a similar loss by the friction of the fluid
in passing the spring-loaded valves, but also decreases the density
of the fluid flUing the cylinder at each stroke, so that the total
weight of gas handled falls considerably below that corresponding
to the swept capacity of the cylinder. Such valves by reason of
their inertia tend also to delay closing till after the reversal of the
motion of the piston at the end of its stroke and thus to cause
a further loss by slippage. For these reasons mechanically-
actuated inlet valves arc generally applied to compressors of
medium and large size. The adaptability of the duplex or two-
crank type of direct-connected air-compressor to varying capa-
city requirements and occasional unusually low speed, together
with the superior economy of its steam cylinders working under
short cut-offs, has brought about its general use except for supply-
ing such small quantities of air as can be delivered by the single
cylinder of the ** straight-line " or single crank tandem compressor
of small size. It is to the common type of duplex compressor with

* Presented at the New York meeting, December, 1903, c»f the American
Society of Mechanical Engineers, and forming part of Volume XXV. of the
Tran»aetion9,

t For further discussion on this topic, consult Tran9action» as folloVs :
No. 884, Vol. XX., p. 967 : ** New System of Valves for Steam Engines, Air En-
gines, and Compressors." E. W. Gordon.
No. 920, Vol. xxiii., p. 151 : " New Valve Gear for Gas, Steam, and Air Engines."
E. W. Naylor.

IMPBOVBMENT IN VALVE-MOTION OF DUPLEX AIR COMPRESSORS. 139

;r"-Kl

' I i ": i

140 IMPBOVEMENT IN VALVE-MOTION OF DUPLEX AIB COMPRESSORS.

I . ? 1. „ y^ .y^ ^J vK T

IMPROVEMENT IN VALVE-MOTION OF DUPLEX AIR COMPRESSORS. 141

Meyer steam cut-off valve gear and mechanically-actuated air-inlet
valves that the construction to be described applies.

2. The main steam valve of a Meyer or riding cut-off gear is set
exactly like any plain slide-valve, being laid out to give proper
steam lead, exhaust opening and exhaust closure or compression,
without regard to the point of cut-off, which will therefore come
somewhere around J or J full stroke. The riding cut-off valve
is operated by an eccentric set either just opposite the crank, or
better, a little back of this position. The air-inlet valves are
usually of rotary type, driven like Corliss steam valves by a rocker
or wrist-plate connected to the valve-arms by short links, or they
may be plain slide valves. In any case the inlet valve must close
as the piston reaches the end of its stroke, and open shortly after
it commences the suction stroke, the lateness of opening being for
the purpose of allovring the air contained in the clearance space to
expand to the pressure of the air in the intake. The required
position of the eccentric operating air-inlet valves is therefore
approximately at right angles to the crank operating the piston, or
more or less back of this position.

3. Air-compressors of the type just described have been regu-
larly provided with six eccentrics to operate the double steam
valves and air-inlet valves of the two sides of the duplex machine.
A moment's consideration of the preceding paragraph will show
that the cut-off valve eccentric of one steam cylinder and the air-
inlet valve eccentric of the opposite cylinder are a little back of the
position opposite the crank on the steam cylinder side, while the
alternate steam and air cylinders have valves to be actuated by
eccentrics, one set back of the position opposite the crank and the
oAer 180 degrees from the first It is only necessary to modify
the arrangement of valve-arms and links of the inlet valves of the
latter compressing cylinder, using precisely similar valves, valve-
arms and other gear except the rockers or wrist-plates, to allow
of driving the air-valve gear of each side of the machine by direct
connection to the cut-off valve rod of the other side.

4. The details of the gear are clearly shown by the illustrations.
Steel tube is common enough to-day to allow of using it for the
right-and-left threaded sleeves of the cut-off valves. Through
each sleeve is passed a smaller solid rod connected at one end by
means of suitable rockers and pins to the eccentric rod, and at
the other end to an arm carried on a cross-shaft. This rod drives
the encircling sleeve, and, therefore, the cut-off valves carried
on it, through the medium of tyro simple split clamps touching

1^ IliPROVEMENT IN VALVE-KOTIOK 07 DUPLEX AIB 0OHPBES8OR8.

IMPROVEMENT IN VALVE-MOTION OF DUPLEX AIR COMPRESSORS. 143

the ends of the sleeve. The cross-shaft passes through bosses on
tlie ccmipressing cylinder casting and terminates in a boss on the
opposite conapressing cylinder, and carries adjacent to the latter
boss the wrist-plate or rocker driving the inlet valves of this side.
This description applies also to the details of the corresponding
gear of the other two cylinders. It happens that there is abso-
lutely no difference between the two. sets of parts except in the
two wrist-plate rockers, and in the fact that one of the long rods
is a little longer than the other because the centre of its rocker is
farther back on the air cylinders.

5. It looks at first sight as if this combination would be trouble-
some to lay out and to adjust in erecting the compressor. There
is really no additional complication, for it is only necessary to
slack off the clamps on the long rods, set up the air-valve motion
in the usual way, locate the eccentrics as required and then set
the sleeve along the rod to give even cut-offs and tighten the col-
lars. The cut-off valves can be changed to equalize the cut-offs
at any time without disturbing the long rod, merely shifting the
clamps as desired, and the air valves may be shifted or reset with-
out disturbing the equality of the cut-offs. The net result of the
arrangement is the saving of two eccentrics, straps, rods and rock-
ers and of the space between or outside the cylinders that would
otherwise be occupied by these parts, at the cost of substituting
a piece of steel tube for* a solid rod, and of enlarging the diameters
of steam chest glands to correspond. The feature of operating
one side of the machine alone in case of necessity is not lost, be-
cause as long as the crank-shaft is not broken and the cylinder
castings remain in position, the eccentric driving the air-inlet
valves of the cylinder which is to be operated may as well be on
one end of the shaft as on the other.

6. The compressor shown embodies a modification of the usual
framing, which has some advantages. The two separate bed
plates are bolted together along their centre lines, and are further
bolted to a single cross member lying under the cylinders. The
strains which tend to work the ordinary duplex machine on its
foundations are thus resisted directly by the cross frame, making
the machine nearly independent of a masonry foundation except
as a mere support. Tie rods between the upper parts of cylinders
and over the guides add greatly to the rigidity of the whole.

7. A number of these machines have been constructed ranging
from 12 to 24 inches stroke, all of which have been shipped with-
out taking apart and leveled upon foundations in a few minutes'

144 IMPBOVEMENT IN VALYE-HOTIOK OF DUPLEX AIB COMPRESSORS.

time. Some of these compressors have been continuously operated
at speeds up to 200 revolutions per minute for months together,
and none of them have developed any objection to the combination
of valve mechanism described, or have shown the tendency of the
usual duplex compressor with independent frames to shift on its
foundations and thus work out of alignment.

TESTS 0# A DQtECrr OONNBOTED BIOHT-FOOT FAH AND ENQINB. 116

No. lOlS.*

TESTS OF A DIRECT CONNECTED EIGHT-FOOT FAN
AND ENGINE.^

BT B. 1. FABWBIX, XBW TORK, H. T.

(Member of the Society.)

1. Thb promoters of the plenum system of heating and ven-
tilatiDg have obtained a very substantial footing among the
paper mills^ because of the moisture which it is necessary to
lemove from the rooms.

For instance, in a room containing two paper machines, each
making 25 tons of paper in 24 hours, there is approximately 100
tons of water which must be disposed of before it condenses on
the roof and trusses, and drips on the machines. Also, in pulp-
grinder rooms, there is a large amount of steam generated which
most be disposed of, or the fog in the room will be so dense as
to impede the work. In most of the other rooms, however, it
is merely a question of heating.

I found it rather difficult to obtain reliable data which would
guide me to an intelligent selection of apparatus for any particu-
lar job. The builders of such apparatus very generously offered
to work out the problem for me, but it is safe to say that in no
case were any two proposals for the same size of apparatus.
"A" would submit a proposition for a small fan, running at
bigh speed, with small air-pipes and a small heater; ^' B " would
offer a large fan to run at a slow speed, with large air-pipes and
a large heater; " C " would perhaps offer a large fan, with large
air-pipes and a small heater. If '^ A " were asked to furnish a

*Preieiited at the New York meeting (December, 1908) of the American
Sodety of Mechanical Engineers, and forming part of Volume XXV. of the
Tnn9aeti(m$.

f For farther references on this subject, see Tranaaetioni as follows :
No. 240, Tol. TiU., p. 818: " Power to Drive Fans."
No. S$4» ToL ix., p. 51.

146 TaSXS OF A DIBEOT CONNECTED EIGHT-FOOT PAN AND ENGINE

large fan to run at a slower speed, he immediately proposed to
enlarge the pipes proportional to the size of outlet of the fan,
and to put in a much larger heater, notwithstanding the fact that
the amount of heat to be delivered was the same. I wish to fur-
ther acknowledge that under these perplexing diflBculties the
representatives of the different fan builders never hesitated to
give me all the assistance which they apparently could, and I
believe that I have finally succeeded in getting some points clear
in my own mind.

2. The question seemed to me to embody three distinct prob-
lems: the air-pipe, the fan, and the heater. The first two are
interdependent, or perhaps we had better say both depend upon
the same assumed data. The starting-point of the problem or
problems is the volume and temperature of air required. In a
factory building, if it is merely a question of heat, a compara-
tively small volume of air at a high temperature is satisfactory.
A change of air as low as 30 minutes is allowable where men do
not stand at their work, and 20 or 25 minutes where they do.
In the grinder-room the change must be as frequent as 15 or
even 10 minutes, depending on the shape of the room, the loca-
tion of the grinders, kind of roof, and other IocqI conditions.
In paper machine rooms, also depending on local conditions, the
required change may be as often as 4 minutes.

It is very easy, with the published tables and data of tests, to
determine the amount of heating surface required to heat the
determined volume of air to the desired temperature. I never
•could see how the size of the fan need affect the amount of heat-
ing surface, provided the volume to be delivered and initial tem-
perature of air were the same in both cases.

3. About tV ounce pressure per square inch will be re-
quired to give the necessary velocity at the outlets of the dis-
tributing pipes. In addition to this there must be as much more
pressure as is necessary to overcome the friction of the pipes,
and herein I find is the principal difference in the practices of
the different builders. Some always figure on 1 ounce pressure
at the fan. Others figure f ounce. Fifteen-sixteenths of an
ounce seemed to me to be an excessive loss, and I usually call
for i ounce at the fan, enlarging the pipe sufficiently to bring
the friction loss well within that figure.

In reducing the pressure required of the fan, we have of course
reduced the amount of work done by the fan; but against this.

TESTS OF A DIBEGT OONNEOTBD EIGHT-FOOT FAN AND ENGINE. 147

saving in work, it is necessary to chai'ge the interest and depre-
ciation on the increased cost of fan and pipe.

4. In order to secure some data for my guidance, I made a
series of experiments on the efficiency of a fan which I purpose
to outline in this paper. The fan was a No. 160, according to
the usual method of designating fan sizes, with a wheel 8 feet
in diameter, and 37 inches wide at the periphery and with one
side inlet. It is shown in Fig. 36. The fan is driven by a direct-
connected steam engine, and discharges into a large chamber
^applying air for combustion to the boilers. This chamber was
left open to maintain atmospheric pressure except in a couple of
tests. The opening of the fan into this chamber was tightly
boarded up, and conical tubes or nozzles were fastened to cir-
cular openings in this board partition. The tubes had a taper of
^ d^rees and, with the exception of the largest two sizes, were
approximately three times the diameter in length. Six different
sizes of tubes were tested, each at 8 or 10 different speeds, ranging
from 50 to 250 revolutions per minute. In addition, a set of ob-
servations at each speed was made, with no outlet whatever.

5. This I realize is not an ideal arrangement, but under the
circumstances seemed to be all that the desired results, would
warrant. The volume of air delivered was determined by means
of a pitot tube located at the extremity of the conical outlet.
We attempted to use an anemometer, but some of the velocities
were beyond the capacity of the instrument and it very soon
proved to be unreliable. With the tapering outlet, it was appar-
ently safe to assume that the air at the extremity had no static
pressure, but that the entire potential energy had been converted
into kinetic energy. The pressure in the fan chamber, as also
the vacuum or suction at the inlet, was measured by a water
column. We found the vacuum at inlet varied at the different
points of the inlet as the gauge was moved from the centre
toward the rim. A number of observations were made, and a
point determined which gave us the mean of all the readings.
The temperature of the air was taken, also the revolutions of the
fan and indicator cards from the engine. This completed the
list of observations taken during each test. Each test was run
for half an hour under uniform conditions, and readings taken
every five minutes. The average results of each test are given
in the table. The volume of air discharged was computed from
the velocity of air as given by the pitot tube, and tabulated in

148 TESTS OF A DIRECT CONNECTED EIGHT-FOOT FAN AND ENGINE.

TESTS OF A DIBEOT CONNECTED EIGHT-FOOT FAN AND ENGINE. 149

column marked "B." The theoretical horse-power was com-
pated from the volmne, determined as above noted, and the press-
ure in chamber as tabulated. The pressure, volume, indicated
horse-power, and efficiency have also been plotted on the two sets
of curves. Figs. 37 and 38. In Fig. 37 the abscissae represent
revolutions of fan, while each curve represents a certain size of
outlet. In Fig. 38 these two items have been interchanged, and
each pressure curve has a different zero line. It is the custom,
I believe, to designate the size of fan outlet in terms of the diam-
eter and peripheral width of fan- wheel, and this I have done in
the present instance.

6. A number of interesting facts stand forth very clearly upon
an examination of these curves. Theoretically, the volume
should vary directly as the speed of the fan with a given size of
outlet, the pressure as the square of the speed, and the horse-
power as the cube of the speed. The curves show that the ratios
are not quite those stated. Up to a certain point the volume
corves are very nearly straight, but at the higher speeds they
seem to fall off. This falling off was more marked with the
large openings, as was also a large increase of the vacuum at
inlet. This loss in the volume delivered is unquestionably due
to the throttling of the air at inlet. The maximum efficiency of
the combined unit was secured at 142 revolutions per minute,
when the pressure was i ounce (this, of course, only applies to
the particular fan tested, and it is fair to presume that if the
fan had had an inlet on each side, the throttling would have
been less and the most efficient speed might have been higher).

7. The curves show very clearly, however, that in the selec-
tion of fan we should choose large sizes running at moderate
speeds and developing a low pressure. We may apparently run
the same fan at a lower speed and lower pressure so that it will
deliver the same volume of air with a considerable saving in
horse-power. This should be done, if in any case the conditions
prevent the selection of a largei:^fan. As I will illustrate in a
few moments, however, the larger fan will do the work* more
efficiently.

It has been frequently stated that up to a certain size of outlet,
variously styled the "theoretical outlet," "square inches of
blast," and "capacity of wheel," any change in size of outlet
makes no change in the pressure, and a variation in volume and
horse-iK)wer directly proportional to the size of the outlet, and

150 TESTS OF A DIRECT CONNECTED EIOIIT-FOOT FAN AND ENGINE.

that further enlarging results in a decided drop of pressure and
falling oflf in the rate of increase of volume and horse-power.
The curves show that up to about .29 D. W. the pressure drops

Fig. 37.

but slightly. The volume and indicated horse-power rise very
nearly by straight lines, but beyond that point there is a sudden
and rapid change, the efficiency also reaching a maximum at
this same point.

TESTS OF A DIRECT CONNECTED EIGHT-FOOT FAN AND ENGINE. 151

8. As being probably pertinent to this point, I wish to call
attention to the fact that the area of the fan-blade is equal to
.2885 D. "W. The efficiency curves show that but slight varia-

Fio. C8.

tions from this " theoretical outlet " should be permitted. Mr.

Snow states that for general practice the square inches of blast

D W
is not far from ' . I understand that the usual width of pe-

152 TESTS OF A DIRECT CONNECTED EIGHT-FOOT FAN AND ENGINE.

riphery of an 8-foot fan, as built by the makers of the fan tested, is
41 inches instead of 37 inches. This would make the area of fan-
blades approximately .32D. W., which might increase theeffective
area, or '* theoretical outlet," to a like quantity and bear out the
statement made by Mr. Snow. This '* theoretical outlet," as
has been stated by Mr. Snow, is not to be understood to be the
actual sizes of the outlet of the fan's casing, but to be the size of
opening which will oflfer a resistance equivalent to the sum of all
resistances of distributing pipes. If in any given case we are
able to state what this equivalent outlet is to be, we can then
select the size of fan which will give us the desired volume of
air at the desired pressure.

9. In Fig. 39 I have reproduced the eflSciency curves and have
drawn another curve showing the relation existing between the
size of outlet and the ratio of air velocity to peripheral velocity of
wheel. It is not safe to reason too much from the concrete to the
abstract, but comparing this curve with similar ones constructed
from the tests of a small pressure blower, as given in Mr. Kent's
handbook, and from the guaranteed results of 'one blower-maker
on his standard fans (presumably computed from tests which are
not available to the public), I think it safe to say that the curve
of the tests may be applied to different sizes of fans with reason-
ably satisfactory results.

10. To illustrate the advantages that have already been men-
tioned, we have only to notice, for instance, that the fan tested,
when running at 200 revolutions per minute, will develop 1 ounce
pressure when delivering a volume of 32,000 cubic feet per min-
ute, at an eflSciency of 51 per cent, and requiring 17.2 indicated
horse-power to run it. The same fan running at 161 revolutions
per minute will deliver the same volume of 32,000 cubic feet at
a pressure of J ounce and an efficiency of 41.7 per cent., but
requiring only lOJ horse-power. While the efficiency in the
first ^instance is considerably higher, it should be pointed out
that the actual horse-power required is 0.7 more. This repre-
sents the actual cost of delivering 32,000 cubic feet of air into
the room under the two pressures noted. Again, this fan run-
ning at 142 revolutions per minute wiU deliver 25,800 cubic
feet of air at a pressure of i ounce, and at an efficiency of
54 per cent. If, however, it is desired to deliver 32,000 cubic
feet of air at J ounce pressure and at maximum efficiency, it
will be necessary to use a larger size of fan, the determination

TESTS OF A DIRECT CONNECTED EIGHT-FOOT FAN AND ENGINE. 153

of which will illustrate the use which I have made of these
tests.
I am not prepared to say that for fans with inlet on one side

only, .29 D.W. is as large a *' theoretical outlet" as should be
used. However, it seems to be evident that where inlet may be
had on both sides, a wider fan may be used, and possibly further
experiments may show that a wider fan may be used with a single

154: TESTS OF A DIRECT CONNECTED EIGHT-FOOT FAN AND ENGINE.

inlet. Assuming that .29 D.W. is to be used, we have for the
volume of air delivered,

Q = ,29 D.W. Vy

in which V is the air velocity corresponding to the assumed
pressure.

11. The velocity corresponding to the pressure of i ounce
is 3653.8. From the foregoing equation we find that,

32000 _
^•^^"3653.8 X. 29" ^^•^•

If W= -, then 2)2 = 30.2 r. The proper value of " r *' was

not determinable from these tests, but probably has been deter-
mined by the builders' experiments. It is apparently between
2.3 and 2.7. In the fan under test, .

r = 2.595.

It is desirable to use standard diameters and vary the width
within certain limits. In the present case the size of fan selected
should be 8^ feet diameter by 42.6 inches wide at periphery.
For the speed of fan, we have from the lower curve in Fig. 39 for
.29 D.W.,

-^ = 1.025.

From this equation the peripheral velocity,

F 3653.8 ofc.-^ j;-
p = YMK' ~ ^^^^ ^^' P^** ^^^'

This is equivalent to 133.5 revolutions per minute of an 8i-foot
fan. By the efficiency of 54 per cent, shown by the chart, this
would require 8.125 horse-power, a still further saving over the
8-foot fan running 161 revolutions per minute.

12. There is another problem upon which I should be glad to
have discussion and enlightenment. In designing an induced
draught plant some time since, I proposed to install two fans,
which, under normal conditions, would both run at slow speed.

TESTS OF A DIRECT CONNECTED EIGHT-FOOT FAN AND ENGINE. 155

either of which in an emergency could be speeded up to do the work
of both. One representative informed me that he did not think
it could be done practically, and after I insisted that it be worked
oat, two 10-foot fans, 24 inches wide at the periphery, were
oflfei*ed. After some discussion and computations, two 10-foot
fans 54 inches wide at periphery were purchased. I based my
coatentions on the tests which I have described, although I am
aware that the conditions when handling gases at a high tem-
perature are somewhat different. I do not see, however, that
these changed conditions prevent the use of a fan in the way
I described. To illustrate, the following data may be taken
from the curves of the 8-foot fan tested. With an outlet of .265
D.W., and running at 150 revolutions per minute, the fan will
deliver 25,000 cubic feet of air at a pressure of j\ ounce, requir-
ing 7.8 horse-power, and give an efficiency of 53 per cent. This
same fan with an outlet .53 D.W., running at 225 revolutions
per minute, will deliver 50,000 cubic feet of air at the same
pressure of ^ ounce. In the latter case the horse-power is
40.8, and efficiency but 26.5 per cent. This efficiency is very
low, but is not to be considered in an emergency. The only
question which requires particular attention is an arrangement
which will allow a change of the outlet, or its equivalent resist-
ance, from .265 D.W. to .53 D.W., and this was very easily
accomplished in the case mentioned. It is necessary in many
industrial plants to provide against a shut down, due to any
ordinary accident, but I fail to see the necessity of putting in
two fans, each sufficiently Jarge to do the entire work under
normal conditions, at maximum efficiency.

I have not entered into this discussion with any feeling akin
to the old Quaker's who is reported to have told his wife that
^* all the world is queer except thee and me, and thee is a little
queer." But I have hoped to provok(^ a discussion by those
who have had better opportunities for studying the problem
than I have, and can give us information not obtainable, at least
convincingly, from these tests.

In closing I wish to acknowledge the able assistance and pains-
taking care of Mr. C. W. Wilder, who made most otthe obser-
vations and computations for me. But for my confidence in his
ability, I should have hesitated to present these results to the
Society.

156 TESTS OF A DIRECT CONNECTED EIGUT-FOOT FAN AND ENGINE.

C
m

Oh
o

*)a93 lad iCaaapos

«0 O « 90 C« l0 1-4 'QOOft

^9<^?S»^

-auoq paAJMqo

jdMod
-asjoq (VOHdioaqj.

-sip i}« JO iaaj ajqu.)

K-*-ff

r7 + .9

'saqon; = q
IB piwq ^llooiaA

00 d O C9 O O 5

11 W ^ "* »o •« -o

o o» 55 35 55 o6 lo 25 ^ op

BOtiOOOO'^lOXQDgf

SSSSSg5SS3

»-• r- ©# 00 CO WOO W«0

SSiligisS?

•ogoe-tr"*?

5S9

i^isiiiiis

'jaqnraqo a{ Mnasdjj Q

•%^n\ %v rannoBA *^

i iiiiisig§s

I I=|ss|li55

•mm jad BUoniqoAaj
'paads

otiooeotocoottocDto

^1-

§?ses58s

etotw^oDoo 'i-io

T-i T-< 1-1 o> ■^' eo o o 0^

sssssssssss

i-<^i-i ef e< e« ed^oj'^'^io'

iislSiislli

»l§isil§i§l

ISigiill^gl

giiisiiiiss

jisiiHrsigi

« C ■V 00€

§iigsis§iii

ii^ii§iiigi

•"Tr^'ofefodeoco"^'^'**©

TESTS OF A DIRECT CONNECTED EIGHT-FOOT FAN AND ENGINE. 157

«9SI«er-ioeoQ0 9tiat« acoo9ao^9««^eDa •(« oaic^otaaeo o<t>io«D«oo

;S?$SSSS$$S

S^S;;^9$S^ :7

8?^liSIS$

ss»^s«!;s

sot" »-» cb 3d S

SggSSJltSSn

^i-«0«9*^«00

i§iiiiii.^s

5§ig§ssiii§

^»*0O'*t-t-X«

«-ii-i«-«e9e990v«oi«

v^ctoteoto

•-• 11 00 ■** t«^

|Siilii?iil

2f 5S5S;S'S'Si5

i22gg8Bi§i|

iiiiil

:&§SSS5§S2

oSSaSSSSSxcS

sssn^s

ss^sss

snsssssss

-^Ot Ot •PofoO V Tf -<> lO

11 r^ « of W'otf « « 00 -^ ^

«-Ii-«'©»"«0OCO

i-H*i-Jof««0O

SiiiifSigg

iSlssgpgg

«i r-i ^ « M 9»

siiiiiisfc

llSSgs

n00^b-O»»-«

ilgISi

i§Siigiii§

iiigiii^^is

iliiis

mmm

m%m^

siiiiiisii

S<M':*t«i^oooaot«o!

Sillil

SSslg?

iSss§giii«

IH*§|gSI§|g

;iis§§

iss^ifc

=ug^i§§i§

Si5iSlie-ii

iillSSgr-glfl

Si^Sii

^m%u

i§§§i§i§§

Ifiiilfilr-

I i-«©xooooe'^ao"*o

§ii^g§s§i

^iiiiiif!

OiietSleO"^

iSglil

§ii;|§i

§iigig

11 n »fefo»«ed«'«*"v ^

§i§SI§

1-I^r^Of « 90 00 V«d O

xiaoxo(eo'^.ico«^

•"S- 0» «0 1- CO o

r.ni-nt.'^i

eo^xe»©t»coox

S|S2

S^§SS|

S^S§§Sg||

^'11

^ ft

ft-

158 TESTS OP A DIRECT CONNECTED EIGHT-FO.yT FAN AND ENGINE.

DISCUSSION.

Mr. E. S. Farwell* — Since the preparation of this pai>er, the
question has been raised of the relation said to exist between the

O M

U, x

I a

r bo ■«

> c

VA<

UUM A

T INLE

4000
8000

12000

leooo

s"\

s Y

20000

'*"*j^

24000

)

28000

^^J

82000

-I
>

86000

N;:

40000

44000

48C00

52000

B6000

IterMii^KA

Fig. 40.

vacuum at inlet and the cubic feet of air discharged per minute.
If the data were correctly taken^ there should be a fixed rela-
tion between these two quantities. That relation is illustrated

* Added after adjournment.

TESTS OF A DIRECT CONNECTED EIGHT-FOOT FAN AND ENGINE. 159

in Fig. 40. Aside from a few unquestionably erratic points
which have been connected to the proper curves by dotted lines,
1 thirtk the curves show a very satisfactory relation existing be-
tween vacuum at inlet and cubic feet of air discharged.

The curves for tests E and F are certainly very interesting,
and it may be pointed out that in these tests the vacuums at inlet
were larger and the probable errors in reading of water gauge
were proportionately less. The point which I attempted to
bring- out in paragraph 6, viz., the limitations imposed by having
a single iulet, I do not think could have been presented so clearly,
if vacuum at inlet had not been read.

160 A SERIES DISTILLING APPARATD8 OP HIGH EFFICIENCY.

No. 1019.*

A SERIES DISTILLING APPARATUS OP HIGH
EFFICIENCY.

BT W. F. M. Q0B8, LAFATSTTB, INDIAlfA.

(Member of the Society.)

1. This apparatus is designed to purify water or other liquids
by distillation. It can be used, also, in the concentration of
liquids carrying soKds in solution, the recovery of which is desired,
It consists of an arangement whereby the supply of liquid to be
evaporated passes through a succession of chambers, in each of
which it is gradually raised in temperature, and in all but the first
of which a portion of the supply is vaporized, the process continu-
ing until all has been changed to vapor. Heat is supplied the
system at the chamber having the highest temperature only, from
which vapor starts in a return circulation, that generated in one
element of the apparatus serving as the source of heat for the
element next lower in temperature, the liquid steam increasing in
volume as it passes the successive elements imtil, finally, all is
discharged from the chamber of lowest temperature. The liquid
thus discharged is entirely the result of condensation. An ap-
paratus embodying this conception, designed to distil 500 gallons
of water an hour with an efficiency of approximately 60 pounds
of water per pound of coal, is shown by Fig. 41, and a section of
a single element entering into its construction, by Fig. 42.

2. The action of the apparatus may be best described in connec-
tion with the diagrammatic sketch Fig. 43. Several chambers.
Ay B, C and Z?, each containing a central tube, are connected by
suitable piping. The first chamber, A, is entirely filled with
liquid and constitutes a preliminary heater. The remaining cham-
bers, from B to D inclusive, are partially filled with liquid, and
from each of these, evaporation occurs. The heater and the cham-

* Presented at the New Fork meeting, December 1903, of the American
Society of Mechanical Engineers, and forming part of Volume XXV. of the
TransaciUms.

A SERIES DiSTILLtKG APPAEATUS OF HIGH EFFIOIENOT. 161

162 A SERIES DISTILUNO APPARATUS OF HIGH EFFIOIEKCY.

jiKim'iliiig Ef ii)|Mnm

ll*'l>roiillji4}if,»nitbb«li -^

^ IdcKlBi OlArt

Chamqef^ a.

e IT AHUICAH lAMC KOn C«.,H-T.

Fig. 42.

A aERIES DISTI*ING APPARATUS OP HIGH EFFICIENCY. vl63

ber B operate under one pressure, and each succeeding chamber,
as C and D, under higher pressures. Heat is supplied to the
liquid in each chamber by vapor or a mixture of vapor and liquid
in the central tube, the external source of heat being Hy which
inaj be considered a small boiler having a closed circulation, steam
being delivered by a vertical pipe, (?, condensed in the central tube
of the chamber D and returned to the boiler H.

3. The -water to be evaporated is delivered into the bottom of
the chamber A, and is gradually heated by contact with the central
tube, as it arisee to the top, from which point it passes to the

Fig. 43.

bottom of the chamber B. Here its temperature is further in-
creased and a portion of it is vaporized, the remainder being
taken up by the pump jP*, and delivered to the chamber C, of
higher pressure, where its temperature is further increased and
where another portion is vaporized. That not vaporized in is
taken up by pump F^ and delivered to chamber 2?, which is under
a still higher pressure, where it is entirely vaporized. The return
circulation begins with the vapor given off in the chamber Z?,
which passes off by the pipe J and becomes a source of heat for
the chamber (7, being condensed by the time it reaches the bot-
tom of chamber C. From this point it is conveyed to the reducing
pressure valve Ly where its pressure is reduced to the pressure of
the vapor within the chamber C, From L the stream passes on
to the mixing valve Jf , where it intermingles with the steam de-
livered from the chamber C, the combined stream of liquid and
vapor passing through the central tube of the chamber B, serving

164 A SERIES DISTILLING APPARATUS OF HKH EFFIOIEKCT.

as its source of heat^ being condensed to liquid form and^ finally,
being delivered to the reducing valve L\ where its pressure is
again reduced, and from which point it passes to M^, where it
mingles with the vapor given off by the chamber B, after which
the combined streams of vapor and Kquid pass through the central
tube of chamber A, are cooled and finally discharged.

4. In the process thus briefly 'outlined, the following facts are
to be noted:

a. That the ingoing stream of liquid is continually increasing
in temperature, and that the return current of vapor is continu-
ally diminishing in temperature, the arrangement being such that
it is entirely possible to have a fixed difference between the heat-
ing medium and the medium to which heat is imparted throughout
all portions of the apparatus; also, each particular portion of each
element retains a fixed temperature.

6. The losses of heat which occur from the apparatus are those
of radiation and those which are represented by the differences in
the temperature or condition between the liquid supplied and the
condensate delivered. Heat thus lost must be supplied from the
source of heat and nothing more.

c. It is conceivable that the temperature of the discharge may
be so reduced as to approach very nearly the temperature of the
supply; also, that as ideal conditions are approached, radiation
losses may be diminished, and, at the limit, losses on both of these
accounts cease; that is, the cycle of the apparatus is one of maxi-
mum efficiency.

5. For satisfactory operation, the total range of temperature to
which the apparatus is subjected is divided equally between the
several stages of the process, the temperature of each chamber
being controlled by the pressure which is maintained upon it,
which in turn is regulated by valves on outlets. For example, if
the apparatus shown diagrammatically by Fig. 43 is assunxed to be
operated from a temperature of 360 degrees in the boiler II, to
a final temperature in chamber B of 212, then the temperature
of the water in chamber C would be maintained at 262, and that
of the water in chamber D at 312, the difference in temperature
for each transmission being in this case 50 degrees. This value
also measures the difference in temperature between the mixture
within the central tube and the liquid around the same in each
chamber. The pressures upon the several chambers, etc., assum-
ing the liquid operated upon to be water, would be, for the boiler ff.

A SERIES DISTILLING APPARATUS OF HIGH EFFICIENCY. 166

sr t.N-i sip ^Ss s*

166 A SERIES DISTILLING APPARATUS OF HIGH EFFIOIENCT. ^

145 pounds; the chamber D and pipe J, 65 pounds; the chamber
C and pipe «/^, 25 pounds ; the chamber B and pipe J , atmosphere;
the pressure in each case being controlled by valtes upon the dis-
charge pipe.

6. The action thus far described in general terms is susceptible
of careful analysis, some of the results of which appear in Fig. 44.
This figure is a diagram of a five-stage apparatus, upon which are
noted the pressure, temperature and condition of mixture of the
circulating streams in their course through the system; also, the
weight of mixture transmitted from one chamber to another for
each pound of vapor delivered by the chamber of highest tem-
perature, F. Other assumptions underlying the values given are
as follows :

a. That the liquid distilled is water.

6. That the difference of temperature between each step in the
process is 33 degrees; that is, that the temperature of each suc-
ceeding chamber is 33 degrees higher than that of the chamber
next below, from which it follows that in any given element the
temperature within the central tube is 33 degrees. higher than the
temperature of the surrounding liquid.

c. That the temperature of water supplied is 62 degrees.

d. That the discharge from A is at atmospheric pressure; that
is, at a pressure of 15 pounds, absolute, and a temperature of 213
degrees.

e. That 2 per cent, of the heat supplied each chamber is lost
by radiation.

7. A more detailed presentation of analytical results is reserved
for later paragraphs, but it may here be noted that the efficiency
of the apparatus depends upon the number of chambers employed,
and that its characteristic feature is to be found in the fact that
it so conserves the heat supplied it that the process may be suc-
cessfully extended through a long series of steps.

8. Considering now the details of the apparatus, it is to be ob-
served that the process thus far described involves a separate
pump for each of the several evaporator chambers. Thus, in
Fig. 43, there are provided for the evaporator chambers iJ, C
and D the pumps F^ F^ and F^. This necessity arises from the
fact that at each stage the feed is passing from a lower to a
higher- pressure, and the requirements of the cycle necessitate
that the water supplied to each pump shall have the tempera-
ture of the chamber next below before being forced to the

A SEBIBS DISTILUKO APPABATUS OF HIGH EFFIOlEKOY. l67

succeeding one. While this requirement is not serious when
the niunber of evaporator chambers is small, it leads to objec-
tionable complication when the series is extended. In the prac-
tical working out of the design, therefore, a system of feeding
is employed which allows the use of a single pump, without
impairing the thermodynamic cycle, the modified arrangement
being inferior to that already described only in the fact that
it adds slightly to the mechanical work necessary to move the
feed. Its single feed-pump delivers water at a pressure suf-
ficiently high to supply all evaporator chambers, including that
of highest pressure. Water delivered by the pump is passed
through a succession of heaters connected with the evaporator
chambers in such a manner that the feed emerging from each
heater has the same temperature as it would have had if it mingled
with the water of the chamber, while branch pipes are employed
to convey the feed to each individual chamber. The course of
the piping in its external appearance is shown by Fig. 41. It
will be seen that the feed enters the base of the preliminary
heater, is delivered therefrom at a point near the top, is conveyed
to a small heater over the second evaporator chamber, thence to
the heater over the third evaporator chamber, and thence on to
the last heater over the sixth evaporator chamber. Branches are
taken off after each heater to supply the evaporator chamber with
which the heater is connected. The control of this supply is best
indicated by Fig. 42, showing the float-valve regulating the admis-
sion of water to the evaporator chamber. It will be remembered
that the pressure above the float-valve is greater than the pressure
in any of the evaporator chambers, so that the feed enters when-
ever the valve is off its seat. Referring more particularly to the
heater (Fig. 42), it will be seen that this is similar in construction
with the main evaporator chamber. The mixture of vapor and
liquid which is to supply heat to the main evaporator chamber
first passes through the tubes of the heater, the feed water circu-
lating around the tubes.

9. Concerning the maintenance of the apparatus, it should be
obser^'ed that the boiler from which the heat is supplied (Fig 43,
H) is subject to a closed circulation. No renewal of water is re-
quired, and hence there can be no trouble at this point from in-
crustation. Concerning deposits of solid matter in other portions
of the system, it is to be noted that only distilled water and vapor
circulate within the tubes (Fig. 42). This is true of the heater

168 A. SEltlES D18T1LL1KQ Al»^AEATU8 OF HIGH EFFICIENCY.

as well as of the evaporator chamber. The solid matter which
may be deposited within the chamber around the tubes may be
disposed of by use of a blow-off, by cleaning through manholes,
or by the complete removal of the nests of tubes from the shell
of the chamber. In the construction of the apparatus the central
tube is threaded to the tube-plate at both ends. All other tubes
are threaded at their lower ends only. The presence of the
central tube maintains the relative position of the tube-plates
and permits all the tubes of a chamber to be withdrawn together.
10. Having defined the character of the apparatus, attention
may be given to matters affecting its performance. In general,
three different classes of the apparatus are to be dealt with, and for
convenience Class A wiYL be designated as representing an appa-
ratus all portions of which are operated at pressures above the
atmosphere; Class B as representing an apparatus all portions of
which are to operate below the pressure of the atmosphere, and
Class C as representing an apparatus a portion of which is to
operate at pressures above the atmosphere and a portion at pres-
sures below the atmosphere. Class A may be regarded as a high-
pressure apparatus, Class B a low-pressure apparatus such as
might be operated by. the use of exhaust steam, and Class C a
high and low pressure apparatus. Practical and convenient ranges
of pressure and temperature applying to each of the three classes
thus designated, when designed for the distillation of water, are
as follows:

TABLE I.

DbSIO NATION.

Absolute Prbssurs.

Tbmpebatitrb F.

Class.

As to Pressure.

Highest.

Lowest.

15
10
1.0

Highest.

Lowest.

A
B
C

High pressure.
Low pressure.
High and low pressure.

115

838
212
888

212
102
102

The relations shown are merely chosen for purposes of illustra-
tion. So far as the working of the apparatus is concerned, any
other practical limits might have been chosen.

11. The weight of water which will be distilled for each pound
of saturated steam supplied from the source of heat for an appar-
atus of the high-pressure type, working under the conditions of

A SERIES DISTILUNG APPARATUS OF HIGH EFFIGIENOT. 169

pressure set forth in Table I., and assuming no loss to occur by
radiation or in the form of vapor from the delivered stream, will
be as follows:

TABLE

II.

EIGHT OF Distilled

Water per

Pound

OF Steam Used.

Namber uf Evaporator

Pounds of Water per Ponnd

Chambers.

of steam.

2.8

3.6

4.4

5.2

5.9

6.5

7.1*

7.6

12. The weight of water distilled for each pound of steam sup-
plied is nearly the same for all three classes as above defined. An
approximate relation applying to all classes is represented by the
formula, W = 0.85 N, in which W is the number of pounds of
water distilled for each pound of steam supplied, and N is the
Dumber of evaporator chambers. If the boiler which serves as
the source of heat delivers ten pounds of steam per pound of coal
burned, the output of distilled water per pound of coal will be ten
times the values given in Table II.

13. Results appearing in the preceding Table II. are in some
cases subject to correcti6n to cover losses due to the presence of
vapor in the discharged stream. In the action of the machine, all
condensation is the result of the cooling action of the incoming
stream of liquid feed. Under certain conditions, this will be insuf-
ficient, in which case a portion of the issuing stream delivered
from the apparatus will l)e vapor and will in part disappear as it
emerges from the pipe. The extent of this loss, if any, decreases
as the nmnl>er of chambers is increased. In apparatus of the Class
A it entirely disappears when seven evaporator chambers are used;
but in types B and C there will be a slight loss even when as many
as ten evaporator chambers are used. The percentage of the total
weight of the discharged stream which will be delivered in the
form of vapor is as follows :

170 A SERIES DISTILLING APPARATUS OP HIGH EPFIOIENCY.

TABLE IIL
Pbrcbntaob of Discharged Stream Lost bt Vaporization.

Class of Apparatus.

Nomber of Evaporator

Chambers.

14. The extent of heating surface in each element (one evapor-
ating chamber and its attached heater) needed for a given capacity
is not materially affected by changes in the number of elements in
the series. Assuming the pressure ranges set forth in Table I.,
and assuming each foot of tube surface to transmit 400 thermal
units per degree difference of temperature per hour, there will
be required in each element of an apparatus designed to distil
1,000 gallons of water per hour an amount of heating surface as
follows :

TABLE IV.

Si^UAHB Feet op Tubs Surface in each Eleitent per 1,000 Gallons of
Water to be Distilled.

Class of Apparatos.
A (High Pressure).
B (Low Pressure).
C (High and Low).

Tabe Surface.
174
188
113

15. The total amount of tube surface required for any appa-
ratus of the capacity stated will be found by multiplying the num-
ber of evaporator chambers it is to contain, plus one, by the value
assigned its class in the preceding statement. The unit added
covers the large heater (1st Heater, Fig. 41). For example, if it
is required to design a seven-chamber apparatus of the A class to
distil 1,000 gallons of water per hour, then the total number of
square feet of heating surface for the whole system will be 8
X 174 or 1,392 square feet.

16. The size of the small feed-water heaters varies with the

A SERIES DISTILUNQ APPARATUS OF HIGH EFFICIENCY.

171

quantity of water which each must handle, and this diminishes
with each step that the feed is advanced, the percentage of the
total feed handled by each heater being as follows:

TABLE V.

PsaCBNTAGB OF THR TOTAL PeED PaBSINO EACH HeATER.
(Xnmbera assigned beaten agree with those given in Fig. 1.)

Designation of Heater.

Nomberof

Braporator

Chambers

la Series.

(Nnmber of heaters one less than namber of evaporators.)

100

•

100

62
^64

100

17. From values in the preceding table, and the kno^^Ti capa-
city of a proposed apparatus, the weight of water in pounds which
will pass each heater can be ascertained, after which the tube sur-
face of each small heater may be found by use of the following
formula, A = 0.00173 TF, in which A is the required area of tube-
surface in the heater and W is the number of pounds of water
passing the heater per hour. On the basis of this formula, it can
be shown that the area of tube surface for the several small heat-
ers of a seven-stage apparatus, designed to distil 1,000 gallons of
water per hour, under the pressure range of Class A, Table I.,
should be as follows :

TABLE VI.
SquARB Feet of Tube Subface iv Small Heater.

Designation of

Square Feet of

Heater.

Heating Surface

2(1

4.8

6.8

4th

8.8

5th

10.6

6th

12.4

172 A SERIES DISTILLING APPARATUS OF HIGH EPFIOIENOY.

This will serve to indicate the significance of the small heater as
a part of the general design. Since all are relatively small, the
action of the apparatus as a whole can not be greatly affected if
all are made the same size. If in the preceding example all were
made equal to the mean value obtained by calculation, the surface
of each would be 8.6 square feet, and the area assigned the several

Pig. 45.

evaporator chambers (Table IV.) may be reduced by a like
amount.

18. Having now determined the proportions wliich must be
given the several members of the apparatus, it should be stated
that the arrangement can not always be that set forth in Figs. 41
and 4:2. In a low-pressure system, or in a high-and-low pressure
system (classes B and C), some modification in the mechanical ar-
rangement shown by the diagrams is necessary. Thus, the cham-
bers which are low in the scale of pressures must be so arranged
that the water of condensation from one chamber will flow freely

A SERIES DISTILLING APPARATUS OF HIGH EFFICIENCY. 173

to the chamber next lower in pressure. That is to say, it will be
necessary to place one element above the other. The reason for
such an arrangement is that the pressure difference between each
chamber is not sufficient to overcome the head of water equivalent
to the distance from the bottom of one chamber to the top of the
next Just how many chambers will need to be so arranged can
only be stated when. the conditions of the individual case shall
have been defined. Obviously, the cycle will not be interfered
with or the working of the apparatus changed if all the elements
are arranged in a vertical series instead of a horizontal series.

19. To the foregoing should be added the fact that the practi-
cability of the cycle described has been well established by the per-
fonnance of a small seven-chamber apparatus, which is shown by
Fig. 4S. The view makes the machine appear more complicated
than it really is, since the heaters are made up of pipe fittings. It
shows also the apparatus with most of the covering removed,
whereas in service all portions of the machine were covered with
non-conducting material. By the aid of this apparatus it has been
shown that vaporization takes place and the whole action proceeds
as has been described, even when the temperature difference be-
tween adjacent chambers is less than ten degrees, and by its use
facts already presented, with reference to capacity and efficiency,
have been substantially checked.

In conclusion, the writer finds pleasure in acknowledging the
important assistance rendered both in the analytical study and in
the experimentation which followed it, by Prof. R. S. Miller and
Mr. Charles Ducas, Junior Members of the Society.

DISCUSSION.

Mr, IJ. n, Suplee, — I should like to ask wherein this arrange-
ment of evaporators differs from the multiple-effect evaporating
pan system of Rillieux, as used in the manufacture of sugar. I
believe that some of the vessels of the United States navy are
equipped with multiple-effect evaporators which are practically
the same as those used in sugar works, and, so far as I can recall,
the arrangement is practically the same as that described in the
paper.

Prof, D. S. Jacobus, — It is stated that the losses of heat which
occur are those of radiation and those represented by the differ-
ences in the temperature or condition between the liquid supplied

174 A SERIES DISTILLING APPARATUS OF HIGH EFFICIENCY.

and the condensate delivered. In testing a large Yaryan triple-
effect evaporator, I found that, in order to get rid of air which
would otherwise accumulate, we had to lead some steam from
each effect directly into the one below it by partly opening by-
pass valves provided for the purpose. If the air were allowed
to accumulate, it would reduce the capacity of the evaporator.
This naturally lowered the eflSciency. JV^ill Professor Goss
kindly tell us whether he has noticed any similar action in the
operation of his apparatus?

Prof. W, F. M. Goss. — ^Responding to the question of Mr.
Suplee, I would say that in all other multiple-effect evaporating
systems with which I am familiar the condensation from each pan
or chamber ie drawn out separately, cooled and delivered. The
presence of the cooler leads to large losses of heat, and the series
can not be made to include very many effects. In the apparatus
under consideration^ the entire discharge, not of vapor only, but
of condensate as well, of one chamber is carried on to heat the
chamber of the next lower temperature, and at the end of the
series the distillate of all chambers is delivered and cooled in the
form of a single stream.

Our experiments have not developed any difficulty of the kind
referred to by Professor Jacobus. The apparatus illustrated in
Fig. 45 has been kept in continuous operation for periods of twelve
or fourteen hours, with no diminution in capacity or efficiency.

Professor Jacobus. — ^If the apparatus were operated with a
vacuum at the last effect, the difficulty with the air might have
been experienced. In the tests of the Yaryan* evaporator there
was a high vacuum at the last effect, and the capacity was con-
siderably reduced if it was operated an hour or so without open-
ing the by-pass valves.

Professor Goss.* — It is true, as has been suggested, that all of
our experiments have been with a high-pressure system. We have
not used a vacuum. A study of the course of the feed and of the
distillate through this apparatus is, however, reassuring, since the
distillate from all sources is constantly leading on in a single
direction, and will, I think, carry with it a reasonable amount of
entrained air. For this reason, I do not anticipate trouble of the
sort described, at whatever pressure the apparatus may be used.

It has been suggested that the value of the paper would be in-

* Author's Closure under the Rules.

A SERIES DISTILLING APPARATUS OF HIGH EFFICIENCY. 175

creased if there were added a statement concerning the cost of dis-
tilling water by means of the apparatus described. The cost
necessarily depends somewhat upon local conditions. For each
dollar cost per ton of coal, a seven-chambered apparatus of the
high-pressure type requires 7.1 cents' worth of fuel per thousand
gallons of watef distilled. Assuming such an apparatus to work
300 days in the year^. and to cost $3 per foot of tube surface, the
interest and depreciation charges at 10 per cent., the first cost
amounts to something less than 6 cents per thousand gallons, mak-
ing the total cost of distilled water, excluding attendance, which
may be little or much, 13 cents per thousand gallons. A general
statement of these facts is as follows :

Cost in cents per thousand gallons = 7.1 x cost of coal per
ton in dollars + 6.

Since the apparatus requires no cooling water, no costs can
arise on this account, a fact which, under some conditions of ser«
vice gives the series apparatus an important advantage over
others of more simple form.

The following tabulated statement constitutes a more elaborate
presentation of the facts affecting cost as based upon an apparatus
of a thousand gallons per hour capacity :

1. Namber of evaporating chambers 4 7 10

2. Gallons of water distilled per year, assnming

800 working days of 24 Lours each 7,200,000 7,200.000 7,200,000

3. Pounds of coal per year 1,670,000 1,019,000 790,000

4. Cost of coal at one dollar per ton, dollars. . 855 510 895

5. Area of tube surface in plant, feet 770 1 ,892 1,914

6. Cost of complete plant on the basis of three

dollars per foot of tube surface, dollars. . 2,210 4,176 5,743

7. Annual interest and depreciation, 10 p. o. . 221 417 57 »

8. Annual total coat, excluding attendance. . . 1,076 927 070

9. Fuel cost per thousand gallons for each

(dollar cost per ton of coal, cents 11.8 7.1 5.0

10. Interest and depreciation charges per thou-
I sand gallons, cents 3.1 5.7 7.9

11. Fuel, Interest and depreciation per thou-
I sand gallons, coal at one dollar per ton,
' cents 14.9 12.8 13.4

12. Fuel, interest and depreciation i>er thou-
sand gallons, coal at two dollars per ton,,
cents 26.7 19.9 18.9

13. Fuel, interest and depreciation per thou-
sand gallons, coal at three dollars per
ton, cents 88.5 27.0 24.4

176 SULPHUROUS ANHYDRIDE (SO,) PRESSURE TEMPERATURE CURVE.

No. 1090.*

THE PRESSURE TEMPERATURE CURVE OF SUL-
PHUROUS ANHYDRIDE {SO,).

BT EDWARD F. MILLER,. BOSTON, XAB8.

(Member of the Society.)

1. The use of sulphurous anhydride as one of the working
vapors in the " Binary Heat Engine " or " Waste Heat Engine ''
has made evident the need of more complete tables of the proper-
ties of this vapor, especially at the high pressures.

2. A series of articles by Prof. R. H. Thurston printed in the
Journal of the Franklin Institute, 1902, explains fully the prin-
ciple of this engine.

3. A paper by the writer, read before the New England Water
Works Association, printed in the 1902 proceedings and reprinted
in Engineering News of November 27, 1902, gives an account
of the engine at the Technische Hochschule at Charlottenberg
with results of tests made on the engine.

4. Volume XII. of Transaclions of A. S. M. E. contains the re-
sults of an experimental determination of the latent heat of SO,
by Professor Jacobus.

5. Messrs. Flowers and Walton, of Cornell, tabulated the work
of the various experimenters on SO and from this tabulation
constructed tables of the properties of this vapor. These tables
are the most complete of any the writer knows of.

6. The work presented in this paper was carried on under the
direction of the writer by Mr. D. D. Mohler, a senior in the course
in Chemical Engineering at Mass. Inst. Technology.

7. The apparatus in which the SO, was vaporized consisted of
an aluminimi bronze cylinder with hemispherical end and cover, 7^
inches inside diameter and 10 inches in length inside of jcover.

♦ Presented at the New York meeting (December, 1903) of the Ameriran
Society of Mechftnical Eogineers, and formiug part of Volame XXV. of tbe
Transaction$,

SULPHUROUS ANHYDRIDE (SO,) PRESSURE TEMPERATURE CURVE. 177

This cylinder was placed on end inside of a copper kettle 15
inches in diameter and 15 inches tall, which was open at the top.
The kettle was covered on the outside with magnesia to a depth of
2 inches, and the bottom rested on an electric stove.

8. Cylinder oil filled the space between the kettle and the cylin-
der to a depth of 5 inches.

9. Immersed in the oil there was a heating coil of 25 feet of No.
24 B. & S. gauge iron wire and a stirrer with vanes turned 90 as
to drive the oil from the top downwards. This stirrer was driven
'>y a ^ horse-power motor.

10. The cover of the aluminum bronze cylinder was provided
^th a gauge connection, a filling valve and a thermometer well
reaching down into the liquid SO

11. Temperatures of the SO, were determined by an Alvergnait
millimeter thermometer graduated to iV of a degree Fahr.

12. Pressures were measured by a gauge of large diameter pro-
vided with a steel tube made specially for this purpose.

13. This gauge could be read to tV of a pound.

14. The SO, used was found by chemical analysis to be 99.3 per
cent, pure with .3 per cent, of CO^ and .4 per cent. air.

15. In filling the aluminum bronze cylinder with SO, the air
was first exhausted to a vacuiun of 28 inches. Liquid SO^ was then
run in till the pressure inside the cylinder was the same as that
of the atmosphere. The cylinder was then exhausted again to
28 inches vacuum and more SO, supplied. The operation was
repeated a third time.

16. After the air had been removed eight pounds of liquid SO,
were run in. This amount brought the level of the liquid SO, a
little above that of the cylinder oil in the outer kettle.

17. The oil could be heated sufficiently in from 10 to 15 minutes
ly the resistance coil to cause an increase of pressure of 5 pounds
inside the cylinder.

18. After heating to the desired temperature the current was
turned from the coil and the electric stove under the kettle was
used to supply sufficient heat to make up for that lost by radiation.

19. By this means the pressure could be maintained nearly con-
stant for a considerable space of time.

20. A series of preliminary observations gave sufficient data to
enable one to plot the curve with considerable accuracy.

21. The final observations, covering a continuous period of 24
hours were taken as follows: At low temperature 20 to 25 readings

178 SULPHUROUS ANHYDRIDE (sO,) PRESSURE TEMPERATURE CURVE.

were taken at each pressure at 15 second intervals. The ther-
mometer readings for each set were practically constant: when the
higher pressures were reached, where the temperature change is
slight for a considerable change of pressure, it was found that the
thermometer lagged; the gauge showing a change before any
movement of the mercury could be detected.

22. To allow for this lag of the thermometer readings were be-
gun when the gauge had reached within iV of a pound of the max-
imum pressure and were continued at 15 second intervals till the
pressure had fallen to -^ff pound below the maximum pressure.

23. It was assumed that the lag of the thermometer on a falling
pressure and falling temperature would be the same as the lag on
a rising pressure and rising temperature.

24. In some cases as many as 40 observations were taken on one
determination.

25. The observations corrected for the lag of the thermometer
and for errors in the .thermometer and in the gauge are given
below :

TABLE I.
SuMif ABT OF Corrected Observations.

£ H

£ 5

£ S

£ i

0,0
§

68.20

49.41

181.98

138.70

171.00

248.61

198.71

852.12

74.88

55.35

139.11

154.40

172.58

249.65

200.01

858.01

81.04

61.77

142.86

160.92

176.51

262.71

201.21

868.80

87.05

6i.80

144.06

166.45

178.97

270.96

202.88

868.76

92.22

76.07

146.81

173 20

180.88

276.78

208.05

871.84

98.87

82.56

148.65

178.46

188.32

287.50

204.97

881.18

103 12

89.59

151.18

184.59

185.01

294.73

206 49

388.04

106 20

94.72

154.87

198.12

187.51

803.26

207.24

892.48

109 62

99.77

ir,6.63

200.01

189.24

811.26

208.26

898.08

112.82

104.57

158.29

204.80

191.29

317.21

209.80

405.89

115.05

109.01

161.94

215.58

192 58

323.64

210.41

409.72

124.72

125.75

164.22

222.78

194.19

880.54

211.14

416.74

128 74

133 29

166.71

229.21

195.23

387.25

129.95

137.17

168.53

236.08

197.14

844.70

26. The writer has plotted both these observations and those
representing Regnault's curve on plotting paper 11 feet by 4 feet

SCLPHXJROUS ANHYDRIDE (SO,) PRESSURE TEMPERATURE CURVE. 179

6 inches and after drawing a smooth curve through the points has
read from this curve the pressure corresponding to each degree.

27. At high pressure the work may be in error possibly one
j>ound, at lower pressure the error is much less.

28. The following table will enable one to compare the results
obtained by the various experimenters:

TABLE II.

ill

III

47.6
66.4
66.4
77.6
90.4
104:5

120.4
137.9
157.1

B « _

ill
'I

47.9

56 6

66.7

78.0

90.6

104.7

109.9

120.5

187.9

157.6

149

158

167

171.5

17«

194

209.8

212

218

302

• eg

•582

o - C

8 * t

177.8
200.8
226.1

258.5

PI
III

210.4

265.9
.880 8

408.9

610.9

1.050.3

t i> o

lae

Sirs

180. C

251.7
830 ! 8

a 2 *i
IP

178.0

£8.

179.6
203.8
230.7
245.2
260.5
831.1
402.0
418.0

29. Uegnault experimented between the temperatures — iO de-
grees Fahr. and 149 degrees Fahr.; Sajotschewski between 122
degrees Fahr. and 302 degrees Fahr., determining 8 points. This
curve is above the others up to about 200 degrees Fahr. where it
apparently crosses; Blumckc between 95 degrees Fahr. and 209.3
degrees Fahr., determining 5 points.

180 SULPHUROUS ANHYDRIDE (SO,) PRESSURE TEMPERATURE CURVE

^20

400

aoo

340

8S0

aoo

200

: !>RESSURE-TEMPERATURE
aJRVEloF

sOlphur dioxide.

240

g220

' 200
180
IfiO
140
120
100
80
60
40

20 40

MUler,B.r.

80 100 120 140 160 180 200

Temperature Am.Bmnk am* o».,.v. r.

Fig. 46.

SULPHUBOUS ANHYDBIDB (SOj PRESSURE TEMPERATURE CURVE. 181

TABLE UI.

Saturated Sulphuboub Anhtdridb.

Tbxfskatukxs ahb PiuuflURxs CoRBSspoNDiNo AS Rbad fbom thi Plot ; NO Attimft Haob

TO CORBBCT YaLUSB BY MaKINO UbB OF A TaBLX OF SBCOMD DiFFIRKNCKS.

£ S

1^1

^ <

182

-40

3.1

11.2

80 7

70.8

189.9

175

267.0

-89

3.2

11.6

81.4

71.6

188

141.9

176

260.5

-88

3.3

11.8

82.1

72.8

134

148.9

177

264.0

-87

3.4

12.1

82.8

74.1

185

146.0

178

267.6

-36

3.6

12.4

83.5

75.4

186

148 2

179

271.2

-35

8.7,

12.7

84.2

76.7

137

150.5

180

274.9

-84

3.8

18.1

84.9

78.0

138

152.8

181

278.6

-83

4.0

18.4

85.6

79.8

139

155.2

182

282.4

-83

4.1

18.7

86.4

80.7

140

157.6

183

286.8

-31

4.3

14.0

87.1

82.1

141

159.9

184

290.2

-SO

4.8

14.4

87.9

88.5

142

162.8

185

294.2

-29

4.5

14.8

88.7

100

84.9

148

164.7

186

298.2

-38

4.7

15.2

89.4

101

86.3

144

167.1

187

802.2

-27

4.9

15.6

40.2

102

87.7

145

169.5

188

806.2

--26

16.0

41.0

108

89.1

146

171.9

189

810 2

-25

5.1

16.4

41.8

104

90.6

147

174.4

190

814.8

-24

5.2

16.8

42.6

105

92.1

148

176.9

191

318.4

% -23

5.4

17.2

48.4

106

98.6

149

179.5

192

822.6

-22

5.5

17.6

44.8

107

95.1

150

182.1

193

826.8

-21

5.7

18.0

45.2

108

96 7

151

184.8

194

331.1

-20

5-9

18.4

46.1

109

98.3

152

187.4

195

885.4

-19

6.0

18.8

47.0

110

99.9

158

190.0

196

889. H

-18

6.1

19.8

47.9

111

101.5

164

193.7

197

844.2

-17

6.3

19.7

.48.8

112

103.1

155

195.4

198

848.7

-16

6 5

20.1.

49.7

118

104.7

156

19d.2

199

a'i8.2

-15

6.7

1 28

20.6

50.6

114

106.8

157

201.0

200

857.7

—14

6 9

1 29

21.0

51.6

115

108.0

158

2a3.8

201

362.8

-13

7-1

21.5

52.6

116

109.7

159

206.6

202

367.0

— 12

7.8

22.0

58.6

117

111.4

160

209.6

208

871.8

-11

7.5

22.5

54.6

118

118.2

161

212.4

204

876.7

—10

7.8

28.1

65.6

119

115.0

162

215 3

205

881.6

—9

8.0

23.6

56.6

120

116.8

163

218.8

206

886.6

—8

8.3

24.2

67.7

121

118.6

164

221.4

207

891.7

—7

8.5

24.7

68.8

122

120.5

165

224.5

208

896.7

— 6 1

8.7

26.8

59.9

128

122.4

166

227.6

209

401.8

-^ 1

9.0

25.9

61.0

124

124.8

167

280.7

210

407.0

-4 1

9.3

26.6

62.1

125

126.2

168

283.9

211

412.5

—3

9.5

27.1

68.2

126

128.1

169

237.1

212

418.0

-2

9.7

27.7

64.8

127

180.0

170

240.8

—1

10-0

28.2

65.5

128

181.9

171

248.6

10.8

28.8

66.7

129

:83.9

171

246.9

10.0

29.4

67.9

180

185.9

178

250.2

10.9

80.1

69.1

131

137.9

174

258.6

182 SULPHUBOUS ANHYDBIDE (sO,) PRBS8UBK TEMPERATURE CURVE.

DISCUSSION.

Mr. S. A. Moss. — Investigation of Professor Miller's table of
final results, Table III., seems to indicate that he has given more
weight to Eegnault's values than to his own in the region where
the experiments overlap. In this region Professor Miller's pres-
sures, as shown by the unsmoothed values given in Table I., are
higher than Eegnault's. Would it not be better to give a table
of the smoothed values actually obtained by Professor Miller
instead of Table III., which gives values from 68 degrees to 149
degrees lower than he obtained?

It has been interesting to compare Professor Miller's values
vntii those which would have been predicted by Kamsay's and
Young's general law for vapor pressures. This law, which is not
as generally known as its importance deserves, gives a means of
computing the vapor pressure corresponding to any temperature,
if a few values of corresponding pressure and temperature aro
known.

In a modified form (see Physical RevieWy Vol. XVI., No. 6,
June, 1903) the law is as follows: Let Tx be the absolute tem-
perature corresponding to any vapor pressure for substance x,
and Tw the absolute temperature corresponding to the same pres-

sure for water vapor, as given by Steam Tables. Then •=- =

a 7= b where a and b are constants, different for each sub-

stance. The values given by Professor Miller in Table I. follow
the law exactly up to about 157 degrees. The constants as com-
puted for this region from Table I. give as the equation for SO^y

^=1.6667 ^-.00036.

Absolute zero is taken as — 459.5 degrees Fahr. If Ramsay's
and Young's law holds true for all values, this equation gives the
saturation temperature for any pressure whatever if we know
the saturation temperature for water vapor (steam) at that pres-
sure.

As stated, the values given by Professor Miller in Table I.
follow this law exactly up to 157 degrees. Beyond this point there
is a slight departure, the pressures given by the law being lower
than those given by Professor Miller. This departure is noth-

SULPHUROUS ANHYDRIDE (SO,) PRESSUB£ 'WaiPBRATURE CURVE. 183

ing at 157 degrees and gradually increases. For 407 pounds
pressure, Professor Miller gives 210 degrees, while the tempera-
ture predicted by the law is 215.2. This discrepancy indicates
either that Eamsay's and Young's law does not hold for the higher
pressures of SO, or else that Professor Miller's higher pressures
are somewhat too great

Professor Miller *— The values given by Eegnault between 68
degrees and 149 degrees were given equal weight with points
determined by these experiments.

If these had been disregarded, the pressures at the lower end
of the curve would have been from .1 to .2 pound higher. As
Kegnault's values were used below 68 degrees, it seemed best to

draw one smooth curve through all the points.

I have read with much interest the article by Mr. Moss in

Physical RevieWy Vol. XVI., No. 6, and had intended to make

a qalculation to show the agreement. I can offer no explanation

as to the cause of the variation cited.

♦ Author's Closure under the Rules.

184 THB prroT tubb.

No. IMl.*

THE PITOT TUBE.

BT W. B. OREOOBT, KXW OBLKAMB, LA.

(Member of the Sociecy.)

1. The object of this paper is to call attention of the niHembers
of the Society to a simple, efficient instrument for measuring the
yelocitiefi of fluids.

This instrument was invented by Pitot in 1730, the form first
used by him is shown in Fig. 47. It consisted of a glass tube,
bent at right angles, one end of which was placed in a stream of
moving liquid, the end facing the direction of motion squarely
with an orifice less in area than the tube. The impact due to
velocity forced the liquid up the vertical portion of the tube
to a height h and gave a measure of the velocity according to
the law V = ^2gh. A second tube open at the bottom, was
added later; the two tubes being placed in a groove in a tri-
angular prism of wood. Fig. 48 is a copy of the drawings of the
tube as described by Pitot. f E is the impact opening of this tube,
placed at one angle of the triangle forming the cross-section
of the prism. Suitable scales FO and LI were made on the
metal slide which is provided with clamps so that it could be
raised or lowered as the ends of the tubes were placed at the
desired depth; from these scales were obtained the difference of
level of the water in the two tubes and direct readings of veloc-
ity respectively.

2. The Pitot tub^was mentioned in a paper read before the
French Academy of Sciences by M. Dubuat in 1753, and the sug-
gestion offered that the tubes be made of tin, and that they contain
floats carrying a rod moving in front of a scale from which the

* Proflented at the New Tork meeting (December, 1908), of the American
Soeiety of Mechanical Engineers, and forming part of Volume XXV. of the
Drantiietioiu.

t Butoire de VAeadimiie d$$ ScUneet, 1782, page 876.

THB PITOT TUBB.

186

heights could be read. The instrument was improved and used by
Darcy and Bazin ; Fig. 49 shows the form used by them. It con-
sists of two tubes, one drawn to a fine point and pointing up
stream, the other with small openings on top and bottom at the
point a as shown in Fig. 49. The first of these is the impact tube
and the second the pressure or static tube; the tops of the two tubes
are connected so that a partial vacuum formed above the liquid
in them will cause the liquid in both to rise by the same amount

Pio. 47.

to a height convenient for reading. The instrument being in
place, the air is exhausted, having opened the stop cocks R and fi',
until the liquid is raided to the desired height when R is closed.
In using the instrument to detcnnine velocity, more or less vibra-
tion occurs in the two columns depending somewhat on the uni-
formity of the velocity being measured, the size of the tubes, their
openings, etc. Maximum and minimum readings may be observed
and a mean computed, or a mean reading may be determined
directly from the instrument. If the openings are small, the
vibrations will be small ; partially closing the cock R^ has the eCFect
of causing the instrument to average the readings. Finally R^
may be quickly closed at the mean reading and the difference of
level read with deliberation. The researches of Darcy and Bazin

186

THE PITOT TUBS.

F L

G I

D B

ilm. Am* J«M« Ck.,jr. T.

Fio. 48.

THB PITOT TnBE.

187

Am.MmmH^UO».,ir.r.

Fid. 49.

190

THB PITOT TUBB.

that the form of the impact tube did not influence the constant
of a Pitot tube in case the opening be placed at the centre of a
surface of revolution.

10. These tubes having nozzles 4, By C and D of Fig. 50 were
then placed in front of a catamaran and hauled through still water
in an open canal, the velocity being obtained from the time re-
quired to pass over a measured course, and the height from the

1 !

•

SnhtailwkM.

f M S A •
Oeatlmeten

Fio. 50.

surface of the water, to which the liquid was forced, accurately
measured from the surface of the canal. The constant ^ to be
used in the formula v = ^ V^ff^ as obtained from these experi-
ments was 1.0053 for nozzle JD.

11. Experiments were next conducted which showed that with a
properly designed static opening the constant of a tube is prac-
tically unity. The form of static opening as used by Mr. White is
similar in principle to that used by Darcy and Bazin. An ordinary
^-inch gas pipe about five inches long was drawn to a point and a
^inch hole drilled through its sides about four inches back from
its pointed end, in a horizontal plane. Careful calibrations of tubes
having this form of static attachment showed an average constant
of unity well within the error of observation. Other tubes were
experimented with, in which there was a suction action on the static
side ; this being added to the head due to impact made the head,

as read, greater than the head due to velocity or 5- and therefore

gave a coefiicient tp less than unity in the formula v = ^ V^ffh,

THE PITOT TUBE. 191

12. Mr. White sums up the results of his researches as follows :

(1) That an impact tube, whose impinging surface is one of
revolution, converts velocity head into static head exactly accord-
ing to the law v = ^2gh whatever the pressure of the surround-
ing fluid.

(2) That only pressure openings which give the true static head
of water should be used in connection with the point of a Pitot
tube. That is to say, that only tubes which have unity as their
coefficient should be used.

(3) That Pitot tubes whose constants are unity in open canal
ratings will remain unity, whatever the pressure of the liquid.

Mr. White had very few experimental data to support conclu-
sion (8).

13. A voluminous paper * of great merit by Messrs. Williams,
Hubbell and Fenkell appeared in 1902, the title is " Experiments
at Detroit, Mich., on the Effect of Curvature Upon the Flow of
Water in Pipes." This paper, together with its discussion, is one
of the most valuable contributions to the subject of the Pitot
tube, as all measurements of velocity were made with various
forms of that instrument. A large portion of the work con-
sisted in the rating of these tubes under various conditions; sev-
eral different forms of the instrument were used. Among the
authors' summary of results accomplished are the following :

A. '^ The invention of the oil differential gauge, by which it
is possible to observe differences of head in closed conduits under
pressure with as great a degree of precision as is attainable in
open conduits with the hook-gauge."

B. " The invention of a form of Pitot tube which can be in-
serted in a water main without the aid of special devices, other
than the tools possessed by every water department ; by the aid of
which a competent observer may obtain gauging with as great ac-
curacy as has yet been attained with any other measuring device,
except the graduated tank and weighing scale."

C. " The determination of a ratio .84 between the mean and
maximum velocities of water flowing in closed circular conduits,
under normal conditions, at ordinary velocities; whereby observa-
tions taken at the centre under such conditions, with a properly
rated Pitot tube, may be relied upon to give results within 3 per
cent, of correctness."

* " TransaeHom American Societj of avil Engineers," Vol. XLVII., April,
1908.

192 THB PITOT TUBE.

D. " The presentation of a series of coefficients for application
to the different fluids used in the fluid differential gauges, by
which the observations so taken may be conveniently reduced to
equivalents in water."

E. " The demonstration of the fact that ratings of Pitot tubes
made by dragging these instruments through still water in open
troughs do not conform, within any reasonable limits to those
obtained when the instrument is stationary in moving wat^r in
a closed conduit."

F. " The demonstration that Pitot tubes must have their co-
efficients determined, whether they consist of point opening alone
or both point and pressure openings."

G. " The demonstration that under some conditions, in straight
pipe there is a difference of pressure at different points around
the circimiference of the same cross-section."

//. " The derivation of the ellipse as the approximate form of
the normal curve of velocities in straight circular pipes."

14. As was to be expected the discussion of the paper brought
forth many valuable contributions on both theoretical and experi-
mental sides. Among those who discussed the Pitot tube are to
be found the names of many of the highest authorities on hy-
draulics. The paper with its complete discussion must be read to
be appreciated; only a few points, which have a direct bearing
on the subject in hand, will be mentioned.

15. In the final discussion of this paper by its authors, after
carefully weighing the contributions and criticisms of the various
men who discussed it, we find the following :

" The weir, the current meter, the water meter, and the nozzle
are the only other devices to be compared with the Pitot tube,
and the results obtained with the first three are rarely if ever
better than those presented herein, and two of them are not
directly applicable to the flow of water in pipes. The nozzle may
perhaps give somewhat better results, but it is limited to compara-
tively small streams. It seems therefore that the writer's conclu-
sion, that the Pitot tube, properly rated in the hands of a skilled
observer, is as accurate as any other device for measuring water in
a closed conduit, except the scale and tank, is fairly sustained."

16. " The writers, therefore, believe that conclusion E is sus-
tained, and that an instrument should be rated under conditions as
nearly as may be similar to those under which it is to be used, and
that dragging an instrument through still water does not conform

THB PITOT TUBB. 193

with this condition when the apparatus is to be used to measure
ranning water, either in closed or open channels. In the light of
present knowledge, if instrument can only be rated by dragging,
the writers would prefer the coefficients established by dragging in
opposite directions in running water to those obtained by dragging
instill water."

17. "In view of the data presented by Mr. White and of the ex-
periments of Messrs. Saph and Schroder, and Adams and Wilson,
wherein instruments of certain forms were used with ring piez-
ometers, as well as the experiments of the writers regarding
obliquity it seems to be demonstrated satisfactorily that conclu-
sion F was too sweeping, and the writers agree as pointed out
by Mr. Ferris, a proper form of point opening will have a coeffi-
cient of unity by the formula A =-^ when combined with a ring

piezometer under normal or nearly normal flow."

18. Again in discussing result G they say, under head of
"Pressure variation," "Conclusion (?,that the pressure in a straight
pipe is not always the same at all points around the circumference
of the cross-section, is based first upon the experimental evidence
in the 16-inch pipe investigation. No discussion has been sub-
mitted containing direct additional evidence upon this particular
point All the Pitot-tube work shows that the components of vel-
ocity parallel to the axis of the pipe vary from point to point
throughout the cross-section, and that changes of that component
of velocity take place at each individual point under distorting in-
fluences. If these components represent or are proportional to,
the total velocity in each part of the fluid, it follows from the T^w
of Conservation of Energj-, since all the water in the cross-section
was originally started in its course under the influence of the same
head, that if velocity head be increased, pressure must be de-
creased, unless the elevation of the stream above datum be de-
creased, whether a single particle or the whole stream be con-
sidered, and therefore, that the pressure must vary throughout
the cross-section, a conclusion which is confirmed by the moat
obvious interpretations of a large number of experiments. The ex-
periments of Messrs. Adams and Wilson, and Saph and Schroder,
as well as some of those of the writers, prove pretty conclusively
that a properly formed point combined with a ring piezometer
gives readings the summation of the deducted velocities of which
amount to the true discharge, on the theory used by the writers

194 THB PITOT TUBS.

that A = -5—, without introducing any coefficient, as has been

demonstrated in the case of a jet and in open channels by Mr.
White. This being true, the acceptance of varying pressure
throughout the cross-section, which pressure varies inversely as
the velocity leads to some complications, if one undertakes to

maintain that the point effect is — -. Another interesting circum-
stance is the reading given by two conical points, one receiving
the impact of a current in a closed pipe and the other directed
down stream. Theoretically it might be expected that the differ-

ence between these readings would be 5— and this alleged fact

is the basis of the claim for a patent granted for such device some
years ago, but, experimentally, or practically, nothing of the kind
happens with his apparatus."

19. " Kegarding the question of the proper formula for the Pitot
tube, the writers are not yet entirely satisfied, and while fully
appreciating Mr. Frizzell's effort to clear the matter up, they yet
prefer to leave the question where they left it before, that, prac-
tically, it makes little difference in the reduction of results whether

«* •«'
a tube reading actually is ^ or -. The discussion presented by

Mr. Seddon gives excellent reason to believe that there is a large
account of internal forces to be balanced before the laws of flow
will be fully understood, and until these forces have been more
fully investigated, too rigid laws had best not be laid down.''

20. In the discussion of this paper there is a contribution by
Edward 8. Cole in which he describes a series of independent ex-
periments conducted by him. A device, which he has named a
Photo-Pitometer, was invented by him and used to obtain a con-
tinuous record of the readings of a Pitot tube, extending over a
period of time. To accomplish this result clockwork, lamp and
photographic paper were used. For descriptions see original
paper.*

21. It appeared that the oil differential gauge had been invented
and used by independent investigators in at least five different
localities and that in one case had it been patented.

22. In quoting from an extensive work like the one just re-

* Vol. XLVIL TroMaetianB American Society of Gvil EDgineers, AprU, 1903,
page 275.

THB PITOT TUBB. 196

f erred to, there is somei^ danger of conveying a wrong impression
by merely giving those parts to which it is desired to direct atten-
tion. The writer wishes to disclaim any such intention and to state
that the particular parts quoted have been selected because they
Lave a direct bearing on the theory and use of the Pitot tube
and are along the lines on which it is wished to discuss some fur-
ther results which possibly throw additional light on the subject.

23. During the fall and winter of 1902 the writer was consult-
ing engineer in a series of tests of the Hydraulic Dredges of the
Mississippi River Commission.

24. By means of these dredges a navigable channel is main-
tained in the river during low water. Large centrifugal pumps,
driven by direct connected steam engines and capable of maintain-
ing a mean velocity of from 15 to 22 feet per second in the dis-
charge pipes, are used as dredge pumps.

25. The series of tests, which extended over several months, in-
cluded efficiency tests of boilers, engines and dredge pumps besides
tests of accessory machinery such as " jet pimips " used to loosen
the material to be removed in dredging. Some of these " jet
pumps " were of the reciprocating steam pump type while others
were centrifugal pumps driven by direct connected engines.

26. A complete report in detail of these tests can be found in
the Report of the Chief of Engineers of the U. S. Army for 1903.
It is in the report of F. B. Maltby, U. S. Assistant Engineer,
Member of the American Society of Civil Engineers, Superin-
tendent of Dredging Operations for the Mississippi River Com-
mission, and forms a part of the report of Captain W. B. Ladue,
Corps of Engineers of the U. S. Army, Secretary of the Mississippi
River Commission.

27. The general method of these tests was outlined by the
writer who was present while some of them were made. The de-
tails were worked out and the tests conducted by Mr. Maltby, who
deserves much credit for a work of great magnitude in which
success was attained only by untiring labor and careful attention
to details.

28. The results are aU of great interest, however. In the dis-
cussion which follows reference will only be made to those parts
of the work which have a direct bearing on the subject in hand,
viz. : the Pitot tube.

29. The problem of measuring the mean velocity of the water
in the discharge pipes was of prime importance as the amount of

196 THB PITOT TUBB.

water pumped and the eflSeiencies of pumps both depended on
the accuracy of measuring this velocity. The quantity of water
in some cases exceeded 120 cubic feet per second. The diameter
of discharge pipes were approximately 32 inches in some cases
and 34 inches in others. Seven dredges were tested; the method
used must be applicable to all. Had a weir been used it would
have been available for only a short time, owing to a rapidly
rising river. A Venturi Meter would have introduced a resistance
in the discharge pipe, and rendered the conditions quite dissimilar
to those of ordinary running, which were desired for these tests.
The velocity to be measured at the centre of the pipe was in some
cases as great as 25 feet per second. The loss of head at the
throat of a Venturi Meter, while small at low velocities, would
be enormous at the velocities used even if a meter considerably
larger than the size of pipe were used.

30. A measuring barge into which the water could be deflected
for a brief interval of time, and measured, had been used in test-
ing these dredges in 1897. A complicated mechanism was re-
quired to deflect the stream of water and the short interval in
which the discharge was measured, together with the uncertainty
of getting the time exactly were the objections to this scheme.

31. All these methods, besides being expensive, are difficult to
apply, partly because the discharge pipes of some dredges are
half submerged in the river, the pontoons being so designed —
while in other cases the pontoons support the pipes 30 inches
above the water.

32. The Pitot tube method was recommended and used and it
is believed that the results obtained are as accurate as could be
had by any of the other methods suggested; besides ease of ap-
plication, cheapness, and the fact that the pumps could be tested
under conditions exactly similar to ordinary running, are all in
favor of this method. The tubes used were designed by Mr.
Maltby, they were made in the machine shops aboard the dredges
as were the gauges which were used in connection with them.
Nine tubes were constructed, but all except Nos. 1, 3, 8 and 9
were discarded because of apparent defects.

Description of Pitot Tvhes.

33. Tube No. 1 consists of two pieces of J-inch brass tubing, en-
closed in a pipe lA inches outside diameter. One of the small

THE PITOT TUBE. 197

brass tubes is bent at the end below the end of the outside en-

closmg pipe, through 90 degrees and the plane of the opening

made truly parallel with the upright pipe to form the impact

opening. The other brass tube is brazed to a solid brass piece,

circular in section and placed at right angles to the tube, the

upstream end has a sharp point which is even with the opening

of the impact tube and below it; it has a iV-inch hole drilled on

each side connecting with the interior of the upright tube, and

fonns the static opening: ordinary :J-inch air cocks form the

upper side of these tubes; ^-inch rubber tubing about 4 feet

long is used to connect these air cocks to the gauges.

34. Lead or cement is used to hold the small tubes in place in
the outer tube. A stuffing box was used through which the outside
tube slides ; it is screwed into a hole tapped in the top of the pipe
with a 1^-inch standard pipe tap. A handle at the top of the tube
carefully set at right angles to the plane of the impact opening
was used to move the tube vertically from top to bottom of the
pipe in traversing and enables the tube to be kept in alignment
with the axis of the pipe.

35. Tube No. 3 is similar to No. 1, with the exception that the
impact point is below the static point.

36. Tube No. 8, and in fact all the different tubes have the same
size vertical outside pipe to permit the use of the same stuffing
hoxes. At its lower end it is joined at right angles to a 1-inch
pipe about 18 inches long, drawn down to a point of the proper
size to admit a J-inch brass tube which is brazed fast into it f orm-
iiig the impact opening. The small tube runs inside the larger
pipe to the top, where it terminates in a J-inch air cock. The
horizontal pipe has three i^-inch holes drilled in its side, form-
ing the static openings; the interior of the pipe is connected
through the upright and one handle to a :J-inch air cock.

37. Tube No. 9 was made as nearly like No. 8 as could be done
ky a skilled mechanic.

It was very soon apparent that the tubes Nos. 1 and 3 did
iiot have a constant ^of unity in the formula v=z g} ^/2gh.

38. It was believed that tubes No. 8 and 9 would give a value
of unity, but to prove this point was another matter. It has been
shown conclusively by Mr. White that an impact point similar to
Aat of tubes Nos. 8 and 9 would register the level correspond-

uig to a given velocity according to the law ^ = 5- when the

198 THE prroT tube.

tube was dragged through still water in an open canal. Messrs.
Williams, Hubbell and Fenkell, as we have seen, agree that a
" point combined with a ring piezometer gives readings, the sum-
mation of the deduced velocities of which amount to the true

discharge " on the theory used by them " that ^ = s^ without

introducing a coefficient. If then it can be shown that these
tubes when calibrated in running water in an open channel ^ve a
constant 9> of unity in the formula v = ^y2jrA and also shown
that the static readings of the tubes correspond to the readings
of the ring piezometer, it would seem that the question of the
value of the constant may be settled in this way. Furthermore
it has been claimed that the constant of a tube rated in an open
channel is different from that obtained by rating in a closed con-
duit under pressure. If this is true, tubes Nos. 8 and 9 being
exactly alike and having the same constant when used simultane-
ously to traverse two sections of a discharge pipe at different
points where the mean velocity of the water is necessarily the
same, but the static pressure very different will show decided
differences in their traverse and in the consequent mean velocity.
We shall see what was obtained by the experiments.

39. Tube ^o. 9 was rated in running water having a velocity of
about 3^ feet per second by comparing it with floats; about 80
floats were run and 600 observations made.

40. The tube was suspended in the Mississippi River with the
point about 12 inches below the surface, facing the current. An
inverted U-tube of glass was used, the air being partially ex-
hausted from the top to bring the water surfaces to a convenient
height for reading.

41. The floats were 24 inches long; the time was observed in
passing over a course 40 feet long, the tube being placed about
one-third the distance from the upper end of the course.

42. In the table which follows the velocity by floats is in each
case the mean of ten observatons, and that given for the Pitot
tube is the mean of about 75 observations :

No. Velocity of Floats. Velocity by Tube.

1 8.316 8.420

2 3.284 3.279

8 8.617 8.898

4 8.419 8.868

6 8.405 8.502

6 8.212 8.228

7 . 8.862 8.854

8 8.882 8.218

Means 8.866 8.344

THB PTTOT TUBB.

199

The means show a coeflScient of unity within about one-third
of 1 per cent.

43. Tubes 8 and 9 were placed in the same discharge pipe 50
feet apart, and three sets of observations made; the tubes were
then changed about and three sets of observations taken again,
with the following results. Each set is the mean of ten readings.

Velocity b7 Tube No. 8.
26.37
24.16
25.77

Velocity by Tube No. 9.
24.60
26 60
26.14

Means.

25.86
26.37
26.55

26.268

IteTeried Poaltioae.

26.19
24.84
26.02

26.048

The means agree within about eight-tenths of 1 per cent.
44. To test the accuracy of velocity measurements by these two
tubes and determine whether or not they were affected by static

}CHs

net

£ U.

hDi

fCDi

£ E

*8tL(

•

Tubi

9 in

8 ••

POfI

ooti

te.i

, 19

72.

350

ft. I

€tUf4

M T

tbea

— o

%—

VEi

OCI

K, /

T. P

E8 I

t£

■^

-J

"=5

•^

)

Ul.

-5

-_j;:^

-»=

f — -

•««i

r9 0t

8ta

tic 8

Id0€

fTu

^Mi

. 0.

- 4.

12/

•Of

Wat

,8.

'19.

78"

FiQ. 61.

pressure, tube 8 was placed in the discharge pipe of the second
pontoon of the U. S. Dredge Epsilon, where the average static
pressure was 19.78 feet of water, and tube No. 9 was placed in

200

THB PITOT TUBE.

the ninth pontoon, 350 feet from tube No. 8, and where the
average pressure was 4.12 feet of water. Traverses were made
by taking observations simultaneously at different points across
the pipe. The mean velocity determined by tube No. 8 from 170
observations, was 22.321 feet per second, wjiile that determined
by tube No. 9 was 22.351 feet per second, a difference of one-
tenth of 1 per cent. The mean velocity was obtained from the
traverses in a vertical plane, by dividing the area of the pipe into
ten equal areas the boundaries of these areas being circles con-

Fio. 58.

centric with the axis of the pipe; the average of the velocities
found in these ten areas was used as the mean velocity for the
whole cross-section.

45. A plotting of the traverse referred to is shown in Fig. 51.
A comparison was made of the static pressure indicated by tube

No. 9 with that obtained by means of piezometers in the sides of
a 32-inch pipe in the following manner :

46. Four ^inch cocks were placed in the sides of the pipe as in-
dicated in Fig. 52 at the points of a, 6, d and e. Great care was
taken to have the axis of these cocks exactly normal to the axis
of the pipe; their ends projected inside the pipe slightly, and after
being screwed into place these ends were filed off carefully to
present a perfectly smooth surface, flush with the inside of the
pipe. When each of the piezometers was connected in turn with

THB PrrOT TUBB.

each other through a differential gauge no difference in pressure
could be observed. Tube No. 9 was then inserted into the same
vertical section as the piezometers, the point of the tube being
at the centre of the pipe. The static side being connected suc-
cessively with each piezometer, through a differential gauge, no
differences of pressure could be observed. The pressure of the
four piezometers and the static side of the tube was exactly the

same.

47. In Fig. 53 are shown four traverses of the discharge pipe of

AV£

fi8&

r 19/

k PI

TUB

e
+

000,

r//<M

DIM

r^{

§t P> mtd

fi2,
3

Tub*

8.
8

^Tfi

tury

»»

ft
It

9
8

1 1

6 it i

7 i

^ 1

^ I

i^ 9 '

) ]

^ 5

i i

b J

t ^

I A

s s

■^•<

s^s

^"^

lo&

rw^

jj^

L*^

^!^

p*r

_,^

1
o

3»*=

S'lta

?fi>

-ft

-»

%<h^.r.

Fie. 58.

the TJ. S. Dredge Epsilon ; the velocity curves and the static press-
ure are given. The static pressures are the actual pressures at
the j)oints of observation; it will be seen that they are greater
at the bottom than at the top of the pipe by the amount of the
static pressure due to a head equal to the diameter of the pipe.
Irregularities ai:e no doubt due, in part, to the practical impos-
sibility of keeping steam pressure and consequently the revolu-
tions of the engines and pumps absolutely constant. Each point
plotted is in general the average of ten readings, considerable
time being required to complete a traverse, as much as an hour in
soiue cases.

202 THB PITOT TUBB.

48. These readings were obtained by using two ordinary U-
gauges and observing impact and static pressures separately. The
work required by this method is much greater than when a differ-
ential gauge is used and the difference of level of the mercury col-
umns only is taken. The latter gives readings from which velocity
can be computed, but does not show the way in which the pressure
is distributed across the pipe. In case separate static and impact
readings are desired, two U-gauges must be used, and while one
side of the mercury column in the U-tube is open to the atmos-
phere, the side connected to the tube has a solid column of water
resting on it. Great care has to be exercised to expel all air, and
to be sure that the water column is solid before readings are taken.
Knowing all the dimensions of gauges, and consequently the
relative height of the mercury columns, with reference to the
point being investigated, the absolute pressure of that point is
easily computed. With the differential gauge, when velocity only
is wanted, the two ends of a U-tube containing mercury are con-
nected to the impact and static openings of the tube. Care must
also be exercised in getting solid columns of water on both sides
in this case before reading; the true difference of pressure in this
case, is that due to the differences of level of the mercury, dimin-
ished by a pressure due to an equal height of water.

49. After the ratio of velocity at the centre of the pipe, to mean
velocity, had been determined by repeated traverses, observations
were taken at the center only, while testing the main pumps. The
value of this contstant varied somewhat, due to local conditions
at the points where traverses were made.

50. Referring again to Fig. 53 a semi-ellipse has been drawn,
using dotted lines; it will be seen that it represents a fairly good
average of the observed velocities; this is more apparent when a
larger number of traverses are plotted.

51. It is seen from these results that the static pressure does not
vary across the section of a straight pipe in which there are great
differences of velocity parallel to the axis of the pipe, except as
affected by gravity. This shows conclusively that the sum of
static and velocity heads for various points in, the same cross-
section of a straight pipe, is not a constant quantity if we under-
stand by velocity head, that due to the velocity parallel to the
axis of the pipe. This ought not to destroy our belief in the
Law of the Conservation of Energy. All the energy possessed
by a particle of water at any point of the cross-section of a straight

THB PITOT TUBE.

, ^tV-

N^ %.

No. 5.

No. 2.

No. 3,

Fig. 54.

204 THB prroT tubb.

pipe IS either energy or position, pressure or motion. Since the
pressure energy istjonstant across a given section and the energy
of motion parallel to the axis of the pipe varies, it follows that
there must be energy of motion other tiian this. Unfortunately
we have no way of measuring the velocity of the whirl of the
particles of water at various points in the cross-section. If this
could be done, undoubtedly it would be found that the energy
due to velocity of whirl is greatest at the walls and least at the
centre of the pipe, and that the energy possessed by a particle of
fluid at one point in the cross-section is equal to that possessed by
any other particle in the same section.

52. It is impossible to convert this energy due to velocity of
whirl into useful work ; of course it is finally converted into heat.

Tube Nos. 1 and 3 were used in important tests; they were
rated by comparing with tube 'No, 8 by placing them in the same
discharge pipe 50 feet apart ; three sets of observations of ten
each were taken, then the positions of the tubes reversed and three
more sets taken. The coefficients of tube No. 1 was found to be
.930 and of tube No. 3 .8915. These values were used in reduc-
ing observations taken by these tubes.

53. The cut shows the discarded tubes.

Tube No. 2 had an impact opening similar to tubes 1 and 3.
The static opening was a vertical brass tube, with lower end cut
off squarely, the axis of the tube being at right angles to the
cttrrent measured. This tube gave the greatest amount of suction
at the static opening, found in any of the tubes. The suction
was so great that when the impact side was made to face down
stream the reading of the impact side was still greater than that
of the static side.

Tube No. 4 was tube No. 2 with the lower end of the static
tube plugged up and with openings on the side.

Tube No. 5 had impact and pressure points similar, but point-
ing in opposite directions.

Tube No. 6 was very similar to tube No. 1, except that lai^e
tubes were used, the impact being filled with a plug having a hole
^-inch in diameter.

54. No rating was made to determine the coefficient of these
tubes, except No. 6, as a search was being made for a tube having
a coefficient of unity, and it was soon evident that this was not the
case with any tubes, except Nos. 8 and 9.

55. The use of the Pitot tube is by no means confined to meas-

THB PITOT TUBS.

205

uring the velocity of water and liquids, but has been used to meas-
ure the velocity of air and gases. Professor Cai^nter in his work
entitled " Heating and Ventilating Buildings," page 41, describes
a tube to be used to measure the velocity of air. In the same
volume, page 45, reference is made to the Prussian Mining Com-
mission which, in ^884, by means of a large gasholder which con-
tained 70,000 cubic feet, investigated several questions relating

OpenOaire

Fio. 55.

to the measurements of the velocity of air. To the question:
" Can the Pitot tube be applied practically for measuring the
speeds of air, and, if so, what formula should be used for calcu-
lating the speed and quantity of air ? " and affirmative answer is
given and a formula to be used.

56. The instrument has also been used to measure velocities of
air in a series of tests of the greatest refinement conducted by
Capt. D. W. Taylor, U. S. N., at the Experimental Model Basin,
Washington Navy Yard. Captain Taylor used several tubes in
the same cross-section so arranged that the mean velocity was the
arithmetical mean of the several velocities given by the tubes.

57- Pitot tubes were adopted for these tests after extensive ex-

THB PITOT TUBB.

I
I

rn^l

-ffi

op e

THE PITOT TUBE. 207

periments with anometers. Static and impact pressures were ob-
tained separately by means of instruments of great precision.
The form of tube used is shown in Fig. 56 which gives a section
through the centre of the tube. The static openings consist of
two slots on opposite sides of the tube ; one of these is shown in
the cut. It was assumed that the constant <py for this tube^ is unity
in the formula v = tps^J^gh.

58. An account of this work will be published in the near
future by Captain Taylor.

59. The three final conclusions of Mr. White, in the discussion
of his paper were :

(1) " That an impact tube, whose impinging surface is one of
revolution, converts velocity head into static head exactly accord-
ing to the law v = v^2grA, whatever the pressure of the surround-
ing fluid."

(2) "That only pressure openings which give the true static
head of water should be used in connection with the point of a
Pitot tube. That is to say, that only tubes which have unity as
their coeflScient should be used." #

(3) " That Pitot tubes whose constants are unity in open canal
ratings will remain unity, whatever the pressure of the liquid."

60. Experimental data confirming the first have been given.
The second is obvious on the basis of economy and general

desirahility.

The third has been confirmed by the experiments with tubes
^'^os. 8 and 9.

61. Finally, the writer believes that the Pitot tube is an instru-
ment hy means of which fluid measurements, whether of liquids or
gases, may be made with as great accuracy as with any of the
ordinary devices used. It is inexpensive and easy of application
~-is capable of being used for both high and low velocities, and
Diay be used to measure velocity of flow in pipes without mate-
rially changing the normal condition of flow; furthermore, it is
specially suited to many cases where other devices would be im-
possible of application. If properly constructed it requires no
rating.

62. These facts should make the Pitot tube a popular instru-
Qient with engineers who have tests to conduct of pumps, turbines
and blowers. To those interested in the complicated problems of
kydraulics this instrument offers advantages not possessed by any
other device.

THE PITOT TUBE.

SUPPLEMENTAL BT PROF. 8. W. K0BIN80N, COLUMBUS, O.

The Pitot Tube velocity instrument is one so exact, so simple,
so inexpensive, and yet so easy of application, that too much can-
not be said in its favor. Mr. Gregory's paper as well as several
others, in which the Pitot Tube plays an important part, indicate
that the excellent qualities of that instrument are at last beiug
generally recognized. It is probably the most nearly perfect
instrument yet discovered for the measurement of velocities of
fluids and it is equally applicable to both gases and liquids. It
applies to liquids in open channels, or to all fluids from orifices
or through conduits, at low or high pressures, and gives results
with remarkable precision. Thus there is no necessity for tedi-
ous calibration or for the use of complicated coefficients. Its
simple construction admits of its being made of glass tubing and
rubber hose.

Being thrown by these reflections into a reminiscent mood, the
writer recalls that in 1873 he isonducted a series of experiments
with the Pitot Tube instrument to measure the velocity of flow
of air issuing from an orifice, and to determine the dimensions
of the jet. These tests may be of some interest because they
are believed to have been the first applications of the instrument
in question to the flow of gases. (See Van Nostrand's " Engin-
eering Magazine," 1886, page 91.) The instrument used at that
time was made from glass drawn down to a fine tube end for the
Pitot tip, and so arranged that it could be moved from side to
side or lengthwise of the jet. Thus the precise form of the lon-
gitudinal or cross section of the jet, and its velocity, at any
point, were studied. (See Van Nostrand's Magazine, as above.)
It was shown at that time that if the tip of the instrument was
placed through the orifice to the inside of the tank and then
withdrawn along the middle of the jet that the pressure shown
by the instrument was always the same tiU the vena contracta
was passed; thus illustrating the fact that, as the internal pres-
sure of a particle of fluid at the tip diminishes, the stored energy
increases, and that, as the potential energy due to pressure falls,
the actual energy due to motion rises, the sum being constant.
These experiments also showed that the Pitot Tube " will exactly
indicate the pressure or head due to velocity when the statical
pressure is eliminated as is done in the double tip." (See illus-

THE PITOT TUBE. 209

trated test and proof in Van Nostrand's " Engineering Mag-
azine " for 1886, page 94, figure 5.)

Although the writer made occasional use of the Pitot Tube
instrument at other times both in the flow of gases from gas
wells and water (as in 1877, the flow of the Sangamon River
described in " Van Nostrand's Magazine," of 1878, page 258,
and in 1893, the Flow of Water from the Castalia Springs in
Ohio), yet his most important experiments were made in 1885,
when he was called upon by the State Geologist of Ohio to
devise a system for measuring the flow of gas wells, then being
drilled in considerable numbers in that State. (See Van Nos-
trand's " Engineering Magazine," Volume 35, page 89, and also
" Geological Survey of Ohio," Vol. 6 for 1888, page 548.) In
working on this matter a careful series of tests was made in the
laboratories of the Ohio State University, with which institution
he was then connected. In these tests both single and double
tips were used in connection with a receiver from which air issued
from different-sized openings. As in the air-tests of 1873, the
pressures were studied with tips within the receiver and at a dis-
tance of from ope to two diameters outside of the orifices, as
well as at intermediate points. It was then concluded that the
Pitot Tube Instnmient is a thoroughly reliable one for deter-
mining the pressure or dynamic head to which the velocity is
due, and that the original notions of Pitot were correct, including
that of the value of the coefficient of correction being unity, a
j>oint that all investigators have accepted as true, certainly after
making a few experiments. Indeed it is wonderful how the
simplest Pitot Tube Instrument will produce perfectly accurate
results. This simplest of instruments is easily made with care,
the essentials being all confined to the very tip ends. That is : a
good Pitot tip is preferably of cylindric form for some distance
back from the tip end, and preferably reamed out to a sharp
edge all aroimd, the latter being square with the cylindric body
of the tip end. It is best a body of revolution, six to ten times
as long as its diameter. Applied to a long pipe with a running
fluid within, a second tip is preferably introduced, which tip
may be formed like the other except. to plug up the forward end
hemispherically and cut a hole directly through square, and at
about half the length. Otherwise, though the two tips as above
are the most reliable, a double tip may be employed as shown
in a cut in the report of the Geological Survey of Ohio for 1890,
page 281.

210

THE PITOT TUBE.

Another interesting point brought out in these experiments is
that relating to the temperatures of gases. From theoretical
grounds it might be expected that in the impact of gases against
the open end of a tube the temperature and density would be
restored to that of the receiver as well as the pressure. For
bivestigating this matter an " encased thermometer " was used.
(See cut in Van Nostrand's " Engineering Magazine/' Vol. 35,
page 100.) This thermometer when properly exposed to the jet
would retain a constant temperature equal for frictionless ori-
fices to that of the receiver, while a naked thermometer similarly
exposed would drop from twenty to twenty-five degrees, partly
due to expansion and partly due to the impact against the ther-
mometer.

So finally it was concluded that "when a fluid flows from a
higher to a lower pressure through a frictionless orifice, the por-

FiQ. 57. Fig. 58.

Patented Aagnst 83, 1808-^481,810.

THE PITOT TUBE.

211

tion caught in a cnp mouth will be restored to its original con-
dition as to pressure, temperature and density." This truth
follows from the fact that in gases p.v.t. == a constant.

Having decided to adopt the Pitot Tube Instrument for the
measurement of gas wells, the writer designed and patented in
1892 the two instruments shown in the accompanying cuts taken
from the pamphlet sent out with the instrument (one for per-
manent installation, and one for changing from place to place,
as in expert work), and brought out formulas and tables for their
use in open wells or in pipe lines under great ranges of pres-
sures. These instruments and tables have since come into very
general use and the results from them are very confidently
acc^ted. That this confidence is not misplaced may be seen
from the following table of Pitot Meter results, which when com-
pared with other meter results show very close agreements.

Tabls or Rebxtltb of Natubal Qas Pitot Tubb Mbtbb Mraburembkts

COMPARKD WITH VaKIOUS SiMULTANBOUS OTBSR MbTBR MEASUREMENTS.

Diameter of
Pipe Line,
in inchee.

6
8
6
6
6

6
6
6
6
3
6
6

lOf

Gage PreMore in Pipe Line.

13.6oaDce8..

10.4 ** ..
9.2 •* ..
9.2 " ..
7 poands . . .
7 •* ...
7 " ;.

19.2 ounces .
19.2 •* ..

19.2 - ..
8.8 '* ..
7.0 " ..

13.3 •• ..

Means

11.5 poands .
12.2 •'
18.8 "

24.6 ••
20.5 "
28.0 "
16.2 »•
21.0 "
87.0 "

Cubic Fbit pib Uoub.

By Pitot Tube.

12,005
4,025
15.848
15,836
29,702
84,590
29,460
20,611
19,175
17,738
8,924
13,825
12,797

.17,578

159,100
150,000
163,700
244,000
200,700
167,800
188,000
1,517,850
1,128,000

By Gas Meter.

11,970
4.480
14.938
14,938
28.900
88,000
82.082
20,980
19.068
17,990
4.307
12,720
12.310

17,500

142,550
140,200
162,960
247,000
197,800
169.900
165.900
1,516,320
1,102,000

10 meisared on 10" line 12,841 ,281

8 branch from above 7,898,921

62dbraDch 4,962,818

Sum of two branches 12.861,784

212 FLEMINO FODB-VALTE ENOINB.

No. 1<»3.*

CONSTRUCTION AND EFFICIENCY OF A FLEMING
FOUR-VALVE ENGINE DIRECTLY CONNECTED TO
400 KW. GENERATOR.

BT Bm/Aimr T. ALLBN, HABUnUBO, PA.

(Member of the Society.)

1. The purpose of this paper is to describe the general conitruc-
tion of a new type of four-valve stationary engine, and the results
obtained from it. The efficiency of this type seems to exceed that
obtained from much more elaborate constructions.

The machine was designed to effect not only the highest effi-
ciency, so far produced, but also to attain this result under wide
variations of load, which latter condition is usual in the majority
of cases.

Before considering in detail the performance of the medium
speed compound four-valve engine, which is the subject of this
paper, it will not be out of place to give a brief explanation of
the motives leading to its final development, and the methods used
in its construction.

2. The object in view in the production of this engine, was the
combination of the advantages of the most economical slow speed
Corliss engines, with the very desirable features of compactness,
better rotative speed, closer regulation and the more efficient
methods of lubrication, possessed by the high-speed automatic
engine.

3. In order to realise the economy of steam consimaption of the
best Corliss engine practice, the sharp cut-off and more perfect
steam distribution, attained by the use of vacuum dash pots and
other accelerating devices common to Corliss engines, was at first
considered absolutely essential and in the earlier stages of the
development the best f orjns of these devices obtainable were used.

4. It was found that with an improved form of detachable cut-

♦ Presented at the New York meeting (December, 1908) of the American So-
ciety of Mechanical Engineers, and forming part of Volume XXV. of the Tram-
aettam.

nmOKG TOtm-TALTB ENGIKB.

213

SPECIMEN CARDS.

ITEST AT ABOUT -i- LOAD.

FROM H.P. CYLINDER.

Crank End

80 Spring. M.E.P. 15.6

Head End

80 Spring. M.E.P. 10.6

FROM LP. CYLINDER.

Crank End

20 Spring. M.E.P. 1.6

Head End

20 Spring. M.E.P. 1.75

Am.hmnk Holt Cb.^.r.

Pig. 59.

214

FLBKING FOmtpVALVB ENGINE.

SPECIMEN CARDS.
TEST AT ABOUT -f- LOAD.

FROM H.P. CYLINDER.
Crank End
80 Spring. M.E.P. 5a8

Head End

80 Spring. M.E.P. 43.2

FROM LP. CYUNDER.

Crank End

20 SpRiNa M.E.P. 5.7

AlUn,B.T,

czz

Head End

20 Spring. M.E.P. 5.9

Ami.Mmnk Sm Ck^. T.

Fig. 60.

7IAMIN0 FOUR-VALYE ENGINB.

215

g^BOMEN CARDS>
TEST AT ABOUT w LOAD.

FROM H.P. CYLINDER.

Head End

80 Sprinq. M.E.P. 53.8

Crank End -

80 8PRiNa M.E.P. 43.8

PROM LP. CYLINDER.
Head End
20 Spring. M.E.P. 6.5

Crank End

20 Spring. M.E.P. 6.5

jtm.r»mk .v«r« c».jf r.

Fia. 61.

216 FUSHIKG FOUR-VALVB EKGINB.

off, such engines could be made to operate fairly satisfactory at
speeds considerably in excess of that generally used on the Corliss
engine; but yet the resultant speed was not as great as was desired
for some purposes, especially electrical, and the design of the valve
gearing still possessed many disadvantageous features.

6. After exhaustive investigations it was finally found by sub-
stituting for the detachable form of cut-off, a peculiar arrange-
ment of bell cranks and levers, that a satisfactory amount of
acceleration could be given to the valves at the points of admission
and cut-off, and this by angular motion only, and unencumbered
by elaborate cut-off devices.

6. This accelerated motion, combined with the advantage ob-
tained by making the steam valves triple ported, produced a more
satisfactory operation at higher speeds than was possible with the
other devices, besides being absolutely noiseless in operation and
requiring less care and attention for maintenance.

7. This form of valve gearing being positive in action and de-
pendent upon a variation of the travel for the different grades of
expansion, made the use of the shaft governor possible, with all
its accruing advantages of speeds and regulation.

8. To secure successful and continuous service for long periods
of operation the use of hardened valve gear pins and phosphor
bronze boxes of ample proportion was essential, as well as the best
material and workmanship obtainable.

9. It was also important, in order to insure thorough lubrication
over the long periods of operation as mentioned above, that the
oiling be accomplished by automatic means, which is the case
in this engine, all the bearings being thoroughly and efficiently
lubricated by a system of self-lubricatijn which requires no atten-
tion on the part of the operator, other than replenishing the oil
at long intervals.

10. The result of combining the features above mentioned has
been the production of an engine possessing many important ad-
vantages as the following record of the performance of a medium
speed four-valve engine of this type will show.

This is, I consider of unusual interest on account of the many
unique features of the design, the somewhat unusual proportion
of the cylinders and the exceptional results obtained.

11. The engine was built by the Harrisburg Foundry & Machine
Works, and is of the tandem compoimd style, directly connected
to an electric generator for the purpose of furnishing power for the
operation of paper mill machinery combined with electric lighting.

FLEMIKO FOUR-VALVB ENGINE.

217

SPECIMEN CARDS*
TEST AT FULL LOAD.

FROM H.P. CYLINDER.
Crank End
80 Spring. M.E.P. 70

Head End

80 SPRiNa M.E.P. 69.8

FROM L.P. CYLINDER

Crank End

20 Spring. M.E.P. 9.6

Head End

20 Spring. M.E.P. 9.8

AmBank IhU C*.,lt.r.

Pio.

2ia

FLEMING FOUB-VALYE ENGINE.

SraOMEN CARDS*
TEST AT ABOUT 1^ LOAD.

FROM H.P. CYLINDER.

Crank End

80 Spring. M.E.P. 77.4

Head End

80 Spring. M.E.P. 73.4

FROM LP. CYLINDER.

Alkn, B.T.

Head End

20 Spring. M.E.P. 10.8

Fig. 68.

^M. Ami Jhtt Ok JV. r.

MJSMING FOUE-VALVE ENGINE.

219

Its nominal capacity is 500 horse-power at a speed of 150
revolutions per minute, 150 pounds pressure at the throttle and 26
inches vacuunL The cylinders are so proportioned as to give
the high ratio of 1 to 7.33 following the style advocated by Mr.
Geo. I. Rockwood, the general dimensions being given in table
number one.

12. No steam jackets are used, but a vertical tubular reheating
receiver is placed in the steam passage between the two cylinders,
steam being admitted to the high pressure cylinder through triple
ported valves of the Corliss type working in chilled iron bushings,
the governing being accomplished by a centrally balanced inertia
shaft or wheel governor. This governor is so constructed that it
is practically balanced in all positions, being made with two inertia
arms, the centers of gravity of which move in harmony with each
other about the center of rotation, the balancing feature avoiding
surging or violent action under all conditions of operation. This

90 lOO15O9OO88OaOO85O4OO45O6OO6SO0OO
Indicated Horse Power

^M.T. CURVE OF STEAM CONSUMPTION ^^w*<.a.jr.r.

Fio. 64.

<>perate9, by means of bell cranks, the steam admission valves
of the high pressure cylinder only, and is so arranged as to de-
crease the lead at the earliest point of cut-oflF. The steam valvea
o' the low pressure cylinder are controlled by a fixed eccentric,
^ arranged that the cut-oflF in that cylinder can only be varied
^^n the engine is not running, and remains constant, imder all
^^nditions of load and pressure.

13. The exhaust valves of both cylinders are of the Corliss type
operated by a single eccentric through the medium of a peculiar
arrangement of rocker arms and bell cranks.

14. In making the tests the water of condensation from the ex-
haust was weighed at the discharge of the condenser, which was of
the surface type. The steam used by the reheater was discharged
from a trap and condensed in a coil and weighed, the quantities
being included in the results given in the tables.

9LEMINO FOUR-TALVE EKOINS.

FLEMING FOUR-TALVE ENGINE.

221

222

VLBICINO FOUB-YALYB KNGINE.

15. The steam was taken from the horizontal water tube boilers
and contained from A to 1^ per cent, of moisture.

The tests were made with about 160 pounds pressure at the
throttle, with the exceptions of two, one of about f load with
a pressure of about 130 poimds, and the other at a time when but
a small portion of the mill was in operation. The load under the
latter conditions being only about i the rated capacity of the en-
gine, a pressure of about 90 pounds was being carried at this time.

16. The trials were of rather short duration, but every precau-
tion was taken to insure accuracy. The indicators, which were
four in number, were attached with short pipes, one to each end of
each cylinder, and operated by a positive pantograph motion with
a light rod and short cord connections, the diagrams being taken
at very close intervals.

17. The speeds given are the average revolutions per minute ob-
tained from the total, registered by a recording speed counter.

TABLE I.
Gbkeral Dimensions of Enginb.

Higb pressure cylinder, diameter 15 incbefl.

Low " ** *' 40i "

Stroke 27 *•

Diameter piston rod, H.-P 2H "

44 u it T pjCrankend « **

^••*^|Head " 3H •'

Batio of areas of cylinders 1 to 7.88

Clearance H. -P. cylinder : SM%

L. P. " 4.67%

Constant for 1 lb. M. E. P. and one revolution H..P. cylinder. .0237

«i «« «4 tt t< <( "LP ** .174

18. Table 11. is a condensed report of the general results ob-
tained under the varying conditions of load and pressure, as fol-
lows: —

TABLE n.

a 9

O M
^1

iF^

|2||i|

Test at about |
load

2
8
2
5
2

155.18
151.25
152.88
150.00
148.89

89.72
129.9
149.4
152.0
158.0

26.5
26.0
26.0
25 9
25.5

44.84
167.78
176.19
249.96
267.24

42.78
153.81
172.09
251.59
286.25

87.07
821.54
348.28
501.55
558.49

14.42

Test at about t
load

18.59

Test at about ^^
load

12.83

Test at about
full.rated load

Test at about lj\.
load

12.66
12.73

FLEMING FOUR-VALYB BKOINB.

FRICTION CARDS.
FROM H.P. CYLINDER.

Crank End

80 Spring. M.E.P. 8.6

Head End

80 SPRiNa M.E.P. 6.

FROM LP. CYLINDER.

Crank End

20 Spring. M.E.P. 0.5

Head End
20 Spring.

M.E.P. 1.1

Frioxion cards taken with brushes on commutator and fields excited.

Total horse-power 37.67 or about 7.6^ of rated load of 600 H.P.

Combined efficiency of unit about 86.4 at full load.

Aamm,M.T,

jM.j0iaA(tCu^.r.

Fxo. 67.

224

FLEMING FOUR-VALVB £KOIKE.

OVERLOAD CARDS.

FROM H.P. CYUNDER.

Crank End

80 Spring. M.E.P. 82.0

Head End

80 SPRiNa M.E.P. 87.4

FROM LP. CYLINDER

Average horse-power h.p. cylinder 295.09

" " " L.P. " 330.42

Total horse-power 625.61
Fie. 68.

• ck.,if.r.

PLBMIKG FOmt-VALVB ENGINE. 225

19. Figs. 59, 60, 61, 62 and 63 are the indicator diagrams taken
throughout the tests, the cards in every case being those repre-
senting the average load for the complete run. Fig. No. 64 is the
diagram of efficiency or steam consumption curve.

Fig. No. 65 is of the diagrams from -h load combined, the cards
being from the crank end of the cylinders.

Fig. No. 66 is the combination of the diagrams from the
full load test, the cards being taken from, the head end of the
cylinders.

20. These combinations are made from the cards shown in Figs.
61 and 62, and are accompanied by tables giving the measurements
taken from the diagrams. The perfection with which the cards
from the W load match the theoretical expansion curve, is worthy
of note. This feature, however, is not so well carried out on the
combined card of the full load. Fig 66.

21. I considered it of interest to add Figs. No. 67 and 68 also.
The former gives the diagrams obtained from the engine when
rumiing the dynamo, without doing any work, excepting the fric-
tion of the combined unit. The latter, Fig. 68, showing a set of
diagrams obtained under an over-load of about 25 per cent.

No steam consumption tests were made under these loads, the
cards simply serving to show the steam distribution under these
conditions.

22. A comparison of the results obtained at the different loads
re?eals some very interestng features; a very important one being
the slight difference in the quantity of steam consumed per in-
dicated horse-power per hour under the various conditions of load
and steam pressure ; representing a curve of economy closely ap-
proximating a straight line. This will be seen by reference to
the diagram of efficiency. The difference between the highest
and lowest steam consumption being only 2.09 pounds.

23. It will also be noticed that the distribution of work between
the two cylinders is nearly uniform under all loads, up to the rated
capacity, after which the tendency is for the low pressure cylinder
to do the greater proportion.

24. As before stated, there is no variation of the point of cut-off
in the low pressure cylinder, the setting of the valves remaining
the same during the complete series of tests; and it may be that
the throttling action of the governor by decreasing the lead and
initial pressure in the high pressure cylinder under the light loads,
contributed largely to the imif ormity of the results.

FLEMING FOUR-VALVE ENGINE. 227

25. It will also be noticed that the best economy was obtained at
about three-quarters load; this I consider due to the fact that the
reheater was sufficiently large to superheat the steam passing to
the low pressure cylinder up to this point. On the load being
increased some accumulation of water was noticeable in the gauge
glass, which would after a time evaporate and pass through the
low pressure cylinder. This failure of the reheater to perform
its proper functions on the increased loads, is clearly indicated
in Fig. 66, and in all probability had some tendency to impair
the efficiency of the apparatus under heavy loads.

26. In conclusion, therefore, it seems apparent that on account
of this particular test and from others corroborating it, made at
other times, the following more important and rather new prin-
ciples are fairly established :

First: That as a prime mover the elaborate dash pot or other
accelerated cut-off devices used in present Corliss Engine practice
are unnecessary complications and unwarranted when comparing
results.

Second: That the centrally balanced, direct-acting fly wheel
device serves its purpose to better advantage than the indirect fly
ball governor.

Third: That there is better warrant for shorter strokes and
moderately high speeds than for longer strokes and resultant
lower speeds, notwithstanding the element of clearances.

Fourth: That self -lubrication without additional apparatus re-
quiring attention to secure it, enters as an improvement in net
efficiency, to say nothing regarding maintenance.

Fifth: That an engine of the described design, although of
marked improvement in point of simplicity will rather exceed
than equal the more elaborate practice heretofore established at
normal load and excels comparable prime movers in a marked de-
gree where the work is of a widely fluctuating character.

Sixth: That considering a resulting decrease in the cost of
foundations, building, floor space, and, in electric practice, gener-
ators, due to better speeds, the design described determines its
importance from the standpoint of investment.

DISCUSSION.

Prof. R. C, Carpenter. — The results of the tests of the Fleming
4-valve engine described in Mr. Allen's paper show very creditable

228 FLEMING FOUR-VALVE EKOtKB.

results compared with an engine of similar dimensions operating at
the relatively high speed of 150 revolutions per minute. I consider
it doubtful, however, if the tests cited are sufficient to establish
a water rate for ordinary cases as uniform as that shown under
the peculiar conditions of the various tests cited. The tests show
four results obtained when the engine carried a load more than
five-eighths of its rating, and a single test when the engine was
loaded to only one-sixth part of its rating; this single test, when
plotted with the results obtained for higher loads, forms a nearly
horizontal curve, as shown in Fig. 64 of the paper, and indicates a
remarkable uniformity of steam consumption for wide variations
in loading. By computing the total steam used per hour and con-
structing a curve with total indicated horse power as abscisssB
and total weight of steam as ordinates, some interesting relations
are shown which are not developed in the paper.

The total indicated horse power and total steAm per hour
are shown in the following table:

Total steam per hoar.

Lb$.

Lb:

87.07

1255.5

89.72

821.54

4869.7

129.9

848.28

4294.8

149.4

601.55

6849.6

152.0

558.49

7046.9

158.0

The curve showing the relation between total steam per hour
and total power developed is, as shown by the diagram, Fig. 70,
which is noted, practically a straight line, A By which can be rep-
resented approximately by the equation

Total steam per hour = 200 -h 12.3 (H. P.)

from which we find

Steam per I. H. P. per hour = tt-^o "^ l^-'*

H. Jr.

I have plotted a great many engine tests in this manner, and
have found that an engine controlled by a throttling governor
invariably gives a straight line curve, which fact I think was first
pointed out by Mr. H. H. Willans of England, and has sometimes
been characterized as Willans' Law. On the other hand, an en-
gine controlled by an automatic governor has for its characteristic
a curved line approximating that shown in dotted lines C D E in
the figure submitted.

FLEMING FOUR-VALVE ENGINE.

229

The governor used on the Fleming engine was of the automatic
type, but the tests which were submitted were made with differ-
ent steam pressures; consequently the general effect of the tests
with light loads would approximate that obtained with a throttling

8000

I B

gtooo

)

|gooo

•

//^

•^4000
snon

Bela

lonb

tweei
idLH

ToU
P.de

Steal
reIop«

a per

lour

flOOO

Toto

Steal

iper

lOUT"

-aoo+

12J(L

i.PO

1000

Stei

mpei

LH.P

perl

our—

4-12^

100

900

VJfcC

800. 400

Total LH.P.

Fig. 70.

600

000

governor applied to the high-pressure cylinder, i.e., a low .steam
pressure would give a later cut-off than would have been experi-
enced had the pressure been high. This, in my opinion, accounts
for the fact that in the curve which I submit the result of the test
with the low load falls in a right line with the other tests. While
this is purely accidental, it has the effect of making the steam

230 FLEMING FOUK-VALVE ENGINE.

consumption less at a light load than would have been the case
had the steam pressure been maintained at 150 pounds for the
entire series of tests.

It is my impression that with a constant steam pressure, this
engine will show about the same variation in steam consumption
per unit of power as the Corliss engines or other good engines of
the automatic type.

Mr. RocJcwood. — ^It appears that the purpose of this paper, as
stated in paragraph 1, is not quite the same thing as the impres-
sion gained by reading it. New principles of design are claimed,
but none is described; and while a novel form of positively-driven
Corliss valve gear is hinted at, its details are not given, and the
feeling of the reader is that the author is claiming broadly, as
a novelty, the use of the shaft governor as a means of controlling
the cut-off valves. The same thing may be said of the references
to the use of a system of automatic lubrication. Also, the
" efficiency ^' — ^if by that is meant economy of steam — ia mis-
takenly claimed to exceed that of other stationary 4-valve en-
gines of the high ratio compound type, especially with variable
loads, for the best performance of this engine is more than one
pound of steam per indicated horse power per hour in excess of
fhe best recorded performance of a high ratio compound engine
of the odinary slow-speed Corliss type. I think, therefore, that
the claims of paragraph 26 — ^with the possible exceptions of the
second and the sixth— &hovld be omitted, and that the construc-
tion which it was the stated object of the paper to describe should
actually be described fully, instead of merely hinted at. These
strictures have reference rather to the form and claims of the
paper than to the engine itself, about which enough is made clear
to excite one's interest and, possibly, one's approval of its pecul-
iarities, _

The chief thing about the design of this engine which, broadly
considered, is still unusual is tfie use of the high cylinder ratio.
It is now about twelve^ years since the first slow speed compound
stationary engine with an exaggerated cylinder ratio was built
and tested. The results were published in Volume XHI. of the
Transactions. The accuracy of these tests and their sufficiency
to prove the superior value of the high ratio over the common
ratio of three or four to one was not at once believed by many
prominent engineers, for the reason that the general theory of
the steam engine seemed to them to be against the possibility of

FLEMING FOUH-VALVE ENGINE. 231

the truth of any such results. At that time the best acknowledged
performance of the compound engine was between 14 and 15
pounds per indicated horse power per hour, and even those re-
ports of triple-expansion pumping-engine tests, wherein was
claimed a steam consumption of between 12 and 13 pounds, were
at first considered to be little better than " fairy tales." \t at
that time an engineer had reported to this Society a paper like the
one we are now considering, he would have been received with
entire incredulity if not with scorn.

A distinguishing characteristic of these high ratio compounds
is the phenomenon of drop. It was believed then, as it is now,
that drop entails a loss of work by free expansion; and, without
going further into the total effect of its use than that, there was a
singular and unanimous determination on the part of all writers
on the steam engine to discourage the toleration of any drop
whatever in compound engines. The fact is now, however, ap-
preciated that the high ratio compound is much more economical
at light loads than is the engine with a low cylinder ratio, and
the tests submitted in the paper are a further proof of this. The
history of the progress of the high ratio idea is interesting, for it
bears on the question of how much drop is desirable. Between
1885 and 1890 several large steamers of the Ley land line had
their old style compound engines, which used steam at 90 pounds
boiler pressure, converted into high ratio compounds with steam
at 150 pounds pressure. The new engines of the steamship ^^ Alge-
rian," for instance, were found on trial to equal the economy
of triple-expansion engines of the same power. Cut-off occurred,
however, at the unusually early point of one-quarter of the stroke
in regular operation, with the result that these engines worked
with no more drop than is usual in marine engines of the triple-
expansion type cutting off at three-quarters of the stroke, as such
engines generally do or even at a later cut-off.

But this system of so altering old ships received something of
a setback in 1892, when a test was made on the steamship
"Iveagh," both with and without the use of the intermediate
cylinder. The actual figures obtained on these trials showed a
saving of coal of 30 per cent. Although these tests seemed on the
face of the matter to be fair comparative tests, the engines and
boiler pressure being identical in both cases, yet in reality they
ouly proved what might have been anticipated, that in the high
ratio compound very excessive drop — such as where there is prac-

233 FLEMING FOUR-VALVB ENGINE.

tically no expansion in either cylinder before release — is very
wasteful. It does not do, in other words, to have all the expan-
sion take place in the receiver. In the case of the " Iveagh " the
drop was more than three times what it was on the engines of
the " Algerian,'' being as much as 100 pounds.

On the other hand, a reasonable amount of drop is accom-
panied by a distinct net gain of economy at all loads. This is
because the waste of heat resulting in all engines from condensa-
tion of steam and its subsequent reevaporation at the moment
of release, without the performance of work, is undeniably re-
duced. In the stationary 4-valve type of engine, the amount of
drop in high ratio compounds is of small extent compared with
what took place in the engines of the "Iveagh''; it never ex-
ceeds 30 pounds, and usually is from 15 to 25 pounds, and hence
these engines do remarkably economical work, and there are in
operation this minute — as the result, I believe, of the publication
of the tests referred to in Volume XIII. — several hundre4 thou-
sand horse power of them.

The theory upon which the engines of the " Algerian '' were
designed was, evidently, a wrong one. It was thought that the
increased waste due to cylinder condensation, which the greater
range in pressure permitted in the first cylinder, would be ob-
viated by preventing the increased temperature range naturally
accompanying it by the introduction between the two cylinders
of a cylinder containing a spiral sheet of copper, and called a
" heat retainer." This device was actually fitted to several
steamers before the absurdity of it became apparent by actual
thermometer tests. Nevertheless, had the designers and owners
of those engines not fully expected that the natural increase in
the range in temperature, due to the enlarged range in pressure
in the high-pressure cylinder, would have been nullified by its
agency, they would have abandoned the high ratio idea without
experiment. My idea, on the other hand, was and is, that there
would be no increase in cylinder condensation due to the omis-
sion of an intermediate cylinder, and that moreover there would
be certain practical gains eflFected by fattening the combined dia-
gram.

It is of interest to compare the performance of the Leavitt
pumping engine at the Chestnut Hill Reservoir, Boston, with
that of the high ratio Cooper Corliss compound mill engine at
Providence, R. I. With identical boiler pressures, ratios of ex-

FLEMING FOUR-VALVE ENGINE.

pansion and degrees of vacuums the steam consumptions per in-
dicated horse power per hour were also identical, or 11.2 pounds.

I can corroborate the straight line shown for steam consump-
tion at variable loads out of my own experience. I should, I con-
fess, Uke to see some proof that on this short stroke engine the
low-pressure clearance does not exceed 4.6 per cent. I have never
seen a Corliss engine that would give less than 6 per cent, to 7 per
cent, clearance even with relatively much longer strokes. Also,
while the design of this engine looks, in the illustration, prac-
tical and satisfactory, I cannot see that it is any less complicated
than if it had dash pots and a fly-ball governor. I believe that
other things being equal except the speed of rotation — that is,
with the same piston speed — the long-stroke engine will beat the
short-stroke engine every time. Without doubt, however, the ad-
vantage of the long stroke is small, and out of consideration of
economy of first cost of all parts of the unit, the short stroke is
to be preferred.

Mr, C. V. Kerr. — ^I would also like to have the author give
the data of that re-heater. If we had the square feet of heating
surface in the re-heater, the pressure under which the steam is
condensed, and the weight and temperature of the water con-
densed, we could make some interesting calculations on the per-
formance of this engine that we cannot make now with the data
at hand.

Mr, J. A. Seymour. — ^I wish to take exception to the conclu-
sions reached in the paper concerning the advantage of a short
stroke as regards economy. I have found from the results of a
series of tests with engines, varying in stroke from 24 to 66
inches, but otherwise similar in all respects, that a moderately
long stroke means a very considerable gain in economy as com-
pared with a short stroke, such as that of the engine described
by Mr. Allen, and that with p, further increase in stroke beyond
a certain point, the *gain becomes much less. That as good
economy cannot be obtained with engines of short stroke as with
those of moderately long stroke is recognized generally by en-
gineers having experience with both types. The remarks just
made by Mr. Rockwood show that he has found this to be so.

I have mentioned this series of tests, which were made in vari-
ous locations, with and without superheating receivers, some with
saturated and some with superheated steam, and in each case
conducted jointly by purchaser and builder to determine as to

284 FLEMING FOUE-VALVE ENGINE.

fulfilment of contract guarantees, merely because engines of
identical style and make are seldom built with such widely vary-
ing strokes and these tests, therefore, the results of which are all
very consistent, aflFord an unusually good basis of comparing the
effect which length of stroke has upon the economy of an engine.

While I agree in regard to the advantage of higher rotative
speeds, I think the paper is wrong in its conclusions concerning
the economy to be secured with short strokes, and that the results
given need confirmation.

Mr, Allen.* — There are some of these questions which have been
asked that are pretty hard for me to make reply to at once. But
first I wish to make a few remarks in corroboration of the results
given in this particular test. Some two or three years ago I was
called upon to make a series of tests upon a pair of engines which
"were of the same design exactly as this engine. This pair of en-
gines was running at a speed of 225 revolutions per minute; the
steam pressure was 125 at the throttle; the vacuum was about
26 inches, and the rated capacity was 300 horse power. It did not
have any reheating receiver or any steam jackets. I think I can
recollect, at least approximately, the figures obtained, and I will
cite them. On the test made at one-quarter load, the results were
18.43 pounds. On the tesl made at one-half load, the results were
17.45 pounds. On the test made at three-quarter load, the results
were 14.97 pounds. On the test made at about the full rated load
of 300 horse power, the result was 15.22 pounds, and on the 25
per cent, over load 15.33 pounds. These figures represent the
actual weight of feed water pumped into the boilers and without
any corrections for moisture. If you will compare the figures
given, the highest and lowest, you will notice that in this case also
the difference is only about 3.46 pounds. The ratio of this engine
was 1 to 5^. It had a high-pressure cylinder of 13 inches. Low
pressure thirty and five thirty-secondths, and the stroke was 17^
inches, running at a speed of 225 revolutions per minute. I have
made other tests of this particular type of engine which cor-
roborate the tests made in this paper.

With reference to the remarks of Professor Carpenter regard-
ing the variation in steam pressure, I wish to say that the gov-
ernor is so arranged on the shaft of this engine that by giving it a
certain amount of angular advance, the lead, as the paper states,

* Author's closure, under the Rules.

FLEMING FOUR-VALVE ENGINE. 235

is reduced at the earlier points of the cut-oflF. If you will measure
the initial pressure given on the card of the sixth load in this paper,
you will notice that it is far from being 90 pounds in the cylinder,
and that the throttling action is very apparent.

It has been my experience that this same engine running with a
higher pressure does not give any higher initial pressure in the
cylinder on the light loads than it would if the pressure at the
throttle were low. So that, outside of the fact that there is a little
more heat in the steam chest, I do not believe that the higher pres-
sure would have made any difference in the result.

Mr. Kent asked for further data regarding the design of the
engine. I will add that to the paper before it is finally published.

Regarding the clearance, the valves of this engine are of rather
peculiar construction, and I really do not think that the engine
can be compared with a Corliss engine — that is, not in the same
sense that the discussion seems to imply. The clearance between
the low-pressure piston and the head was only about one-eighth of
an inch. The head was turned and polished, as also was the piston,
on both sides, the crank end being cast in solid. The percentage
of clearance given is that percentage which has been carefully
calculated from the drawing and checked up by two or three dif-
ferent persons.

Regarding the condenser, this was of the surface type — I can-
not remember the name of it. It was operated independently of
the engine. There was no test made of the amount of steam con-
sumed by the condenser.

With reference to the reheater, steam direct from the boiler
was passed through the reheater and carried off by means of a
trap. I am sorry to say that through an error on the part of the
man looking after the water coming from this reheater, he neg-
lected to take the temperature of it. So I cannot give you that
<J«ta, but the weights I am absolutely certain are correct; that is,
the weghts of water consumed by the reheater. These I have in
the original tests, but not in this paper. That data of course I
could put in the final printed matter also.

In response to numerous requests for more complete details of
the construction and further information of the test of the 4-valve
engine, which was the subject of my paper, I have considered it
advisable to add the drawing Fig. 71, showing cross section of
valves and cylinders, and Fig. 72 showing the plan and elevation
of the engine in detail.

286

FLEMING FOCB-VALVE ENGINE.

FLEMINO FOUR-VALVE ENGINE.

237

238

FLEMING FOUE-VALVE ENGINE.

The drawing of the cylinders, Fig. 71, was made for a smaller
engine, but the design and construction of the valves is the same
as on the engine from which the test was made.

It will be noticed on referring to this cross sectional drawing,

Fig. 73.

that the steam valves of the high-pressure cylinders operate in re- ,
movable cast-iron bushings, as described in the paper, and that the
ports are arranged in such a manner that the three edges are
opened at one time. This arrangement of triple ports combined
with the rapid angular motion, obtained from the use of the
peculiar arrangement of bell cranks, gives a very quick ad-
mission and cut-off. The use of the bushings gives a ready means
of renewal when repairs are necessary, as the cages can simply be

FLEMING POUR-VALVE ENGINE. S39

removed and new ones forced in, in a very short time, obviating
the necessity of boring out the holes as must be done in the case
of the Corliss engine.

The steam valves of the low-pressure cylinder are also triple
ported, but have a different arrangement of steam passage and
do not operate in bushings. We do not consider this necessary in
this cylinder, as the valves are constantly working at the same
travel, insuring thorough lubrication, and as the surfaces are large,
the amount of wear is very slight after years of operation.

The exhaust valves in both cylinders are single ported, the
valves being virtually plug cocks which close in such a manner as
to eliminate entirely the space inside the valve from the clearance
volume^ The valves of the high-pressure cylinders are operated
by means of our centrally balanced inertia governor, which is
clearly shown by Fig. 73. This governor is the same as is used
on all engines of the Fleming system.

The steam valves of the low-pressure cylinder are opei*ated by
means of an independent eccentric, which is practically fixed to
the shaft, but has a point of suspension similar to that used on the
governor, and is arranged with a screw by means of which it can
be moved across the shaft, and the point of cut-off varied in the
low-pressure cylinder without changing the lead. This can be
done only while the engine is not in motion.

The eccentric next to the main bearing controls the exhaust
valves in both high and low-pressure cylinder, by means of a
peculiar combination of rocker arms. These rocker arms are all
made of opeii hearth steel castings, and those which operate the
steam valves are keyed to the shaft at each end next to the rocker
shaft bearings.

This shaft has a running fit in two bearings attached to a
bracket which is bolted to the side of the bed. Between these
two rocker arms at the extreme end of the shaft is located the
rocker arm which connects to the steam valves of the low-pres-
sure cylinder, also the rocker arm which operates the exhaust
valves of both cylinders. These rocker arms run loose on the
shaft and are babbitted inside the hole to avoid cutting. They
are also provided with large bronze adjusting shoes to take up
the wear at this point.

All the pins used throughout the valve gearing, and in fact
on the whole engine, are made of steel, case hardened and ground,
and the connections in the valve gearing are of phosphor bronECj

240 FLEMING FOtm-VAtVE ENGINE.

provided with means of adjustment so that wear can be taken up
without removing the rods from their pins. All the bell cranks
and valve arms are made amply strong of malleable iron to guard
against possible breakage.

There is a disconnecting valve for each pair of valves where
the reach rod connects to each set of bell cranks. This discon-
necting device is arranged in such a manner that by simply
throwing a lever, a cam is disengaged from the rod, which is then
free to move through the pin connected to the bell crank, so that
by means of a starting bar each pair of valves can be readily
tried before the engine is turned over under steam.

The valve gearing is most substantial throughout, all the parts
being exceptionally large and heavy. The brackets of the steam
valves are connected together with links and ream-fitted bolts
in such a manner that the two steam brackets of each cylinder
form a rigid truss and are made to distribute the strain in an
admirable manner.

The same system of self lubrication is used on these engines
as is used on all the Fleming engines. The cylinders are neatly
covered with cast iron lagging, ground and polished. The space
between being filled with the best quality of non-conducting
material. The connection between the high- and low-pressure
cylinder is arranged in halves, and attached to the cylinders by
means of cap bolts, so that in case it is necessary to examine the
interior of the low-pressure cylinder the connection can be readily
removed, thus allowing ample room to move back the head and
examine the cylinder or make adjustments to the pi^on.

The piston of the low-pressure cylinder has a bull ring of the
phosphor bronze type, which is arranged in such a manner that
in case of wear the piston rod can easily be raised. The use of
phosphor bronze in the bull ring has proven very satisfactory
and insures long wear without cutting.

The drawing of the plan and elevation of the engine, Fig. 72,
shows more clearly the style and construction of the valve gearing
than does the cut. Fig. 69, in the paper read before the Society,
which cut was made from a photograph of the engine.

In order to show more clearly the arrangement of the re-heat-
ing receiver and its connections on this particular engine, I have
considered it of interest to add drawing. Fig. 74, which is a
foundation plan of this particular engine.

It will be noticed that the re-heating receiver is of the vertical

FLEMING FOUR-VALVE ENGINE.

241

h- =^'8--

•-^8 Tii^"* — .T^T^sr**

£48 PLEMIKG FOUB-VALVE ENOIKE.

type. It was built by the Wheeler Surface Condenser Co., and
contains a series of vertical tubes, one inside the other, arranged
after the manner usual in the constniction of their surface con-
densers.

This re-heating receiver contains 250 square feet of heating
surface, and the writer is of the opinion that the falling off of
the economy on the full load is largely due to the fact that this
surface was insufficient for this size engine. The amount repre-
sents about i square foot to the horse power at the rated capacity
of the engine.

In reply to question asked by Mr. C. V. Kerr with reference
to the re-heater. The steam entered the re-heater at 150 pounds
pressure, and was taken therefrom by means of a Bundy trap
and passed through a coil of pipe immersed in a tank which was
constantly running full of cold water. This tank was placed
above another tank on a pair of scales, so that when the trap dis-
charged its contents, all the steam passing through the re-heater
was allowed to flow into the tank on the scales where it was
weighed.

As mentioned by the writer, there was no temperature taken
of this water, but it is fair to say that it was not above 100 degrees.

The amount of water used by the re^heater under the different
loads given in the table of the test Xo. 2 of my paper is as follows:

Test at i lo id. steam used per I. H. P. per hour 1.26 lb8.

(t «( i i. .« .< .• *t •« <» "f oi •«

B J. *»»

«f «i T. «. 4» <* •< it it .. 02 "

** "full ** '• ** ** .54 "

tt tt^^i^ it tt .« *i .. .1 *. 54 u

Referring to remarks made during the discussion in which ex-
ception was taken to the conclusions reached in the paper concern-
ing the advantage of the short stroke engine as regards economy,
the writer would say that he has been called upon to make tests
of a great number of engines similar to the one described in the
pajK^r, and it is fair to say that, in each and every case, he has
found the economy to exceed that of the Corliss engine, either
medium or long stroke, under similar conditions.

The results cited by the writer on the engine of 300 horse-
power, with a stroke of 17^ inches, speed 225 revolutions per
minute, were at most economical points less than 15 pounds, with
125 pounds of steam and 25 inches vacuum and without any
re-heating receiver.

FLEMING FOUH-VALVB KNOINB. 343

The writer has also in mind another test which he was called
upon to make on a simple four valve engine of this type, size
15 X 15; speed, 180 revolutions; horse-power, 135; steam pressure,
120 pounds at the throttle, non-condensing. The results obtained
from this engine were 23.4 pounds per indicated horse power per
hour, which is as good as the best obtained from either long or
medium stroke Corliss machine or other engines of similar design.
Referring to Professor Carpenter's part of the discussion:
the writer's experience has been that, with the governor advanced
as is done on the Fleming four valve engine (thus decreasing the
lead to zero at the lighter loads), in controlling the speed of the
engine, the initial pressure must be throttled as the steam enters
the high-pressure cylinder, especially on the light loads; if this
were not the case the engine would run away, therefore the
lighter the load the less the initial pressure in the cylinder. This
is true no matter what the pressure is at the throttle, so had all
the tests been made with the same pressure at the throttle, in the
writer's opinion the results could not have varied greatly. The
engine carries a certain load at a certain speed, and as the regula-
tion is very close for all changes of load, the initial pressure is cut
• down in order to maintain a speed practically constant under
varying loads.

Referring to Mr. Rockwood's remarks during the discussion,
the clearance in the low-pressure cylinder of this engine was
determined, as mentioned by the writer, by carefully calculating
the volumes from drawings.

The pistons ran \vithin about ^ inch of the cylinder head. The
work of calculating this clearance was very carefully done and
was gone over by several different people, so that the writer is
confident that the calculations are correct.

A reference to the cross sectional drawing, Fig. 71, will show
that the valves lay very close to the cylinder, and that there is no
clearance in the exhaust valves at all, they being single ported,
and the only portion of the steam valves which entered the clear-
ance is the narrow part through the valve.

It will be remembered in the ordinary form of Corliss exhaust
valve there is quite an amount of space in the valve which must
be taken into the clearance, this possibly has something to do
with the comparatively small amount of clearance in the cylinder
of this type and stroke as compared with what can usually be
obtained with a Corliss form of valve.

214 FLEMING FOUR-YALYE ENGINE.

The claims made in the paper regarding the economy of the
engine are intended to mean that the net efficiency of an engine
of this type is higher than can be obtained from the long-stroke,
slow-speed engine, because of the combination of low steam con-
sumption per horse power; more perfect regulation on account
of the use of the shaft governor; economy of oil on account of
self lubrication^ reduced amount of floor space and foundation
as compared with the slow speed engine, and if direct connected
very much reduced cost of generator on account of the high
rotative speed.

A COMPACT GAS ENGINE: BEAM TYPE. 245

No. 1093.*

A COMPACT CAS ENGINE: BEAM TYPE.

BT O. B. MOBOAM, WOROSSTKR, MASS.

(Member of the Society.)

1. Having had an experience of 35 years in the manufacture
and u.se of gas-producers and producer-gas, it has been the pleasure
of the writer to feel something of the possible future of the gas
engine and to endeavor to add something to its development and
perfection.

2. This has been done collaterally to other and more pressing
demands but always with unabated confidence and interest.

3. In common with others the enormous gaseous wastes of the
blast furnace and coke ovens has been a tempting field for re-
search and exploitation. Our esteemed fellow member, Mr. H.
H. Campbell, in his " Manufacture and Properties of Iron and
Steel," strikingly calls attention to the fact that while the blast
furnace is primarily a producer of iron, it is a gigantic gas pro-
ducer as well.f

4. Like the dependence of the steam engine upon its boiler for
uniformity and quality of steam so is that of the gas engine upon
the producer for suitable gas. Nor is such gas easily obtained.
Blast furnace and producer gas, especially the former, contain
large amounts of dust which comes from the ores, limestone and
ash, and which is driven out with the blast. This dust has been the
bete noir of the gas engineer in attempted utilization of the gas.
Clogging the checker work and flues when used for reheating,
impeding combustion when burned under boilers, and cutting and

♦ Pres«»nted at the New York meeting; (December, 190S) of the American
Society of Mechanical Engineers, and forming part of Volume XXV. of the
Tramaetiom,

f Mr. CampbeU has made a valaable suggestion in his ** MetaUurgj of Iron
and Steel." (See page 1*24.) He says : ** It is quite possible that the exhaust
inifleB from the jras engine can 1>e profitably employed to heat the blast in the
stoves of the blast furnace.'.*

246

A COMPACT GAS EXGINE: BEAM 1T?E.

A COMPACT GAS ENGINE: BEAM TYPE.

247

grinding vital parts of the machine when used in gas engines.
Various devices for " washing " or eliminating the dust have
been devised, when, in the year 1900, German engineers discov-

FiG. 76.

ered the value of passing the gas through a simple centrifugal
fan blower injected with a small spray of water. The result has
been exceedingly promising and when perfected will don bf less
solve the troublesome problem.

248

A COMPACT GAS ENGINE: BEAM TYPE.

2
£

A COMPACT GAS ENGINE.* BEAM TYPE.

249

5. This en^ne, as illustrated in this paper, operates on the
two-cycle principle with an explosion at every downward stroke
of the piston. Two compressors, one for gas and one for air,
furnish the engine with its charges of gas and air, and provide air
for scavenging. The exhanst is through ports in the cylinder
which are uncovered by the piston near the end of its downward

r=5?

—

t ee

■.

HM.

I oo

k ao

-t"

J-

••■

■>

•

- ;

•v

•

A i

•1

IteV

If M

' >

*m^

FiQ. 78.

s>troke; thus avoiding the use of water-cooled exhaust valves and
the mechanism which would be required to operate them. It can
abo be operated on the four-cycle system by slight changes.

6. The engine herein describeil has been designed with special
reference to the use of such blast furnace and producer gases,
the working beam type of engine and bent lever beam have been
chosen for the following reasons: —

7. It allows the cylinders to be placed upright, the balancing of
the pistons and their connecting rods; it subjects the cylinder

250

A COMl'AOT GAS ENGINE: BEAU TYPE.

A COMPACT GAS ENGINE: BEAM TYPE. 251

walls to much less pressure from the pistons, reducing the fric-
tion to a minimum. During the first half of the working stroke
the average pressure of the piston on the walls of the cylinder
is 100 pounds per square foot of surface. Figure 78 shows
dia^ammatically a striking comparison between the beam
type of engine and the ordinary horizontal type with trunk piston
direct connected to crank shaft. Two cases of the latter type are
illustrated, one with the connecting rod of a length of 5 and the
other 6 cranks respectively. The abscissse represent points in
the working stroke while the ordinates represent total pressures
in pounds on the cylinder wall due to the angularity of the con-
necting rods. The distance between each of the horizontal cross
lines is equivalent to 1,000 pounds, and the distance between the
vertical cross lines to two inches of piston stroke.

8. The bent lever beam also gives us the least pressure against
the walls of the cylinder during that part of the working stroke
when there is the highest speed of the piston, and also facilitates
the lubrication of the piston and the piston connecting rod pins.

Compactness,

9. The John Cockerill Co., of Seraing, Belgium, who were pio-
neers in the construction of large gas engines using blast furnace
gas, purpose exhibiting a 3,000 horse-power gas engine at the
St. Louis Exposition in 1904. An official statement gives the
floor space to be occupied by this engine as 85 feet x 45 feet
(8,825 square feet). A pair of engines of the beam type arranged
for blast furnace duty with a capacity of 3,000 brake horse-power,
although of the same power as the Cockerill engine require a
space of only 24 feet x 32 feet (768 square feet) one-fifth the
space.

10. Each of the inlet valves of the engine is a combination of the
mushroom or poppet valve and cylindrical valves with ports for
gas and air, all the valves being arranged on one stem. The
cylindrical valves have radial partitions which facilitate the mix-
ture of gas and air before entering the cylinder, thus securing
prompt ignition at the first part of the working stroke and a cor-
responding high efficiency. The air ports and poppet valves are
o]>ened in advance of the gas ports so as to thoroughly scavenge
the cylinder. Figure 79 shows the inlet valve in sec-
tion in three characteristic positions. The left hand figure
represents the valve closed, the central figure in a scavenging

252

A COMPACT GAS ENGINE: BEAM TYPE.

Fio. 80.

A COMPACT GAS ENGINE: BEAM TYPE. 253

position and the right hand figure the position of the valve when
fully open. The proportion of gas and air can be varied while
the engine is running by rotating the valve slightly upon its axis
by means of the handle shown in the illustration.

11. The valve gear is of the releasing or Corliss type, with dash
pots to secure quiet closing. The valve gear is operated directly
from the working beam without eccentrics or cams, which reduces
the lubrication and care usually required in steam and gas
engines. Figure 80 shows the valve gear as adapted for use
in a two-cylinder engine. The piston of the left hand cylinder
has completed its downward stroke, the scavenge lever during the
latter part of this stroke has raised the valve to the position
shown in the illustration, which allows the trigger to engage with
the valve slide and continue the raising of the valve, opening the
gas ports, as the piston starts on its upward stroke.

12. This trigger is under direct control of the governor which
effects through its connections the tripping of this trigger at
varying points in the stroke, as determined by the load on the
engine. This tripping allows the valve to return to its seat, the
dash pot preventing any shock due to closing. The right hand
valve gear is shown in its highest position, the trigger having
been tripped by the action of the governor.

13. Each cylinder of the engine i-s provided with two ignition or
spark plugs so that in case of failure the current can be s\vitched
from one to the other without stopping the engine. The timing
of the ignition, so that the spark may come before passing the
center, at the center, or at any required part of the working
stroke is accomplished by a movable disc which can be changed
instantly to any required position when the engine is running at
full speed.

14. Fig. 75 sho^^s a two-cylinder engine of the beam type with
two vertical single acting motor cylinders and one vertical double
acting blowing cylinder of 1,600 brake horse-power. The con-
necting rods work on the same crank pin and are at an angle of
90 degrees, an arrangement of recognized merit.

15. Fig. 76 represents the interior of a power station with over-
head travelling crane. The group of engines shown is made up of
units of 1,500 horse-power each, using blast furnace gas and
arranged for furnishing blast for the furnaces.

16. Fig. 77 shows an elevation partly in section of the engine
without blowing tube.

254

A COMPACT GAS ENGINE: BEAK TYPE.

17. Fig. 81 is a connection diagram for a beam and a direct con-
nected type of engine. It shows the amount of angularity of the
connecting rods and the points in the stroke at which the angu-
larity is a maximum.

18. These then are the lines along which my efforts have been

made. They are brought to the attention of the Society with the
hope that they will give some measure of interest and enthusiasm
in this highly important field.

19. If the wastes of the old generation are the profits of the new
then large dividends are waiting us in the squandering of the
past.

A COMPACT GAS ENGINB: BEAM TYPE. 255

20. The fearful wastes of blast, and coke furnaces; the enormous
power in them awaiting capture and control must impress any
one at all sensitive to manufacturing economies.

21. May this new motor assist in bringing quick subjection, har-
ness a new and better power for the wants of man, and so obey
the divine conmiand " to subdue and have dominion."

22. In closing, the author desires to express his thanks to his
assistant, Mr. A. J. Gifford.

DISCUSSION.

Prof, S. A. Reeve, — I should like to refer to the statement made
in the paper, quoting the remark of Mr. Campbell where he de-
fines the gas-engine as " a cannon with its projectile fastened to
the crank-shaft," * to say that that quotation voices a view of the
inc?chanics of the gas-engine which I find to be quite common,
and which I yet believe to be a totally mistaken view. The com-
mon idea is that when a gas-engine piston starts upon its stroke
a blow is struck against the crank-pin. Now we know, from
our simplest forms of construction, such as the hammer, that
there must be velocity of motion destroyed in order to develop
a blow; but in the gas-engine there is no such thing. At the
moment of explosion no motion is developed, to be destroyed
in striking the blow, and hence there can be no impact. What
lost-motion there may have been in the engine-connections has
already been taken up by the compression of the charge, and
when the explosion-pressure is formed in the cylinder it comes
on the piston-head decidedly more gently than it does in the
case of the steam-engine; because in the steam-engine, when the
valve is opened, the steam rushes in at a very high velocity, there
is motion, which is destroyed against the piston, and there is a
blow struck. But in the gas-engine there is no such thing. The
pressure is developed, without motion, directly upon the piston-
head, and the only reason why we have trouble with main-
bearings, etc., is the fact of the enormous pressure. I think that
if we attempted to design steam-engines working, within a single
cylinder, from 400 pounds per square inch of initial pressure
down to atmospheric pressure we should meet with very much
more trouble than is found in the gas-engine.

* Bditob. — The Author omitted this from his paper in revising it.

256

A COMPACT GAS ENGINE I BEAM TYPE.

There is, thus, in gas-engines, a very great range of fluid-
pressure during each stroke, and a corresponding need for allow-
ance for it in designing them. This allowance, in Mr. Morgan's
plan, has been accomplished by providing a very much increased
weight of reciprocating parts, to take up, by its inertia, this
excessive initial pressure. In addition to those parts present
in the ordinary gas-engine: the piston and the primary, single-
acting connecting-rod, there enters into this design the beam
and the secondary, double-acting connecting-rod. In the first
place, there are two pistons and primary rods, whereas the most

Fig. 82.

common design of gas-engine has only one of each. Secondly,
the beam itself is quite a heavy piece of metal; but its moment
of inertia is not so strikingly great as might appear at first sight
to be the case; its radius of gyration is so small. Thirdly, the
main connecting-rod does present a good deal of inertia, but no
more so than any other connocting-rod for carrving the same
amount of power. You ^v^ll note the parts on this diagram Fig. 77.
The forces at work upon the different pins in the engine have
been carefully worked out, and I can show you, from the results
of this work, the uses of these inertia-forces for modifying the
effective forces at work upon the crank-pin. They are displayed

A COMPACT GAS ENGINE! BEAM TYPE. 257

in Fig. 82. At AB you see represented the center-lines of motion
of the two pistons. AB h the axis of one cylinder; CD is the axis
of the other. EF is the arc of motion for the first wrist-pin, and
Gil is the other. Frpm an initial projected indicator card — which,
I might say, was made the worst possible for the ends desired,
with a perfectly vertical explosion-line rising to a peak some
50 pounds per square inch higher than would usually or properly
occur — was calculated the effect of the combined forces upon
the different pins. You will notice that there is no reversal
of the forces except at dead centers, and that the forces are
very much more constant on the crank-pin than they are in the
original fluid-action upon the pistons.

The forces acting upon the main wrist-pin are shown by the
curves KL. There is a very heavy initial force at the beginning
of each stroke, but the peak shown there is more or less hypothet-
ical in character and is much exaggerated over actual probability.
During the rest of the stroke the pressure upon the pin, going in
either direction, is fairly constant.

The diagram at the right shows the final net resultant
forces at work upon the crank-pin, at each ten-degree divi-
sion of its revolution, including fluid-pressure, inertia of
all parts and the weight of the main connecting-rod.
Those forces coming to the pin along the conneeting-rod are
shown by arrows in the direction of its axis. Compounded with
them are the axial and vectoral inertia and the weight of the
connecting-rod, to give the final resultant. In the diagram in
the right hand upper corner are displayed the tangential effects
of these same net resultants. In order to avoid confusion of
lines their magnitude is displayed by radial arrows extending from
the circumference outwardly; but their true direction is tangential,
at the point whence the arrow rises.

I think that it has been urged that engines of an explosive
type having heavy reciprocating parts cannot be run at high
piston-speeds. This engine has a piston-speed of 810 feet per
minute; but to show what it would do at even higher piston-
speeds I have had projected here by Mr. Gifford a diagram
showing the effective forces upon the crank-pin under a piston-
speed of 1,000 feet per minute (exhibiting Fig. 83). This peak
at the beginning of the stroke is, as before, due to more or less
hypothetical conditions of operation. At this point in the
stroke the pressure upon the crank-pin becomes almost zero.

258

A COMPACT OAS engine: beam ttfe.

This drop in pressure at this point might have been avoided by
a modification in the design had it been one of the objects in
view to attain a high piston-speed. Those o'rdinates, are marked
off at 10,000-pound intervals. This diagram has been gotten up
to show that an engine of this type, with a very heavy triangular

>»m|lm. I

"*S,

** »

»»oa

JUA^

^«gSfie>u».J

«o»

»^

1 _^

iB"-«

»■-»

!••■■«

»• ti

IO-«4

»*-a

»• a

lO' }

ro'" ii

W t

10* M

!•• H

»• a

**• M

»• »40' MO- «a

Fio. 88.

beam and three connecting-rods, can be operated at a piston-
speed at high as 1,000 feet per minute without any reversal of
the forces acting upon the crank-pin, except at or near dead-center.
Mr. B. H, Thwaite. — ^I had no idea that I should be called into
a discussion on the subject of high power gas engines at this
meeting. It has been impossible for me to read the paper that
has been presented to you with the care it deserves; and, as I
have promised your President that^ I would prepare a paper for
your Society at your next meeting, I think you will perhaps
appreciate what I have to say upon the subject of gas engines
better if you wait until that time. Professor Reeve has already
answered one of the main mistakes in this paper, namely, the
impression that the gas engine is an explosive engine. It is
nothing of the sort. In the early days of the free piston that
statement might possibly have been correct. Xow we obtain with
gas of 100 to 120 B. T. U. a card whicih is quite different, and one
that produces a more satisfactory crank-turning effect. Instead of

A COMPACT GAS ENGINE: BEAM TYPE. 259

driving, direct on to the dead centre, I catch the maximum im-
pulse about 20 degrees above the neutral line, giving remarkably
good thermodynamic and uniform rotative results. The first
card of the free piston engine using gas of 700 to 800 B. T. U.,
and no compression or dilution, does give the impression of an
explosive engine, but now the modern gas engine is an essentially
slow combustion one.

I notice that Mr. Campbell makes a suggestion, in the foot-
note of the first page of Mr. Morgan's paper, that '^ it is possible
that exhaust gases from the gas engine can be profitably em-
ployed to heat the blast in the stoves of the blast furnace." I
think the thermal value of the exhaust gases is so low as not to
be worth anything. In our engine it is not much above 450
degrees Fahrenheit, and if you divide that temperature by half
you will see that for all practical purposes it will be of no use
for hot blast stoves of blast furnaces.

ilr. Morgan drew attention to the fact that Mr. Campbell
says that the blast furnace gas is a primary producer of iron
and also a gigantic gas producer. That is true. But the ordi-
nary producer gives a gas of a thermal value of 150 to 200
degrees, whereas blast furnace gas has a thermal value of no
more than 120 British thermal units, and sometimes as low as
80 B. T. U., a vast difference; indeed you must highly compress
it to find its thermodynamic value. All you have to do is to
compress it enough and then you will fire it with absolute or
mechanical certainty.

I notice also that Mr. Morgan speaks about the difficulty of
using blast furnace because of the dust it contains, which he
truly says has been the bete noir of the gas engineer. Some
nine years ago I killed this black beast and I have never had
any trouble with it since. In every instance of the proper
application of my system it has been a decided and continuous
success.

As to the centrifugal system referred to by Mr. Morgan,
that is and has always been more or less associated with my
plant. Other people are simply trying to do with the fan, and
some modification of my system, but if you do it that way you
^11 not only decrease the thermal results appreciably, but you
^1 have it to seriously increase the depreciation per cent.
The Cockerill engine, which is being constructed now by the
West Yorkshire Iron Company, England, is being equipped with

260 A COMPACT GAS ENGINE: BEAM TYPE.

my system for treating and purifying the blast furnace gases. I
would ask Professor Keeve if we are to understand that this
design has ever been carried out practically, or is it merely an
academic design?

Mr. Reeve. — ^It'is already in construction.

Mr. Thwaite. — I am not going to criticize this design severely,
because I always look with sympathy upon a new mechanical
development — and I hope this one will have a thorough trial; I
notice that Professor Keeve has already answered one point
which occurred to me as a difficulty, and that is the question of
the piston speed. When I saw this design I certainly thought
he would have some difficulty in getting the ideal piston speed
to secure the best thenual results. There is a certain piston
speed, i.e., 800 to 900 feet per minute, which enables you to get
the best thermal results out of the gas engine; and if Professor
Reeve tells me that you can get the speed, why, then you will
get a good thermal effect.

Then I notice that he has in the design a vertical blowing
cylinder. I have had a vertical blowing cylinder for my quarter
crank blowing engine all the time, and this gas engine of mine
works admirably and has given the greatest possible satisfaction.
I work it at a speed of 1:^ revolutions, and I get a remarkably
steady air pressure all the time. There is no more variation
than one-eighth of an inch of the mercury gauge, and the engine
works constantly night and day and for thirty-six days without
a stop.

The final part of Mr. Morgan's paper, which commenced with
a great deal of interesting philot^ophy — and that reminds me
that you have in this meeting gentlemen who are really philos-
ophers and who are really very witty, for yesterday you had
scientific reasoning gilded wdth humor, especially in the paper
presented by Mr. Barth. Finally, I find in Mr. Morgan's
paper this statement: "If the wastes of the old genera-
tion are the profits of the new then large dividends are w-aiting
us in the squandering of the past." I quite agree with that
statement. I believe, gentlemen, that it is your duty as en-
gineers to give some attention to this question of the internal
(combustion engine. I know you will be able to improve upon
it, and you have here a subject that is deserving of your most
serious attention; it is a tree of knowledge that deserves your
care, and I believe you will eventually reap satisfactory fruit

A COMPx\CT GAS ENGINE: BEAM TYPE. 2f>l

from it. Mr. Morgan also speaks of the fearful power wastes
of blast and coke furnaees. You must bear in mind that at
present you cannot continue operations when the demand falls
for pig iron, but with the new source of profit from the sale or
use of power the blast furnaces may be kept in operation, which
under the present conditions would be blown out. Certainly
blast furnace gas is absolutely ideal for generating power by
direct combustion, and is the best potential powder next to water-
falls in existence.

(lentlemen, I thank you for your kind attention.

J/r. //. //. Suplee. — ^I desire to call attention to what appears
to be an error in the paper regarding the action of the old
explosive gas engine. Mr. Campbell described the gas engine
as a gun whose projectile was connected to a crank shaft, but
that statement is not quite correct if he was referring, as I sup-
pose, to the old Otto & Langen engine. In that engine, the
immediate predecessor of the modem four-cycle engine, the
piston was not positively connected to the crank-shaft, the design
being what is termed a " free piston " engine, the piston rod
heing a rack gearing into a pinion on the crank shaft and con-
nected to it by a ratchet and pawl. The piston was undoubtedly
a projectile, but when it was shot upwards it was not connected
to the shaft, the power stroke being made by the downward
stroke of the piston, this stroke being due to the atmospheric
pressure on top of the piston, a vacuum being formed under-
neath after the explosion.

^^- Thwaiie, — I was only referring to the explosion on the
piston. That is what I had reference to.

^l^' W, II. Morse, — I am interested in this discussion because
1^ seems to advocate the vertical as against the horizontal
eyKnder. As a matter of history, I think the Westinghouse
Company some two or three years ago developed a vertical
^^^ne of comparatively high power. If any Westinghouse men
^re present I should like to inquire why it is that of late in a
number of papers we see illustrations of a 1,500 horse-power
double acting tandem engine wdth horizontal rather than vertical
<*ylinders.

ilr. E. A. Uehling. — There is no question before The Mechan-
ical Engineers and the iron manufacturers of this country of
greater importance than that of the utilization of blast furnaces
gas in internal combustion engines. Some time ago I took the

A COMPACT GAS ENGINE: BEAM TYPE.

pains to determine, as nearly as might be, the power that is
now wasted in the manufacture of iron. The method of pro-
cedure and results obtained were published in the Stevens In-
stitute Indicator in a paper called *' The Blast Furnace as a Power
Plant." Taking as near as possible the average practice of this
country and the output of pig iron of 1902 as the basis of cal-
culation, I found that there are in excess of one and three-
quarters millions of horse power going to waste in the manu-
facture of pig iron in the United States. Since then 1 have had
the opportunity of investigating the gases of several plants, and
I find that my figures are too low. In my calculations I found
that there w^ere, over and above the power requirements of the
blast furnaces themselves, and allowing suflScient gas for heating
of the blast, 840 horse power per ton of iron produced per hour,
available for outside purposes. In one plant, for instance, that
I investigated where they operate four blast furnaces and a large
steel works, if all the gas available, after deducting what is
necessary to heat the blast, were utilized direct in efficient gas
engines, there would remain over and above the 20,000 horse
power about 18,000 horse power for sale. The available sur-
plus power wall differ at different plants because it depends on
the class of iron made, the kind of fuel used and the character
of ore smelted; but it is in no case less than 800 horse power
per ton pf iron manufactured per hour.

There seems to have been, especially in this country, great
hesitancy in recei\dng this idea of utilizing blast furnace gas
direct in gas engines, and on the first casual glance it would seem
as though there could not be much in it, because it is a very lean
gas, having only less than ^ the calorific power of water gear;
but, as Mr. Thw^aite has just stated, it is none the less an ideal
gas for internal combustion engines, and when you come to
figure it through to the explosive mixture it is not a lean gas,
i.e., the heat effect does not differ so much in the explosive mix-
ture in the engine as might be expected from the difference in
calorific powder. For instance, take an illuminating gas of a
heat value of 764 British thermal units per cubic foot, when
mixed with the theoretical air required for complete combustion
it has only 101.8 heat units per cubic foot. An average blast
furnace gas with only 87.5 British thennal units per cubic foot
makes a theoretical niLxture of 68.9 heat units per cubic foot.
Thus we see that while the illuminating gas has nearly 9 times

A COMPACT GAS ENGINE: BEAM TYPE. 263

the calorific power of the blast furnace gas, when mixed with
sufficient air for complete combustion it has less than 1.5 the
heat value of a similar mixture of the latter. Adding the neces-
sary excess of air brings them still closer together. Illuminat-
ing ^as requires nearly 10 times the air necessary for blast fur-
nace gas; besides, the elements in illuminating gas do not lend
themselves to high compression, because you are liable to get
premature explosions; whereas the elements in blast furnace gas
not only lend themselves, but it is a necessity that they should
be highly compressed in order to make the explosion a certainty;
this is largely due to the high percentage of CO,, which is a
great retarder of combustion.

Jfr. F. R. Jones. — I wish a little more information; I wish
we might know where this horse power is from — whether it is
the heating power of the gas or the great power of the engine?

Mr. Uehling, — This rating is done by taking the heat units of
the gas, determined by calorimeter and checked by analysis, and
assuming an efficiency of twenty-five per cent, in the engine.
Much better has been done than that. I believe it has been
done up to thirty-two, and I think some tests have been made
where thirty-five has been realized, but I have taken twenty-five
per cent, as a basis, which I think is in every way conservative.

264 TESTS OF A COMPOUND ENGINE USING SUPERHEATED STEAM.

No. 1034.*

TESTS OF A COMPOUND ENGINE USING SUPER-
HEATED STEAM.

BY PROF. B. 8. JACOBUS, HOBOKKX, H. J.

(Member of the Society.)

1. The engine on which the tests were made was located at
the Millboume Mills in Philadelphia. It is of the Rice and
Sargent horizontal, cross-compound condensing type especially
designed for using highly superheated steam. A Schmidt super-
heater is used for superheating the steam.

2. The tests were made to determine whether certain guar-
antees were fulfilled which were made by the builders — ^the
Providence Engineering Works. The work was done conjointly
by Mr. A. C. Wood and the writer, Mr. Wood representing the
purchasers of the engine, and the writer the manufacturers. Mr.
A. S. Vogt, Mechanical Engineer of the Pennsylvania K. R.

•Co., at Altoona, detailed two experienced men from his testing
department to assist Mr. Wood, and with other observers and
computers, furnished by Mr. Wood, and the staff of the writer,
it was possible to check all important data by securing duplicate
records. The writer wishes to state that he is greatly indebted
to Mr. Wood for his valuable services and for his hearty co-
operation in the work.

3. The high pressure, cylinder of the engine was furnished
with double beat poppet valves and the low pressure cylinder with
Corliss valves. The inlet and exhaust valves of each cylinder
were operated by separate eccentrics. The inlet valves were
closed by vacuum dash pots and the exhaust by a positive motion.
A portion of the highly superheated steam furnished to the en-
gine passed through a coil in the receiver between the high and

* Presented at the New York meeting (December, 1908) of the American
Society of Mechanical Engineers, and forming part of Volume XXV. of the
Traruactions,

TESTS OP A COMPOUND ENGINE USING SUPERHEATED STEAM. 265

the low pressure cylinders. There was a by-pass valve operated by
the governor which regulated the amount of steam passing
through the coil, a greater amount being made to pass through
when the engine was heavily loaded thaa when it was lightly
loaded. There were no steam jackets. The exhaust steam passed
from the engine to a Worthington jet condenser ujonnected with
a Worthington vacuum pump.

4. The steam used by the engine was furnished by an Edge
i Moor water tube boiler. From the boiler it passed to the top of

the Schmidt superheater which was set up at the side of the
boiler. The saturated steam entering the superheater came in
contact with the coldest gases as they left the superheater. From
the upper coils the steam passed downward through a pipe out-
side of the superheater to the bottom of the lowermost set of coils
of the superheater and thus, when only partly heated, was made
to pass through the coils which were nearest the fire. This pre-
vented the overheating of the lowermost coils of the superheater,
which would have taken place had the steam been made to pas»
downward all the way through the superheater so as to bring
the hottest steam near the fire. From the lowermost set of coils
the steam passed upward and back and forth through the coils
until it was finally led off from a header directly below the top
set of coils and conveyed to the engine. Steam to drive the
vacniiin pump and for other purposes in the mill was furnished
^7 a separate boiler.

5. The principal dimensions of the engine, measured when
hot, were :

Bore of Cylinders 10.07 and 28.03 inches.

Length of Stroke 42 inches.

Diameter of Piston rods 8.00 and .3.60 inches.

6. The average clearance volumes computed from measure-
ments made on the engine and from the working drawings were
4.1 per cent, for the high pressure cylinder, and 5.8 per cent, for
the low pressure cylinder.

7. After the tests with superheated steam were completed a
test was made with saturated steam. The writer was not present
at this test but was represented by his associate, Prof. F. L. Pryor.

8. The principal results of the tests are given in detail in Table
I. From this it may be seen that the water consumption per hour
l>er indicated horse-power with superheated steam was 9.76

266 TESTS OF A COMPOUND ENGINE USING SUPERHEATED STEAM.

Teak No. J. May 27, J903.
Superheated Steam

Horse Power 474.5

Steam Pressure at ilirottle 141.2

Superheating at Tlirottle Deg.F. 852.5
Vacuum at Engine 25.11

rl40

140-1

100-

Jlfm.3mmk AM* Ck^r.

Fig. 84.

TESTS OP A COMPOUND ENGINE USING SUPERHEATED STEAM. 267

PresBore at ThroCfle

Test No. U May 27, 1901

Superheated Steam*
Hone Power 474.5

Water per Hour per Horse Power 0.70 lbs.

Superheating at the Throttle 8B8.5*'Fahr.

Superheating entering High Preesure Cylinder 879.srFahr.

Condensation

^iR.Aiia AMt G>.^.r.

Fig. 85.

pounds at 474.5 horse-power, 9.56 pounds at 420.4 horse-power,
and 9.70 pounds at 276.8 horse-power. The pressure of the
steam at the engine throttle was slightly over 140 pounds per
square inch, and the superheating at the engine throttle from
about 350 degrees to 400 degrees Fahr. The vacuum measured
near the engine ranged from about 25 to 26 inches of mercury.
The heat consumption per minute per indicated horse-power,
according to the standard recommended by the Civil Engineers
of London, where the engine is charged with the heat in the

TESTS OP A COMPOUND ENGINE USING SUPERHEATED STEAM.

steam at the throttle valve and is credited with returning the
feed water to the boiler at the maximum possible temperature
that could be obtained by a feed water heater, in the exhaust
pipe, was 205.0 British Thermal Units at 474.5 horse-power,
203.7 British Thermal Units at 420.4 horse-power, and 208.8
British Thermal Units at 276.8 horse-power.

9. With saturated steam the water consumption was 13.84
pounds per hour per horse-power at 406.7 horse-power, and the
heat consumption 248.2 British Thermal Units per minute per
indicated horse-power. In obtaining this figure the steam con-
densed through radiation of the steam pipe leading from the
boilers to the engine has been deducted in order to make the
figures for the heat and coal consumption strictly comparable
wdth those given for the superheated steam, which are based on
the temperature of the superheated steam at the engine throttle
valve. In this test the vacuum was poorer than in any of the
others, being 24.5 inches. It must also be borne in mind that
the engine was built -for superheated steam, and the cylinder
ratio was not what the makers would use for saturated steam.
In computing the heat consumption for this test an allowance is
made for the drip from the high pressure steam main at the coil
in the receiver. This drip amounted to 280 pounds per hour over
what could be accounted for by the radiation of the steam main
from the boiler to the engine. This amount of the high pressure
steam was, therefore, condensed in the receiver coil, and imparted
heat to the receiver.

10. Tests were made to determine the economy of the boiler
and of the superheater. The coal used was run of mine, Georges
Creek, Cimiberland. The equivalent evaporation from and at
212 degrees Fahr. per pound of dry coal for the boiler alone
was 10.30 pounds, for the- superheater it was 8.97 pounds, and
for the boiler and superheater combined it was 10.07 pounds.
The heat of combustion of the coal determined by means of a
Mahler bomb calorimeter was 14,060 British Thermal Units per
pound. The efficiency of the boiler based on the heat of the coal
was 70.7 per cent., of the superheater 61.6 per cent, and of the
combined boiler and superheater 69.2 per cent. In computing
the efiiciency of the superheater the specific heat of the super-
heated steam was taken at the vahie found by Regnault for
atmospheric pressure or .48. At higher pressures this value is
probably too low. On the other hand the amount of superheat-

TESTS OP A COMPOUND ENGINE USING SUPERHEATED STEAM. 269

ing given by the mercury thermometers used in the tests is
greater than that which would have been given by an air ther-
mometer, and this introduces an error in the opposite direction
from that which probably exists through taking the specific heat
of the steam at .48 and the one error therefore, tends to counter-
balance the other.

11. The coal consumption may be obtained directly from the
heat consumption of the engine already given by multiplying by
60 and dividing by 9,725 for the superheated steam tests, and by
0,947 for the test with saturated steam, 9,725 being the heat im-
parted to the boiler and the superheater in British Thermal Unite
per pound of coal burned in the tests with superheated steam and
0,947 that imparted to the boiler in the test with saturated steam.
When computed in this way any loss of heat by radiation in the
pipe leading from the boiler to the engine is eliminated, which is
a necessary condition in the present case as the pipe is exception-
ally long. On this basis the coal consumption per hour per indi-
cated horse-power for superheated steam was 1.265 pounds at
474.5 horse-power, 1.257 pounds at 420.4 horse-power, and 1.288
pounds at 276.8 horse-power. For the test with saturated steam
the coal consumption was 1.497 pounds per hour at 406.7 horse-
power. In the figures just given for the coal consumption the
coal used to drive the independent vacuum pump is not included.
The fipres, therefore, represent what would have been obtained
per indicated horse-power if the vacuum pump had been driven
from the main engine.

12. The coal consumption forms a more accurate basis for
comparing the efficiencies of engines using saturated and super-
heated steam than the heat consimfiption, because when the coal
consumption is obtained it allows for any difference in the econ-
^^y of the combined boiler and superheater from that of an
ordinary boiler. Furthermore, the results obtained for the coal
consumption are practically independent of the value employed
for the specific heat of the steam which is not the case with the
heat consumption. For example, if the specific heat was taken
at .6 instead of .48 in computing the results which have been
given it would cause a difference in the coal consumption of less
than one-half of one per cent., whereas it would cause a differ-
ence in the heat consumption of about 3^ per cent.

13. In the test with saturated steam the coal consumption was
about 19 per cent, greater than that obtained at the same power

270 TESTS OP A COMPOUND ENGINE USING SUPERHEATED STEAM.

Tat No. 2. June t% J903.
Sttperiieated Steam*

Horse Power 430.4

Steam Pressure at Engine 142.4

Superheating at Throttle Deg.F. miJi
Vacuum at Engine

140-1

AtmoBpherlc

wlm.AraJbAM»C>.^.r.

Fio. 86,

TESTS OF A COMPOUND ENGINE USING SUPERHEATED STEAM. 271

Pre— qre at Throttto

Superheating
i 154° Fahrenheit

TtttNo.2. June t% J903.
Superheated Steam*
Hone Power 430.4

Water per Hour per Horse Power 9.66 lbs.

Superheatliig at the Throttle 874.:f Fahr.

Superheating entering High Pressure Cylinder 296.6fFahr.

^4T-

_0.7S» Condensation .

Am.3mnk Uttt Ot.^. T.

Fig. 87.

with superheated steam. This does not form a proper basis of
comparison, however, for the two reasons already stated, which
are that the engine was not built for saturated steam and that
the vacuum in the test was not as high as it should have been.
An engine of the same general type built for saturated steam,
but of about twice the power, was tested by the writer at the
Brooklyn plant of the American Sugar Refining Company and
the results were embodied in a paper presented by him at the Sara-
toga meeting. The most economical heat consumption obtained

272 TESTS OF A COMPOUND ENGINE USING SUPEBHEATED STEAM.

Tat No. 3. July J7, J903.
Superiieated Steam*

Horse Power 278.8

Steam Pressure at Engine 145.8

Superheating at Throttle Deg.F. 806.8

Vacuum at Engine 86J»

Atmospheric

Atmospherio

Atmospheric
L. P. - C. E.
19.971b. Spring

Awufi«n» Stt 09.^.r.

Fig. 88.

TESTS OP A COMPOUND ENGINE USING SUPEEHEATED STEAM. 273

PrMgore mt ThrotOe

Twt No. 3- July J7, (Ml

Superheated Steam.
Uonne Power

Water p«r llour per Bore« Power
Sapcrbei^ln^ ttt the Throttle

SnporUeatiiitf eiitcring lll^h ppessur© Cylinder 3Qa.:fFahr.

275.8
9, 70 lbs.

:^hlt% Condensatton

Am.BmHkNH»C».^.r.

FlO. 89.

was 222.7 British Thermal Units per minute per horse-power.
The vacuum in this case was exceptionally high, being over 28
inches. If the vacuum had been about 26 inches, as in the pres-
ent tests, it is probable that the heat consumption would have
been about 230 British Thermal Units per minute per indicated
horse-power, and the coal consumption on the basis already given
1.42 pounds per hour per horse-power. As the minimum coal
consumption with the engine using superheated steam was 1.257
pounds per hour per horse-power there is a gain in favor of using

274 TESTS OF A COMPOUND ENGINE USING SUPERUEATED STEAM.

Test No. 4. Jtsly H, )903.
Saturated Steanu

Bone Power
Steam Pressure at Engine
Saturated Steam
Vacuum at Engine

40Cr
145.1

rl40

-100

-60

Atmospheric

140-1

Aw^MamkJtk»$Ck.^r.

Fig. 90.

TESTS OF A COMPOUND ENGINE USINQ SUPERHEATED STEAM. 275

Pressare at Throttla

Tcit No. 4. July 24> J903.

Saturated Steam.

Horse Power 40C.7

Water per Hour per Horse Power 18.84 lbs.

:=a 12.0^

Fig. 91.

superheated steam of .163 pounds per hour per horse-power, or
the larger steam engine using saturated steam is ahout 13 per
cent, less economical than the smaller one using superheated steam.

Methods of Condn<^ting the Tests.

14. The water was weighed in an iron tank of about 1,000
pounds capacity placed on a platform scale, and it was then
emptied into a lower tank connected with the suction pipe of a

276 TESTS OF A COMPOUND ENGINE USING SUPERHEATED STEAM.

geared boiler feed pump driven by the main engine. All steam
and water connections, where there was a possibility of leakage
occurring, were either broken or blanked off.

15. Indicators of the Star Brass Manufacturing Co., having
outside springs, 'were used on the high pressure cylinder. Tabor
indicators of the ordinary pattern were used on the low pressure
cylinder.

16. Mercury columns were used to measure the vacuum both
in the exhaust pipe near the engine and near the condenser. A
spring vacuum gauge was also used near the condenser which was
calibrated in order to determine its correction.

17. Mercury thermometers placed in thermometer wells were
used for determining the temperatures. Where the temperatures
were not too high mercury^ was used in the wells, but if the mer-
cury evaporated, melted solder was employed. In every case the
error due to exposing a portion of the stem of the thermometer
was determined by noting the graduation to which it was im-
mersed and estimating the temperature of exposed part of the
mercury in the stem by tying a second thermometer against the
stem.

18. Readings were taken at intervals of 15 minutes through-
out the tests. The water record was balanced up at the end of
each hour. The feed pump was kept running uniformly and
was not stopped in balancing up the water. The engine was run
for some time before starting each of the tests in order that the
temperatures should become practically constant. The steam
leakages were carefully estimated, and in one case a leakage test
was made on the boiler and the results confirmed the amount
which was estimated. The steam leaks amounted to 40 pounds
per hour as a maximum.

19. The engine test of June 19th was run at the same time as
the tests of the boiler and superheater. Arrangements were
made for a continuous engine test of twelve hours or more, but
six hours after starting the engine test the water rate suddenly
increased. It was found that a gasket at the receiver head which
had been leaking slightly during the first part of the test, and
the leakage of which had been estimated at three pounds per
hour, was leaking much more than at first. It was impossible to
estimate the amount of the increased leakage, and the engine
test was therefore discontinued at the end of six hours. As the
leak did not affect the tests of the boiler or superheater these

TESTS OF A COMPOUND ENGINE USING SUPERHEATED STEAM. 277

tests were made of 14 hours duration. The readings of the
amount of feed water indicated uniform conditions during the
engine test, the average rate per hour for the first two hours
being 4,026 pounds; for the second two hours, 4,012 pounds, and
for the last tw^o hours of the test, 4,016 pounds. This shows that
the duration of the test was ample for securing accurate results.
20. The mean effective pressures and indicated horse-powers
are given in Table II. Table III gives the averj^e data and
results of the boiler test, Table IV those for the superheater
test and Table V the data used in computing the superheating
or the percentage of condensation existing in the cylinder at
various points in the stroke. The average indicator cards and
the combined diagrams for each of the tests are shown in Figs.
1 to 8 inclusive. The amount of superheating or of priming
existing at various points in the expansion lines is marked on the
combined diagrams. Zeuner's formula for suj>erheated steam
is used in computing the amount of superheating.

278 TESTS OF A COMPOUND ENGINE USING SUPERHEATED STEAM.

•-3

M5CO COOOO

OQ0Ot>-;Di-i

oi ?o c^i t>I CO '^*
-g; 35 1^ X »2 <N

rHCO»oooa

05 ^ OC t* 00

OCOOU5 05C1

go ^' QC ^ »C QQ
X l^ ?0 CO CO «-H

CO "^* CD t>! --^

d .^

ocs ^x^

'^t^'^iOO^

'^icc^t^ci

2H?5^^S ISiliS iigS§

2 o a

9)15 o

.2 22 ® £- ^ 3
d) era c 2 !-
S m 3 = g g

. k. ^ 3

a« C je &> u 'S

O o ^

""(1) £ c 8 g

«c.Egg|
ES£22|

»c CO* r>^ 06 05 d

^ ^-^bC

S -*^ c: S c s
1-i c4 CO -^ to CO

C.5;g

£££

S ^

J2 aoua

3*^-fiJ=^

.£ s-cS-c
s «d c S c=

t>^o6o>o^

TESTS OF A COMPOUND ENGINE USING SUPERHEATED STEAM. 279

s gs

280 TESTS OF A COMPOUND ENGINE USING SUPERHEATED STEAM.

•-^

Tj« 00

r^.

d CO

TJ4

S ?;

.2 si

»c

CO 00

^S'^

«§•

00 o

•S.5'0

i i

Isf

s ?

°.

l|a

oi ci

o o

*-4

1-^ I-l

l°s.

^ j:

^ oi4

^*"o

to 00

t>^

< 1

I-l

1-^

1-H

■

^ ^

•d

to 00

r>^

H '

Oi t^

OS OS

»-< -M

f-4

i:^*

^ o

•d

£«

W3 00

r^.

t>. 05

d d

i-i

»-< »-<

»H

3 s

I-l

S §8

C^l N

00 00

t^ t^

> .

00 00

i^ r^

• 1

WXJO

•OM

»H

CI CO

TESTS OF A COMPOUND ENGINE USING SUPEBHEATED STEAM. 281

•

TABLE UI.

Average Data and Results of Boiler Test.

Made June 19 and 20. 1903.

1. Size of grate; length, 7 feet; breadth, 9 feet, 6.5 inches; area,

square feet 66.8

2. Water heating surface in square feet 3,332.

3. Superheating surface in souare feet 350.

4. Ratio of total heating surtace to grate surface 55. 1 : 1

5. Horse-power of boiler as rated by builders 370.

Total Quantities.

6. Length of test, hours 14

T TP • U4 r • * ^ 1 r Total fired 7.205.

^' K*^!l!^ r.^^l w J Correction due to thickness of fire at

burned during test, ^ end of test 158.

inpounds. [ Corrected weight 7,047.

8. Weight of dry coal burned during test, in pounds 6,823 .

9. Total weight of water evaporated during test, in pounds 58,585.

10. Equivalent weight of water that would have been exaporated

from and at 212 degrees Fahr., in pounds 70,302.

1 1 . Weight of ashes, in pounds 717.

Average Quantities.

12. Steam pressure in pounds per square inch above atmosphere. . 147.4

13. Temperature of feed water in degrees Fahr 66.6

U.Ten>peratureoffluega«e«indeg„^Fahr.jAet^Xa^i„K:: m.

15. Temperature of boiler room in degrees Falir. j ^y^ bulb 71 6

• y, -r* u^ • • 1 r-^ *- ^ In back connection 0.1

16. Draught mmches of water j overthefire 0.1

17. Percentage of moisture in coal 3. 18

18. Percentage of ash, based on dry coal 10.51

19. Barometer, in inches of mereuiy 29.82

Economic Results.

20. Factor of evaporation 1 . 200

21. Weight of wat^ evaporated per pound \ promind at 2i2del* ' ' ^'^^

of fuel, mcluding moisture, m pounds I ^^ p^^^. ^ ^^

22. Weight of water evaporated per j Actual 8.59

pound of dry fuel, in pounds / From and at 21 2 degrees Fahr. 10 . 30

23. Weight of water evaporated per j Actual 9 . 59

pound of combustible, in pounds ) From and at 21 2 degrees Fahr. 11.51

24. Weight of diy coal burned per hour per square foot of grate sur-

face, in pounds 7.30

25 Calorific value of the dry coal in British Thermal Units per

pound 14,060.

26. Calorific vahie of the combustible in British Thermal Units per

pound 15,510.

27. Efficiency of boiler (based on the combustible), in per cent 71 .7

28. Efficiency of boiler including the grate (based on the dry coal),

in percent ^. 70.7

29. Water evaporated, in pounds per square foot of total heating

surface per hour, from and at 212 degrees Fahr 1 . 36

30. Average horse-power developed based on standard recom-

mended by A. S. M. E., viz., 34.5 pounds of water evaporated

from and at 212 degrees Fahr., per hour 145.7

TESTS OF A COMPOUND ENGINE USING SUPERHEATED STEAM.

TABLE IV.
Average Data and Results op Superheater Test.

1. Heating surface in square feet. 642.

2. Grate surface in square feet 4

3. Duration of test in nours 14

4. Pressure of steam furnished to superheater in pounds per

square inch above the atmosphere 147 .4

Temperatures in Degrees Fahr, by Mercury Thermomeler.

5. Steam entering the superheater 365.6

6. Steam leaving the superheater 809. 1

7. Amount steam was superheated 443.5

Total Quantities.

8. Steam passing through the superheater in pounds 58,025.

9. Heat imparted to the steam in British Thermal Units 12,352,000.

10. Total moist coal fired, in pounds 1,473.

11. Total dry coal consumed m pounds 1,426.

12. Coal burned per square foot of grate surface per hour in

pounds ." 25.5

13. Heat imparted to the steam in the superheater per pound

of coal burned, in British Thermal Units 8,662.

14. Equivalent evaporation from and at 212 degrees Fahr. in /

pounds per pound of dry coal 8 .97

15. Heat of combustion of the dry coal in British Thermal

Units per pound 14,060.

16. Efficiency of the superheater based on the heat of com-

bustion of the coal in per cent 61 .6

Combined Economy of the Boiler and Superheater,

17. Actual evaporation of the boiler per pound of dry coal in

pounds 8.586

18. Coal burned in su]>erheater in pounds per pound of steam

passing through it .02458

19. Coal burned in superheater in pounds per pound of coal

burned at boiler. Item 17 x Item 18 .211

20. Actual evaporation of the combined boiler and super-

heater, Iteml7 + (1 + Iteml9) 7.090

21. Factor ojf evaporation of the combined boiler and super-

heater 1 .421

22. Equivalent evaporation in pounds per pound of dry coal,

from and at 212 degrees Fahr., for the combined boiler

and superheater 10.07

23. Efficiency of the combined boiler and superheater based

on the neat of combustion of the coal, in per cent 69 . 2

TESTS OP A COMPOUND ENGINE USING SUPERHEATED STEAM. 283

TABLE V.

Data us£d in Computing the Amounts op Superheating at Various
Points in the Expansion Line of Indicator Diagrams. Average
FOR SIX Diagrams.

Weights of Steam per Stroke in Pounds,

I
May 27 June 19

Date of Test ;

Passing through cylinders of engine

Retained in clearance space of high pressure
cylinder

Retained in clearance space of low pressure

cylinder ! 0.0345 0.0319

Total contained in high pressure cylinder dur-^ j

ing exp^ansion

Total contained in low pressure cylinder dur-
ing expansion

May

0.3739 0.3272

0.0619 0.0606

0.4358, 0.3778
0.4084 0.3591

July 17
0.2182

0.0299

0.0357

0.2481

0.2539

July 24
0.4362

0.0693

0.0489

0.5055

0.4851

284 TESTS OF A COMPOUND ENGINE USING SUPERUEATED STEAM.

TABLE W.^Continued.

Datk of Test.

111

verage lentttli of

dicator cards in

Inchea.

II
it

Length Including

clearance, to the

exptineion curve,

in inchcD.

III lip ml

^^- '^^1^

<£t^^

H.P.

May 27

Superheated.

4.97

120

2.09

4.67

528

187 deg.

•♦

110

2.21

4.94

497

163 "

«<

«•

2.60

5.81

460

140 *•

««

3.20' 7.16

419 '116 "

*♦

«*

4.14 ! 9.28

354 ' 73 •*

LP.

2.03 1 14.88

292

44 **

t«

2.42, 17.74

249

14 *'

2.99 121.92

225

3 "

1*
l<

' **

14
10

3.61 126.46
4.83 35.41

4.1p.c.
6.3 *•

June 19

HP.

4.18

120

1.47

4.50

495

154 deg.

110

1.59

4.88

487 152 **

1.88

5.76

454 1134 "

««

2.30

7.06

408 |105 '*

3.00

9.20

348 67 *'

37.8

3.71

11.37

294

30 "

««

L.P.

4.16

1.87

18.63

282

47 "

2.28

22.71

247

25 "

2.75

27.39

0.7 p.c.

3.65

36.36

3.8 '•

July 17

HP.

4.18

120

0.91

4.25

448

107 deg.

•<

*•

110

0.96

4.48

420

85 "

»<

1.14

5.33

393

73 "

1.38

6.45

344

41 "

1.83

8.56

298

18 "

*«»

2.71

12.66

6.1 p.c.

LP.

2.08

29.16

242

*«

2.72

38.14

195

2 "

3.72

52.16

1.6 p.c.

July 24

Saturated.

4.17

115

1.33

3.05

20.0 "

100

1.52

3.49

19.9 "

1.67

3.84

.... 19.8 "

2.10

4.82

.... 20.8 "

2.88

6.62

.... ,20.6 "

3 60

8.27

.... 19.5 "

LP.

4.18

1.77

12.99

.... 115.8 "

n
tt

2.00 14.68
2.55 18.71

.... 15.2 "
14.1 "

3.80 27.89

.... 12.6 "

TESTS OF A COMPOUND ENGINE USING SUPERHEATED STEAM. 286

DISCUSSION.

Mr. George I. Rockwood. — I have not seen an advance copy
of this paper by Professor Jacobus, so in what I have to say I
cannot attempt to speak with any preparation at all. The fig-
ures speak for themselves, however, and suggest some questions.
For instance, in two tests — numbered 2 and 4 — one with and the
other without superheat, the load being substantially alike in
each case — being 420 horse-power with superheat, and 406 with
saturated steam — the steam consumption per horse-power per
hour is p:iven as 9.5 pounds and 13.8 pounds respectively. The
difference represents an apparent saving of over 30 per cent, due
to superheating, which is so much greater than usual as to need
explanation. Now, as the pa})er deals with the economy of
steam obtain(^l by the use of superheaters, it may be in order
to (liscnas other sides of th(» general question of the actual econ-
omy of coal effected by their use.

A factor which greatly affects their value is the manner in
which the regulation of the superheat is accomplished. To illus-
trate, I lately visited a turbine station in Rhode Island and saw
there one of these independently-fired superheaters at work, and
in this ease the upper doors were just a little open and the ash-
pit doors open too; this was to prevent too hot a fire, as with
the fire doors closed th^ temperature of the steam would be too
hot for safe use.

Another case of the same sort of temperature regulation is
found in the provision of a hollow bridge wall and a register
on the outside face of the brick setting, for the purpose of cool-
ing the gases as they come from the fire. It is obvious that such
means of temperature regulation involve a loss of heat out of
proportion to any saving effected in the engine or turbine. Its
use is permissible only in cases where the steam exhausted from
an engine is to be used in manufacturing processes after being
.superheated.

The most e<;onomical way, and, indeed, the only economical
way, to preserve the temperature of the steam at a constant
point is to keep the weight of steam flowing through the super-
heater proportionally the same as the weight of coal burnt on the
superheater grates. One way I have employed of getting around
the difficulty is to provide a " bleeder " pipe at the engine throttle

286 TESTS OF A COMPOUND KNGINE USING SUPERHEATED STEAM.

and to open it when the load is light or the engine shut down.
In this case there are 3,000 horse-power of boilers and only ten
per cent, of the steam is made at a high pressure — 200 pounds
per square inch — for use in the engine. The rest is used at 100
pounds pressure in heating water in a dye-house. The " bleeder"
simply diverts some of the steam intended for use in the engine
into the dye-house pipes after it has passed through the super-
heater. In any bleachery or dye-house or steam-driven paper
mill such a method of regulation would be effective. The thing
to be avoided is too hot a fire under the superheater with partial
load, as this results in burning out the superheater tubes and
ruining the engine or turbine.

Mr, R. H. Rice, — It may be interesting to give some further
particulars of the construction of this engine which is utilizing
very high temperatures of steam without difficulty — in fact, with
no more trouble than any of the ordinary saturated steam en-
gines of the same builders.

The double beat poppet valves have seats surrounded by the
inlet steam in such a way that the expansion of the seat is equal
in extent and effect to that of the valve — thus overcoming com-
pletely the characteristic defect of ordinary designs of their type
of valve, namely, excessive leakage at any temperature other
than the particular one at which they were originally ground.
The inlet valves are driven by the ordinary trip gear of the
builders, with vacuum dashpots, wath th^ addition of a simple
linkage which controls the closure of the valve independent of
the extent of closing motion imparted by the dashpot, and thus
prevents slamming or partial closure of the valve. The exhaust
valves are actuated by a system of links devoid of cams, always
in connection with the eccentric except when hand-actuated at
starting or stopping, and which keeps the valve stationary during
the forward stroke, as is necessary when using the poppet type,
and all joints are adjustable for wear.

The stuffing-boxes are on long necks to take them well away
from the superheat, and the piston-rod stuffing-boxes have
metallic packing provided with water jackets which, however,
have never been used. The piston packing consists of two simple
cast-iron spring rings with joint plates.

The horse power cylinder is so designed that the working
portion of its band is a simple cylinder without ribs, all connec-
tions to the cylinder, such as valve elects, lagging bosses, inlets

TKTS OF A COMPOUND ENGINE USING SUPERHEATED STEAM. 287

and exhaust gauges, etc., being at the ends. The clearance of
the poppet valve cylinder has been kept fully as small as the
equivalent Corliss-valve cylinder, in contrast with foreign prac-
tice, where it is usually much greater.

The only trouble noticed with lubrication was a smoking due
to the carbonization of the animal or vegetable constituents of
the original oil used. On notifying the oil makers of this trouble
they at once produced an oil which eliminated all complaint.

The operation of the superheater has proved to be simple: in
fact it is easier to run than a boiler, since the pyrometer dial is
the only thing needing to be watched. Fire is never built in the
superheater without a flow of steam through the coils, under
which conditions there is no sign of deterioration. The tem-
perature is readily regulated, even when the engine is shut down
for changes in the mill, which happens once or twice in 24 hours
in regular operation. If the shut-down is for more than a few
minutes a small flow of steam is secured by cracking the throttle
valve and allowing a little steam to blow through the engine,
bat for short stoppages this is not necessary. The pipes, cyl-
inders and receiver are covered with 3 inches of a standard
magnesia covering over pipes and flanges.

Lines 13, 14, 15, Table I, page 6, of the paper shows interest-
ingly the control of the temperatures of steam entering the two
cylinders by the governor-controlled valve of the reheater, test
Xo. 3, with light load, showing an inlet temperature of 672 de-
grees to high pressure cylinder and 353 degrees to low pressure
cylinder, while test No. 1, with heavy load, shows 634 degrees
to high, and 408 degrees to low. This variation of temperature
inversely with load in high cylinder is believed to be necessary
with high superheat to protect it from burning the oil. In this
engine the regulation is effected by giving to the low pressure
cylinder the superheat which has been taken out of the steam
going to the high cylinder.

Although the cylinder ratio is different from that usually
adopted in this country for saturated steam, it nevertheless cor-
responds with European practice, and considering the low
vacuum and the absence of jacket, the steam consumption found
— 13.84 pounds — is not an excessive figure.

In comparing the coal consumption of this engine and that
reported by Professor Jacobus at the Saratoga meeting, it should
be noticed that the best heat consumption of the latter engine

288 TKSTS OF A COMPOUND ENGINE USING SUPERHEATED STEAM.

was obtained with so low a m. e. p. that the engine, if propor-
tional for this m. e. p. at full load would be excessive in cost on
account of its considerable cylinder dimensions; so much so as
to make its cost comparable to that of a superheated steam
engine and its superheater complete, whereas the superheated
steam engine shows a saving of about 12 per cent, of coal, which
is therefore obtained vnih only a small additional investment.

Tests of engine constructed on this system in England with
a vacuum of about 28 inches of mercury show a steam consump-
tion of 9 pounds per indicated horse power, and it is probable
with a load of about 875 horse power (which could not be ob-
tained in this case owing to the arrangement of the mill), and
ail equally good vacuum, that the same figure could have been
obtained in this case.

In regard to the superheater at Newport, I am somewhat
familiar with that installation. The superheater was made of
sufficient capacity to superheat the steam for 2,000 kilowatt
capacity of turbines, and it is now superheating steam for less
than 200 kilowatts capacity. It is, therefore^ operating under
entirely abnormal conditions.

Mr, E, T. Child, — I would like to ask Professor Jacobus what
was the ratio of cylinders in the Brooklyn engine? And is this
• ratio of 3 to 1 in the Millbourne Mills engine the blast for super-
heated steam? It is often desirable to use superheated steam in
engines already built and with cylinder-ratios adapted for satur-
ated steam. Will these larger cylinder-ratios make the engine
illy adapted for superheated steam?

Professor Jacobus, — What is your last question? I did not
quite catch it.

Mr, Child, — Is this ratio of 3 to 1 distinctly based on the use
of superheated steam, and, if you have an engine built for satur-
ated steam, say, 4 or 5 or 6 to 1, is that illy adapted for use with
superheated steam?

Professor Jacobus, — The ratio in the Brooklyn engine was
about 4 to 1. In regard to the last question, more experiments
are needed before it can be answered. Perhaps Mr. Rice can
give us some more information on this point as he is the one
who decided to use the 3 to 1 ratio.

Mr, Rice, — In adopting this ratio we adopted the foreign
practice; and I presume that ratio was adopted abroad for two
reasons: First, that it corresponds closely to their saturated steam

TESTS OF A COMPOUND ENGINE USING SUPERHEATED STEAM. 289

practice; and, secondly, by having a large, high-pressure cyl-
inder they are able to use more superheat in that cylinder.

Mr, A. E. Cltiett. — I would like to inquire whether outside
of the cylinder-ratios any data is available regarding the amount
of superheat that could be applied to an ordinary compound
engine of about the same size as the one under discussion, with-
out endangering the engine in any way?

Mr. Rice. — I cannot answer this, because it depends largely
on the character of the load, the point of cut-off in the cylinder,
and various other elements; but in general a temperature in
existing engines of over 500 degrees Fahrenheit would give
trouble.

Prof. W. F. M. O088. — It has seemed to me that the use of
superheated steam in connection with multi-cylinder engines is
somewhat illogical, and I am interested to know whether Pro-
fessor Jacobus regards the results of his tests as a clear justifica-
tion for such a combination. The large amount of cylinder-con-
densation, and the consequent low efficiency which attend the
action of the single-cylinder engine may be reduced in several
different ways. One way is to multiply the number of the en-
gine cylinders, and another way is to preserve the engine in its
simplest form and employ a superheater. Either plan tends to
the same result. In either case, also, the improvement is at the
cost of added complication in the design of the plant. In one
case, this complication takes the form of added cylinders, in the
other of an independent device in the form of a superheater.
My point is that by following either one of these lines of develop-
ment to the exclusion of the other, results can easily be obtained
which approach closely those given by the compound super-
heating plant described by Professor Jacobus. If this is true,
where is the argument in favor of adding to* a plant complica-
tions of two different sorts. If, for example, a compound engine
will not give the desired degree of efficiency, would it not in
most cases be better practice to endeavor to better the perform-
ance by extending the series to a triple expansion typo, rather
than by introducing a device requiring so much attention as a
superheater? Or, if it has been decided to employ superheated
steam and to meet the problems incident to such practice, then
why not seek the full advantage of its use by reducing the engine
to its simplest form, using a single cylinder only.

I do not wish to be understood as criticising in the least the

290 TESTS OF A COMPOUND ENGINE USING SUPERHEATED STEAM.

excellent paper vrhich Professor Jacobus has presented. I
merely raise the question as to whether the combination which
is presented in the engine he describes is one which is likely to
find favor in future practice. In view of his very careful study
of the subject, a statement from Professor Jacobus as to the
arguments which go to sustain the peculiar combination of ele-
ments which the engine he describes presents would be most
valuable.

Professor Jacohtis* — In reply to Professor Goss will say that I
feel that the gain in economy secured by superheated steam may in
many cases justify its use. It is true that with saturated steam
we may step from the simple engine to the compound, and from
the compound to the triple with a gain at each step, but it also
appears that in any type of engine there may be a gain due to
superheating. The main point to be determined is whether the
additional gain in economy due to the superheating will more than
offset the interest on the increased capital investment, and the
item for depreciation and repairs of the superheater. Mr. Rice
has pointed out that if a compound engine using saturated steam
is proportioned to give the best economy the mean effective pressure
would be so low as to lead to large dimensions, and that when^so
constructed its cost is comparable to that of a superheated steam
engine and its superheater complete. One thing is certain, and
that is that the advisability of using superheated steam depends
on the cost of the coal used. If coal is expensive it may pay to
use superheated steam, whereas with a cheaper coal it may not. I
hope to go into this matter of costs more carefully and presents the
results at a later meeting.

Mr. Geo. II. Barrus kindly pointed out after reading my paper
that by basing the heat consumption on the temperature at which
the superheated steam is furnished to the engine, I have neglected
any difference that there may be with saturated and superheated
steam in the radiation of the steam mains leading from the boiler
to the engine. Any excess of radiation with the highly super-
heated steam over that for saturated steam, should be charged
against the superheated steam plant and the saving that is shown
for the superheated steam should, therefore, be somewhat less
than I have given. Taking the Millbourne Mills engine there
would be a difference in the radiation of the steam mains with

* Author's Closare under the Rules.

TESTS X)F A COMPOUND ENGINE USING SUPERHEATED STEAM. 291

saturated and superheated steam, which would affect the economy
by about one per cent. The steam main of this engine is, however,
excessively long, being over twice the length that might ordinarily
be used. It is fair to say, therefore, that where the steam main
is of the ordinary length and is covered in the same way for satu-
rated and superheated steam, the results for the heat consumption
with superheated steam, computed as has already been explained,
should be discounted by about one-half of one per cent. ; a quantity
which is well wathin the range of accuracy of the tests. As the
eovering^ for superheated steam is usually heavier than for
saturated steam, this small difference due to the radiation of the
steam main wdll be still further diminished. The mains carr\ing
superheated steam at the Millbourne Mills are 6 inches in diameter,
and are covered with a double layer of asbestos and magnesia
covering. The inner layer is asbestos fire felt 1| inches in thick-
ness, and the outer magnesia strips of about the same thickness
held in place by wire and plastered with asbestos cement. The
average outside diameter of the covering is 12^ inches.

292 STANDARD UNIT OF REPBIOERATION.

No. 1095.*

STANDARD UNIT OF REFRIGERATION.

BT 3. C. BSRTSCH, ATLANTA, GA.

(Member of the Society.)

1. Since three-quarters of a century ago, mechanical refrigera-
tion has been in practical use. For the first period of about fifty
years it was applied most exclusively for making ice and cooling
liquids in small quantities only. But during the past twenty-five
yeat's its applications have increased so wonderfully and on such
a large scale that it must now be taken care of as a special line of
business by the refrigerating engineer, and it can no longer be
handled in connection with any other branch of mechanical engi-
neering.

2. Kefrigeration is now successfully applied in numerous indus-
tries, either as a valuable assistance or as a most necessary part
for obtaining quicker and better results in the manufacture of
many articles.

3. As an exclusive process mechanical refrigeration is used for
making ice, preserving perishable goods, controlling fermentation,
fractional distillation, assistina: in mining and for many other
purposes. It will sooner or later be applied to our dwellings and
public halls, not only to replace the ice in the refrigerators, but to
furnish also cooler ^nd better air during the hot season of the
year.

4. In spite of the many years of practical application, and in
spite of the thousands of refrigerating machines in operation all
over the civilized world, no standard unit of refrigeration has as
yet been established. We have to-day no commonly adopted pro-
portions or uniform rules to measure the actual work produced by
the refrigerating machine save the adoption of 284,000 British
thermal units as an equal to the work accomplished by melting one

♦ Presented at the New York meeting, December, 1903, of the American
Society of Mechanical Engineers, and fonning part of Volume XXV. of the
TranBocHons,

STANDARD UNIT OF RETBIOEEATION. 293

ton of 2,000 pounds of ice of 32 degrees Fahr. to water of 32 de-
grees Fahr. But even this value is not properly applied, inasmuch
as it is correctly used for calculating the work equal to ice-melting
only; as soon as ice-making is to be calculated, two tons of ice-
melting are considered as equal to one ton of ice-making, regard-
less of the fact that under ordinary condition the making of one
ton of ice aniounts only to about 410,000 to 420,000 British ther-
mal units instead of 568,000 British thermal units.

5. The absence of a standard unit, proportions or rules is the
cause of much confusion and many disputes. Everyone is at lib-
erty to choose any rating he sees fit, and as each rating or guaran-
tee can be substantiated under certain assumptions, the purchaser
of a machine is at a loss to know what he is getting for his money.

6. We find to-day the capacities of machines based on a con-
densing water of from 60 up to 75 degrees; compressor displace-
ments of from 6,500 up to 8,500 cubic inches per ton-minute, and
guarantees for power required of from one up to two horse-power
V^T ton refrigeration in 24 hours.

7. Realizing that such conditions can not be tolerated much
longer, and that uniform rules for this special branch of engineer-
ing are badly needed, the Southern Ice Exchange, in convention
at Atlanta, Ga., last February, took the lead in appointing a com-
Diittee for standardizing the proportions of the machinery and
apparatus needed for ice-making and refrigeration.

8. Some other and similar associations did likewise, and the Ice
Machine Builders' Association took up this matter also, in order
to assist in the establishing of uniform regulations for the benefit
^f all concerned.

9. Guided by the high standing this Society holds throughout
the engineering world, I desire to invite your hearty cooperation
on this important subject, hence this paper.

Ammonia Compression System.

10. The overwhelming majority of refrigerating machines in
actual operation are of the ammonia compression system, which
should therefore be considered first. Afterwards the anmionia
absorption system, the carbonic acid compression system, vacuum
air system and others could be considered along the same lines.

11. With the ammonia compression system pure anhydrous
liquid ammonia is evaporated under a low pressure within a system

294 STANDARD UNIT OP REFRIGERATION.

of pipes. Such a vaporization is only possible if heat is supplied
and such heat is furnished by the substance — air, water, brine,
etc. — surrounding said pipes, whereby consequently the substance
becomes refrigerated and the ammonia takes the form of vapor.
The office of the refrigerating machine is to compress this vapor
and to discharge the compressed gas into the condenser, where it
is condensed — ^liquified — ^for further use in the refrigerating sys-
tem.

Properties of Ammonia.

12. The properties of the ammonia in the state of liquid and
vapor has been calculated by Professor De Volson Wood. They
are well known and universally accepted as correct.

13. We know exactly what the ammonia can do under certain
conditions, and as the temperature, pressure, volume and weight
of same are of fixed relations to each other, we can calculate two
of these properties as soon as two of them are known.

Standard Ton of Refrigeration,

14. The quantity of refrigeration is now expressed in tons of
ice-melting.

As the latent heat of ice is 142 British thermal units, it takes
for melting or making one ton of 2,000 pounds of ice of 32 de-
grees the quantity of 2,000 X 142, or 284,000 British thermal
units.

The correct and better expression which would answer for all
purposes would, therefore, be :

" One ton of refrigeration is the latent heat absorbed or set
free by melting or making one ton of ice of 32 degrees to or from
water of 32 degrees Fahr."

Quamiity of Ammonia Evaporated,

15. To transfer the heat of 284,000 British thermal units by
means of mechanical refrigeration a certain amount of liquid am-
monia must be evaporated, and to arrive at the actual amount of
ammonia needed we must consider the unavoidable losses of heat
during the process of vaporization.

16. First. The liquid ammonia coming from the condenser has

STANDARD UNIT OP REFRIGERATION. 295

a much higher temperature than the one at which it evaporates.
The difference between these temperatures is a loss which must be
deducted from the heat of vaporization due to the boiling point
of the ammonia in the refrigerator.

Second, The vapor leaving the refrigerator becomes heated on
its way to the compressor, and especially by coming in contact with
the walls and parts of the compressor, which are heated to a cer-
tain extent during the compression period. This heating of the
vapor causes expansion, and consequently a reduction of the
weight due to its pressure.

Third, It is impossible to build a compressor which discharges
every particle of the compressed gas. Whatever is left in the
cylinder at the moment the piston begins the suction period will
reexpand and fill up a certain space, thus reducing the volume of
the entering vapor, or, in other words, the displacement of the
compressor.

Fourth. More or less loss is caused by leaks through valves and
piston, radiation of the different parts of the system, imperfec-
tion of the insulation and pipe covering and impurities in the
ammonia itself.

17. The first loss is an exactly known quantity, and is, there-
fore, accounted for separately in all calculations. But all the
other losses can not be measured correctly and independently
from each other, wherefore they are accounted for in form of a
certain per cent, of efficiency of the compressor.

Efficiency of Compressor,

18. Prof. J. E. Denton stated in Transactions A. 8. M. E., voL
xii. : " . . . An important deduction from the measurements
by meter of the quantity of ammonia circulated, compared with
the weight of ammonia accounted for by the displacements of
the compressors, is that the latter falls 25 per cent, short. . . .''
Such a shortage of 25 per cent., or an efficiency of the compressor
of 75 per cent., has not as yet been proven incorrect, and as
actual practice agrees indeed closely with such an efficiency, it
can safely be considered as correct.

19. Having now the quantity of heat equal to one ton of re-
frigeration, and also the loss of heat and the efficiency of the com-
pressor as undisputed facts, it would seem a very simple matter to
find the quantity of ammonia which must be actually evaporated

STANDARD UNIT OF REFRIGERATION.

under certain conditions in order to produce one ton of refrig-
eration.

20. But just the choice of these certain conditions upon which
the standard unit shall be based has so far prevented an agreement
between the parties concerned.

21. For the establishment of a standard unit of refrigeration it
is necessary to agree on the speed of the machine and on the tem-
peratures or pressures which shall form the bases for all calcula-
tions. As soon as such bases are found and accepted, the entire
question is settled.

Speed of Machines.

22. Up to the present time the most serious trouble is the great
difference in the speed of the machines.

Some of the builders use good practice and base their machines
on a piston speed of from 125 feet for the smallest up to 300 feet
for the largest machines.

23. Others* are not so particular, and use a speed of from 150 to
450 feet per minute. Still others don't care at all for any propor-
tions, but simply increase the speed of one and the same machine
if a larger capacity is needed, just to meet competition regardless
of the efficiency or the life of the machine.

24. Such conditions are not known in any other branch of me-
chanical engineering. Standard regulations like those accepted in
steam and electrical engineering should also be possible in re-
frigerating engineering.

25. Recently it has been proposed to adopt 50 revolutions per
minute as the speed for all machines regardless of size. Such an
effort is arbitrary and against the principles of modem engineer-
ing. It would mean for the refrigerating machines as much as
a proposition of 80 revolutions per minute for all steam engines
or 500 revolutions per minute for all dynamos.

26. The speed must necessarily be expressed in feet piston
travel per minute, and the number of feet must be chosen with due
regards to the fact that all compressors have poppet valves instead
of positively mechanically governed valves like the steam engine.

For this reason the speed of a compressor can never be as high
as the speed adopted for the steam engine.

27. I have been in favor of a speed of 250 feet as a standard.
But after a careful comparison of the speed of well proportioned

STANDARD UNIT OP REFRIGERATION. 297

machines in actual practice, I think it will be necessary to divide
the speed in several classes, in order to prevent too great a de-
parture from the present common practice.

28. All machines with a stroke of 8 inches and less are of little
or no importance and could be excluded. The machines above 8-
inch strokes could be divided into four classes, for which the stan-
dard piston speed per minute should not exceed:

180 feet for machines with a stroke up to and iDcluding 12 inches;
240 " ** *• '* *• - •* over 12 and including 24 inches ;

o/w\ I* (I «< <• «< «< *< 24 '* ** 86 *'

aeO ** '* •• *' " '• *' 36 inches.

29. A comparison of this proposition with the present rating of
the manufacturers of two of the leading machines, shows the fol-
lowing table of speed:

298

STANDARD UNIT OF OEFBIOEBATION.

S £

S S

!<:

S S

f! 2

S iS

s s

s ?.

g ^-

'^ s

s 1

" Q

s s

n ^

»r

s s;

S i2

8 2

«j

5 p

"3 £

^ (S

«f

o. «

>»

•3a|

'pjvpnvif

8,J

ling

pd404

dojj

STANDARD UNIT OP REFRIGERATION. 299

TempercUnre vs, Prewwte.

A glance at any ammonia table must convince us that the tem-
perature and not the pressure is the basis for all calculations of
the capacity of a refrigerating machine. This fact is not at all
a matter of choice, but it is dictated by natural conditions.

The change in the temperature of the atmosphere and of the
water supply is the cause of changes of the ammonia pressure, but
a change in the latter can never produce a change in the former.

All the properties of ammonia are based on the absolute tem-
perature from which the different pressures are derived.

And yet, in spite of this fact, most all calculations of the capa-
city of refrigerating machines are based on the suction and con-
densing pressures, which are often expressed in fractions of one-
hundredths of a pound, which of course can not be read on any
gauge.

The consequence of such a faulty method is that the exact
knowledge of the relations between temperature and pressure are
seldom fully understood by those who operate this class of ma-
chines, and that it is possible that for a place having a condensing
water with 85 degrees a machine is furnished of which the capa-
city is based on a temperature of 60 degrees.

The present method of allowing two and even more tons of re-
frigeration for one ton of ice-making capacity is also caused by
the disregard of temperatures, because the boiling point of the
ammonia for ice-making nmst be much lower than the one taken
at present as a basis for refrigeration. I know of a case where a
machine of 150 tons refrigerating capacity had to be worked up
to its full capacity to produce 05 tons of ice.

The ammonia tables in practical use give for each degree from
minus 40 up to plus 100 the absolute and gauge pressures, volumes
and weights of the ammonia, but for only two even numbers of
pounds gauge pressure (76 and 87 pounds) the corresponding tem-
peratures, and all the others must be calculated or assumed.

This evil can be easily corrected by basing all calculations of
the capacity on the

Temperature to be produced, and

Actual temperature of the condensing water.

300 STANDARD UNIT OF REFRIGERATION.

Temperature to be Produced.

From zero required for sharp freezers up to 60 degrees for
certain rooms, the range in temperatures is a very large one.

The standard should be a fair average of the temperatures most
needed, and this is no doubt the temperature required for ice-
making, combinations of ice-making and cold storage, and re-
frigerating plants with brine circulation.

A temperature of 15 degrees Fahr. as a basis would therefore
be most suitable for these purposes, and would give a boiling point
of zero and a suction pressure of about 15 pounds. It could be
applied for about 90 per cent, of all plants, and the balance of 10
per cent, for which temperatures below or above 15 degrees are
required could be based on especially prepared equations or tables.

Actual Temperature of Condensing Water.

Different parts of the country furnish water of a great dif-
ference in temperature. Water as low as 50 is found at some
places in the North; an average of 65 degrees in the East; and
from 75 to 85 degrees in the West and South. But comparatively
few plants can depend on an abundance of water, which is clearly
demonstrated by the rapidly increasing demand of cooling towers.

The temperature of water of the greatest part of this country
is from 72 to 78 degrees, and wherever a cooling tower is used it
will furnish a water of about the same temperature, whether the
initial temperature was 60 or 85 degrees, some extreme conditions
not counted.

A condensing water of 75 degrees as a basis would certainly be
in line with the great majority of actual conditions. A few de-
grees more or less than 75 could not change conditions very mate-
rially, as a difference of 5 degrees does not amount to more than
about 1 per cent, more or less capacity, which is small compared
with a factor of 75 per cent, as the basis for the eflSciency of the
compressor.

With a condensing water of 75 degrees and a modern condens-
ing apparatus, or with a special liquid cooler in connection with
an old style condenser, a liquid ammonia of 80 degrees can be
easily obtained.

8TANDABO UNIT OF REFRIGEBATION. 301

Displacement of the Compressor.

The displacement of the compressor is dependent upon the con-
ditions already mentioned. Taking as a standard basis 284,000
British thermal units as the heat value of one-ton refrigeration;
15 degrees as the temperature to be produced ; zero as the boiling
point of the ammonia to be evaporated; 555,500 British thermal
units as the heat of vaporization of one pound at said boiling point,
and 0.1107 pound as the weight of one cubic foot of vapor at said
boiling point ; 80 degrees as the temperature of the liquid am-
monia, and 75 per cent, as the efficiency of the compressor, the
standard displacement per minute per ton of refrigeration in 24
hours.

284,000

(555.50 - 80) X 0.1107 x 1440 x 0.75)
or 5 cvhicfeet (8,640 cubic inches).

= 4.992

Standard Unit of Refrigeration.

Kecapitulating my propositions, the requirements for a standard
unit of refrigeration would be as follows :

1. 284,000 British thermal units as the latent heat of 2,000
pounds of ice constitute one ton of refrigeration.

2. The efficiency of the anmionia compressor is 75 per cent of
the theoretical capacity.

3. The limit of piston travel in feet per minute shall be

180 feet for strokes up to and including 12 inches ;
240 *• •* " over 12 ** ** 24 "

300 •• '* " •* 24 '* " S6 ••

360 " " " •* 86 inches.

4. The temperature to be produced 15 degrees Fahr., and the
boiling point of the evaporating anmionia is zero.

5. The temperature of the condensing water is taken at 75
degrees Fahr., and the temperature of the liquid ammonia at 80
degrees.

6. The displacement of the compressor must be 5 cubic feet,
or 8,640 cubic inches, per minute for each ton of refrigeration
in 24 hours.

300 STANDARD UNIT OF KEFRIGERATION.

Temperature to he Produced.

From zero required for sharp freezers up to 60 degrees for
certain rooms, the range in temperatures is a very large one.

Actual Temperature of Condensing Water.

STANDABO UNIT OF REFRIGEBATION. 301

Displacement of the Compressor.

The displacement of the compressor is dependent upon the con-
ditions already mentioned. Taking as a standard basis 284,000
British thermal units as the heat value of one-ton refrigeration;
15 degrees as the temperature to be produced; zero as the boiling
point of the ammonia to be evaporated; 555,500 British thermal
units as the heat of vaporization of one pound at said boiling point,
and 0.1107 pound as the weight of one cubic foot of vapor at said
boiling point; 80 degrees as the temperature of the liquid am-
monia, and 75 per cent, as the eflSciency of the compressor, the
standard displacement per minute per ton of refrigeration in 24
hours.

284,000 .

(555.50 - 80) X 0.1107 x 1440 x 0.75) " '

or 5 cubiefeet (8,640 cubic inches).

Standard Unit of Hefrigeration.

Recapitulating my propositions, the requirements for a standard
^it of refrigeration would be as follows :

1. 284,000 British thermal units as the latent heat of 2,000
pounds of ice constitute one ton of refrigeration.

2. The efficiency of the ammonia compressor is 75 per cent, of
the theoretical capacity.

3- The limit of piston travel in feet per minute shall be

ISO feet for strokes up to and including 12 inches ;

240 ' over 12 »* ** 24 **

300 '* " *' •* 24 •* *• 86 ••

360 " •* " •* 86 inches.

*• The temperature to be produced 15 degrees Fahr., and the
boiling point of the evaporating ammonia is zero.

5. The temperature of the condensing water is taken at 75
degrees Fahr., and the temperature of the liquid ammonia at 80
degrees.

6. The displacement of the compressor must be 5 cubic feet,
or 8,640 cubic inches, per minute for each ton of refrigeration
in 24 hours.

302 8TANDABD UNIT OF BEFBIOERATION.

Hence we have

0.75 X 1,440 X (655.50 - 80) x 0,1107 x 5 . .„_ .
284;000 = 1.0008 ton,

representing actually one practical ton of refrigercUion.

Ice-Making Capacity.

It may be interesting to find the relations between one ton of
refrigeration and one ton of ice-making under the very same con-
dition as taken for the standard unit of refrigeration.

With a cooling water of 75 degrees, the water coming from the
re-boiler with 212 degrees must be cooled before it can enter the
ice-making apparatus, and it can therefore be taken again at a
temperature of 80 degrees. By means of evaporating anmionia
it must be cooled to 32 degrees, and after being frozen the ice
must be cooled down to 15 degrees. To simplify matters, we
will take the specific heat of the ice also at 1 instead of 0.5, and
the work to be done would then be the transfer of the sensible
heat of 2,000 x (80 - 15) = 130,000 British thermal units, in
addition to the latent heat of 284,000 British thermal units.

The total of 414,000 British thermal units would then consti-
tute the heat value of one ton of ice-making, and one practical
ton of refrigeration would be equal to *

0.75 X 1,440 X (555.5 - 80) x 0.1107 x 5 _ ^^^^

4i4;;6oo ^-^^^

or 0.68 ton of actual ice-making capacity.

The following Table of Capacities give a comparison of the
builders' rating with the actual capacities according to my propo-
sition for a standard unit.

It proves conclusively that the machines mentioned are not
based on uniform rules. While the capacity of most of the single-
acting machines comes very near within the results from applying
my proposition for a standard unit, the capacities of most of the
double-acting machines show very great differences.

The true proportions between refrigerating and ice-making
capacity hold good for the single-acting machines, but not for the
double-acting machines, which for several good reasons fall short
in actual practice, which fact is plainly demonstrated by the
builders' rating for ice-making capacity.

STANDABD UNIX OF REFRIQEBATION.

303

3 § 1

a s 9 §

$^ n

S 8 S 1

8 f

1 i

s s s 1

« s

.•3

1 g

? S 8 g

s s

- S K 1

9 d

o 5

i 1

= S S f

ss »

©»

s 1

s^ 5 8 1

5:; fi

§ s

• s s 1

" 00 '

"»

^ S s a

"^ <o

§ § 1

i g 1

a 9 9 I

» S 3 1

S i

2 §

'--

_*r —

* a s §

8 S

S 1

S S 3 g

5 R

; i

5 S 8 1

«y

1
'1

2 s e 1

-4P

i i

- i2 S 1

S 5

2:»

I-

• a a 8

* 3

i !

~ s g 1

«o

1 8 i

-i g

^ a

a ^

a —

•4
Q

1
1

1 I

til,

5 « 1 1

a IP

[li

ill

^ &

1 -Su|isu a,ja

l>liufl

1 -pjHpUH

48 p^^odojj 1

304 STANDARD UNIT OF REFRIGERATION.

Conclusion.

It has been my endeavor to show that the pressures as a basis
for calculation are in place and necessary for computing the power
required, but that for computing the capacities of refrigerating
machines the temperatures are the only correct basis.

As this subject is of great interest to a great number of indus-
tries, I beg leave to suggest that a committee may be appointed
consisting of members of this Society actually engaged in re-
frigerating engineering, and that such committee may have the
power to cooperate with all the committees already appointed
by the associations of builders and owners of ice and refrigerating
machinery for the sole purpose to establish a Standard Unit of
Eefrigeration.

DISCUSSION.

Mr. Gardner T. Voorhe^s, — In reply to Mr. Bertsch's paper
on the Standard Unit of Refrigeration I ^nsh first to briefly re-
view the efforts that have been made in this line.

In my work as a Refrigerating Engineer, for a long time I
have been greatly impressed by the great differences of opinion
regarding the capacity of refrigerating machines.

In a series of articles that I publislied in Ice and Refrigeration
of May, June, July, August, September and October, 1902, en-
titled '^ Analyzing the Compressor," I endeavored among other
things to indicate a basis for accurately determining the capacity
of such refrigerating machines.

My correspondence with the prominent builders of such
machines at that time so impressed me with the great diversity
of ratings for capacity that I detennined to try and have a stan-
dard unit ton for refrigerating machines adopted by those most
interested.

It was my intention to have personally presented this subject
before the Society at its last winter meeting, but my duties in
connection with the St. Louis World's Fair Refrigeration Ex-
hibit at that time prevented my being at the meeting as these
same duties have at present prevented my preparing a more
elaborate paper.

I brought the question to the attention of the Society in letters
of December 3 and 22, 1002, and asked that a committee be ap-

STANDARD UNIT OF REFRIGEBATION.

305

pointed to consider the question. In my letter of December 2 2d,
1 suggested the following as a basis on which to attempt to frame
rules for the standard unit ton for all classes of refrigerating
machines.

First: As referred to compression machines, the use of the
factor (cubic displacement of compressor per revolution per ton
of refrigeration).

Second: For absorption machines, cubic displacement of the
liquor pump; per stroke, per ton of refrigeration, with accom-
panying difFerence in percentages of ammonia contained in the
strong and weak liquors.

I personally brought up the question at the last February
meeting of the Southern Ice Exchange at Atlanta, Ga., with the
result that I had the honor to be appointed chairman of a com,-
mittee of five from that association to consider the subject and
report.

Through the active efforts of Mr. J. F. Kickerson, the pub-
lisher of Ice and Refrigeration, and Mr. J. 0. At wood, manager
of the National Ammonia Co., similar committees were appointed
at the last March meetings' of the Southwestern Ide ^lanufac-
turers' Association at Dallas, Texas; The Indiana Ice Manufac-
turers' Association at Indianapolis, Ind., and the Western Ice
Manufacturers' Association at Kansas City, Kansas.

Since the above committees were appointed the Ice Machine
Builders of this country have organized an association and have
devoted some little time to discussion and tests that may help to
^tablish such a unit.

Numerous articles bearing on the subject have been published
from time to time in the columns of Ice and Refrigeration by
Prof. J. E. Siebel, Eugene T. Skinkle, J. C. Bertsch, A. E.
Siebert and myself. For the benefit of those who may be inter-
ested to review those articles I give the following references to
the pages of Ice and Refrigeration:

Month.
March . .

May ..

Jane . .
July ..

Page.
88d-«8e.

194

16-18

1908.
Anthor.
G. T. Voorhees

loe Machine Baildere. .

E. T. Skinkle

Subject.

Introduction of Rubject at South-
ern Ice Exchange Meeting.

Proposed rule for Standard Unit
Ton.

Criticism of Builders Unit by edi-
torial. .

Capacity of Compressor, with
^bles.

306

STANDABO UNIT OF R£FBia£RATION.

1908.-<ContInued.)

Month.

Page.

Author.

Subject

Aug. . .

. 47-60 . .

. G. T. Voorhees

. . General review of sabject. Rales
suggested and carves for capa-
city. Criticism of Skinkle's
tables.

Aug. . .

. 50-51 ..

. A. Siebert

. Criticism of Skinkle's tables and
proposed new formula.

Sept. ..

.110 ..

. G. T. Voorheea

. Diagramatic representation of am-
monia in a compression cycle.
Criticism of Siebert's formula.

. 101-102. .

. E. T. Skinkle

. Reply to criticisms of Voorbees
and Siebert.

«(

. 10^108. .

. Prof. J. E. Siebel

. . Criticism of Siebert's formula.

«(

.103

. A. Siebert

. Criticism of Voorhees' articles.

Oct. ..

.129-181..

. G. T. Voorheefi

. Reply to Skinkle's criticism with
curves showing Voorhees*, Skin-
kle's and Siebert's results and
correction to apply to page 110
of September article.

. 125-128..

. J. C. Beptach

. General discussion of subject, with
tables, together with discussion
of Voorhees', Skinkle's and Sie-
bert's articles.

«»

. 181-182..

. A. Siebert

. .* Reply to criticism of Siebel, Voor-
hees and Skinkle.

.140

. Ice Machine Buildera. .

. . Brief notice of tests for unit ton.

Nov. ..

. 195-196..

. G. T. Voorhees

. Criticism of Bertsch's article.

A brief summary of the points discussed are as follows: —
Superheating effect, formulee for capacity, rules for capacity,
general curves and tables.

The rules proposed by me in these articles were as follows : —

For Ammonia Compression Machines.

First: A standard unit ton of refrigeration is 284,000 British
thermal units.

Second: A refrigerating machine must operate continuously
for 24 hours to do refrigeration equal to its rated capacity.

Third: A refrigerating machine shall operate at fifty revo-
lutions per minute when rated at its standard capacity.

Fourth: The standard displacement of an ammonia compressor
of one ton capacity is five cubic feet per minute.

Fifth: The approximate displacement per minute of a com-
pressor per ton of refrigeration at a back pressure other than 15

STANDARD UNIT OF REFRIGERATION. 307

pounds per square inch gauge and a temperature of water to the
condenser other than 75 degrees Fahr. is 5 cubic feet plus or
minus .18 cubic feet for each pound that the back pressure is
below or above 15 pounds and plus or minus .001 cubic feet for
each degree that the temperature of water to the condenser is
above or below 75 degrees Fahr.

Referring now to Mr. Bertsch's valuable paper, I ofFer the
following criticisms:

The statement, article 14, " One ton of refrigeration is the
latent heat absorbed or set free by melting or making one ton
of ice of 32 degrees to or from water of 32 degrees Fahr." would
seem to be more concise if stated as follows t

One ton of refrigeration is the measure of the heat taken up
or given out by melting or making 2,000 pounds of ice of 32
degrees Fahr. to or from water of 32 degrees Fahr. This removes
the objection of synonymously using (ton of refrigeration) and
(latent heat).

" Given out " seems to be more appropriate than " set free "
and *^ 2,000 pounds " obviates the question as to whether a short
or long ton is meant.

For the latter part of articlq 15, the phrase: "and to arrive
at the actual amount of ammonia needed, we must consider the
unavoidable losses of heat during the process of vaporization,"
seems more correct if it reads : " and to arrive at the actual
amount of ammonia needed we must consider the unavoidable
gain or loss of heat to or from the system during the process of
refrigeration.^^

In the second clause of article 16, only the eflfect of a dry
compressor is noted; whereas a wet compressor would have addi
tional weight of ammonia introduced through the expansion valve
to counteract the superheating effect during compression.

Article 19, after speaking of the 25 per cent, loss in capacity
of compressor in article 18 the statement that the efficiency of
the compressor is an undisputed fact seems erroneous in two
respects.

First: That the use of the word efficiency is not correct. It
might be worded as follows, " and also the loss of heat and the
effective capacity of the compressor as undisputed facts."

Second: That the effective capacity of the compressor is a very
much disputed fact. I admit that the work and tests of Professor
Denton are the best data we seem to have at the present time,

308 STANDARD UNIT OF REFRIGERATION.

and that probably the 25 per cent, loss of capacity is correct
for certain back and condenser pressures. But I firmly believe
that future Experiments will show that the effective capacity of
the compressor is a variable quantity depending on the back and
condenser pressures, the nature of the compressor (wet or dry)
and the area and nature of the surfaces that the refrigerant comes
in contact with between the time that it leaves the cooler until
it is trapped in the compressor.

1 believe that the loss of capacity will be more than 25 per
cent, for back pressures less than 15 pounds gauge and condens-
ing water hotter than 75 degrees Fahr. and less than 25 per cent,
for back pressures more than 15 pounds gauge and condensing
water colder than 75 degrees Fahr.

My theory closely follows the cylinder condensation phenomena
of the steam engine and seems to be borne out by general experi-
mental data.

Articles 25, 26, 27 criticise my suggestion for the adoption of
a standard number of revolutions for the compressor and suggest
a sort of sliding scale of piston speeds.

I believe the general niles to be adopted should be as simple
as possible, and I suggest that if a standard number of revolutions
can be agreed upon that it will much simplify the final rules.
For instance, the variation in piston speed suggested by Mr,
Bertsch is 100 per cent., while from an inspection of the tables
published in his paper on page 7 it is evident that there is not such
a large variation in the number of revolutions.

It is not necessary to maintain the arbitrary proportion between
the diameter of piston and length of stroke as in these tables. It
is evident that where the length of the stroke is twice the diam-
eter of the cylinder (as is the case \vith many up-to-date machines)
that a fixed number of revolutions as 50 would give just the same
displacement as a greater number of revolutions and a smaller
stroke for the same diameter of piston.

Why not, as new patterns are gotten out, make the smaller
machines with extra long strokes and the larger machines with
shorter strokes so that a standard number of revolutions may be
adopted ?

My argument as advanced in favor or revolutions in place of
piston speed is as follows, from page 49, August, 1003, Ice and
Refrigeration,

" Why should not a compressor run at a high piston speed like

STANDARD UNIT OF REFRIQEBATTON. 309

a steam engine ? The reason is that the valves of a steam engine
are positively mechanically governed and must act positively at
each revolution of the engine. A compressor usually has poppet
valves which are not positively mechanically governed, but depend
for their action upon a difference of pressure of a gas on the op-
posite faces of the valves. Such valves have an appreciable inertia
and can only reciprocate a limited number of times per minute
without requiring an excessive difference of pressure on the two
faces of the valve. Such a difference of pressure is out of the
question where low back pressure is to be maintained in the
cylinder during the suction stroke.

" A too rapid reciprocation of the piston will either cause the
suction valve to so act that the pressure in the cylinder is much
less than that in the suction pipe, or else the rapid motion of the
piston will not give the valve time enough to overcome its inertia
and change the direction of its motion, with the result that the
valve will never be fully open or fully shut. The result of such
action evidently greatly cuts down the capacity of the compressor
so that it would not do as good work at quite high speeds as it
would at slower speeds.

" We all know from the steam engine that the limit of possible
piston speed in the compressor is never reached so far as operat-
ing the piston and stuffing box are concerned. It seems to me
that the piston speed should not enter into the question. The
vital question seems to me to be : How many times per minute
can the valves be made to reciprocate to advantage ? If I owned
and operated a compressor I would, from my experience, set the
limit at fifty revolutions per minute.

" Whatever the number of revolutions finally settled upon as
being advisable it is e\'ident that it will apply equally well to a
large or a small machine so long as the machine has equal valve
area for equal weights of gas pumped at equal pressures. A con-
sideration of this may show a marked difference not only in the
different makes of machines but in different sizes of the same
makes of machines; and may lead in the future to different pro-
portions between diameter and stroke so as to give larger valve
areas."

In addition to the above I will repeat a remark that was made
to me by a very successful refrigerating machine builder, " I
find it cheaper to install ample compressor displacement rather
than to install extra expansion coils, condensers, ice cans, etc."

310 STANDARD UNIT OF REFRIGERATION.

There is a whole volume of common sense in the above state-
ment and a careful study of its results would, I think, lead to a
conclusion that a slow running machine is a good investment both
for the builder and the purchaser.

Mr. Bertsch implies that we should adopt temperatures rather
than pressures because temperatures are given in even numbers
in the ammonia tables while pressures are usually expressed as
fractions. This seems to be no argument, for, if desired, ammonia
tables can easily be calculated with pressures in whole numbers
and temperatures generally in fractions.

Mr. Bertsch asssumes that there should be a difference of 15
degrees Fahr. between the boiling point of the refrigerant and
the temperature of the substance to be cooled. A difference as
great as this is not found in modern practice where shell or double
tube brine coolers are used, for with such coolers a minimum
difference of temperatures of from 8 degrees Fahr. to 5 degrees
Fahr. is usual while 10 degrees Fahr. difference is a maximum.

Referring to the rules, the first one is not \\t11 worded, while for
rule two, I suggest that the word efficiency should be changed to
effective capacity. Rule three does not seem to be simple enough.
Rule four is too arbitrary and not based on standard conditions to
give the best economy: I advise back pressure as a basis to fig-
ure on.

In obtaining the factor 0.68 as the multiplier to be applied to
the tons refrigerating capacity to obtain the tons ice making
capacity, no account has been taken of the gain of heat by radia-
tion to the ice making tanks or of the loss of ice in thawing from
the cans or plates.

I wish to call attention to the rule adopted by the Ice Machine
Builders' Association as published on page 194 of the May, 1903,
number of Ice and Refrigeration which reads: "It is acknowl-
edged by the Ice Machine Builders' Association of the United
States, here assembled: that, in the operation of refrigerating
machinery it requires the evaporation of 27.7 pounds of anhydrous
ammonia per hour at a pressure of 15.67 pounds above atmos-
phere, condensing pressure to be taken at 185 pounds above atmos-
phere, to produce an effect equal to the melting of one ton of ice
per twenty-four hours, and that the capacity ratings of refrigerat-
ing machines should be figured on this basis."

This rule seems to be framed much as that for the standard
boiler horse-power and it is good as far as it goes, and its form

STANDARD UNIT OF REFRIGERATION. 311

would be ideal if it were as convenient to measure ammonia as
it is to measure water in a boiler test, or if the question of the
speed of the machine were considered. Metering anhydrous
anamonia is very uncertain and requires specially constructed
meters. I have used the same meter on some tests that Professor
JJenton used on his tests and I found that unless great care w^as
taken in properly installing the meter, much gas would be gener-
ated in the liquid pipe and in the meter which would make the
meter buzz around like a top with results that were anything but
reliable.

Furthermore, it would seem to me to be very questionable as
to whether anhydrous ammonia could be successfully measured
without using special apparatus and the services of an expert whose
services should not be required in so simple a matter.

The Ice Machine Builders have shown great interest in this sub-
ject and have started to make tests on one make of machine with
the evident purpose of gathering data upon which to base the
standard unit ton.

I think it unwise to attempt to derive data that is supposed'
to stand for all time from a single make of machine, especially
if that machine is erected for the sole purpose of making
tests. I think data should be obtained from all recognized
standard makes of machines and that tests should be made at the
places where the machines are in actual commercial operation.

Each machine builder can single out one or more of his best
machines and have a careful expert test made by his own experts
under the supervision of some reliable and impartial expert.

A general review of these results should give the necessary
data for the final action, and then every builder will have had
a chance to have his say and to submit the indisputable value
of his machine as shown from verified tests.

At first thought I feel sure that many of the machine builders
will object to this method as they seem to have a mistaken idea
that their competitors would get the better of them and that the
public in general would learn too many things that they should
not know. However, I believe that a careful consideration will
convince most of the builders that such a course would have many
advantages in bringing to light the best points of all machines
so that all could profit thereby in discarding old and undesirable
types and gradually adopting the better and more modern methods.

Let the public know more about ice machines and remove the

312 STANDARD UNIT OF REFRIGERATION.

general impression that there is something very unusual in their
construction and operation and the public will very shortly want
more machines. »

It is my intention to -offer every facility for such tests to be
made upon the several different types of machines to be exhibited
at the St. Louis World's Fair next year, and I will be glad if all
the recognized experts in this or any country will get together and
suggest methods for and help conduct such tests.

In closing I wish to apologize for taking so much time and
touching on points not covered in Mr. Bertsch's paper.

My only excuse is the great interest I take in this subject.

Mr. Thos. Shipley, — It is because I do not believe that some of
the statements made in the paper under discussion should be al-
lowed to go on record as facts, that I felt called upon to make
these remarks.

It is the fact that no unit of refrigeration has been adopted by
the engineers engaged in the manufacture or operation of ice
*'and refrigerating machinery, and it is also the fact that every
engineer so engaged has felt the need of the adoption of some
unit upon which the commercial rating of ice and refrigerating
machinery could be based.

The unit, if it may be called such, must be based upon an
adopted back or suction pressure, as upon this back or suction
pressure depends the conditions under which the ammonia is
evaporated and the work done.

I speak of ammonia as the refrigerant; it is the one that is
almost universally used, and also because to attempt to adopt a
separate unit for every refrigerant would be an endless job and*
not advisable at tliis state of the art.

When once a standard back pressure has been adopted, then the
unit has been arrived at; the back pressure has been the bone of
contention and must be agreed upon before anything else can be
done.

Mr. Bertsch makes the statement that the efficiency of a com-
pressor is 75 per cent, and bases this statement on deductions
made by Professor Denton years ago, and he further states that
this efficiency has not been proven incorrect.

In this I must correct Mr. Bertsch; 75 per cent efficiency is
incorrect, and has been proven so conclusively, especially in the
recent tests made at York by a committee selected from the

STANDARD UNIT OP REFRIGERATION. 313

manufacturers of ice and refrigerating machinery. The three
members who directed the test were Geo. Richmond, representing
the De La Vergne Refrigerating Machine Co.; N. H. Hiller,
representing the Carbondale Machine Co., and myself, represent-
ing the York Manufacturing Co.

The fourth member, Mr. Theo. Vilter, of the Vilter Manufac-
turing Co., was unable to be present at the tests.

The three active members of this committee are all members
of this Society.

The compressor we used developed an eflBciency of 83 per cent,
under 15.67 pounds back pressure.

The plant used in making the tests was put up for that pur-
pose, and every means possible was taken to guard against errors.

A mercury column was used to determine the back pressure,
as we found gauges were not reliable for the purpose. We also
weighed the liquid ammonia,' as it was impossible for us to get
a meter which would handle liquid ammonia accurately.

The plant was of sufficient size to warrant accuracy, and the tests
were run for 6 days consecutively, hence the compressor efficiency
obtained can be relied upon.

The compressor was operated at an average of about 70 rev-
olutions per minute.

As to the proper speed at which a compressor should be oper-
ated, I will say that from tests which are now being carried on
on the same test plant under my direction, it has been shown
that the efficiency of a compressor increases as the revolutions
increase.

The speed tests were made for each 10 revolutions from 40 to
100, the compressor being 18-inch stroke.

There is no question whatever that the efficiency of the dif-
ferent types of machines varies, hence the adoption of a standard
compressor displacement per ton of refrigerating would not be
possible any more than it is possible to adopt a standard amount
of steam per horse-power for all types of engines.

Prof, S, A, Reeve, — The writer wishes in the first place to
commend the enterprise of Mr. Bertsch in bringing this subject
before the Society. It is one which has long needed attention.

In the second place, he would suggest the advisability of sub-
mitting this matter to the consideration of a committee, for the
formulation of the Society's views as to the adoption of some such
standard unit. To that end he would suggest that a committee

314 STANDARD UNIT OF REFRIGERATION.

be appointed by the President, to report to the Society at its next
annual meeting.

Thirdly, he would suggest to such committee, if one be ap-
pointed, the need for a standard method of expressing the
efficiency of a refrigerating or ice-making machine as well as for
a standard unit of refrigeration; for we have no recognized methpd
now. To start the discussion, he would suggest, tentatively, the
following method:

Let Q be the standard unit of refrigeration, supposedly the
284,000 British thermal units suggested by Mr. Bertsch.

Let the lower limit of temperature be the zero-point (F) sug-
gested by Mr. Bertsch or, in general, any absolute temperature
Ti which might be adopted in its place, or which might be left
undecided for variation to suit each case.

Let Tj be the upper limit of temperature, that of the condensing
water, either the 536 degrees absolute (75 Fahr.) suggested by
Mr. Bertsch, or the particular temperature applying in any given
case.

Then the least amount of energy, measured in British thermal
units which could possibly accomplish the unit of refrigeration
would be, in general,

^j;^^ 284,000 =(?,,

or, supposing the adoption of Mr. Bertsch's standard tempera-
tures, Qw = 46,204 British thermal units, or very close to J
horse-power per ton capacity.

In any given case let the aciual energy absorbed in producing
the standard unit of refrigeration be qw, which will always be
some quantity larger than Qw. Let the efficiency of any such
a case be expressed as

This method, it is true, takes no cognizance of the fact that
not even a perfect ammonia-machine could ever hope to reach
100 per cent, efficiency. For this reason the writer would per-
sonally prefer to see the expression for the efficiency of the
machine referred to that of a perfect ammonia-machine, which

STANDARD UNIT OF REFRIGERATION. 315

would always have an efficiency less than that expressed by (Tj —
T,) -T- T|. Biit because he finds that the average engineer looks
askance at and will not adopt a quantity so complex in its com-
putation as the efficiency of the perfect ammonia-compression-
uiachine, he suggests the use of the simpler expression just given.

Mr. 8. H. Bunnell. — The adoption of an arbitrary capacity
unit is much like certifying officially that all machines have the
same efficiency. The condition imposed on the makers of re-
frigerating machinery is that required of all constructors of
machines — that they shall be able to perform what they promise.
To do this satisfactorily generally means that the speed of ma-
chines must not be too high, the consumption of power too great,
or the liability of accident too imminent. If the manufacturer
can construct an apparatus of superior efficiency with out sacri-
ficing essentials such as these he should have the benefit of his
efforts.

The high acting slow speed American compressor with pistons
running to practical contact with heads, must necessarily displace
more ammonia per stroke than the double-acting machine of
same cylinder volume, but the latter may make up for clearance
and valve loss by saving in friction losses on account of avoiding
the idle stroke. An arbitrary unit based on cylinder dimensions
alone favor small and inefficient machines by giving them a
rating higher than they deserve.

The builder designs his com^iressor with regard to the uni-
versal requirement of maximum efficiency at minimum expense,
and selects speed, dimensions, style, materials and workmanship
in accordance with his judgment. He must not build or run
his refrigerating compressor on other than satisfactory com-
pressor lines. But the ratio of useful eflFect to theoretical
capacity depends on his ability as designer and constructor, not
on mere cylinder volume. The purchaser of a refractory plant
wants useful effect, and generally requires a practical demonstra-
tion, with due regard to the power consumption and other ex-
penses. If a constructor wants to furnish cylinders with a larger
clearance in order to obtain certain advantages in design of
valves, the provision of an arbitrary capacity unit gives him the
right to claim more for his machine than it can do in com-
parison with some single-acting machine without appreciable
clearance. Neither 75 per cent, or any other figure can cover
all cases correctly.

316 STANDARD UNIT OF REFRIGERATION.

We have already two useful units of refrigeration; one, that
defined by Mr. Bertsch and universally used in America, and the
other, the arbitrary ice-making unit, the ton of ice-making capa-
city, equal to two tons of refrigerating or simple cooling capacity.
The ice-making unit is based on practical results; for the ice
made is weighed after melting from cans or cutting from plates,
and some of the refrigeration is dissipated in leakage through
tank walls and other losses besides direct water cooling. The
proposed ice-making standard could only be approached in large
plants.

A practical and absolute standard has been adopted by the
Ice Machine Builders' Association of America. It may be com-
pared with the definition of paragraph 6 of the recapitulation.
With a condensing pressure of 185 pounds gauge, the displace-
ment of the compressor must be such that it will pass to the con-
denser .462 pounds of saturated ammonia vapor evaporated at
15.67 pounds gauge pressure per ton per minute. I do not see
the force of the point made against pressure readings and in
favor of temperatures. Pressure gauges must be used for
safety's sake, and may well be supplemented by thermometers;
but since pressures and corresponding temperatures of saturated
ammonia are shown side by side on the gauge dials their relation
is always apparent. Fractions of pounds or degrees are of no
importance in practical design and operation.

Mr, Wm. T. Magruder, — Will the author please inform us
why he prefers to use (paragraph 14) " 142 British thermal
units " as the latent heat of ice, when the more generally accepted
figure is 144 British thermal unitB.

Professor Jacobus. — This subject would be made much clearer
if the paper were divided in two parts; first, that relating to
standard units of refrigeration ; and second, that recommending
some standard way of rating the capacity of refrigerating ma-
chines. If this were done the first part would be very simple, and
the standards now in common use, — the ice-melting capacity and
the ice-making capacity, — ^would probably be all that would need
to be considered. When we come to the second part, however,
and attempt to specify what shall be the standard rating of a *
refrigerating machine, we have a complicated problem to deal
with. Some classes of machines will do better work relatively
to others at high temperatures of refrigeration and others at low ;
some are handicapped by warm condensing water to a greater

STANDARD UNIT OF REFRIGERATION. 317

extent than others, and it will be very hard to say what set of
conditions will be fair for all machines or whether any set of
conditions can be selected which will be fair for all.

Finally, even if a standard set of conditions are selected on
which to base the rating of machines, it will be necessary to know
the actual conditions under which a machine is to be operated
before we can estimate the work that the machine should be able
to accomplish.

Mr, J. C. Bertsch* — The discussions have furnished some val-
uable statements, and I am much pleased that this subject has
met with such great interest.

Mr. G. T. Voorhees places liimself with many of his criticisms
in the attitude of a corrector by using hair-splitting methods whicli
.-hould have been omitted. At the present stage of the question
it is immaterial whether we say "Absorbed" or "Taken up:"
" set free " or " given out;" " efficiency " or " effective capacity,"
etc.

While I endeavor to fit the Standard Unit as much as possible
to the present conditions, Mr. Voorhees desires to change most of
the existing conditions to make them fit his proposed rules. If
he would study manufacturing methods and machine shop prac-
tice, he would surely realize the impossibility to change the exist-
ing proportions of the machines so radically; to make new am-
monia tables, and so on.

Mr. Voorhees proposed first to consider pressures only in cal-
culating the refrigerating capacity. My article on that subject
in Ice and Refrigeratioriy of October, 1903, induced him to
" amend " his rules, and he adopted from my proposition the tem-
perature of condensing water for his condensing pressure. Now,
if temperatures are the proper thing for one side, why shall they
not be proper for the other side of a system ?

The fact that difFerences of from 3 to 5 degrees between the
refrigerant and, the substance to be refrigerated are sufficient
when modem apparatus are used does not justify the making
of such a difference the standard. At least 90 per cent, of the
existing and future plants are and will be of the " old-time brine
tank style," for which the Standard Unit must also fit. Besides,
if such small differences should be adopted, then a compressor
displacement of 5 cubic feet per ton is much too large, or an
efficiency of 75 per cent, is much too high.

* Author's Closure, under the Rolee.

318 STANDARD UNIT OF REFRIGERATION.

A boiling point of zero would produce temperatures of from
3 to 5 degrees, which are far below the practical average. But
taking 15 degrees as the actual average temperature to be pro-
duced, a boiling point of 10 degrees would have to be assumed^
and the displacement of the efficiency would be as follows :

1. Duplacement

284,000

(549.35-80) X 0.1383 x 1440 x 0.75 "" ^'^^^ ^^^^"^ ^^^

2. Efficiency.

284,000 - ft Ai QQ

(549.35-80) X 0.1383 x 1440 x 5 "^ 0.6133.

Such a displacement of efficiency does not at all agree with
actual practice, and is at variance with all, even with Mr. Voor-
hees's own proposition of a Standard Unit.

Mr. Thos. Shipley has made a full confession in favor of my
proposition, while he intended to oppose the same. His principal
remarks were:

The efficiency of a compressor is 83 per cent, instead of 75
per cent.

The back pressure has been the bone of contention

Gauges are not reliable for measuring back pressures-

All of this Mr. Shipley found out by conducting a test with
a machine put up for that special purpose, and where " every
means possible was taken to guard against errors." This test
was made by three experts and for a period of six days.

I beg to ask now every fair-minded engiiieer to answer the fol-
lowing question : " If a machine, built and put up with all possible
care, tested by three eminent experts, with all the assistance to
their command regardless of cost, does not give a higher efficiency
than 83 per cent., what can be expected of a machine built and
put up in competition with ten or more other makes, operated
by an ordinary engineer, and oftentimes by some one who knows
just how to hold an oil can, run for twenty-four hours per day
during a period of about six months, without any appliance to
properly control the proper charge of ammonia, running short on
condensing water and working against many other unfavorable
conditions, — what efficiency can be expected of such a machine
representing the average outfit? "

At least 50 per cent, of all the machines in commercial opera-
tion will not give an efficiency of 75 per cent., even if they were
all of the single acting type. But when it comes to the double

STANDARD UNIT OF REFEIGEEATION. 319

acting machines, with the entire system full of oil, eflSciencies
of from 50 to 60 per cent, will be a fair average.

If actual tests shall be made for finding a Standard Unit, then
let us test machines in commercial operation and under natural
and normal conditions^ such as the public can afford to main-
tain. But testing especially built machines at the manufacturer's
place by experts and under the most favorable conditions can
be called anything but Standard.

Mr. Shipley is in favor of back pressure, but he admits that
gauges are not reliable for measuring same. He thinks that upon
the back pressure depends the work done.

The logical order of question and answer is simply this : " What ■
work shall be done ? " " Producing a temperature of 15 degrees."
Well, then we must carry a back pressure of about 15 pounds.

We see on this simple example that the work to be done is
the first consideration, and all other conditions, even the back
pressure, must be arranged accordingly. The public knows ex-
actly what temperatures are needed, but nobody cares for the
pressures carried. Hardly 5 per cent, of all the ammonia gauges
are correct for any length of time, which is confirmed by Mr.
Shipley's own statement, but the cheapest thermometer does not
differ more than one degree. To repair a gauge takes much time
and is expensive, because it must be sent to the factory, but a
thermometer can be bought in every little town. All the work
of a refrigerating machine is controlled by the temperature, and
the owner or manager of an ice plant, cold storage house or brew-
ery will certainly not order his engineer to keep certain back
pressures, but they simply ask for certain temperatures. These
are some of the reasons why I propose to make the temperatures
the foundation for a Standard Unit, and not the pressures of the
ammonia.

Let the public know what a machine can do and how the work
can be controlled, instead of covering defects with rules and
terms which can not be understood by many.

The remarks of Mr. S. H. Bunnell are ansr^vered by the fore-
going with the exgeption of the item with reference to gauges
having the corresponding temperatures marked on the dial. We
all know, and Mr. Shipley is the authority for the fact, that am-
monia gauges are not reliable, as most of them are from 1 to 15
pounds out. But if the readings of the pressures are incorrect
and unreliable, then the corresponding temperatures are also

320 STANDARD UNIT OF BEFRIGERATION.

wrong, and without using a thermometer to ascertain the true
state of affairs, nobody knows the conditions under which the work
is done.

The question of Professor Wm. T. Magruder I will answer
thus : " Since the quantity of 284,000 British thermal units is
generally adopted as the cooling effect of one ton, or 2,000 pounds,
of ice, the latent heat of ice is accepted as being 142 British ther-
mal units."

BOILER PLiVTE, RIVET STEEL, STEEL CASTINGS AND FORGINGS. 321

No. 1096.*

SPECIFICATIONS FOR BOILER PLATE, RIVET STEEL,
STEEL CASTINGS AND STEEL FOROINGS.

Professor Spangler. — You may remember the conditions under
which this Committee was appointed, but to make it entirely
clear I would like to go into the history of it just a little. There
is a society known as the American Society for Testing Mate-
rials, which was the outgrowth of the International organization
of which we have heard a great deal at meetings of this Society.
Committee No. 1 of that Society prepared a series of specifica-
tions, and Mr. Webster, at the request of Mr. Hutton, presented
these specifications at a meeting of this Society, and asked that
a committee be appointed on this particular subj/Bct. A, com-
mittee of five was appointed, consisting of Mr. Cramp, Mr.
Kent, Mr. Morison, Mr. Waitt and myself. In the usual way
copies of these specifications were sent to various members of
the Society, with the usual result — that is, in a few cases, after
writing two or three letters, replies were received. The Com-
mittee decided to submit, at this time, a report to the Society,
subject to revision, asking that the report be sent to all members
of the Society, that something like a full written discussion from
members who are interested in the subject might be had, and
that a revised report be formulated at some future time.

It seems to me to be the proper procedure that, after this
Society has finished whatever work it' may decide to do, the
report, together with the report of all the committees of other
societies that may be working on the subject, should go back to
Committee No. 1 — that is, any report that we might make
should be rather an advisory report than an attempt at a finality.
This Committee No. 1 is the Committee which will finally,
I believe, formulate specifications under which work of this sort
is to be done.

With this as an introduction, your Committee would resp^t-
fully report as follows:

♦ Presented at the New York meeting (Decern l>er, 1908) of the American Society
of Mechanical Engineers, and forming part of Vohime XXV. of the TransacUona.

322 BOILER PLATE, RIVET STEEL, STEEL CASTINGS AND FORGINGS.

SPECIFICATIONS FOR BOILER PLATE, RIVET STEEL, STEEL
CASTINGS AND STEEL FORGINGS.

This report is sent out subject to revision, and the Committee asks that
written discussion be sent to its chairman that the results may be incorporated
in the final report to be presented at the New York meeting of the Society.

The Committee to which was referred the question of specifications for boiler
plate, rivet steel, steel castings and steel forgings, reports that it has used the
specifications prepared by the American Branch of Committee No. 1 of the
International Association for Testing Materials, of which Mr. Wm. R. Webster
is Chairman, as the basis of its work, and the changes hereafter noted are recom-
mended in these specifications.

1. That the maximum sulphur in flange or boiler steel be reduced from
.05 to .04.

2. That the tensile strength be specified as stated in the table with an allow-
able variation of 5,000 pounds. That fire box steel be specified at 55,000 pounds
instead of 57,000 p>ounds per square inch. That the determination of the yield
point for ordinary grades be omitted.

3. The tensile strength of castings has been modified, the specified
value desired being stated, and the variation, 5,000 pounds, being allowed.
The values, as re^mmended by Committee No. 1 , and by this Committee, are
as follows: —

Com. No. I's. RecommeDded by

Minimam. Committee.

Soft 60,000 60,000 ± 5,000

Medium 70,000 70,000 ± 5,000

Hard 85,000 80,000 ± 5,000

4. The elongation in 8-in. is stated instead of in 2-in. and an increase in elonga-
tion of 25% is called for on the 2-in. specimen.

For a 2-in. specimen from castings the corresponding elongations are:

Recommended by
Com. No. 1. this Committee

Soft 22% 20%

Medium 18% 17.5%

Hard 15% 15%

5. That the 8-in. specimen be made the standard specimen and the 2-in. to
be used only when it is inconvenient to use the 8-in.

6. That nickel steel forgings and oil tempered forgings be not included in
this specification, because the present state of the art does not warrant general
specifications being drawn for these materials.

7. That for soft or low carbon steel forgings the chemical requirements be
not over .06 phosphorous, and .05 sulphur, instead of .10 phosphorous and
.10 carbon.

8. That for "carbon steel not annealed" the t^rm *' medium steel" be used,
and that the sulphur be reduced from .06 to .05 per cent.

9. That, wherever it is desirable that the elastic limit be determined, an
extensometer be used, and that the elastic limit be taken as ''that point at

BOILEB PLATE, RIVET STEEL, STEEL CASTINGS AND POROINGS. 323

which the elongaUon in 8-in. per 1,000 pounds of added stress per square inch
first exceeds four ten-thousandths of an inch." ♦

10. The remainder of the specifications of Committee No. 1 are recommended
for adoption, and are here re-arranged.

Standard Specifications for Steel Boiler Plate, Rivets, Castings and

foroinos.

Process of Manufacture.

Boiler Plate and Rivet Steel shall be made by the open hearth process.
Castings and Forgings may be made by the open hearth, crucible, or Besse-
mer process.
Castings may be annealed or unannealed as specified.

TennU Tests,

Test piece — ^The standard test specimen shall be eight inches (8") gauged
length. The standard shape is shown in Fio. 92.

PARALLEL SEOTION

NOT LESS THAN 9"~

Vi" ::l"':i'^

5 •

-18^

Mg. 92.

Width of specimen along the parallel section shall be li inches, whenever
possible.

Thickness of specimen shaU be one-half inch or over, whenever possible.

Plates — Two opposite sides shall be the rolled surfaces if not over f-inch thick.

Rivets — Rivet rounds and small rolled bars shall b.e ^ested full size as rolled.

Castings and Forgings — Specimen may be planed parallel sided or turned
parallel for not less than 9 inches in length, the smaUest dimension being }-inch,
if possible.

When it is inconvenient to use the standard test specimen the specimen
may be made as shown in Fig. 93. In every such specimen the elongation
in two inches will be 25% greater than that specified for the standard specimen.

Number of Test Specimens.

If a 'tensile specimen develops flaws or breaks outside the middle third of
its gauged length, another may be substituted.

* The "apparent elastic limit," suggested by Prof. J. B. Johnson in his "Ma-
terials of Construction," and restated by William Kent in Transactions of Amer-
ican Institute of Mining Engineers, 1903.

324 BOILER PLATE, RIVET STEEL, STEEL CASTINGS AND FORGINGS.

Plates — One from each plate as it is rolled.
Rivet Rounds. — Two from each melt.

Castings and Forgings — Depending upon the character and importance of
the piece.

Location of Test Specimens.

Castings — ^A test piece shall be cut cold from a coupon to be molded and
cast on some portion of one or more castings from each melt or blow, or from
the sink-heads (in case heads of sufficient size are used.) The coupon or sink-
head must receive the same treatment as the casting or castings, before the
specimen is cut out, and before the coupon or sink-head is removed from the
casting.

Forgings — The test specimen shall be cut cold from the forging or full-sized
prolongation of the same parallel to the axis of the forging and half way be-
tween the center and outside, the specimens to be longitudinal, t.e., the length

Fig. 93.

of the specimen to correspond with the direction in which the metal is most
drawn out or worked. When forgings have large ends or collars, the test
specimens shall be taken from a prolongation of the same diameter or section
as that of the forging back of the large end or collar. In the case of hollow
shafting, either forged or bored, the specimen shall be taken within the finished
section prolonged, half way between the inner and outer surface of the wall of
the forging.

Bending Tests,

Bending tests may be made either by pressure or by blows.

Cold bending tests are to be made on the material in the condition in which
it is to be used. For a quenched bending test the specimen shall be heated
to a light cherry-red as seen in the dark, and quenched in water, the tempera-
ture of which is between 80° and 90° Fahrenheit.

Test Specimen.

Plates — One and one-half inches wide and if }-inch or less in thickness with
opposite faces rolled. If over f-inch thick, specimen may be reduced to ^-inch.
Edges are to be milled or planed.

BOILER PLATE, RIVET STEEL, STEEL CASTINGS AND FORGINGa 325

Rivet Rounds — Tested full size as rolled.

Cattinga and Forgings — Specimen one inch by one-half inch.

Number of Test Specimens.

Plates — One cold bending and one quenched bending specimen from each
plate as it is rolled.

Rivet Rounds — ^Two cold bending and two quenched bending specimens for
each melt.

Location of Specimen.

Castings and Forgings — As specified for tension specimen.

Chemical Analysis.

Turnings from tensile specimen, drillings from tensile or bending specimen
or drillings from small test ingot may be used for chemical analysis.

For locomotive fire box steel check analysis may be required from the tensile
*P^imen of each plate as rolled.

Drop Test.

^ test to destruction may be substituted for the tensile test, in the case of

^'^^ or unimportant castings, by selecting three castings from a lot. This

, ^ «hall show the material to be ductile and free from injurious defects, and

Citable for the purposes intended. A lot shall consist of all castings from

the same melt or blow, annealed in the same furnace charge.

Percussion Test.

^rge castings are to be suspended and hammered all over. No cracks^
naws, defects, nor weakness shall appear after such treatment.

Homogeneity Test for Fire Box Steel*

^^rnple taken from a broken tensile test specimen, shall not show any
sw^^ Beam or cavity more than one-fourth inch (J") long in either of the three
lectures x>btained as described below.

portion of the broken tensile specimen is either nicked with a chisel or
grooved on a machine, transversely about a sixteenth of an inch dV") deep,
in three places about two inches (2") apart. The first groove should be made
on one side, two inches (2") from the square end of the specimen; the second,
two inches (2") from it on the opposite side; and the third, two inches (2")
'Tom the last, and on the opposite side from it. The test specimen is then
pwt in a vice, with the first groove about a quarter of an inch (J") above the
jaws, care being taken to hold it finnly. The projecting end of the test speci-
men la thg,^ broken off by means of a hammer, a number of light blows being
^^» and the bending being away from the groove. The specimen is broken
by the other two grooves in the same way. The object of this treatment is
to open and render visible to the eye any «eams due to failure to weld up, or
to foreign interposed matter, or cavities due to gas bubbles in the ingot. After

326 BOILER PLATE, BIVET STEEL, STEEL CASTINGS AND FORGINGS,

rupture, one side of each fracture is examined, a pocket lense being used if nec-
essary, and the length of the seams and cavities is determined.

Branding.

Every finished piece of steel plate shall be stamped with the melt number,
and each plate, casting or forging and the coupon or test specimen cut from
it, shall be stamped with a separate identifying mark or number. Rivet steel
may be shipped in bundles securely wired together with the melt number on
a metal tag attached.

Variaiion in Weight.

The variation in cross section or weight of more than 2J per cent, from that
specified will be sufficient cause for rejection, except in the case of sheared
plates, which will be covered by the following permissible variations:

Plates 12} p>ounds per square foot or heavier, up to 100 inches wide, when
ordered to weight, shall not average more than 2} per cent, variation above
or 2i per cent, below the theoretical weight. When 100 inches wide and over
5 per cent, above or 5 per cent, below the theoretical weight.

Plates under 12i pounds per square foot, when ordered to weight, shall not
average a greater variation than the following:

Up to 75 inches wide, 2} per cent, above or 2i per cent, below the theoretical
weight. 75 inches wide up to 100 inches wide, 5 per cent, above or 3 per cent,
below the theoretical weight. When 100 inches wide and over 10 per cent,
above or 3 per cent, below the theoretical weight.

For all plates ordered to gauge, there will be permitted an average excess
of weight over that corresponding to the dimensions on the order equal in
amount to that specified in the following table:

Table op Allowances for Overweight for Rectangular Plates When
Ordered to Gauge.

Plates will be considered up to gauge if measuring not over yj^-inch less
than the ordered gauge.
The weight of 1 cubic inch of rolled steel is assumed to be 0.2833 pound.

Plates \4nch and over in thickness.

Width of Flat*.

Thickness of plate.

Up to 76 inches.

75 to 100 inches.

Over 100 Inches.

Inch.

Per cent.

Over t

BOILER PLATE, RIVET STEEL, STEEL CASTINGS AND FORGINGS. 327

Plates under \ inch in thickness.

Width or Platb.

ThickDMs of plate.
Inch.

Up to 50 inches
Per cent.

i up to A

A " i

10
Si

50 inches and above.
• Per cent.

12i

Finish.

All material must have workmanlike finish.

Plates must be free from injurious surface defects and laminations.

Castings must be true to pattern, free from blemish, flaws or shrinkage cracks.
Bearing surfaces shall be solid and no porosity shall be allowed in positions
where the resistance and value of the castings for the purpose intended will
be seriously affected thereby.

Forgings must be free from cracks, flaws, seams or other injurious imper-
fections, and must conform to dimensions.

Inspection^
The inspector representing the piu*chaser shall have all reasonable facilities
afforded to him by the manufacturer to satisfy him that the finished material
is furnished in accordance with these specifications. All tests and inspections
shall be made at the place of manufacture, prior to shipment.
Respectfully submitted,

H. W. Spanoler, Chairman.

Chbxical Pbopbbtiss.

Physical Propbbties.

Bbmdino.

Stbbl.

Phos-
phorus
(not over),
per cent.

Sulphar
(not
OTcr),

per cent.

Manganese,
per cent.

Tensile

strength,

lbs. per sq. in.

(Allowable

TsrIation,

±5,0C0lbs.)

Boiler Plate a
Rivet:
Extra soft. . .

Fire box...]

Flange or (
boiler (

Forcings:

Soft

Medium

High

.04
Acid, .04
Basic, .03
Acid, .06
Basic,. 04

.06
.06
.04

.04

i .04

.05
.05
.04

.30to.5()
.30 to. 50

.30 to. 60

60,000
55,000

60,000

60,000
70,000
80,000*

28t
26t

25t

22
16
18

35
30
35

Flat.
Flat.

Flat.
i"

180
180

180

180
180
180

Castings. (When physical requirements are not specified carbon must be less
than 40 per cent, and phosphorus less than .08 per cent.) :

Soft
Medium .
Hard ...

.05
.05
.05

60,000
70,000
80,000

16
14
12

30
25
20

1"
1"

120
90

* For carbon steel, to be annealed and having no diameter nor thickness greater than
lO inehee. allow a reduction of 1.000 pounds for each additional inch in diameter or in
thiclcness of section.

t For material over ( inch thick deduct 1 per cent, for each i inch excess. For material
onder A inch thick deduct 2\ per cent, for each x^ inch decrease.

328 BOILER PLATE, RIVET STEEL, STEEL CASTINGS AND FORGINGS.

The Committee submits this as a tentative report, and asks
for it the careful consideration of the members of the Society
who are interested in the subject.

Mr. Henning. — From what Professor Spangler has said I am
simply amazed. There are statements made in the report which
cannot be supported. The determination of a very important
property indicated by the " yield point " has been dropped be-
cause it has become the custom in our mills to run machines at
such speed as to make it impossible to determine it. Now, I am
going to stand and fight for this, the determination of this point,
until I am dead. It is time to put a stop to such preposterous
audacity. I tfell you, gentlemen, as engineers, that we should
rather determine the permanence and the actual strength of all
machines and structures, not by the ultimate resistance, the
breaking point, but solely by the location of the yield point,
that point at which the material begins to change its shape per-
manently. A lathe, a machine, a bridge or boiler, once it begins
to change its shape permanently, is ruined. It has become the
custom in this country to run testing machines at such spee^J
that no one can tell whether the beam is floating at zero and in
dicating the load that is transmitted to the test piece, and I ar?
ready to prove that in court or anywhere else. Under such con
ditions it is absolutely impossible to determine the yield poiht ox
any other facts. The elastic limit is something we need not tall'
about, because it is difficult to determine, except by the mosf
sensitive apparatus. The method here described is absolutely in
accurate. I will tell you why. When you determine the one
thousandth of an inch of elongation it can only be done by ap-
plying a load to the test piece and taking a reading by very deli-
cate apparatus ; it must read to the ten-thousandth of an inch in
order to get accurately the thousandths of inches. When you take
a reading and stop the load and then reload that material, it begins
to stretch slightly, but the yield point will thereby be raised.

I wish to prevent such a report going into print. What I am
stating are well known facts.

Therefore, I do not want such specifications proposed when
there are methods for determining the yield point accurately —
by simply running the testing machine at a proper speed. I re-
peat, that by running a machine as rapidly as stated, no one can
know whether the beam is kept floating by the loads applied or
by inertia, and I object most strongly to such statements appear-
ing at this late date in a report of this Society.

BOILEE PLATK, RIVET STEKL^ STEEL CASTINGS AND FORGINOS. 329

DISCUSSION.

Mr, Chis. C. Henning. — Proposed specifications treating of the
subject named in the report of this Special Committee, and prac-
tically identical with it were proposed for discussion at the Boston
meeting of this Society, and there and then received a rather
thorough discussion which was not controverted.

At the Saratoga meeting the chairman of this Special Com-
mittee made a verbal report which again called forth criticism
which has not been proven to be incorrect by the Committee in
its present report.

On the other hand the chairman makes complaint that he re-
ceived but slight assistance and scant courtesy from the supposedly
interested membership, by reporting as follows : " In the usual
way copies of these specifications were sent to various members
of the Society, with the usual result — that is, in a few cases, after
writing two or three letters, replies were received." It may be
necessary to point out at this time and place that the Committee
was appointed to deveflop neither new specifications nor new
methods of testing, but merely to evolve from existing knowledge
and specifications a new set based on those in use, from which
^ould be eliminated their incompatible differences or incon-
g'Tiities, or clauses which had become unsuitable or useless, and
^^ the other hand to bring all the requirements up to date.

In order to do this a committee should be composed of engin-
eers who are intimately familiar with the subject submitted to
them; they should have all specifications before them; be famil-
wr with the design and construction of the products covered by
tbese specifications, and should also be in close contact with the
steel works and shops in which the steel is made, the work done
^nd the material tested.

Only such engineers can properly prepare specifications which
will be generally acceptable.

In spite of this complaint about lack of co-operation, the com-
mittee failed to avail themselves of previous criticism of the
specifications which they used as a basis for their work.* The
first criticism that I must make is about the laxity and indefinite-
ness of the language, and the errors and misunderstandings
which this necessarily introduces, because if there is anything an

* See pages 642-667, Vol. XXIIL, Transaetiona, A, 8. M. £.

330 BOILER PLATE, RIVET STEEL, STEEL CASTINGS AND FORGINOS.

acceptable specification should do, it is to specifically specify
detaily to avoid accidental or deliberate misinterpretation.

On the second page of the report which refers to castings, it is
stated that 2-inch test-pieces s>hould only be used "when it is incon-
venient to use the 8-inch." Now it is a well-known fact that it is
never inconvenient to cast one or more 8-inch test-pieces with
the casting when only one or several of them are made. It can
only be inconv^enient to the manufacturer when it does not suit
him to do so and in no other case.

These specifications are to be followed not after the work has
been done, but are furnished with the call for bids and are sup-
posed to be followed from beginning to the end of the work.
Hence there caAnot possibly be any excuse for failure to provide
8-inch test-pieces gated from the castings at the time the latter
are poured.

On the third page of the report which refers to boiler plate test-
pieces, it is prescribed that the " width of specimen along the
parallel section shall be 1^ inch whenever possible." This is
simply an absurd concession! Every boiler plate is many inches
wide, never under 24 inches, and as the specimens are cut out of
the crop ends sheared off the full width of plates there never can
be any difficulty of obtaining specimens of the prescribed width!
It is always possible to obtain specimens 1 i inches wide in the
parallel part^ except when the rolling mill does not want to do so ;
nowadays tests are invariably made at the rolling mill. Again it is
specified that : " When it is inconvenient to use the standard test
specimen, it may be made as shown in Fig. 93," which means
that a specimen of ^-inch diameter and 2-inch gauge length may
be used. How obliging the committee intends to be to the larg-
est mills who roll the largest plates f-inch thick and over! Xo
such test-pieces as shown could be cut from thinner plates —
just examine the dimensions! On the fifth page of the report a
'* drop test " is suggested for " small and unimportant castings."
Now what is the use of testing this class of castings at all? More-
over there is not one drop-test apparatus in existence at any steel
foundry in this country proper for this kind of test. At only a few
can very large drops be found, ample to break up very large cast-
ings before recharging them in the furnace. A hand hammer or
sledge would seem to me to be more appropriate for the purpose.
But what is the use of testing " unimportant castings " at all ?
Do standard specifications ever refer to unimportant material?

BOILEK PLATE, RIVET 8TEEL, STEEL CASTINGS AND FORGINQS. 331

On the other hand, the " percussion test " is prescribed for ** large
castings " ; these " are to be suspended and hammered all over/'

Martens * defines what a " percussion test " is supposed to be
by those familiar with it — it is a very different test from that
referred to in the specifications which is everywhere — the world
over — known as the " hammer test."

The hammer test again is only useful in case of small castings,
and does not have any effect on large ones. How then can it
be of any service whatever for the purpose proposed ?

Again, how can crushing castings in the large drops prove
them to be " suitable for the purposes intended." Such lan-
guage in specifications is rather ingenuous !

On the fifth page, under " Homogeneity Test," it is stated that
** a sample from a broken tensile test specimen shall not show
any single seam or cavity more than ^ inch long in either of
the three fractures obtained as described below." Now it is
well known that every firebox sheet is sheared on all sides, hence
if there are any defects in any sheet a competent examination of
its edges will always reveal them and should cause its rejection.
Even if the plates containing pitting or gas holes in the plates
showed only under a magnifying glass this would be ample cause
for rejections! What is the use of looking for defects ^ inch long
with a magnifying glass, I should like to know. They can be
seen by the naked eye at a distance of five feet! Let us exam-
ine the method proposed for finding such J-inch defects and at
the same time note carefully the language used which prescribes
the use of a " pocket lens " for finding J-inch defects!

" A portion of the broken tensile specimen is either nicked
with a chisel or grooved on a machine, transversely about a six-
teenth of an inch (tV'O deep, in three places about two inches
(2^^ apart. The first groove should he made on one side^ two
inches (2'^ from the square end of the specimen ; the second,
two mches (2") from it on the opposite side ; and the third, two
inches (2") from the last, and on the opposite side from it. The
test specimen is then put in a vise, with the first groove about
a quarter of an inch (^") above the jaws, care being taken to hold
It firmly. The projecting end of the test specimen is then
broken off by means of a hammer, a number of light blows being
^^, and the bending being away from the groove. The speci-
men is broken by the other two grooves in the same way. The

* " Martens's Handbook of Testing Materials/' pp. 291-29a

332 BOILEfi PLATE^ RlVET STEEL^ STEEL CASTINGS AND FOUGINQS.

object of this treatment is to open and render visible to the eye
any seams due to failure to weld up, or to foreign interposed
matter, or cavities due to cas bubbles in the ingot. After rup-
ture, one side of each fracture is examined, a pocket lens being
used if necessary, and the length of the seams and cavities is
determined."

It will be noted that the material is to be either " nicked with
a chisel or grooved on a machine, transversely about tV of an
inch deep," and the effect of this nicking on the steel is not taken
into account and is supposed to be inappreciable in either case,
even when plates vary in thickness from ^ inch to over 1 inch thick-
ness.* Everyone knows that very different results are produced.
But let us look further. It is specified that the grooves are to be
made in three places, two inches apart; ^' the first groove on one
side two inches from the square end of the specimen ; the second
two inches from it, on the opposite side ; and the third two inches
from the last and on the opposite side from it."

It is at once evident that two nicks will come, when following
the instructions, two inches from the square end of test piece and
opposite each other, and the third four inches from the end,
while it may be clear that this was not at all intended to be the
case.

But let us proceed and examine the method further which
prescribes that the specimen shall be *^ broken off by means of a
hammer, a number of light blows being used, after it has been
put in a vise, with the first groove about a quarter-inch above the
jaws, with the additional wise caution, '* care must be taken to
hold it firmly."

Let us remember that the ends of these test pieces are from ^
to one inch thick and two inches wide, and only three inches long
according to Fig. 92, or f-inch diameter and one inch long ac-
cording to Fig. 93, when the former shape " is inconvenient."

Now I will challenge anyone to break off a piece of fire-box
plate one inch thick by two inches wide, scored iV inch deep, as *
prescribed, when clamped firmly in a vise, by means of light
blows of a hammer. This is a ridiculous and absurd direction
which must be patent to all. This done, it is prescribed that
" the specimen is broken hy the other two grooves in the same
way." Has such English ever before been used in Standard

* These specifications cover i-inch plates, as they are referred to on seventh
page of ihe report.

BOILER PLATE^ RIVET STEEL^ STEEL CASTINGS AND FORQINGS. 333

Specifications ? The next important point which I am bound to
again criticise is the proposed omission of the determination of
*• yield point/' and definition and method of determination of
*' elastic limit."

It is proposed * " that the determination of the yield point for
ordinary grades be t)mitted.''

The use of the words " ordinary grades " has undoubtedly
been resorted to to ward off criticism.

Let us look at the meaning of the word " ordinary.^^ The
small boy applies this word to a useless dog or to another boy for
whom he has no respect because possessing bad habits and qual-
ities. He has no use for such material. Just the same with
boiler plate. What's the use of determining its yield-point,
when it's just ordinary, no good, worthless! That's about the
idea of the Committee.

But the dictionary defines " ordinary " as follows : " of com-
mon or ordinary occurrence, customary, usual." Hence, as
these specifications are supposed to apply to all usual, customary
or common steel of ordinary occurrence, this word does actually
apply to all good boiler steel except that of extraordinary quali-
ties. Hence the committee proposes to drop the determination
of the yield point in all cases of testing standard qualities of
boiler plate, instead of only of the inferior grades which are
unworthy of consideration, as the Committee would have us
believe.

Xow let us see what yield point is and how it is determined
easily and accurately by simplest means. Our honorary mem-
ber, Professor Unwin, is one gentleman who tells us what the
yield point is and how important it is to determine it.f Un-

* Second page of the report, paragraph 2.

t " 6. T%e Yield Point. — In iron and steel, and in some other roUed or ham-
mered materials, at a sfress exceeding more or less the elastic limit, there occurs
& Urge and almost sadden increase of deformation in the ordinary method of
testing, and the deformation is permanent, or plastic deformation. For greater
stress^ the plastic deformation increases, and it amounts, before fracture is
reached, to many hundred times the whole elastic deformation. The point at
which this almost sadden augmentation of plastic deformation occurs is termed
the yield point or breaking down point. It is obvious that a general plastic
jielding of a structure would ruin it for practical purposes, hence the yield
point seems to fix a limit of stress independent of that determined from consid-
erations of safety against fracture which the working stress should not exceed.
The yield point is raised by loading which exceeds the primitive yield point,
bat it is not usually practicable to raise the yield point of a material artificially

334 BOILER PLATE, RIVET STEEL, STEEL CASTINGS AND FORGINQS.

win says, p. 63, " Next to the yield point the most important
point to observe is the point where the maximiim load is reached."

Professor Martens * is another authority who emphasizes the
importance of the determination of yield point and defines clearly
the difference between it and elastic limit. Martens says, p. 28,
"' The yield point as also the proportional limit are well-defined
points." Again we may refer to our late member, Prof. J. B.
Johnson, who adapted his ideas to those proposed by the French
Commission on Methods of Testing Materials, but unfortunately
did not interpret the statements published in their language cor-
rectly. He translates the French term " la limite d'Elasticite
Apparente " as the *^ apparent elastic limit," and then defines it
as a point which is not at all apparent, but can only be deter-
mined and recognized by most delicate apparatus and with great
care and much labor. The very meaning of the word " appar-
ent " is as follows : " clearly perceived, or perceivable ; easily
understood, evident." The fact is that according to the defini-
tions of Bauschniger, Martens, Tetmayer, Bach, Unwin, and
many others it is readily observable by any careful inspector or
engineer in a mill, and does not require a laboratory equipment
with most highly trained assistants.

Why this method proposed by Johnson for determining the
'* elastic limit " (when his French models distinctly state that
they coined the expression to indicate the " yield- point " for
which they had no word) and re-stated by Kent in Trans. Amer.
Inst. Mining Engrs., 1903, should now be adopted by this
committee for determining a doubtful point very difficult to
ascertain, is difficult to understand. Let us examine what this
method proposed by Johnson and adopted by Kent really means
and leads to. Martens says on p. 30 :

" This is the proper place to call attention to a very important
misconception which is produced by the uncertainty of accurate
definition of the idea of elastic limit, and the existing careless dis-
tinction between proportional and elastic limits and of yield
point."

If you will read the references to Unmn and Martens, given
heretofore, relating to yield point, you will find that adding loads

before using it in a structure, and consequently the primitive yield jwint, due to
the mechanical operations of manufacture, fixes with respect to deformation the
dangerous limit of stress." '* The Testing of Materials of Construction," W. C.
Unwin, pages 7, 62, 98, 99, 250, 805.

♦ '* Hand-book on Testing Material*," Martens, pages 28, 80, 44, 261, etc.

BOILER PLATE, RIVET STEEL, STEEL CASTINGS AND FORGINGS. 335

successively and intermittently always raises it when testing
materials like boiler plate. Another point is this, that the yield
point shows itself very suddenly in many cases within an incre-
ment of load of 250 pounds per square inch.

The plates which these Specifications refer to vary from ^ inch
to over one inch in thickness as the allowable variations in weight
of all of these thicknesses are given in extensive tables. The
proposed increments of load for consecutive measurements of
extension are 1,000 pounds per square inch. Now the sections
of the proposed test-pieces 1^ x i inch to 1^ x 1 inch will vary
from 0.18 to 1.50 square inch.

The necessary load on the testing machines to produce 1,000
pounds per square inch on 0.18 square inch section will be 180
pounds. It is a practical impossibility, therefore, to make such a
t€8t on thin plates with any ordinary machine, and it is impossible
to make it accurately on thicker test-pieces with any machine run-
ning at the speeds customary at the present day in all mills. The
mere starting and stopping of such a machine will add a thousand
pounds more or less to an indefinite degree to my own knowledge,
by the great inertia of machine at such speeds and the inertness
of all operators under such conditions.

Moreover, the time required for such determination of what
is erroneously called " elastic lunit," but should have been called
"yield point," is so great and troublesome that no mill, where
the testing is to be done according to these specifications, would
or could put up with it. It would simply paralyze all the mill-
testing laboratories in the country. Moreover, the yield point
makes itself as quickly apparent on a large as well as on a small
test piece, because when this point is reached the material seems
to break down instantaneously.

The fundamental rule for' making all tests of materials is to
^^n the machine so as to add equal increments of load continu-
ously in equal intervals of time. The above method is abso-
lutely opposed to this fundamental rule necessary to obtain uni-
form and comparable results.

There is one simple and accurate way of determining the
yield point, which should be done in every case — as the yield
point is the most important point to be fixed according to all
authorities, for the proper design of all machines and structures
subject to varying loads.

This simple and accurate method is to use a pair of finely

336 BOILER PLATE^ RIVET STEEL^ STEEL CASTINGS AND FORGINGS.

pointed dividers, with a reading glass mounted on one leg to
show accurately what is happening under one of the points.

The dividers are set to any distance about eight inches. When
the dividers have then been set on the test piece with the point
of one leg in-a punch mark, a fine scribe-line is made on the test
piece with the other. As the loading then proceeds, stretching
at a uniform rate will become apparent as the scribed line moves
away from the point of the dividers. Instantly the yield-point
is reached the rate of extension increases with remarkable speed
in a striking manner and the operator then takes a reading of
load without any disturbance or interruption in the operation of
the test

This method is quite within the capacity of all those usually
employed in making routine tests in shops and mills, and gives
accurate results wherever desired.

There is another suggestion I should like to make to the com-
mittee as it has had such difficulty in obtaining assistance in pre-
paring these proposed standard specifications.

I refer to Figs. 92 and 93, in which account has been taken
of metric dimensions* after this Society last spring decided to
have nothing to do with the metric system, as it was no good at
the present time, in fact being hardly in general use anywhere.

These figures show metric dimension to the hundredths of a
millimetre. A hundredth of a millimetre is not quite .0004
inch, and our mechanics and those using the metric standards
might become perplexed and get these dimensions incorrect to
one of two hundredths of a millimetre. Had not these fractions
better be left off?

Moreover, if the approximate metric measures were substi-
tuted, the principal dimensions of the proposed test-pieces would
be almost identical with those used in continental Europe. But
I have little hopes that our committee will drop a decimal of a
millimetre in the radius of a fillet or in the total lengths and
dimensions as shown. Moreover, there is one positive error in
Fig. 92. There should not be eight divisions of the 8-inch
gauge length, but twenty of 0.40 inch each, which is practically
the same as the centimetre divisions which the Europeans have
adopted. The one-inch divisions are much too large for measur-
ing the proportional elongation of the test-piece, and those 0.40
inch are just about right, and results measured thereon can be

♦ Editor :— These metric dimensions have been omitted in revision.

BOILER PLATE, RIVET STEEL, STEEL CASTINGS AND FOROINGS. 337

readily compared with those obtained in Europe or published in
the most valuable foreign reports, and without introducing any
measurable errors.

It is not possible to prepare one set of Standard Specifications
to cover boiler plate, rivet steel, steel castings and f orgings.

This is demonstrated by the fact that in the report before us
. specifications for tank or sheet steel have become mixed up with
those for boiler plate, as will be seen by comparing the tables of
allowable variations in weight of plates. One table refers to
boiler plate from J inch thickness and over, while the other
refers to all sheets | inch in thickness and under. It is a well-
known fact that such thin plates arc never used in boiler con-
struction, hence any reference to them in this paper is quite out
of place. Much more might l>o criticised in the report as to the
indefiniteness as to how many tests shall be made, what facilities
for making them shall be provided, and who shall provide them.
In these Specifications everything referring to these matters
i3 left entirely to the good will of the steel maker, and no author-
ity is given to the purchaser of the material, who is the first party
to be satisfied.

It is to be hoped that this review of the report will give the
Diembers of the committee cause for reflection, and enable them
^ produce a report more nearly up to the present time in the
field of testing materials and to prepare practical and useful
Standard Specifications, which will serve the objects of the en-
gineer as well as those of the steel maker. The apparent result
of producing those before us has been a rather sad one.
Jir.W. W.Dingee.—

Racine, Wis., U. S. A., November 14, 1903.
N. F. R. HtUtan:

I^ Sir: In response to circular No. 979, I would say that for the past seven
years the J. I. Case Threshing Machine Company of Racine, Wis., have furnished
^ir Purchasing Agent with specifications for all the various material used in
^ business. They have a well equipped Chemical and Physical laboratory
^^ these specifications are prepared and where samples of all material re-
wived are tested to see that they come up to requirements. These specifica-
tions embody the result of our long experience in the requirements of material
for our special purposes and as a result we now have control of many subtle
"ifluences that formerly were not understood and the tendency of which was
to make the life of a manufacturer a burden.

Enclosed are samples of these papers touching the subject matter under con-
sideration and which may be made a part of the discussion.

Yours truly,
(Enclosure.) W. W. Dinoee.

338 BOILER PLATE^ RIVET STEEL, STEEL CASTINGS AND FORCINGS.

No. 307.

SPECIFICATIONS FOR BOILER RIVETS.

The rivets purchased under this specification are to be made
of the best grade of Open-Hearth steel, and should be formed in
solid steel dies, i.e., closed dies.

The rivets must be exact in size and fit the metallic gauges
- furnished by the J. I. Case T. M. Co.

In the event of no gauge being furnished it is understood that
the shape and dimensions are to conform to the blue prints fur-
nished by the company, or to a regular stock sample.

The body of the rivet should be perfectly round, exact in size,
and free from oxide or scale.

The head should be in accordance with the metallic gauge,
blue print, or sample.

The rivets must be free from all injurious defects, be finished
in a workmanlike manner, and fulfill the following requirements:

Physical Test.

Cold Bending Nick Test — Nicked with a cold chisel on one
side of the shank or body, the rivet must bend double (away from
the nick), flat upon itself (i.e., 180 degrees) without breaking.

Note — A hand cold chisel must be used in nicking the test pieces, and a ham-
mer of not over three poimds in weight used for striking the chisel. No rivet
will stand the bending test if nicked deeper than one-fourth the diameter of the
rivet.

Quench Test. — Heated to a cherry-red, quenched in water at
a temperature of 80 degrees Fahr., and then nicked on the side
with a cold chisel, the rivet must bend double (away from the
nick), flat upon itself without breaking.

Dish Test (Cold). — A rivet held in an upright manner or ver-
tical position, under the trip or drop hammer, and flattened by
repeated blows to a circular disk must show no signs of '^ cold
shortness," i.e., must not crack, split or crumble.

Flat Test (Hot). — A rivet heated to a bright cherry-red and
placed in a flat or horizontal position under the hammer, must
show no signs of " red shortness " (i.e., must not crumble), when
flattened to a thickness of one-fifth the original diameter of the
body.

BOILER PLATE^ RIVET STEEL, STEEL CASTINGS AND FOROINGS. 339

Chemical Comjpodtion.

Kivets which will pass all of the required physical tests should
be within the following limits in regard to composition:

Phosphorus should not exceed . 03 per cent.
StUphur " " " .025 " "

SiHcon " " " .02 " "

Manganese " " " .50 " "

Carbon " " " .15 " ."

Remarks. — ^Material will not be accepted which fails to meet
the requirements in regard to size, uniformity and gauge dimen-
sions.

Failure of more than five per cent, of the rivets to pass the
physical tests will be cause for rejectment.

(A certain number of rivets will be taken from each keg or
package for the physical test, and such test decides the acceptance
or rejectment of that package in particular, but if more than half
of the individual tests fail, the whole lot may be rejected.)

Material will also be rejected which shows on analysis:

Phosphorus over . 04 per cent.
Sulphur " .035 " •*

Any alteration in this specification, either in the physical re-
quirements or the chemical composition, must be stated in writ-
ing at the time of contract and signed by the party furnishing the
material, also by the Purchasing Agent.

No. 302.
SPECIFICATIONS FOR STEEL CASTINGS.

Steel for castings may be made by either the Open Hearth,
Crucible, or Tropenas Bessemer Process; preference being given
the open-hearth product.

The castings must be true to pattern, free from external blem-
ishes, blow holes, shrinkage cracks, cold shuts and other injurious
defects.

No porosity shall be allowed in portions where the resistance
and value of the casting, for the purpose intended, will be seri-
ously affected thereby.

340 BOILER PLATE^ RIVET STEEL, STEEL CASTINGS AND FORGINaS.

The castings are to be annealed or un-annealed, as specified by
the purchasing agent under the head of remarks.

If annealed, the castings must receive a proper heat treatment
and be given a sufficient time in which to cool, so that all internal
strains are relieved and the metal assumes its proper degree of
ductility.

Large castings suspended and hammered all over, should not
after this treatment show any defect or weakness.

The castings must be properly cleaned and free from sand,
scale, etc.

Steel castings will be divided into three classes: Soft, Medium
and Hard.

The physical qualities and chemical composition of the differ-
ent grades or classes must conform to the following require-
ments.

Soft GoAit Steel.

Physical Test.

Test pieces for the physical test may be cut cold from a
coupon, molded and cast on some portion of one or more castings
from each melt, or from a sink head, riser or sprue, on such
castings.

The test piece should be fashioned so as to give a gauged
length of two inches and about one-half inch diameter.

A bending test made on a specimen one inch wide and about
one-half inch thick, must bend cold, around a diameter of one
inch, through an angle of 120 degrees without fracture on the
outside of the bent portion.

Tensile Strength must not be less than 58,000 pounds per
square inch.

Elastic Limit must not be less than 27,000 pounds per square
inch.

Elongation measured in 2 inches, must not be less than 22. per
cent., and the reduction in area should not be less than 30. per
cent.

Chemical Composition,

Carbon may range from 0.10 to 0.22 per cent.

Manganefie " " " .20 " .75 " "

Phosphorus must not exceed .08 " "
Stdphur " " " .06 " "

Silicon should not exceed .40 " "

BOILEB PLATE, RIVET STEEL, STEEL CASTINGS AND FORQINaS. 341

Medium Cast Steel.

Physical Test.

Tensile Strength must not be less than 65,000 pounds per
square inch.

Blastic Limit must not be less than 31,000 pounds per square
inch.

Elongation measured in 2 inches, must not be less than 18. per
e^nt., and the reduction in area should not be less than 25. per

cent

Chemical Composition.

Carbon may range from 0.22 to 0.35 per cent.

Manganese " " " .25 *' .80 " "

Pho»j)horu9 must not exceed .07 " "

Sulphur *' " " .05 " "

Silicon should not exceed .35 " "

Hard Cast Steel.

Physical Test.

Tensile Strength must not be less than 80,000 pounds per
square inch.

Blastic Limit must not be less than 39,000 pounds per square
inch.

Elongation measured in 2 inches, must not be less than 12.

per cent., and the reduction in area should not be less than 20.

per cent.

Chemical Composition.

Carbon may range from 0.35 to 0.50 per cent.

Manganese " *' '* .30 *' .85 " "

Phosphorus must- not exceed .06 " *'

Sulphur " " " .04 " "

Silicon should not exceed .30 " "

When no special grade or class is specified by the purchasing
agent, the material may be anywhere within the limits of a soft
and hard steel.

Note. — It will be noticed that the specified chemical composition is not at all
ri^d except in the case of phosphorus and sulphur.

Experience has shown that phosphorus in un worked steel, if very high, pro-
duces extreme brittleness by making the material ''cold short.'*

High sulphur produces ''red shortness/' increases shrinkage, makes the cast-
ing hard, and indirectly causes blow holes.

Manganese stiffens the steel, raises the elastic limit, piartially neutralizes the
effect of sulphur, eliminates the occluded gases, and in a measure prevents blow
holes.

Silicon in the initial charge raises the heat and imparts fluidity; in the final
product it has a tendency to impart rigidity and hardness.

342 BOILER PLATE, RIVET STEEL, STEEL CASTINGS AND F0RQING8.

Aluminium, either pure, or in the form of "Ferro-aluminium" is gener-
ally added to the melted steel for the purpose of quieting the bath, raising the
heat, purifying the metal, and imparting greater fluidity; an excessive amount
of aluminium added to a steel low in manganese, will often produce a segregated
porosity in the material.

Blow holes in steel castings are the most common cause of trouble, and the
defect is seldom discovered imtil the casting breaks.

Porosity or spongy places in castings are also a source of trouble; this may
in a measure be avoided by the use of fillets in sharp comers and angles, and
by changing the form of the pattern so as to relieve the draw.

Under the head of physical test we have allowed an elastic limit of less than
half the ultimate tensile strength, usually however, it will be higher, and in the
high carbons, somewhat closer to the breaking point.

Carbon is not rigidly specified, as noticed in the adoption of the word
''should," which implies that the maker be allowed a reasonable variation in
order to get the desired strength.

We reserve the right to use the drop test or any other reasonable test to de-
termine the quaUty of the material and its freedom from blow holes or other
flaws.

Instruction in regard to annealing, and such changes as the purchasing
agent may decide on in respect to physical requirements, chemical composition,
etc., must be stated in writing and signed by the party in question.

No. 324.
SPECIFICATIONS FOR STEEL RIVET RODS.

Two grades of steel are considered in this specification and
are to be designated as " Extra Quality " (Class A), and " Or-
dinary Quality '' (Class B).

Class A is to be a mild steel of superior quality, made by the
Open-Hearth process.

Class B is to be a good grade of soft steel, made by the Open-
Hearth or Bessemer process, as ordered and agreed upon at the
time of contract.

Both grades of material must be free from injurious defects,
be free from excess of scale, finished in a workmanlike manner,
and conform to the requirements of the physical test and the
chemical composition.

" Extra Quality (Class A)."

Chemical Composition.

Phosphorus must not exceed 0.03 per cent

Sulphur " " '* .025 " "

Afan^ane«e should not exceed .50 " **

Carhon " " " .15 " "

Smwn " " " .02 " "

BOILER PLATE^ RIVET STEEL/ STEEL CASTINGS AND FORGINOS. 848

Material will not be accepted which shows on analysis:

Phosphorus above 0.04 per cent.
StUphur " .035 " "

Physical Tests.

The Tensile Strength should not be less, than 55,000 pounds
or more than 62,000 pounds per square inch.

The Elastic Limit must not be less than one-half the ultimate
strength.

The Elongation, measured in 8 inches, should not be less than
28.00 per cent.

Material will not be accepted which shows a tensile strength
of less than 45,000 pounds or more than 65,000 pounds per
square inch.

"Obdinaby Quality (Class B).^
Chemical Composition,
Phosphorus must not exceed 0.09 per cent
Sulphur " " " .05 " "

ilfaiiyaf)«<e should not exceed .50 " **
Carbon " " " .16 " "

SUicon " " " .05 " "

Material will not be accepted which shows on analysis:

Phosphorus above 0. 10 per cent.
Sulphur " .06 " "

Physical Tests.

The Tensile Strength should not be less than 52,000 pounds or
more than 62,000 pounds per square inch.

The Elastic Limit must not be less than one-half the ultimate
strength.

The Elongation, measured in 8 inches, should not be less than
26.00 per cent.

Material will not be accepted which shows a tensile strength
of less than 45,000 pounds or more than 70,000 pounds per
square inch.

Cold Bending Test, — Class A and B material must bend flat
upon itself (180 degrees) without showing signs of fracture on
the outside portion of the bend.

Quench Test. — Class A or B must show no signs of fracture
when bent flat upon itself, after heating to a cherry red and
quenching in water.

344 BOILER PLATE^ RIVET STEEL^ STEEL CASTINGS AND FORGINGS.

Hammer Test. — Class A, material heated to a bright cherry red
and drawn out under the hammer to a thickness, at the point,
to one-fifth its original diameter must not split, crack or crumble.

Class B, treated in the same manner must not crack or
crumble when hammered to a thickness equal to one-third the
original diameter (rf the rod.

Any alteration in this specification in regard to composition
or physical requirements pertaining to either class of material
must be stated in writing at the time of contract and signed by
the Purchasing Agent and the party furnishing the material.

No. 287.

SPECIFICATIONS FOR BOILER PLATE, FIRE BOX AND
FLANGE STEEL.

Boiler Shell Plates, Front Tube Plates and Butt Strips.

The material purchased imder this specification is understood
to be a superior grade of " Open-Hearth '' steel.

It must be homogeneous in structure, free from blister, cracks
and other injurious flaws or defects.

It must also fulfil all the requirements of the following
physical test:

Physical Test.

Tensile Strength, — The ultimate breaking strain must not be
less than 52,000 pounds or more than 65,000 pounds per square
inch.

Elastic Limit. — The elastic limit must not be less than one-
half of the ultimate tensile strength.

Elongation. — ^The elongation measured an eight (8) inches,
must not be less than the following per cent., according to thick-
ness:

20 per cent, for plate f in. and under.
22 " " " " i to i in. thick.
25 " " " " fin. and over.

Cold Bending Test. — The specimen must bend flat upon itself
(180 degrees), without showing signs of fracture at any part of
the bend.

Quench Test. — A piece of the steel plate heated to a bright
cherry-red color, then quenched in water at a temperature of

BOILER PLATE^ RIVET STEEL^ STEEL CASTINGS AND F0RGIN08. 345

82 degrees Fahr. must bend around a curve equal to one and one-
half (1|) times the thickness of the plate without showing signs
of fracture on the outside portion of the bend.

Chemical ComposiUon.

In order to meet all the requirements of the physical test and
be within the limits of a superior grade of Open-Hearth steel,
the chemical constituents must conform to the following speci-
fied percentage:

Carborit not under 0.12 or above 0.20 per cent.
Phosphorus, not above .04 " "

Sulphur, a u 04 u u

Manganese, " " .50 " "

Saicon, " " * .05 " "

Rejedment. — Material will be rejected which shows:

(1) A tensile strength of less than 50,000 pounds per square
inch or more than 65,000 pounds unless the elongation is 28
per cent, or more.

(2) An elongation less than the quotient of 1,400,000, divided
by the tensile strength per square inch.

(3) Failure to pass the cold bending or the quench test.

(4) Phosphorus above 0.05 per cent. Sulphur above .05 per
cent.

Fire Box and Back Tube Plate,

The material desired under this heading is understofld to be
the best grade of " Open-Hearth " steel that it is possible to
make by modem methods.

Metal of the following composition is desired.

Carbon f 0.18 percent.

Phosphorus, not dho\e .03 ** "
Sulphur, " " .02 " "

Manganese, " " .40 " "
Silicon, " " .02 " "

Material will be rejected which shows on analysis:

Carbon, below . 14 or over . 25 per cent.
Phosphorus, above .035 " "

Sulphur, " .045 " "

Manganese, " .50 " "

SUican, " .06 " "

346 BOILER PLATE^ EIVET STEEL, STEEL CASTINGS AND FORGINGS.

Physical Test

This steel should pass all the specified tests, viz:

Tensile Strength, — The ultimate breaking strain should not
be less than 52,000 pounds per square inch, nor more than
65,000 pounds.

Plates with a tensile strength of more than 65,000 pounds will
not be rejected providing the elongation is 30 per cent, or over.

Elastic Limit. — ^Must not be less than one-half of the tensile
strength.

Usually the elastic limit is about 35,000 pounds where the ten-
sile strength averages 60,000 pounds per square inch.

Elongation. — ^Measured in 8 inches, should not be less than 28
per cent; and in all cases it should not be less than the quotient
obtained by dividing 1,450,000 by the tensile strength per square
inch.

This material must also stand the cold bending test and the
quench test.

It must also be free from seams, cracks, cavities due to gas
bubbles in the ingot, pipe laminations or other defects.

A micro-photograph of the material should show a homogene-
ous structure.

The plate should be of even thickness, straight, smooth rolled
and finished in a workmanlike manner.

Mr. Walter FUnt—

New York City, November 18, 1903.
American Society of Mechaniccd Engineers:

Gentledl^n: I have the report of the committee which was appointed to
make out specifications for boiler plate, rivet steel, etc., and I have looked the
matter over quite thoroughly. It seems to me that Mr. Henning's point is well
taken. Surely every one must admit that when a member of a bridge or boiler
or any other structure is loaded beyond the "jrield point," it will never again
perform its duty in that structure, and the other members are then called
upon to perform a duty for which they were not designed. I will, therefore,
say that the determination of the "yield point'* certainly ought to be retained,
and every care taken to determine it with accuracy.

Yours very truly,

Walter Flint.

L, S, Randolph. — This discussion seems to be open to criticism
for the following reasons:

First. The grouping of the specifications for steel castings
and boiler plate is a very decided mistake. This material is not
used by the same departments of the shop, or are they made in
the same shop, and it is a useless complication to have one specifi-

BOILER PLA%E^ RIVET STEEL^ STEEL CASTINGS AND FORGINOS. 847

cation covering two classes of material so widely different as
boiler plate and steel castings. It is doubtful whether there is
any reason for the riveted steel going on the same sheet with
boiler plate.

Second. The minimum requirements should be lower with
possibly a different nomenclatum carrying it very soft and giving
55,000 pounds ultimate tensile strength plus or minus 5,000
pounds. This allows material to come under this head which
is needed for certain classes of fire-box work. Of course, the
corresponding change in the elongation being made.

Third. The specifications are a little ambiguous as to the size
of test specimens, the indication being that 2^ inch test speci-
mens can be used on boiler plate, although this may be intended
to castings and forgings only. One of the most important con-
siderations in testing materials is to have the specimens of uni-
fonn size. If there is much work where the large specimens
cannot be obtained, a small specimen should be adopted as
standard and should be held to.

Fourth. It should be stated that the facilities for making
tests, etc., should be furnished by the manufacturer free of cost.
This is what is being done at the present time.

Fifth. I agree with Mr. Henning in the dropping of the yield
point It would be much more sensible to drop the ultimate
tensile strength and leave the yield point in even if we had to
slow down our machines somewhat. There is little doubt in the
mind of the writer that much of the inspection to-day is simply
perfunctory. The specifications should be drawn up primarily
so as to insure the furnishing of first-class material, and should
then be eased off so as to get everything that is absolutely neces-
sary in the qiifility of material desired at a minimum price and
at the minimum expenditure of trouble by the manufacturer.
The manufacturer, however, should be made to come to the
specifications, and not the specifications to the manufacturer.

Sixth. The drop test and percussion test are, in the opinion
of the writer, hardly necessary for the boiler steel, and the*
specifications are so ambiguously drawn in regard to the two
tests that it is impossible to say whether they are meant for
eastings only or boiler steel also. It should be said that it is
very difficult to have specifications so widely different in uses
wid characteristics under the same heading and not have such
ambiguities "^ occur.

348 BOILER PLATE, RIVET STEEL, STEEL CASTINGS A]ft> FOROINOS.

A, Bement, — A most surprising feature of these specifications
is that the determination of the most important characteristic
is excluded, that of the yield point. I cannot believe that the
members of this Society will allow the perpetuation of this grave
error in the final report of the committee. It is specified that
this omission be made for " ordinary grades," which means, in
fact, all commercial grades, and is almost equivalent to recom-
mending the abandonment of this determination altogether.

I would suggest that requirements for material used for bolts
and studs be included in these . specifications, especially those
employed in engines and similar machinery. With rivets, if
unsuitable material is employed, there is liability of their break-
ing before delivery to the customer. But Avith bolts and studs,
bad material can be used without liability of failure until deliv-
ery of the machine is made to the customer; then failure is quite
liable to be attributed to some other cause.

Prof. 0. Lanza. — ^I should like to make a few remarks, simply
to touch upon two points. Professor Spangler, in proposing to
drop the yield point, claimed that in steel varying from 45,000
to 65,000 pounds tensile strength the other requirements would
secure, in every case, a yield point of at least half the tensile
strength. In making that remark he admits its importance, and
it seems to me that the \-ield point requirement ought not to be
dropped, at any rate, imless that position is indisputably proved,
and I do not believe that it is proved yet beyond the possibility of
doubt.

Another point is that, in considering axles and moving parts
of machinery, which are subjected to alternate stresses, we need
to look after our specifications with reference to the power of
material to bear repeated stresses. Investigationamipon repeated
stress show that the yield point plays some part which is not
very well known yet, and it seems to me important that we should
retain a requirement for it until these relations can be deter-
mined.

The only other remark I desire to make is, that when too
high a speed is adopted in the testing machine, it -is not only the
yield point which is concealed, but various other things which
are of importance — ^matters which concern the tensile strength
also.

Prof. R. C. Carpenter. — I would like to make a remark on this.
It seems to me that the yield point is of importance, and of suf-

BOILEK PLATE, EIVET STEEL, STEEL CASTINGS AND FOROINOS. 349

ficient importance to be retained in these specifications. One of
the important things which I think may result from retaining
the requirement for determining the yield point is that it will
lead to more care in the testing of materials. At the present
time the material tests made by manufacturers are made too
rapidly to be accurate, as I have found out by experience. In-
deed, they are in many cases very far from accurate. This ad-
ditional requirement must require more care, and will give us
much more reliable results, from commercial tests.

n. W, SpangUr. — As the Chairman of the committee has
received but one communication, that of Mr. Bement, relating
to the matter of these specifications, which discussion is here-
with presenter!, there has bet»n no meeting of the committee
since the last meeting of the Society. The following comments
are his own, and not those of the committee.

It is to be remembered that these specifications cover steel
between 45,000 and 85,000 pounds tensile strength having def-
inite chemical and physical properties.

There has been but one objection raised to these specifica-
tions, and that as I understand it is to the dropping of the yield
point (or elastic Jimit) for materials covered by these specifica-
tions. With the other requirements of the specification ful-
filled, a high elastic limit will be obtained. I would be glad to
be referred to any data showing that if all the requirements of
this specification are met, the yield point is not as high or higher
than half the ultimate strength. A careful examination of the
reports of the Watertown Arsenal for 1901 and 1902 shows that
there are 133 tests of steel, mostly castings and forgings (the
nltmiate strength of which would bring them within the limits
of these specifications, and which materials might be tendered
nnder these specifications) having the elastic limit less than one-
balf the ultimate strength. Of the samples one hundred and
mneteen would not fulfil the other physical requirements. As
the chemical composition of these samples is not stated, it is
unpossible to tell from the printed reports whether the re-
mainder would fulfil the specifications or not, but 89.5 per cent,
would be rejected on their physical tests alone exclusive of the
elastic limit and the chemical composition.

The objection made by Mr. Henning that the suggested
method of determining the elastic limit (yield point) is inaccur-
ate because " when you take a reading and stop the load and

860 BOILER PLATE, EIVET STEEL, STEEL CASTINGS AND FOHGINGS.

.<E.L.= 46800
^ Juit. =62400
8''Elong.= 33.3jf
2''Elong.= 47.5j^
jCEed -60.7

TEST OF SOTT STKKL

Machine running continuously after 5000."^
Scale weight moved by hand*

128

137

.28

168

128

125

124

Fig. 94.

118

then reload the material, it begins to stretch slightly, but the
yield point will thereby be raised '^ does not, I believe, apply to
materials that are not loaded above the proportional limit, and

BOILER PLATE, RIVET 8TEI;L, STEEL CASTINGS AND F0RGING8. 351

-4800/perDinch

17660 -400001)er a Inoh

laseo-aooooperainoh

8B«-S0OO0riMrnliidh

44a0-100o/per ainoh

yfCRL.— 46000
<Ult -64000

8"Blong.»=80.8jf
2"Elong.= 47,5j^
Bed. =62.8jt

TEST OF SOFT Si'KKL

Machine stopped at each step. Scale weight advanced
before machine is again started.

124

123

124

125

181

1.00

164

126

Fig. 95.

practically I do not believe that the value of the quantity to be
determined here is affected by the stopping and starting of the
machine necessary for the reading of a micrometer extenso-

352 BOILER PLATE^ RIVET STEEL, STEEL CASTINGS AND F0RGING8.

meter. A series of tests were made by me on a bar of com-
mercial soft steel to test this point. An extensometer having a
magnification of about 150 times (exactly 147 times) was used,

„(E.L.«4Qa00
(Ult -66000
Elong. 8''=»28,8j<
Elong. 2-44.»
Ked.Area-59.(*

10620.

SOFT BESSEMER MACHINERY STOCK.

.20

121

122

124

134

155

127

Fig. 96.

and two of the diagrams obtained, together with the results of
the eight tests made, are here recorded.

In Fig. 94r herewith the scales are horizQutally 147" = 1''

BOILEB PLATE^ RIVET STEEL^ STEEL CASTINGS AND FOEOINGS. 353

80000^

'^^ OPEN HEARTH COMMON SPRING STOCX

Fig. 97.

354 BOILER PLATE^ RIVET STEEL, STlIEL CASTINGS AND FORQINOS.

stretch, and vertically 1"= 4,687 pounds. The irregularity is
due to the fact that the weight on the scale beam was moved by
hand. In Fig. 95 the scale beam was first set at 4,420 pounds,
and the load applied as rapidly as in Fig. 94. When the scale
beam rose the machine was stopped and the micrometers^ were
read. The scale beam was then set at the next load and the
machine started, each jog in the diagram showing one stoppage
of the machine.

While it is true that a few experiments may mean but little,
I know of no other printed data covering the same ground and
submit these results which show that in this case at least the
stopping of the machine a number of times and allowing it to
remain at rest long enough to read the micrometers did not
perceptibly raise the yield point.

I attach also two test sheets, one from soft Bessemer machinery
stock, Fig. 96, and one from open hearth common spring stock,
Fig. 97, showing that before and after the machine had been
stopped there has been in these two cases no change in the gen-
eral direction of the curve. In each of the two cases the
machine was stopped twice to determine the scale of the
diagrams.

Commercial Soft Steel.

^«^^^,S^PP'y*°K Ultimate. Yield point.

No. 6— Step 64,000 46,000

No. 5— Continuous . . 62,400 46,800

No. 3— Step 61,300 47,000

No. 2— Continuous . . 61,500 45,300

X— Step 64,100 45,000

Y— Continuous . . . ! . 63,300 47,500

Z— Step 62,700 47,000

A— Continuous 64,000 47,300

Average step 63,025 46,250

Average continuous. 62,800 46,725

Elongation in

Bedactlon

t in., p. c. 8 In., p. c.

p.c.

30.3 47.5

62.3

30.3 47.5

60.7

29.1 49.0

62.3

31.5 50.0

62.3

Outside marks.

28.9 42.0

64.0

30.1 46.5

60.7

28.4 47.0

62.3

29.8 47.6

61.8

29.8 46.6

62.3

ORDNANCE FOR THE LAND SERVICE. 365

ORDNANCE FOB THE LAND SERVICE.

MAJOR R. BIRMIB, OBDNANCX DXPABTMBNT, UNITBD ITATES ARMY.

1. The development of the modem land and naval ordnance
in the United States is of comparative recent date, and is a sub-
ject for interesting study from a political as well as an industrial
and mechanical standpoint. It is with the latter, however, that
I have to deal, and particularly the mechanical features, although
such is their diversity that an outline only can be attempted.
My subject, generally, will be confined to a brief account of the
growth of the present land armament, the uses to which the
guns and their accessories are applied, and to some discussion of
the more important mechanical principles involved in their con-
struction in connection with experiments.

2. There are two general divisions of artillery f^ land service;
first, the mobile artillery designed for field and siege service,
and used also in the land defense of sea-coast forts, or generally
for oflFensive operations, in which the weight of the material
combined with the specific service in view control the caliber and
design; second, the artillery on fixed mounts as used generally
in sea-coast fortifications, and also in permanent inland defenses
for defensive purposes, in which the caliber and design are not
restricted by weight and may be varied to subserve only the
needs of defense.

3. The different types of artillery are classified as gims,
howitzers and mortars. Guns are used for direct fire with a
superior angle of elevation, usually limited to about 15 degrees.
They fire a single weight of powder charge wth the object of
attaining a maximum of power. Howitzers are pieces of
medium length and velocity, used for curved fire up to about
45 degrees elevation. They apply generally to mobile arma-
ment for field and siege service, and by reason of their relatively

» Presented at a Special Monthly Meeting in New York (Febranry. 1904) of the
Aoaerican Society of Mechanical Engineers, and forming part of Volume XXV. of
the TrauBaeHanB,

3S6 ORDNANCE FOR THE LAND SERVICE.

light weight as compared with guns can be made of larger caliber
and use a heavier projectile than a gun of equal manoeuvring
capacity. This advantage and their supplemental use to guns
is further obtained by ability to deliver the curved fire at rela-
tively short range, for which reduced charges may be used. Mor-
tars are short pieces fired with medium or low velocity and used
for high angle fire with elevations up to about 70 degrees. They
require a fixed platform, and their utility is limited to siege and
sea-coast emplacements. Their firing angle enables complete pro-
tection to be given them in their emplacements from a hostile
gunfire. The projectile carries a large bursting charge, and the
angle of fall enables it to reach the most vulnerable parts of
targets generally. In the case of a battleship this target embraces
not only the relatively thin-armored deck, reached by direct im-
pact, but also the unprotected hull, reached by under-water ex-
plosions. To cover the field within the range of a mortar
different powder charges and elevations are employed, giving suc-
cessive zones of fire.

4. The following tables, which include the various calibers of
standard types in our service, enable a more comprehensive view
to be taken of the diversity of types included in a complete equip-
ment and how they are employed. All of these are rifled, breech-
loading pieces, made of steel and use smokeless powder, except
the model 1886 12-inch mortar, which is composed of a cast-iron
body reinforced with steel hoops. The tables omit the small arm
rifle caliber Gatling or automatic machine guns that are mounted
on tripod or wheel mounts and used both in the field and in fortifi-
cations for combating men and horses.

5. All of the carriages are fitted with hydraulic recoil cylinders
and spring return, except the 3.6-inch mortar and the 3.6-inch
gun field carriage which is of older pattern, dating from 1891.
The 3-inch model 1902, with long recoil of gun on carriage, is of
the most modern type. Modifications of the existing mobile artil-
lery now in progress or proposed are ae follows:

6. Introduce a light field gun to operate with cavalry, having
a caliber 2.38 inches, weight of projectile 7.5 pounds, and muzzle
velocity 1,700 f. s., and a 3.8-inch field howitzer of the same
mobility as a 3-inch field gim, to fire a 30-pound projectile.

7. Eliminate the 3.6-inch field gun, 5-inch field howitzer,
5-inch siege gun, 7-inch siege howitzer, and substitute therefor
3.8-inch rifles firing a projectile weighing 30 pounds, and 4.7-inch

OUDNANCE FOU THE LAKD SERVICE.

357

"3.

111

III
Si

^ ^ a

1 I
ill lill

1 I 1 1 H I

'auoq jad
)q8pM 8up)ai|i

§ §

§ I §

'saejoq jo iaqurn^

'aSau 9Ai)3Aj|a
9)«ui}zoiddy

5 s

I II § § § i

« lo «o « «r of V

'Xli;>o(aA aiszuH

§ §§§§!§

•iq^i^Ai

I^S' SgaS98§

■g

e^ 2 "2 "3 "5 "S "S

^ S S S 9 S S

•Q P« O* CW 01 Ot &

II I II II

a ^ ■ « « « 00

_. "O fl TJ "O "O "O

1 S S 8 S S S

"5 "oi U "oj "oJ "S'

« .d ^ jO ^ ^ J3

]^ 00 OQ CC CO OQ 00

'aSxvqa japMoj

S 8 « a « • o

5 i

'uoq }o qi^ayi

s fe a

a
g

i I

1 |l

«i • .d

I -s -s

5 5 5

x& I& t:.

358 ORDNANCE FOK THE LAND SERVICE.

howitzers of the same mobility firing a projectile weighing 60
pounds; also 4.7-inch rifle firing a projectile weighing 60 pounds,
and 6-inch howitzers of the same mobility firing a projectile
weighing 120 pounds.

8. This plan has the merit of consistency in that the weights
of projectiles are exactly doubled for each increase of caliber.
It would appear that the 3.8-inch howitzer and the field gim of
the same caliber are relatively the least essential and might be
omitted, and the light field gun would be improved by increas-
ing the caliber and the weight of projectile to about 10 pounds,
making a more efficient projectile for both shell and shrapnel
fire.

6 and 15 pounder guns to give 3,000 f. s. initial velocity and
a 6-inch wire-wound gun to give 3,500 or 3,600 f. s. velocity
have been projected.

9. The diversity of calibers of guns is remarkable, indicating
how completely every phase of the problem of attack or defense
is sought to be covered. When we examine further the designs
of modem guns, of the carriages and mounts, sometimes oper-
ated by electricity, the projectiles, fuses, powders and high ex-
plosives, the instruments and methods employed for range find-
ing and for fire control and direction, it is apparent how complex
the science of ordnance has become. Simplicity of means and
design is especially desirable in the military service, which
requires the greatest perfection in training of personnel. More-
over, the material is through necessity generally designed with
high regard for economy in weight and dimensions and withal
is subjected to most severe usage. The present period is marked
by a strong tendency to increased complexity and expense, en-
gendered by the rivalry of nations and made possible by the
advance of knowledge in mechanical appliances. Experience
has taught that this tendency is no more to be successfully com-
bated in military than in civil progress, and we must employ the
common medium of trial to seek out what is good and reject
what is unsatisfactory.

nistorical Sketch of the Devd^jmient of Designs and Trials of
Experimental Material,

10. It is to be remarked that the efficiency of the land arma-
ment existing at the close of the Civil War has not since been

ORDNANCE FOR THE LAND SERVICE.

359

o
o

94«uipcoiddy

poavj pj«q JO non
•«20jjad p9)«mn8B

^§ § §

See .o «r

§ § § §

t^ r^ 00 00

§§§§§§..

« of 0^ • -H^ ^ of

■X8jaa» a(zsiiK

'X){90pA 9[2ZaK

•iq3|aiki

s ^ g s

t> 00 00 o et A

s s :; 8 •-■ "^

S g

§ § ^

of of V

§ §

s i 1 1 1 g

S 8 « 8f ^ »

;<^~ M of

§ § § §

ef 00 of 00

of ei

!§§§§§§

of of of of f-T ^ t4^

I § g

§ i I I § § I §

1 1* e
•S -2 -S

1 ? 5

^ P^

• >

fit

8_-

'aSivqa japMo<i

S'^ 8 : * : SS : : : : : : : : « *

jg: :*^S8898S|gg|5aS

•Moqjoqiaoai

|8 2S9S!S2SS$S$3'*SS

i I i N I

r^ »^ •-" Tl ,«, «J

>4 :^ ^ ;i^ ^ ^'

{ki ^ h pci « «

0? pi plj 05 •£ "C

^ XI .a 43 .d .d

o s ^ a 3 ^

2 2^2^6

i i I

^ S

C C3 o o o

^ ^ a a s

.d ^ .xa .a ^

o o o o cS

fl a a a

s s ^ ^ s

360 ORDNANCE FOR THE LAND SERVICE.

equalled until the present period. In 1866 the artillery of the
land service was deemed very efficient. It comprised, however,
only muzzle-loading guns, and the introduction of rifling, with
its accompanying change from the round to the elongated pro-
jectile and consequent increase of weig'ht of projectile, range
and power generally, had not then reached the larger calibers.
The principal rifled guns were the 3-inch wrought-iron field rifle,
the 4.5-inch cast-iron siege rifle and the Parrott 10-pounder, 30-
pounder, 100-pounder, 200-pounder and 300-pounder rifles made
of cast-iron reinforced with a coiled and welded wrought-iron
band shrunk over the breech. Bronze smooth-bore guns were
represented in the Coehorn mortar, mountain howitzers and the
well-known 12-pounder Napoleon field gun. Cast-iron smooth-
bores constituted the remainder, and comprised 8-inch howitzers
and 8 and 10 inch mortars for siege equipment, with 13 and 15
inch mortars and 8, 10, 15 and 20 inch Rodman smooth-bore
guns for sea-coast armament. The carriages were of simple pat-
tern. Those for field guns were made with the flasks and trails
of wood, and those for the sea-coast guns, which were made of
iron, comprised principally a chassis and top carriage. The
chassis rested upon wheels upon traverse circles enabling the
piece to be moved in azimuth by hand bars inserted in sockets in
the traverse wheels. The top carriage was fitted with eccentric
wheels at the front to provide rolling friction in part for run-
ning into battery and sliding friction for recoil. Data pertain-
ing to some representative pieces of this equipment are as
follows:

Powder Projec- Mnzzle Bleva- -o.^,^

Piece. charge. tfle. Telodty. tion. t?2ff

IbB. Ibe. f.8. degrees. y»~••

5.82.incli8mootli bore mortar 6 17.75 45 1200

Wnch rifle (Parrott) 1.0 10.5 1282 20 5000

4.2-inch rifle (Parrott) 8.5 80. 1298 25 6700

6.4-iiicli rifle (Parrott) 10. 100. 1300 85 8450

18incli smooth bore mortar 20. 228 45 4650

20-iDch smooth bore gan (Rodman) 200. 1080 25 8000

11. The magnitude of the armament existing at that time may
be inferred from the statement that the number of Rodman
smooth-bore guns available was about 425 8-inch, 1,000 10-inch,
305 15-inch and 2 20-inch guns.

12. Great merit was attached to the heavy, smooth-bore guns
for their " " racking " effect, even after the introduction of

ORDNANCE FOB THE LAND SERVICE. 361

annor on ships. Nevertheless, the advantage of elongated pro-
jectiles with their " punching " effect was recognized, and 8 and
12 inch cast-iron rifles were made for trial as eariy as 1861.

13. The period of 20 years from 1866 to 1886 was marked by
the introduction of breech-loading in place of muzzle-loading
guns and the gradual substitution, first, of wrought-iron, and
afterwards, of steel for cast-iron in their construction. This
period may well be styled experimental in this country, except
that it resulted in the production of some 210 8-inch muzzle-
loading rifles converted from 10-inch Rodman smooth-bore guns
by Uning them with wrought-iron or steel tubes and a number
of 3.2-inch steel breech-loading field guns. The 8-inch con-
verted rifle served a temporary purpose, but the steel field-gun,
with its metal carriage (Appendix 26, Report of 1887) using bow
spring wheel brakes to restrain the recoil on firing, has with
some modifications given good service even as late as the Chinese
expedition of 1900. It is now to be replaced, however, by a
quick-firing gun with long recoil carriage, model 1902. During
this period the total amount expended by the Ordnance Depart-
ment to include the first cost of experimental guns and of those
supplied for service, amounted to somewhat less than $1,500,000.
14. Satisfied apparently by the existing armament of 1866,
in connection with the reaction from the expense of the Civil
War, it was not until 1872 that Congress made provision to
attempt a revision of the armament, when the so-called Heavy
Gun Board was appointed. The continued agitation of the sub-
ject may be inferred in noting that the Getty Board was ap-
pointed by act of Congress, approved March 3, 1881; the Gun
Foundry Board by act of Congress March 3, 1883; the Arma-
ment Board by act of July 5, 1884; and finally the Endicott
Board by act of March 3, 1885. During the same time there
^ere reports from the special Senate committee. Senator Logan
chairman, appointed August 2, 1882; the Senate select com-
luittee on ordnance and warships, Senator Hawley chairman,
appointed July 3, 1884; and a similar House committee, with
^r. Randall chairman, July 6, 1884. The Endicott Board com-
pleted its labors in 1886, submitting a scheme of sea-coast arma-
ment which was virtually adopted by Congress in the Fortifica-
tion Act of September 22, 1888, and while modified from time
to time as occasion required, is still being carried out. At this
time, after a long contest before the committees of Congress

862 ORDNANCE FOK THE LAND SERVICE.

with the advocates of different systems, including the use of
cast-iron, it had been demonstrated that the built-up forged-steel
gun recommended by the Ordnance Department could be relied
upon.

15. A somewhat extended description of the experimental
guns of this period and their trials is contained in a paper read
before the military service institution at Governor's Island,
November 26, 1887, and brief reference only will be made to
them here. The Heavy Gun Board of 1872 selected nine sys-
tems for trial, to \vit:

16. Muzzle-loading guns: (1) Dr. W. E. Woodbridge; (2)
Alonzo Hitchcock; (3) Cast-iron guns lined with wrought-iron
or steel tubes.

17. Breech-loading gxms: (1) Frederick Krupp; (2) E. A. Sut-
eliffe; (3) Nathan Thompson; (4) French and Swedish systems.

19. The Fortifications Bill of 1883, embodying the recom-
mendations of the Senate ordnance conmaittee and the Getty
Board, authorized the continuation of the conversion of 10-inch
smooth-bores into 8-inch muzzle-loading rifles, and in addition
the trial of five different systems of gun construction and two
types of breech mechanism, as follows:

20. Built-up forged steel breech-loading rifles with slotted
screw breech closure.

21. Cast-iron breech-loading rifles.

22. Combined cast-iron and steel built-up breech-loading rifles
and rifled mortars of the same system with slotted screw breech
closure.

23. Wire-wound breech-loading rifles.

24. The multi-charge gun.

25. The Mann breech mechanism.

26. The Yates breech mechanism.

27. During the period 1873 to 1882 trials were also made, at
Sandy Hook Proving Ground, with breech-loading field guns and
the Dean 3.5-inch mandreled bronze gun. The Dean gun was
produced in 1877. It was subjected to a firing test which, so far
as it went, proved the good quality of the material, but it was a
muzzle-loading gun made after a design already out of date and
gave inferior ballistic results. This system has been extensively
used in Austria, as proposed by General Uchatius, for field and

ORDNANCE FOB THB LAND 8EBVICE. -363

"lighter siege guns, but was not successful in larger calibers. The
breech-loading field guns that were tested included the Sutcliffe,
Moffat and Krupp breech mechanism. Preference was found
in these tests for the Krupp mechanism, but subsequent tests led
to the final adoption in our service of the slotted screw breech
mechanism.

28. The gun proposed by Alonzo Hitchcock ^as a 9-inch
muzzle-loading rifle designed to be made by butt welding disks
or cheeses of wrought-iron and forming a solid piece to be bored
out for the gun. After nearly three years of labor the project
was abandoned as being too difficult and costly, if not imprac-
ticable, to be fulfilled.

29. The Sutcliffe and Thompson guns were cast-iron breech-
loading rifles of 9 to 12 inches caliber. There is a general sim-
ilarity in their breech mechanism in that the block is rolled to
one side for inserting the charge. The Sutcliffe gun was fired
in all 26 rounds and the Thompson 2 rounds, when the guns were
sent to the Philadelphia Centennial Exposition and not after-
wards tested.

30. The principal feature of the multicharge gim consists in
utilizing the accelerating principle for the action of the powder
upon the projectile. This is sought to be obtained by having a
series of powder charges placed in pockets at intervals along the
bore near the breech which are ignited by the inflamed gases
of the breech charge following up the passage of the projectile
over the opening of each powder pocket into the bore. The gun
was patented by A. S. Lyman, who, in connection with J. R.
Haskell, began experimenting to test the system about 1855.
Guns of 2i-inch, 6-inch and 8-inch caliber have been tested.
The last test was that of the 8-inch steel gun, with one breech
and two auxiliary powder chambers, at Sandy Hook in 1897.
At the second round the metal between the forward chamber
and the bore was crushed in, due probably to premature ignition
of the powder charge in that pocket by passage of the gas in
advance of the projectile.

31. The principle involved in the Mann breech mechanism is
to completely separate the longitudinal from the tangential
strains due to firing a gun. This is accomplished by making the
breech of the gun separate from the gun body and connecting it
by heavy side straps with the trunnion supports on the carriage.
Several calibers of this gim were tested between 1862 and 1884.

364 ORDNANCE FOR THE LAND SERVICE.

The last occasion was a 6.5-incli gun, fired at Sandy Hook in
1884, which was burst at the 24th round.

32. The Yates breech mechanism comprises a couple of con-
cave clamps which open outwards from the breech, and when
closed embrace the breech of the gun exteriorly and support a
solid head gas check or cartridge case for sealing the escape of
gas. The 8-inch rifle was tested at Sandy Hook, 1885-6. The
gun was fired in all 312 rounds, when it was destroyed, by burst-
ing through the body.

CaM-Iron Rifles.

33. In company with the manufacture of the Hodman smooth-
bore gun the manufacture of cast-iron in this country was
brought to a high state of perfection and exhaustive attempts
were made to utilize this metal in the construction of heavy
rifled guns. Seven muzzle-loading cast-iron Rodman rifles of 8,
10 and 12 inch caliber were produced between 1861 and 1869.
When, however, the 12-inch rifle of 1868 was burst at the 27th
round, in 1871, the Ordnance Department reconmiended that no
cast-iron rifles be made for service. Efforts were not, however,
relinquished from other sources. Mr. Norman Wiard, in the
Nut Island experiments, 1873 to 1875, attempted to show the
utility of cast-iron rifles using mitten or subcaliber projectiles
but without any reasonable success. Congress subsequently re-
quired the manufacture of a 12-inch cast-iron breech-loading
rifle in 1883. It was made with 28 calibers length of bore and
gave a muzzle velocity of about 1,750 f. s. with 800-pound pro-
jectile, corresponding to 17,000 foot tons muzzle energy. After
firing 137 rounds the trials were suspended, due to the erosion
of the bore which increased rapidly toward the end of the test
and became so serious as to lead to the conclusion that it would
be unsafe to continue the firing. Again, in 1889, pursuant to
act of Congress, September 22, 1888, the South Boston Iron
Works submitted a 12-inch cast-iron rifled mortar, which burst
explosively on trial at the 20th round.

Converted Guns.

34. A quantity of serviceable muzzle-loading guns dating
from 1874, although of low power, were produced by lining the

ORDNANCE FOR THE LAND SERVICE. 365

Kodman smooth-bore guns first with coiled and welded wrought-
iron and later with steel tubes. In addition to the muzzle-load-
ing guns, several calibres of converted breech-loading guns were
constructed and tested, but without ultimate success. In this
design a jacket made of a heavy steel piece was screwed into the
breech of a smooth-bore gun and adapted to receive the Krupp
breech mechanism. The steel forgings for these alterations
were procured in England^ and chiefly due to the poor quality
of the steel the plans were abandoned. This occasioned an un-
favorable opinion regarding the use of steel in gun construction.
It was difficult for some years afterwards to convince the doubt-
ers that there was taking place a great improvement in the qual-
ity of steel for guns gained by knowledge and experience in its
manufacture. It was equally unfortimate for the Krupp breex;h
mechanism that its first application to large guns in this country
was made in this connection. The slotted screw mechanism was
applied to the subsequent experimental guns and became the
established type in our service.

Comhined VasUlron and Sled Otma wad Mortars.

35. This method of construction, authorized in 1883, was
practically forestalled by the advance in steel manufacture be-
fore the experimental pieces were completed and tested. Four
experimental types were made, including two 12-inch breech-
loading rifles and one muzzle-loading and one breech-loading
mortar. One of the rifles comprised a cast-iron body lined for
about one-half the length of the bore with a steel tube inserted
from the breech. The second rifle, in addition to the half-tube
lining, was reinforced by a double row of steel hooping on the
cast-iron body, extending from the breech to a distance forward
of the trunnion band. The two 12-inch rifled mortars were made
^th a cast-iron body reinforced by a double row of steel hooping.
Of these types the breech-loading mortar only survived for
service construction, but gave place to the all-steel mortar in the
model of 1890. The hooped guns and mortars of this type
exemplified the built-up construction by shrinkage and the
pnnciples of this method were carefully applied in making them.
The manufacture of the 12-inch hooped rifle was preceded by an
experimental construction embodying a complete section of the
gun through the powder chamber; that is, a compound cylinder

366 ORDNANCE FOR THE LAND SERVICE.

forming a counterpart of the gun section. The section of cast-
iron cylinder used in this trial was cut from the body of the gun
casting and the steel parts were of similar material to the forg-
ings made for the gun. The objects accomplished by this means
were the verification of the shrinkages calculated for the gun
and a practical test of the metals on the same scale as the gun
itself.

Cast-Steel Guns.

36. Trials of cast-steel guns were authorized by the act of
March 3, 1887, and two guns of 6-inch calibre were procured
and tested by the Navy Department. One was made of Bes-
semer steel by the Pittsburg Steel Casting Company, and the
other of open-hearth steel by the Standard Steel Casting Com-
pany. The processes of treatment of the castings were left in
the hands of the manufacturers and have not been published-
It is understood that both guns were cast solid and that the
Bessemer casting, after having been bored out, was subjected to
a process of heating combined with interior cooling to produce
a certain degree of initial tension, but the eflFect of this treatment
has riot been ascertained. The open-hearth casting is believed
to have been simply annealed. In the firing tests, which were
made first with reduced charge and then with the normal charge
for guns of this caliber, the Bessemer gun burst explosively at
the second round. The open-hearth gun was unduly enlarged
after firing twelve rounds. The strength of these guns was in-
sufficient for the powder pressure, which amounted to about
fifteen tons per square inch. Specimen tests of the Bessemer
steel gave an elastic limit varying from 43,000 to 55,000 poimds,
and of the open-hearth steel from 30,000 to 40,000 pounds. On
the usual basis of estimate for strength of gun cylinders the
open-hearth gun might have supported with safety repeated
powder pressure not exceeding about ten tons.

37. Another cast-steel gun, of 8-inch calibre, proposed by the
well known and esteemed inventor, Dr. Gatling, was procured
pursuant to act of Congress, June 6, 1896, and tested in 1899.
In the construction of this gun it was attempted to produce
initial tension by mandreling the bore while hot. On trial the
gim burst explosively at the fifteenth round, after being subjected
to pressures not exceeding about 41,000 pounds. One of the
broken fragments, comprising half of the powder chamber,

ORDNANCE FOR THE LAND SERVICE. 367

showed along the ruptured surface three cavities extending from
one to four inches into the body of the metal.

38. The Bofors Company, in Sweden, makes guns of cast-steel
of a first rate quality. It must be understood, however, that
these guns are made with tube, jacket and hoops on built-up
principle and the separate parts are nickel steel castings treated
to improve their quality in essentially the same manner as the
forgings used for other steel guns.

Wire Wound Guns.

39. The modem wire-wound gun is a worthy rival of the built-
up forged steel gun. The same principles are employed in the
construction of both. Dr. \^. E. Woodbridge presented the first
plan for a wire-wound gun in 1850, when a 2.5-inch gun was
made and tested. Three additional Woodbridge guns, of 10-
inch caliber, have also been tested. The first, proposed in 1872,
was a muzzle-loader comprising a thin steel tube wound with
wire and subsequently dipped as a whole into molten solder to
fill the interstices between the wires and consolidate the struc-
ture. This gun on trial was ruptured longitudinally after a lim-
ited number of rounds. The second 10-inch rifle was a breech-
loader made of a cast-iron body wrapped with steel wire. The
test was concluded in 1892 after firing 161 roimds. In common
with other cast-iron bores it suffered great erosion and was laid
aside. The third was a steel 10-inch breech-loading rifle with a
steel tube, a steel jacket of cold rolled bars or staves made to
form a cylinder fitting the tube over about one-half its length
from the breech and steel wire wound over the jacket. The test
of this gun, in 1894, produced four serious longitudinal cracks
in the bore of the tube after firing 23 rounds. These cracks
were generally developed in straight lines following the joints of
the enveloping stave cylinder.

40. The Brown segmental wire gun is still on trial. This con-
sists essentially of a steel tube enveloped by a segmental jacket
which is wrapped on the outside with steel wire. Two guns of
5-inch caliber have been tested; one by firing 192 rounds, in
1894, and the other 300 rounds, in 1899. In the first of these
guns the segmental jacket was formed by twelve wedge-shaped
staves forming a cylinder around the tube or liner but the latter
extended from the breech to a distance in front of the powder

368 ORDNANCE FOR THE LAND SERVICE.

chamber only, leaving the forward portion of the bore to be
formed by the segmental cylinder. In the second gun the
wedge-shaped segments are replaced by curved segments form-
ing the segmental tube and the lining tube extends throughout
the length of the bore. The latter construction is also used in
the 10-inch breech-loading rifle now at the proving groimd. It
is understood that the company has undertaken to submit for
trial a 6-inch gun designed to give a muzzle velocity of 3,500
f. s. with 100-pound projectile.

41. The Crozier 10-inch wire-wound gun consists essentially
of a cylindrical tube of the usual type wrapped throughout its
length with steel wire and an outside jacket utilized to support
the breech block and the gun in mounting. This gun was tested
by firing 275 roimds, and recommended in 1896 — ^page 320,
report of 1896 — ^by the testing board as a suitable gun for
service. The construction of a 6-inch gun on the same general
plan, which may be developed to' give a muzzle velocity of '3,600
f. fl., has been recently undertaken.

Types of Chin Cartridges,

42. The development of types of gun carriages followed that
of the guns. The first improved type of metal field carriage
with bow spring brakes was planned by Lieutenant-Colonel A. R.
Buffington, of the Ordnance Department, who installed a plant
on an economical basis for its manufacture at Springfield Ar-
mory, Massachusetts, in 1887. This type is now being replaced
by the long recoil, model of 1902.

43. The present model of 5-inch siege carriage dates from
1896. It is a wheeled carriage with the gun mounted rigidly in
the trunnion beds at a height suflicient to fire over a high para-
pet. In firing, the carriage is anchored to a pintle on the siege
platform by means of an hydraulic cylinder that absorbs the
recoil. Wedges are also placed behind the wheels. A limber
is used with the carriage for travelling and the gun is shifted to
the travelling beds to distribute the weight on the four wheels.
This type of gun and carriage, with 45-pound projectile, is to be
replaced by one of 4.7-inch caliber, w4th 60-pound projectile,
the carriage for which will provide for long recoil gun on car-
riage to give the same steadiness in firing as the model 1902 field
gun.

ORDNA^fCE FOR THE LAND SERVICE. 369

44. The present 7-ineh siege howitzer carriage, model 1899,
supersedes the first model of 1890. The height of axis of piece
in firing position is reduced in thp later model and the carriage
made about 1,000 pounds lighter, with simpler construction,
improved sights and brakes and improved means for shifting the
piece for travelling. The general arrangement of the recoil
system is the same as in the original carriage. This was de-
signed by Captain William Crozier, Ordnance Department
(Xotes on Construction of Ordnance No. 57, January 12, 1891),
introducing the novel feature at that time of recoil of piece on
carriage for mobile artillery. The piece is restrained on the
carriage by a pair of hydraulic recoil cylinders and counter recoil
springs in addition to the hydraulic cylinder anchored to the
pintle of the platform.

45. The 3.6-inch mortar carriage dating from 1890 is a simple
rigid type fired from a small wooden platform. To avoid over-
turning the carriage when fired with full charges at the lower
elevations, it is necessary to anchor it with a rope attached to
the front of the checks and fastened to some anchorage such as
a strong stake driven into the ground. The first model of 7-inch
mortar carriage dates from 1892 but has been since improved
and modified. The piece is restrained in recoil by a pair of
hydraulic cylinders and is returned to the firing position by
counter recoil springs. Clip circles are attached to the wooden
platform to permit of traversing about a fixed position by means
of a hand bar engaging in the teeth of a rack cut in the rear clip
circle. . •

46. The carriages for rapid fire guns of the sea-coast arma-
ment, which include the guns up to 6-inch caliber, are of bar-
bette and pedestal type, excepting the 6-pounder and also the
0-inch guns on disappearing carriages. A part of the 15-
pounder and 5-inch rapid fire guns have the so-called balanced
pillar mount which enables the gun to be lowered behind the
parapet when not in action and concealing it from view of the
enemy. When raised to the firing position, however, and during
the whole period of action the protection is no greater than the
ordinary barbette mounting.

47. The 6-pounder carriage is a wheeled mount designed in
1896 for the double purpose to use in a fixed position, with
anchorage behind a parapet, and for moving into the open for
firing upon landing parties. The piece has a short recoil of

870 ORDNANCE FOR THE LAND SERVICE.

gun on carriage and is unsteady in firing even when firmly an-
chored. In future construction the 6-pounder guns will be
given a pedestal mount. The 15-pounder rapid fire gun car-
riages have been materially modified since the eariy type of
1897. The balanced pillar construction then adopted for this
caliber as well as for 5-inch rapid fire guns is expensive and
found not to be ae satisfactory in operation as desired. A fixed
pedestal mount was -substituted later. The model of 1902 15-
poimder carriage embodies the latest improvements including
particularly gearing for controlling the traverse of the gun and
improved arrangements of the telescopic sight mounting.

48. The first 6-inch rapid fire guns were mounted on a Buffing-
ton-Crozier disappearing carriage, model 1898, In 1900 an
improved model of barbette carriage, with curved shield, was
designed as a type for 5-inch and 6-inch rapid fire guns on bar-
bette mounts. A number of these carriages are now under con-
struction. Later a decision was reached to restore the disap-
pearing mount to the 6-inch gun and it has been urged also to
likewise mount the 5-inch guns in future to afford the best
protection to the gun and mount and also to the firing party.

Seorcocbst Barbette Carriages.

49. A limited number of the heavier, sea-coast guns are
mounted upon open barbette carriages, designed to be ultimately
protected by front shields as in tfce case of the rapid fire guns
on barbette mounts. The types of heavy barbette carriages
include the 8-inch model 1892, 10-inch model 1893 and 12-inch
model 1892, which are similar in their main features and act
upon the " gravity return " principle. The rails of the chassis
are inclined upward to the rear and the top carriage carrying the
gun is mounted thereon with rollers. The recoil is absorbed by
hydraulic recoil cylinders and the top carriage returns to the
firing position by rolling down the inclined rails of the chassis
under the action of gravity. The loading arrangements for
these carriages, particularly the chain shot hoist, are slow and
cumbersome. The carriages are geared for ready hand power
control in traversing and elevation.

ORDNANCE FOR THE LAND SERVICE. 371

Sechcoast Casemate Ca/rria^e.

50. A 12-inch minimum port casemate carriage was, ordered
from Grusonwork, Germany, 1892, and tested at the Sandy
Hook Proving Ground (Appendix 26, Report of 1895). The
manoeuvring of this carriage is controlled in large part by
hydraulic machinery, the pressure for which is taken from an
accumulator fed by a hand pumping apparatus supplied by six
pumps served by twelve men. A steam pimap may be sub-
stituted for the hand power. The carriage fulfilled the con-
ditions of the trial and was accepted as a type suitable for service.
It can be adapted for guns of 8, 10 or 12 inch caliber. How-
ever, no call has been made for this type of mounting for guns
in our service since no armored casemate forts have been erected.

Mortar Carriages,

51. 12-ineh mortar carriages were amongst the first to be fur-
nished in quantity with the new armament. Two type carriages
were ordered in 1890, one from Easton & Anderson, of the
Raska^off spring return pattern, and the other from Whitworth,
of the Canet pattern. The Easton & Anderson carriage com-
bined hydraulic recoil cylinders with two piles of Belville springs
to return the piece to the firing position. The carriage was ac-
cepted, with modifications, as a type in 1891 and eighty-five of
them have been manufactured for service. Considerable difficulty
was met in manufacturing the Belville springs, and although this
was overcome by careful investigation and tests made with the
Emory testing machine at Watertown Arsenal, it was finally
determined to use coiled springs in place of the Belville.

52. The Canet carriage comprised hydraulic cylinders with
requisite throttling arrangement to check the recoil and a re-
cuperator holding compressed air under an initial pressure of
about 900 pounds per square inch to return the piece to the
firing position. The leakage of air from the recuperator during
the trial was comparatively slight. The manoeuvring, however,,
was slow and the general results of the trial failed to recommend
the type for service.

53. The present type of 12-inch mortar carriage, model 1896,
of which about 300 have been provided, was designed by Captain
W. B. Gordon, Ordnance Department, U. S. A. The first car-

372 ORDNANCE FOR THE LAND SERVICE.

riage was ^ade at the West Point foundry, Cold Spring, New
York, in 1895, and after trial was adopted with the minor modi-
fications shown to be necessary. In this carriage the recoil is
checked by hydraulic cylinders and the piece returned to the
firing position by springs compressed during recoil. A number
of double coil spiral springs are placed under the saddle that sup-
ports the piece at a point about one-third its length from the
pivot connection of the saddle with the racer of the carriage.
The recoil is checked by two hydraulic cylinders with return pas-
sage. The cylinders are trunnioned and hung in brackets bolted
to the upper surface of the racer, one on each side of the car-
riage. The lower ends of the cylinders are connected by an
equalizing pipe. The carriage is geared for traversing and eleva-
tion by hand power. Recent modifications of this carriage have
been found necessary to strengthen the racer and other parts to
support the strain due to the use of full charges of smokeless
powder. Improved quadrants for giving elevations are also
imder trial.

Gun Lift Carriages,

54. The first form of disappearing carriage to be adopted in
the service is the gun lift carriage, which is used in the emplace-
ments of special construction. The gun and carriage complete
are lowered in a shaft and completely concealed in the loading
position, the loading being done through a gallery opposite
which the breech of the gun stands when lowered. The power
for raising and lowering is furnished by hydraulic apparatus con-
nected with an accumulator. The carriage proper, differs but
little from the gravity return barbette carriages, excepting an
automatic arrangement of the recoil cylinders to retain the gun
in the retracted position for lowering into the shaft and loading.
The expense for installation and maintenance of the gun lift bat-
tery, the slow rate of fire of the guns and the later adoption of
the disappearing carriage has led to a very limited installation
of the gun lift carriage proper. A few altered gun lift carriages
have been moimted in ordinary barbette emplacements.

Disappearing Carriages.

55. The following disappearing carriages were made and
tested prior to or concurrent with the present adopted tyi)es of

ORDNANCE FOR THE LAND SERVICE. 873

Buffington-Crozier carriage for 6, 8, 10 and 12 inch high power
sea-coast guns, namely:

Pneumatic lO-ioch carriage, Ist pattern, produced in 1891

2nd " " " 1899.

Gordon ** *' Ijst " •* " 1892.

2nd •* " " 1894.

Howell " " •* ** 1898.

The first pneumatic carriage was furnished under specifica-
tions calling for the piece to be raised to the firing position, its
recoil controlled, the carriage traversed and the charge raised
and loaded in the gun by compressed air power. Also to pro-
vide a hand pump capable of charging the recoil cylinders so
that steam power might be dispensed with, and to provide means
for traversing, elevating and loading the piece by hand power.
This carriage fulfilled the tests prescribed for acceptance, but
its slowness in operating by hand power and general perform-
ance, coupled with the extent of steam plant required, failed
to commend it as a service type. The second carriage failed to
fulfil the tests for acceptance.

56. The first Gordon 10-inch carriage comprised essentially
a heavy bed plate supporting two side frames with a pivot plate
fastened under the bed plate; the whole resting upon a traverse
circle, for the substructure. The top carriage carrying the gun
IS attached to the inner ends of two double crank -arms on each
side, journaled through the side frames and carrying heavy
counterpoise plates outside of the frame. In recoil the crank
arms revolve 180 degrees, the gun being lowered at the same
tune that the weights are raised by revolution of the crank arms.
Two hydraulic cylinders are connected with this motion, in which
the piston rods are pushed to the front and force liquid into an
air chamber to store therein sufficient energy to raise the piece
to the firing position by air pressure on opening a valve. On
trial the carriage was found to be slow in operation but showed
enough merit to warrant the construction of an improved pat-
tern. The trial of the second carriage was fairly satisfactory,
and resulted in an opinion from the testing board that the car-
nage might be used in emplacements where a center pintle
carriage furnishing an all-around fire is desired.

57. The Howell 10-inch is a counterpoise carriage embodying
the plan in which the gun levers, upon the ends of which the
&wn is supported, revolve about a fixed axis, and the counter-

374 ORDNANCE FOR THE LAND SERVICE.

weight is supported at the other ends of the arms by interposition
of an hydraulic cylinder to reduce the shock of suddenly starting
the counterweight. On trial the carriage passed the firing tests
prescribed for acceptance; it, however, weighed nearly twice as
much as the BuflSngton-Crozier and did not equal the latter in
general performance.

BuffingtofirCrozier Disappearing Ca/rriagea.

58. In 1890 Captain William Crozier (now Chief of Ord-
nance, U.S.A.) redesigned a type of counterpoise carriage that
had been proposed some years previously by General BuflSngton
when a Captain of Ordnance. The principal features of the
carriage relating to the disappearing principle comprise a pair
of gun lever arms, with the gun supported in trunnions at the
upper end and a counterweight attached at the lower end. The
counterweight is raised when the gun recoils in firing and be-
comes operative to return the gun in battery upon releasing the
pawls. The gun lever arms are mounted on a horizontal axis,
which is journaled in the top carriage. The top carriage moves
on rollers to the rear in recoil and carries with it two hydraulic
cylinders with throttling bars giving variable openings in the
passage of the piston head so designed as to make the resistance
uniform throughout the recoil. The counterweight aflorda
some assistance in checking the recoil. During the recoil the
trunnions of the gun describe ellipses in passing to the loading
position. Sample 8 and 10 inch carriages of this type were
tested in 1894 and at once adopted for service. The Ordnance
Board, which conducted the tests, reported as to the 8-inch:
" The test of this carriage has demonstrated that it possesses in
a marked degree the properties which should pertain to a dis-
appearing carriage for high power guns. It is simple in con-
struction so that its parts and purposes are easily understood."
Also as to the 10-inch: " The advantages of this system of disap-
pearing carriage, as set forth in the report of the Board on the
8-inch carriage, are confirmed and emphasized by trial of a car-
riage adapted for a gun of much greater caliber and power, and
it is the opinion of the Board that the exhatistive test to which
this system has now been subjected demonstrates that on account
of the simple features of construction involving no valves, pumps
or other complicated appliances, and the fact that by methods
easily understood by the average artillery soldier the operations

ORDNANCE FOR THE LAND SERVICE. 875

of loading and manoeuvring are effected with remarkable ease,
certainty and rapidity, it is worthy of adoption in the service on
all sites except those where an all-round traverse is absolutely
necessary." The all-aroifnd fire model of 10-inch carriage was
produced in 1896. Improved models of the 8 and 10 inch car-
riages for limited field of fire were also produced in 1896, com-
prising a center turn-table instead of a forward turn-table and
rear traverse circle, and live chassis rollers instead of axle rol-
lers. The field of fire was increased from 150 to 170 degrees, and
a considerable saving of counterweight effected a material reduc-
tion in the effort required to haul a piece down by hand.

59. When the model 1896 12-inch carriage was tested and
the time required to fire ten rounds was 16 minutes 57.2 seconds,
it was felt that a notable advance in gun carriage construction
had been made, since it had previously been considered imprac-
ticable to mount guns of this size and power upon disappearing
carriages. An improved model of 12-inch carriage was brought
out in 1897.

60. The 6-inch disappearing carriage, model 1898, includes a
righting platform with handwheels and gearing that enable. the
gunner himself to control the direction and elevation of the gun
from his position at the telescopic sight.

61. Up to this date the carriages were arranged to be operated
in traversing and elevation and for retraction by hand power.
They are now being modified to use electric power for these oper-
ations. The latest patterns of the Buffington-Crozier disappear*
ing carriage, model of 1001, embody all of the desirable improve-
ments indicated by experience and practice had with the car-
riages up to the present time, including the application of a sys-
tem of electric control in connection with an improved method
of sighting.

62. The manufacture of sea-coast carriages proceeded rapidly
after the types were established. At the close of the fiscal year
ended June 30, 1898, 398 carriages had been delivered; in 1899

the number was increased to 605, and in 1903 to 1,000.

' «

Industrial Development Accompanying the Denumd for Modem

Ordnance.

63. The built-up forged steel gun, as is generally known, is
the type of construction used in existing land as well as navy
ordnance. Before discussing the principles of construction of

3Y6 ORDNANCE FOR THE LAND SERVICE.

the type, reference will be made to the industrial development in
this country, which may be said to be a direct consequence of the
demonstrated success of this type of gun in the early eighties.
That is to say, progress in armament Twas waiting for a good gun
and after its adoption the rest foUdwed as a matter of course.
The immense operations of the Navy Department will not, of
course, be forgotten by any one in this connection, but these
remarks must be confined to the influence of the War Depart-
ment, and more particularly the Ordnance Bureau of the War
Department, which is charged with the procurement of the land
ordnance. The first active steps for manufacture of high power
steel guns were taken in 1883, when a field gun was ordered and
the forgings procured for 8-inch and 10-inch sea-coast rifles.
The 8-inch rifle was completed in 1886, at the West Point
foundry, where I was at the time on duty as inspector of ord-
nance. Two years later, by the act of September 22, 188S, a
long period of inaction was terminated and Congress made lib-
eral appropriations for the manufacture of these guns, and each
year since has continued appropriations enabling rapid progress
to be made towards the completion of the plans for sea-coast
defense proposed by the Endicott Board in 1886, as well as for
improving the field and siege armament. The progress shown
in the procurement of sea-coast armament dates, therefore, from
1888.

64. In his annual report of 1903 the Chief of Ordnance, Gen-
eral Crozier, gives the number of guns manufactured which have
been issued or are available for issue to fortifications, together
with nearly an equal number of carriages, as follows:

6'pounder rapid-fire guns 70

15-ponDder rapid-fire gims 119

4-incli rapid-fire guns 4

4.7-incb rapid-fire ArmstroDg guns 84

5-inch rapid-fire (Ordnance Department) guns 82

6-lncli rapid-fire Armstrong guns 8

6incli rapid-fire (Ordnance Department) guns 42

8-inch breech-loading rifles 85

lO-ineh breech-loading rifles 134

12-inch breech-loading rifles 127

12-inch breech-loading mortars 371

Total i^

65. The report of the Secretary of War for the same year
states that the provision of heavy guns in sea-coast fortifications

ORDNANCE FOR THE LAND SERVICE. 877

is far advanced. The most pressing requirements in the way
of material are a further provision of rapid fire guns and the
installation of fire control apparatus. A table is given showing
that about $51,342,800 have been expended for armament, in-
cluding material in process of manufacture, and the amount
remaining to be appropriated for armament to complete the
project is $12,111,775.

66. In the inception of this work the Chief of Ordnance, Gen-
eral Benet, on April 3, 1883, addressed a circular letter (report
of 1883, page 6) to more than twenty of the principal steel
makers in the United States to ascertain existing facilities for
making gun steel forgings and inviting an expression of their
views on the subject. The letter stated the physical qualities
required in forgings and summarized the methods of manu-
facture in vogue in other countries so far as known at the time.
The replies to this letter were, with one or two exceptions,
adverse to taking up the work and introducing new plants, being
no doubt largely influenced by the uncertainty at that time at-
tending its future. The Midvale Steel Company was the first
to respond and undertake the manufacture, first of plain hoop
and then of trunnion hoop forgings, and by improving its
facilities was able in 1885 to accept an order for a complete set
of forgings, tube, jacket and heops for an 8-inch rifle. Mean-
time, in the absence of facilities in this country for making the
larger forgings, in all six gim tubes, three jackets and five
trunnion hoop forgings for different types of experimental
8-inch, 10-inch and 12-ineh guns were ordered from abroad from
Sir Joseph Whitworth & Company ani the Creussot Steel
Works. Meantime also, the Ordnance Department continued
with limited means to procure forgings for field and siege guns
and hoop forgings for sea-coast guns — the latter largely for.
experimental purposes, which produced results of great benefit
to the art of manufacture and knowledge of gun construction.
The Cambria Iron Works undertook a part of these orders but
did not materially enlarge its plant. The net result was that
when in 1888 through act of Congress the Department was able
to give larger orders for forgings, the Midvale Steel Company
and the Bethlehem Iron Works were prepared to fill them for
all the sizes of forgings demanded. Both companies received
remunerative orders and that of the Bethlehem company in-
cluded twenty-three 8-inch, twenty-three 10-inch and fifteen

378 ORDNANCE FOB THE LAND SERVICE.

12-inch sets of complete forgings for guns. Up to June 30,
1903, an approximate estimate of the amount of gun steel forg-
ings procured by the War Department since 1888 from domestic
manufacturers is 25,700 long tons, at an average cost of about
$500 per ton, or an aggregate of about $14,000,000.

67. I think it may be fairly claimed that the thorough and
systematic course of investigation of quality and manufacture
pursued by the Ordnance Department in the incipiency of gun
steel manufacture in this country was most influential in bring-
ing about a speedy and satisfactory state of manufacture and in
making it profitable as well by establishing the excellence of the
built-up forged steel gun. One of the first steps taken was to
test the efficiency of oil tempering as compared with simple an-
nealing. Two hoops of open-hearth steel, about 45 inches
interior diameter and 4 inches thick, were procured from the
Midvale company, one having been annealed, oil tempered and
again annealed, and the other simply annealed. Test specimens
were taken from the hoops and the hoops themselves were
assembled with relatively heavy shrinkage upon cast-iron cyl-
inders. The results (Xotes on the Construction of Ordnance,
No. 25) were, first, to establish the superiority of the oil
tempered and annealed metal on account of its high elastic limit
and great extensibility within that limit, and, second — ^which was
highly important — to establish a striking similarity between the
behavior of the metal in the specimen tests and that of the hoops
as a whole in the shrinkage tests. This, indeed, gave a basis for
all future shrinkage work, since it is upon the tests of detached
specimens that we must judge generally of the physical qualities
of the metal.

68. Next followed shrinkage and specimen tests of trunnion
hoops to determine the quality of metal to be obtained in these
irregular pieces. And then the actual construction of a type
sea-coast gun was preceded by building up a compound cylinder
made to be a complete counterpart of the gun in the section
through the reinforce. The purposes of these experimental
constructions were fully realized, namely: to obtain such data as
could be made available in future construction of the guns; to
determine the behavior of the elementary cylinders in combina-
tion under the theoretical shrinkages previously deduced by a
mathematical application of the formulae and thus test the
theories upon which the formulae are based; to observe the in-

ORDNANCE FOR THE LAND SERVICE. 379

dividual behavior of the elementary cylinders; and, finally, to
determine whether the shrinkages so deduced should be applied
in the future construction of the guns or to what extent they
should be modified for that construction.

69. An interesting discussion of the quality of steel most suit-
able for guns is contained in the proceedings of the Naval In-
stitute No. 40, 1887. The effort was made to show that the
steel which the Army and Navy Ordnance Bureaus were then
advocating was of a grade of so-called high steel with uncertain
strength and properties and that guns should be made of mild
steel with little or no carbon and having an ultimate strength
of from 55,000 to 65,000 pounds. The commercial advantage
claimed was that this grade of steel could then be readily manu-
factured in our own country and a high price would necessarily
be paid for the special steel which the government was demand-
ing. The discussion was convincing in showing that what was
plainly wanted was a metal possessing a high elastic limit with
good ductility and ultimate strength to withstand the pressure to
which a gun may be subjected without deformation. Events
have since shown that the constant demands of the government
for maintaining a high standard of quality and to improve it
wherever practicable have been of great benefit to the domestic
trade since the qualities of high grade steel manufactured for all
purposes in this country are not surpassed elsewhere.

70. The physical qualities prescribed for forgings for guns of
8 and 12 inch caliber and upwards are:

Elastic TcneUe EIongHtioxi i,«„,«.„M^«

limit strength after ^ !S^

ponnds pounds rnptnre w^irlSt

per sq. in. per sq. in. per cent. »^*^ *'*'"*•

Tube 46,000 86,000 17 80

Jacket 48,000 90.000 16 27

Trannion hoope 50,000 90,000 14 20

Cjllndrical hoops 68,000 93,000 14 20

71. The steel is ordinarily made by the open-hearth process
with a mediimi per cent., about 45, of carbon. Steel containing
about 3^ per cent, of nickel is now required for breech blocks and
spindles with an elastic limit of 70,000 pounds. It is also sup-
plied for field guns and the tubes and jackets of 5 and 6 inch
rapid fire guns, the prescribed limit being 65,000 pounds. The
tube and jacket of the 16-inch type gun were made of nickel
steel, since the manufacturers would not otherwise guarantee the

380 ORDNANCE FOR THE LAND SERVICE.

required elastic limit of 60,000 pounds in the metal. The cost
of nickel steel for the present acts as a bar to its more extended
use. The gain in elastic limit is about 30 per cent, over previous
standards.

72. The establishment of the army gun factory at Watervliet
Arsenal was authorized by the act of September 22, 1888. The
tools and equipments were purchased and sufficient machinery
installed in the new building to begin operations in September,
1890, or TOthin twenty-one months from the time ground was
first broken for building. Up to June 30, 1903, this factory,
besides a number of field and siege guns and 12-inch mortars,
had completed fifty-seven 8-inch, one hundred 10-inch, one
hundred and twenty-five 12-inch and one 16-inch sea-coast guns
with forgings furnished by private steel makers. The present
capacity of the shops per annum on a basis of 8 hours' work per
day is equivalent to:

Field guns, 170 ; siege guns, 10 ; siege Howitzers, 10 ; siege mortars, 11; 5-incli
rapid-fire guns, 10 ; 6-inch rapid-fire gnns, 18 ; 10-inch guns, 16 ; 13-inch
guns, 16; 12-inch mortars, 20;

or a total of about 276 pieces of the different calibers. While
a large proportion of the guns have been made at Watervliet
Arsenal a goodly number have also been made at private fac-
tories. All of the 6 and 15 pounder gims have been so procured,
as well as a number of sea-coast guns and a large part of the
mortars. In 1891 the Bethlehem Steel Company, then under
the management of Mr. Frick, made contract ^dth the Depart-
ment to furnish twenty-five 8-inch, fifty 10-inch and twenty-five
12-inch guns complete and at once proceeded to erect the fine
gun finishing plant which to-day distinguishes that company as
the most extensive, complete and best equipped private factory
for the manufacture of ordnance in the country.

73. The Watertown Arsenal is well equipped for the manu-
facture of the modem sea-coast carriages by the government.
Its capacity is, however, limited and most of the sea-coast car-
riages are procured from private establishments. The Rock
Island Arsenal is well equipped for making field and siege car-
riages.

74. The Department has no plant for the manufacture of
powders or steel projectiles. These constitute a large item of
expenditure hardly less in amount than for the guns or carriages.

ORDNANCE FOR THE LAND SERVICE. 381

All material procured from private factories is made after
specifications prepared by the Ordnance Department, and under
the inspection of its officers stationed at the place of manu-
facture. It is besides subjected to proof firing, tests for acx;ept-
ance. By this system, coupled with the work turned out from
the government shops, the Ordnance Department is admittedly,
and justly so I believe, credited with maintaining a high standard
of excellence for its products.

Experiments Made to Determine the Utility of Initial Tension
and Shrinkage in Gun Construction^ Compared with Theory,

75. When a simple, homogeneous cylinder is subjected to in-
terior pressure the surface of the bore sustains the greatest strain
tangentially, while the metal of the wall is strained to a less
degree depending upon its distance from the axis. The elastic
resistance of the cylinder is therefore measured by an interior
pressure that will expand the bore to the point where the elastic
extensibility of the metal is reached. In a cylinder having initial
tension produced by interior cooling or by the shrinkage of
cylinders one upon another, or by wire winding, in either case
producing a compression of the metal at the surface of the bore,
the same law of strains due to an applied interior pressure holds
good as in a simple cylinder, but the outer layers of metal having
been initially strained will be worked at greater tension than
in the case of a simple cy4inder and the aggregate resistance to
interior pressure will be increased. It is necessary, of course,
to observe that none of the outer layers, or indeed any of the
metal in the wall, be strained beyond the elastic limit, but the
maximum resistance of the cylinder will evidently be attained
if each of the indefinably thin concentric cylindrical lamina into
which the wall of the cylinder may be conceived to be divided be
strained simultaneously to the elastic limit. The object of
initial tension or shrinkage is therefore to derive the use of that
portion of the elastic extensibility and work in the outer parts
of the wall in resisting interior pressure which are not utilized in
the simple cylinder. On the other hand the surface of the bore
should not be initially compressed beyond the elastic limit of
the metal. This is the simple theory involved in the tangential
strength of the built-up gun. The necessary longitudinal
strength is readily secured.

384

ORDNAXCE FOR THE LAND SERVICE.

Cf9

Z
O

o
tk
o

H
U

z
<

H
0.

*8J9Ud|)inai JO 91909

ORDNANCE FOB THE LAND SERVICE. 385

to the state of rest of the system is the facility afforded for the
adjustment of shrinkages in different sections throughout the
length of the gun. In making the computations for all parts of
a complete gun, the changes in the number of layers and in
sectional dimensions from part to part render it necessary to
divide the whole length into a number of sections and compute
the resistance, shrinkages and strains for each. These sections
cannot, however, be considered wholly independent, as that
would give rise to a variety of values which would be incon-
venient to apply in practice and would cause undesirable inequal-
ities in strains in passing from one section of the gun to another.
Whilst not overlooking the primary consideration to preserve
wherever practicable the maximimi resistance and in all cases a
sufficient elastic resistance to withstand anticipated pressures,
two general rules are to be observed, namely: 1. To apply as far
as practicable, uniform values for the shrinkages in contiguous
sections where the shrinkage diameters are the same, or nearly
so. 2. To so modulate the curves of compression or contraction
of bore in the state of rest that the final curve will present a
comparatively smooth contour, conforming in general to the
curve of powder pressure and having no abrupt change of ordi-
nates. As it will in general be most necessary to preserve the
maximum resistance in the section of the powder chamber, that
section should be considered first and the values belonging to it
taken to govern others. Under these considerations it fre-
quently becomes necessary for a given section of the gun to
assume certain values, as, for instance, one or more of the
shrinkages, the contraction of bore or other conditions and com-
bine the various equations or transform them to obtain desired
results.

83. The method of division into sections is illustrated in the
figure, Fig. 99, for the 8-inoh experimental rifle finished in
1886.* The radial changes in dimensions of the cylinders due
to assemblage under shrinkage and the magnitude of the stresses,

♦ The namerouB and very short hoops used in this gun were soon changed in
f Qtnre constructions when it was found that the eteel makers could readily pro-
duce longer hoops. In the 12-inch rifle, model 1900, which is 42 feet in length,
there are but two lioops in the " C " row extending from the front end of the
Jacket to the muzzle of the gun, each of which is about 138 inches long. Similarly
the ** D*' row is formed of a single hoop 108 inches long, and the *' A " row of
two hoops only, 91, and 109 inches lonp: ; the single hoop over the breech in the
** B" row is 108 inches long, and the trunnion hoop 28 inches.

ORDNANCE FOR THE LAND SERVICE.

..Hi S \

in a sectional view, are illustrated- in the two figures, Fig. 100,
for 12-inch rifle.

84. Practice has shown a remarkable conformity of results
obtained in the actual manufacture of guns wdth those indicated
by the formulas applied in their construction. One of the earlier
experiments consisted in building a section of gun with 8-inch
bore, composed of four cylinders, shrunk one upon the other.
The displacements of diameters and lengths caused by each

ORDNANCE FOB THE LAND SERVICE.

387

shrinkage were carefully measured and showed in every in-
stance close agreement Avith values deduced by the formulae.
For example, the anticipated total compression of the 8-inch
bore was 0.0129 inch, and its measured compression 0.0131 inch.
The anticipated extension of the exterior of the fourth layer or
outer hoop (diameter, 31.5 inches) was 0.0285 inch and its meas-

ORDNANCE FOR THE LAND SERVICE.

ured extension 0.0276 inch. These culminating measurements
of the series thus showed a difference of less than 2 per cent,
between the actual and anticipated results.

85. In the 8-inch experimental rifle the anticipated compres-
sion of the bore in the section of the powder chamber was .0154
and the measured compression .0156 inch. This gun was first
made without chase hooping. After firing 24 rounds the bore
of the tube at some 15 inches from the muzzle was found to be

MEAtUREO BTREME8 IN SECTION OF HOLLOW POitOINQ TREATED BY INTERIOR COOUNO
AND COMPUTED ELASTIC RESISTANCE TO INTERIOR PRESSURE.

Am. abd JA» Cb.^. t.

Fig. 101.

enlarged .006 of an inch. It was then decided to return the
gim to the factory and extend the hooping to the muzzle. On
turning off the chase metal for hooping the bore contracted when
metal was turned off showing that there existed a zone of com-
pressed metal at the exterior of the tube damaging to its strength
and due to the imperfect treatment at that time in vogue at the
Whitworth works where the tube was procured. The chase
hooping fully re-established the strength of the gim, which has
been fired in all 335 rounds and is still in serviceable condition

ORDNANCE FOB THE LAND SERVICE.

PLATB S. INITIAL TENSION IN HOLLOW CYLINDER.

m g

0-21600

Axts of radii

jKr»i«,Jt.

Fio. 102.

excepting erosion of the bore, which detracts from its accuracy.

86. Fig. 101, with figures showing the measured stresses in a

forging representing sections of a field gun which was treated by

interior cooling for experiment, is taken from a paper read to

888

ORDNANCE FOR THE LAND SERVICE.

ured extension 0.0276 inch. These culminating measurements
of the series thus showed a difference of less than 2 per cent,
between the actual and anticipated results.

MEASURED BTREME8 IN SECTION OF HOLLOW POitOINQ TREATED BY INTERIOR COOUNO
AND COMPUTED ELASTIC RESISTANCE TO INTERIOR PRESSURE.
SEFONE 4|»nEAUHa

^4N.ila«.f .Vtrfa Gt.,:^.f^^

Fig. 101.

enlarged .006 of an inch. It was then decided to return the
gun to the factory and extend the hooping to the muzzle. On
turning off the chase metal for hooping the bore contracted when
metal was turned off showing that there existed a zone of com-
pressed metal at the exterior of the tube damaging to its strength
and due to the imperfect treatment at that time in vogue at the
Whitworth works where the tube was procured. The chase
hooping fully re-established the strength of the gun, which has
been fired in all 335 rounds and is still in serviceable condition

ORDNANCE FOB THE LAND SERVICE.

PLATE $. INITIAL TENSION IN HOLLOW CYLINDER.

a t.

0-21600

Axis of radii

jnrM*^-

Fig. 102.

excepting erosion of the bore, which detracts from its accuracy.

86. Fig. 101, with figures showing the measured stresses in a

forging representing sections of a field gun which was treated by

interior cooling for experiment, is taken from a paper read to

890 ORDNANCE FOR THE LAND SERVICE.

the Philosophical Society of Washington, D. C, May 11, 1895,
and printed in Bulletin Vol. XIII, pp. 27-102. The wall of the
forging was cut into thin concentric rings on four several sec-
tions, and the changes measured on the diameter of each ring
due to its release from the forging, these changes giving a
measure of the strains induced by the treatment.

87. An analysis of the laws governing the resistance of a cyl-
inder with a properiy regulated degree of initial tension pro-
duced by interior cooling shows that a thickness of 0.65 of a
caliber, neariy, will give the strongest tube that can be made
from a given weight of metal of given quality. For this thick-
ness in calibers only will all the fibres of the metal throughout
the thickness of the wall reach the elastic limit from extension
simultaneously under the action of interior pressure. This
particular case is illustrated in Fig. 102. The interior cooling
being supposed perfectly conducted, the metal at the surface of
the bore is compressed to its elastic limit, p = 45,000 pounds
assumed. The curve of initial tension is a continuous one ex-
tending through the neutral point of stress " b " to a tension of
21,600 at " c," on the outer surface of the cylinder. If now this
cylinder be subjected to an interior pressure the stress upon the
metal throughout the thickness of the wall will be developed in
resisting the pressure until an interior pressure of P = 49,950
pounds per square inch is reached. At this stage of the press-
ure the stress curve, which was originally represented by the
initial tension curve, a, b, c, is now formed in the straight line,
d, e; that is to say, the stress upon the metal throughout the
thickness of the wall is at every point equal to the elastic limit of
the metal. The elastic tangential resistance of the cylinder to
interior pressure is 1.11 times the elastic limit of the metal to be
derived from tests of specimens although the wall of the cylinder
is only 0.6473 of a caliber thick or actually 8X0.6473 = 5.18
inches. The same relations would, of course, hold good for a
tube with any given diameter of bore and metal of given elastic
limit.

88. The following conclusions may here be quoted from the
bulletin of the Philosophical Society previously cited.

For cylinders of greater thickness than 0.65 caliber, a state of uniform strain
in the wall will be reached in action and passed before the elastic limit of the
metal is attained, and with increasing pressure this limit will be fully reached
only at the surface of the bore, thus determining the limit of pressure. For such

ORDNANCE FOR THE LAND SERVICE. 391

cjlinders tbe best condition of resistance will be obtained by utilizing tbe fall
limit of compression of tbe metal in tbe initial tension.

Bat for cylinders of less tbickness tban 0.65 caliber, a state of aniform strain
in action eqaal to tbe elastic limit of tbe metal can be attained witb a compression
of bore less tban tbe limit p. Tbe tbinner tbe cylinder tbe less sbould be tbe
initial compression imposed. It follows tbat tbe possible maximum resistance
of sucb cylinders will be obtained by adjusting tbe initial compression witbin
limits. If tbe full limit of initial compression were given, tbe elastic limit of tbe
metal would be reacbed in action at tbe exterior of tbe cylinder sooner tban at
tbe bore.

As a consequence, also, of tbe preceding, tbe resistance of tbe cylinders of
less tbickness tban 0.65 caliber, treated by interior cooling, should be directly
proportional to tbe tbickness. Tbis treatment gives tbe means of imparting tbe
greatest resistance so far known to sucb cylinders.

89. A 3-inch and a 5-inch gun made of single forgings
strengthened by interior cooling have been constructed and tested
extensively with very satisfactory results. A number of the
tubes for 5-inch rapid fire guns have also been made in this way.
The outcome of the process in these tubes has not been en-
couraging for the continuation of this method in view of the
trouble experienced and the number of re-treatments necessary
to obtain the degree of initial tension desired. Care and ex-
perience in manufacture should be able to control this method
and apply it advantageously at least to guns of small and medium
caliber. The difficulty to bo anticipated with guns of large
caliber is in procuring a forging for the gun body in one piece
of uniform quality. This is, however, also a question of experi-
ence and advancement in knowledge of the art, and quite within
the limits of future possibility.

The Modem Field Gun (Figs. 103 cmd lOlt, *).

90. The chief object of improvement in the modern field gun
is to obtain the greatest possible rapidity of aimed fire, and this
has been rendered possible by the introduction of smokeless
powder, since with the front of the battery covered by a heavy
cloud of smoke, as formerly, such rapidity of fire as is now
attained would not have been possible. Improvements for this
purpose have been made in the breech mechanism of the gun, in

* It is regretted tbat photograpbs of tbe model 1902 field gun and limber are
not available at tbis time. Tbese plates sbow tbe material tested in 1001-2 wblch
resembles the adopted type.

392

ORDNANCE FOR THE LAND SERVICE.

»
O

ORDNANCE FOR THE LAND SERVICE.

393

the ammunition^ and especially in the carriage. Improvements
in the mechanism consist principally in a single motion in open-
ing and closing the block, safety devices to prevent firing until
the block is fully closed and locked, and eccentric position of the
firing pin in the block to prevent premature discharge of the
primer i^ the fixed ammunition when the block is closed.

91. Improvements in the carriage constitute essentially a con-

-u

MM ^L /

L A^^J C^m r^K^KKli ' fll

r 1 ■ ^^Hi

Fig. 104. — Limber for Field Gun.

struction which admits of the gim being fired without recoil of
the carriage on the ground, without jump of the wheels, with
practically no disturbance of the aim, together with such an
attachment of sights that the sighting can be continuously main-
tained while the gun is being loaded. Improvements in the am-
munition are a change from separate loading of the projectile
and powder charge to fixed ammunition.

92. In the pattern of equipment now to be replaced the car-

394 ORDNANCE FOB THE LAND SERVICE.

riage recoiled at each round and had to be run to the front, re-
pointed by shifting the trail, and finally the cannoneers must get
clear of the carriage to allow the gun to be fired. All of these
were time-consuming elements.

93. Three experimental carriages embodying recoil of the gun
on the carriage have been tried precedent to the present adopted
type, model 1902. First the model of 1898, with the short
recoil of gun on carriage, and having an axle traverse to give
motion to the gun in azimuth without shifting the trail; next,
another short recoil carriage made in 1900. The distinctive
features of this were hydraulic recoil cylinders with counter
recoil springs in the cylinder, azimuth motion of the gun on the
carriage without shifting the trail, provided by a turn-table, and
a folding spade at the rear end of the trail. Next came the long
recoil carriage allowing a recoil of 44 inches to the gim on the
carriage, made wdth two hydraulic recoil check cylinders and
return springs in the cylinder. The two last-named types were
entered in the exhaustive tests of field material made in 1901
and 1902, together with several types of foreign and domestic
manufacture. The results of those tests led to a combination
of good points found, which have been introduced in the model
of 1902.

94. The gun is made of nickel steel built up with clips that
slide in and secure it to the guide rails of the cradle. A lug
under the breech connects the gun with the recoil cylinder.
The breech mechanism is of the interrupted screw type and
opens with a single motion. The block is bored out to receive
the firing mechanism which is placed eccentrically in the block
so that in closing, the firing pin will not be brought opposite the
primer in the cartridge case until the block is rotated and locked.
A safety device prevents firing until the breech is closed. The
gun may be fired by lanyard from the rear, but is habitually
fired by the cannoneer seated on the right through a firing shaft
attached to the carriage. This shaft is so arranged that it
cannot be operated until the gun has returned to the safe position
in the battery. The advantages of this breech mechanism are
rapidity of fire, great power of extraction and ejection of cart-
ridge case, ease of loading in that the cartridge is not required
to be pushed into place by hand as in sliding block mechanisms,
but the last part of the motion is given by the closing of the
block. The mechanism is well arranged for protection of the

ORDNANCE FOB THE LAND SERVICE. 395

parts from dust and injury, especially blowbacks, is simple and
no tools are required for dismantling it.

95. The projectiles used are shrapnel and high explosive shell
weighing 15 pounds. The gun gives a muzzle velocity of 1,700
foot-seconds, and a range of 6,250 yards with 15 degrees eleva-
tion. A somewhat greater range can be obtained by sinking the
trail. The maximum range is about 7,500 yards. The accuracy
of the gun is excellent at 6,000 yards range, and compares favor-
ably at this range with the 3.2-inch gun, which it replaces, at
4,000 yards range.

The carriage comprises the lower carriage, consisting of wheels
axle and trail, together with a rocker and cradle which contains
the recoil system and in which the gun slides. The cradle rests
on the rocker and is controlled by the elevating gear. The
rocker forms a table also in which the cradle can be moved in
azimuth 4 degrees on either side of the centre to change the
direction of the gun by that amount without moving the trail.
The spade at the end of the trail is fixed in position and has
wide flanges to prevent sinking* in the ground. The recoil equip-
ment comprises the cylinder, piston rod, counter recoil buffer
and springs. The cylinder lies in the cradle and is surrounded
by the springs, and its rear end is attached to the breech of the
gun. The recoil springs are made of thin steel riband rect-
angular in shape coiled on edge. The column of springs com-
prises three similar coils assembled with tension sufficient to
return the piece when fired at 15 degrees elevation. The piston
rod is bored out for the throttling bar and the interior of the
cylinder has three ribs of variable profile to control the recoil by
openings through which the liquid in the cylinder passes. The
variable openings in the cylinder are calculated so as to make the
resistance which the liquid offers plus the resistance of the
springs such that the wheels will not jump from the ground
when the piece is fired at zero elevation. This is accomplished
by making the gravity moment of the system about an axis
through the point of support of the trail greater than the sum
of the moments of the piston pull and the spring resistance about
the same axis. The recoil of the gun on the carriage is 48
inches. The counter recoil rod with one end secured to the cyl-
inder has its free end centred and supported at all times in the
bore of the piston rod. It regulates the velocity of return of
the piece throughout the whole length of counter recoil, thus

396 ORDNANCE FOR THE LAND SERVICE.

obviating the sudden strain and shock in designs where the
counter recoil is unrestrained until the piece is nearly in battery.
As a result of this arrangement the gun can be fired without even
as much motion as would dislodge a piece of money placed upon
the wheel.

96. The shield is made of hardened steel .2 of an inch thick,
in three parts. A small portion of the top is made to fold down
so that it will not project above the wheels while travelling.
With this arrangement the carriage might be overturned without
injury. The shields are tested for acceptance by firing thirty
caliber steel-jacketed bullets with muzzle velocity of 2,300 f. s.
at a range of 100 yards. The bullet must not penetrate. The
shield thus offers protection against small arm fire and shrapnel
bullets.

97. The road brake has a double lever, one in front and on«
in rear of the shield. The first is for the use of the cannoneer
seated on the axle seat as a road brake, and the other is placed
where it can be readily operated in firing the gun to prevent
movement of the carriage especially when fired from a sloping
platform. The brake blocks bear upon the front of the wheel
in the firing position where they are out of the way of the firing
party. When limbered up in travelling on the road this places
the blocks behind the wheels and no mud is collected on the
blocks.

98. The sights comprise a panoramic telescopic sight on the
left of the piece, and a range quadrant on the right. In using
the range quadrant the gimner on the left may devote his entire
attention to giving direction, while another cannoneer can give
elevation at the range quadrant. The piece is, however, usually
aimed by the gunner seated on the left, who has at hand the
elevating and traversing wheels for training the gun. The
sights are attached to the cradle, and since this does not move in
firing the piece the sighting may be continuous. The panoramic
sight enables the gun to be aimed by directing the sight upon
any object off the line of fire, which is especially useful when
any direct target made by the enemy is indistinct.

99. It is found in order to secure steadiness of the carriage
that the time consumed in recoil and counter recoil must be made
about two seconds. The operations of opening, loading and
closing the block, and firing, consume about one second only, so
that practically therefore the maximum rapidity of unaimed fire

ORDNANCE FOB THE LAND SERVICE. 397

from a piece of this description is about three seconds per round,
or twenty rounds per minute. After the trail is set a rapidity
of 10 to 12 aimed rounds per minute can be maintained for
some time. As compared with this, the 3.2-inch gun would fire
not more than about two rounds per minute.

100. The weight of the model 1902 carriage, exclusive of gun,
is 1,308 pounds, as compared with 1,321 pounds for the carriage
of the 3.2-inch rifle, which is remarkable in view of the additional
parts in the new carriage, but shows how much the essential
parts in the old carriage which are still retained in the new could
be lightened and yet retain sufficient strength for service by
introducing the plan of recoil of the gun upon the carriage.

101. Four rounds of ammunition are carried in tubes under
the axle seats to be ready for immediate use, and 36 rounds are
carried in the limber. Fig. 104. The gun and limber complete
with ammunition weighs 3,800 pounds, giving a load of 633
pounds per horse for each of the six horses.

102. The organization of batteries with these gims will com-
prise four guns with carriages and limbers, and 12 caissons and
limbers. The caissons carry 70 rounds each, and their limbers
36 rounds, making a total of 1,432 rounds for the battery, or
358 rounds for each gun.

Types of Guns^ Mounts and Breech Mechanisms.

103. In all the more recent type of guns a continuous move-
ment of breech mechanism in opening and closing the block is
substituted for the separate operations, one to unscrew the block
and the other to withdraw it, formerly required. The principal
parts of the breech mechanism are the spindle with mushroom
shaped head, plastic obturator and steel split rings to seal the
escape of gas; the block with its threaded and smooth sectors
arranged to require a small turning movement for locking; the
parts that operate to turn and withdraw the block; the carrier
(either tray or ring) that supports the block when open; the
safety attachments that prevent firing until the block is closed
and locked, and the firing mechanism.

104. Fig. 105 illustrates the more modern form of breech
mechanism in the 6-inch rapid fire gun, model 1900. This gun
gives 3,000 f. s. muzzle velocity with 100-pound projectile, is
£tted with the Stockett breech mechanism and a firing mechanism

398 ORDNANCE FOR THE LAND SERVICE.

for combination electric-friction primer; and also a loading tray
with automatic movement which rises to support the insertion of
the charge and projectile when the breech is opened, and falls
to clear the way for the block as the latter is closed. The
capacity of the powder chamber in this gun is 2,114 cubic inches,
as compared with 1,278 cubic inches in the model 1897 gun,
which was designed for 2,600 f. s. muzzle velocity.

The 6-inch Bofors gun, Fig. 106, has a standard muzzle velocity
of 2,624 f. s. with 100-poimd projectile. The most interesting
feature of the gim is the breech mechanism, which is opened and

Fig. 105. — Breech Mechanism, 6-inch R. F. Gun. Model of 1900.

closed by a single movement and may be operated automatically
or by hand. The projectile recess is conical in shape, with a
large and convenient opening to the rear for loading, and is fitted
with a loading tray that works automatically. The mounting is
provided with two sets of sights, one on either side, so that one
man can control the direction and another the sighting in eleva-
tion for rapid firing. In trials for rapidity at wll 10 rounds
were fired in 103 seconds with the breech operated by hand, and
10 rounds in 94.6 seconds with the breech operated automatic-
ally. The automatic arrangement is complete for both opening
and closing the block. The opening is effected by the recoil of
the gun in its cradle, which at the same time compresses a spring

ORDNANCE FOR THE LAND SERVICE.

o
B

1-4

400 OUDXANCE FOR THE LAND SEBVICE.

and a catch holds the block open. On releasing this catch the
spring automatically closes the block. This mechanism has
worked well throughout. The automatic feature gives only a
little increase in rapidity of fire, but has the advantage of saving
space for loading and dispenses with the services of one man.
Among the many labor and time saving devices now attempted
to be introduced for the service of guns, which are frequently*
objectionable because of delicacy and increased complications,
the Bofors automatic breech opening and closing device appears
exceptional for distinctive merit and certainty of operation. To
see it in operation one experiences the same sense of relief, to
a degree, as in the use of fixed ammunition instead of separate
loading in the rajud fire field gims. The cannoneers are saved
by so much from violent and rapid exertion, the loading pro-
gresses more smoothly and withal with increased rapidity.

105. This gun has been fired 386 rounds with a number of
pressures exceeding 50,000 pounds, and in one case reaching
nearly 70^000 pounds, derived from charges purposely increased
to test the mechanism or the behavior of various lots of smoke-
less powder to* determine if there existed a critical point of
pressure.

106. Designs have been made, and work is in progress to apply
plans of automatically opening and closing the breech block to
oth?r rapid fire guns and to guns mounted on disappearing car-
riages.

107. The 16-inch gun, Fig. 107, is at present mounted upon
the proof carriage. It has been recently decided to make for
this gun a BuflSngton-Crozier disappearing carriage, and to mount
it at some place on the coast not yet designated. The piece
has been fired eight times, using the first sample lot of DuPont
nitrocellulose smokeless powder made for it, with charges vary-
ing in weight from 450 to 640 pounds and a standard weight of
projectile of 2,400 pounds. Round Xo. 6 with 640-pound charge
gave 2,345 f. s. muzzle velocity with 38,545 pounds pressure and
muzzle energy 91,500 foot tons. Colonel Ingalls' estimate of
the maximum range of this gim, provided it should be elevated
for firing at an angle of nearly 42 degrees, is 20.9 miles, and the
maximum ordinate of the trajectory 30,516 feet, or above 5J
miles. The battle range of the gun with an elevation of 10
degrees, which may be given on the disappearing carriage, should
be about 16,000 yards, or over 9 miles.

ORDXANCE FOR THE LAND SERVICE.

401

402 ORDNANCE FOR THE LAND SERVICE.

108. The proof firing was attended with entire success. At
the fourth round, with a full charge, a velocity of 2,317 f. s. was
obtained with 36,700 pounds pressure. The gun was designed
and the powder sample made to give a velocity of 2,300 f. s. with
not exceeding 38,000 pounds pressure. The results are remark-
able when it is considered that the charge of smokeless powder
exceeded largely any that had been heretofore fired from a gun.

109. The influence of caliber in producing power is shown in
that the muzzle energy of the 16-inch gun is 2.4 times that of
the 12-inch rifle, model 1895, designed for the same muzzle
velocity. The 12-inch rifle, model 1900, with 2,560 f. s. muzzle
velocity, gives a muzzle energy of 45,420 tons, or about one-half
that of the 16-inch gun. In actual firings the 12-inch rifle,
model 1900, has given a range of 13,360 yards, or 7^ miles with
10 degrees elevation.

110. Improvements in the appliances for giving elevation to
12-inch mortars comprise several designs of quadrants perma-
nently fixed to the piece within view of the cannoneer at the ele-
vating wheel, and also a range scale on a wheel attached to the
elevating axle. .The method formerly used, the quadrant placed
on flats on the breech of piece, and requiring the gunner to give
orders to the cannoneer at the elevating wheel to raise or lower
the piece so as to adjust the level of the quadrant was slow and
cumbersome. With the new arrangements the elevation is given
by bringing the pointer to a fixed point on an arc, and does not
depend upon the adjustment of the level bubble. Fig. 108 shows
the 12-inch mortar carriage, model 1896, with improved quad-
rant in place over the right rimbase.

Powders,

mi Smokeless powder only is now manufactured for the
service of all guns, and the black or sphere-hexagonal and brown
prismatic powders are discontinued except to use up material on
hand for target practice. However, a quantity of fine grain
black powder is used as a priming charge in the smokeless
powder cartridge. The present practice makes this priming
charge 2.6 of one per cent, of the smokeless powder charge, and
is the cause of producing a considerable amount of smoke in the
discharge, particularly with the large guns.

ORDNANCE FOR THE LAND SERVICE.

403

404 ORDNANCE FOR THE LAND SERVICE.

112. The form of grain now in use is a multi-perforated cyl-
inder which when burned in the open air gives an increasing sur-
face of combustion until the circles meet in burning away,
leaving 12 to 16 per cent, of the grain in splinters to be con-
sumed with a decreasing surface of combustion. Other forms
in use elsewhere are the strip and the hollow cylinder or tubular
gram. The latter burns away with a uniform surface of com-
bustion, and this, combined with its facility for ease of loading
when made up into cartridges, appears* to make it the most desir-
able form to use.

113. Nitrocellulose powder has generally displaced that con-
taining nitroglycerine. The heat produced is less and larger
charges are required, but this is offset by the decreased erosion
found with the nitrocellulose composition.

114. The advantage of smokeless over the brown powder,
aside from the question of smoke, lies in the fact that it is all
converted into gas, while the brown and black powders giv/j
somewhat more than one-half the products of combustion in
solid residue. All the properties of smokeless powder are not
as yet known, and there is a large field for investigation. Kecent
experiments have shown that certain lots of powder at least have
a critical point of pressure. That is to say, the pressures will
be found to increase regularly according to a certain law "with
the charge up to a certain point, beyond which, if the charge is
increased, very high and abnormal pressures may be encountered.
These experiments have been purposely made with charges
greater than the 8er\ace requirements in order to investigate the
law of critical pressures.

115. Smokeless powder requires a large air spacing, and the
powder chambers of guns to give high velocities in order to hold
the large charges required for this must be made correspondingly
large. There is one size of powder grain best adapted to each
caliber of gun, and nothing is gained by increasing the size or
in effect using a slower burning powder for that gun. A larger
charge of the slower burning powder would be required to give
the same velocity, vdth loss of efficiency. For an equal weight
of charge the gas vnW have a longer path to work over and will
impart more velocity to the projectile the nearer to the breech
the powder is consumed in the bore. The limit for minimum
size of grain and quickness of burning is fixed by the pressure
that the gun will support in the powder chamber, and the limit

ORDNANCE FOR THE LAND SERVICE. 405

for maximum size of grain may roughly be evidenced by the
absence of unconsumed powder on firing.

Projectiles.

116. The present approved form of shrapnel body comprises
a steel case with bursting charge in the base, and the case made
strong enough to eject the balls after the manner of a small gun
without rupture when the bursting charge is ignited. The base
charge gives an increase of about 200 f. s. velocity to the balls
over that of the shrapnel at the point of burst. The cone of
dispersion is about 9 or 10 degrees. In the shrapnel made with
the bursting charge in the head the cone of dispersion is some-
what greater. The velocity of the balls, however, is somewhat
retarded by the burst and the effective range is materially less
than that of the shrapnel with bursting charge in the base.

117. The common form of armor-piercing shot and shell have
capacity to perforate about one caliber of hard-faced steel armor
for the shot, and one-half caliber for the shell. Both are to be
used in our service charged with high explosive. The 12-inch
shot, for example, holds 23 pounds and the 12-inch shell 68
pounds of explosive. They are made of special quality of steel
containing chromium or manganese, and carefully hardened and
tempered. The processes of manufacture are held as trade
secrets by the different makers.*

118. Where armor penetration is not required, as in high ex-
plosive shell for the field and siege guns, the walls of the shell
are made as thin as consistent with necessary strength to with-
stand the crushing effort accompanying discharge from the gun
so that a maximum charge of explosive may be carried.

119. Recently an armor-piercing projectile, with thickness of
wall intermediate between those of the present shot and shell,
has been produced by the Wheeler Sterling Company of Pitts-
burg. This projectile, by virtue of the improvements in
manufacture, has the same armor-piercing power as the thick-

♦ Figs. 109 and 110 show the characteristic bursting eflfect of the two high
explosive compounds that have been adopted in our service. The projectiles
exploded were 12-inch armor piercing shell, which weigh, empty, about 950
pounds, and contain about G7 pounds of Maximite, or 57 pounds of explosive 2>,
with Frankford Arsenal detonating fuze. The shell were buried ten feet in sand
for explosion, 'and the fragments afterwards recovered as shown.

406

ORDNANCE FOR THE LAND SERVICE.

Fig. 109. — Fr.\gmentation, 12-inch A. P. Shell Charged with Maximite.

walled shot and costs about the same as the shot but nearly twice
as much as the present shell. It is designed to replace the two
present projectiles and give a single armor-piercing projectile of
good bursting charge capacity for each gun. The base of the
projectile is cut away to allow the band to be stripped at the rear
in perforating armor and thus diminish the resistance to penetra-
tion due ordinarily to the enlargement of the diameter of the pro-
jectile at the band. The cap is assembled by means of a wire
forced into a groove made partly in the projectile and partly in
the cap after the latter is in place. The cap of the present pro-
jectile is forced and hammered into a groove on the projectile.

120. A striking example of the effect of low temperature on
steel plates was seen in firings a few days since at the Proving

Fig. 110. — Fragmentation of 12-inch A. P. Shell Charged wrm Explosive D.

ORDNANCE FOR THE LAND SERVICE. 407

Ground. 8-inch A. P. shell were fired at a tempered, nickel steel
plate 9 feet x 7^ feet x 3 inches thick, backed by oak timbers.
The weather had remained at a temperature approaching zero for
several days. The projectiles were fired to strike the plate with
angles varying from 20 to 40 degrees between the face of the plate
and the line of fire, to observe the effect at these small angles on
the plate and also to test the relative efficiency of capped and un-
capped projectiles.. At each round the cold plate was badly
cracked, and markedly more so than a similar plate previously
tested in warmer weather. Thereupon two pieces from the same
plate were heated to a temperature of 100 degrees or more and
attacked by similar projectiles. In both cases, although the plates
(pieces) were smaller than before, each piece was dished and bent
by the impact of the shell but there was no cracking. These tests
have so far negatived the report received from English sources
that the uncapped projectile is equal in proficiency to the capped
projectile at very oblique angles of impact. On the contrary it
appears that the capped projectile is decidedly superior.

Accuracy and Endurance of Guns,

121. It is the practice of the Ordnance Department to subject
each gun to an exhaustive test for endurance before adopting it
as a type for construction and introduction into service. All the
heavy sea-coast guns are hooped to the muzzle.

122. The estimated tangential resistance of the 8, 10 and 12
inch guns, that is, the pressure per square inch which can be sup-
ported in the powder chamber without exceeding the elastic limit
of the gun, is about 52,000 pounds, and that at the muzzle about
22,000 pounds.

123. If 'these pressures are exceeded a permanent set of the
bore may be produced, but rupture is prevented by the ductility of
the metal and the guns can support higher pressures. The stand-
ard limit of powder pressure for the charge is 38,000 pounds per
square inch, or about 73 per cent, of the elastic resistance of the
gun.

124. The actual limit of safe pressure for these gims is probably
about 70,000 pounds per square inch. But one case of ex-
plosive rupture of a gun of this type has occurred at the Sandy
Hook Proving Ground aside from experiments with explosives.
This was a 10-inch gun in March, 1899, using an experimental

408

ORDNANCE FOR THE LAND SERVICE.

smokeless powder which proved to be of brittle nature. The
breech of the gun was destroyed. The two pressure gauges regis-
tered 78,000 and 79,000 pounds per square inch.

125. On another occasion the breech block of a 5-inch gun was
stripped under an estimated pressure of 70,000 pounds. 6-inch
K. F. gjm No. 1, model 1897, sustained a pressure of 86,000
pounds as registered by the gauge, which resulted in wedging the
breech block. Various calibers of the guns, including the 12-inch,
have been repeatedly subjected to pressures of 50,000 pounds,
and in exceptional cases considerably higher pressures without
serious effect.

126. The type guns of heavy caliber have sustained the fol-
lowing number of rounds for endurance, namely:

Experimental 8-incb gun No. 1, West Point foundry 335 rounds

'* 8-inch gun No. 1, Watervliet Arsenal type. . . .890 rounds

** lO-inch gun No. 1, Watervliet Arsenal type. . .292 rounds

** 12-incli gun No. 1, Watervliet Arsenal type. . .265 rounds

and the types of 12-inch mortars, models 1886 and 1890, respec-
tively, 398 and 399 rounds. Barring erosion, all the guns are
intact except the second 8-inch, in which longitudinal cracks ap-
peared in the tube near the muzzle after 388 rounds.

127. Plans have been prepared for relining the 10-inch gim,
No. 1, by a half tube inserted from the breech, which will remove
the eroded part of the bore and should render the gun fit for a
duplication probably of its first record.

128. The following accuracy targets, see also Fig. Ill, form
part of the record of the guns named:

Rounds
nambered.

Range.

DeviatfoD from
centre of impact.

, GON.

Ff.
1.94
1.44
1.84
1.64

Ft.
0.88
2.61
0.64
1.68

Rectangle

Containing

BhoU.

10 inch B. L. R., W. A., No. 1 (8) 97 to 104

** *• *♦ •' (5) 283 to 287

10.lncb B. L. R. Crozier wire wound (5) 271 to 275
12.inch B. L. R., W. A., No. 1. . . . (5) 216 to 2>0

Yd^.

1,760

8.000

1.760

3,000

Ft.
1.28
2.98
1.95
2.85

Ft.
8.4 X 4
8.2 x4.4
3. X 7 2
6 8 X 5.2

129. It will be observed that three of the targets were made,
beginning with round 216 in the 12-inch gun, round 271 in the

ORDKAKCE FOB THE LAND SEBTICE.

409

ld'B.L. Rllto, Cro>lOTiny«
Wound. IUas«17t0 7<U.

^ M.H.D. 044 **
( UJ). IM ••

10*B.L.IUfl«,W.4.K<>.L
IUBff»17W7dt.

p»

!•
8 r

imL

(

8.4

— -

«—

— *

•.•*!

t?3

'rf

!"•

1
jr."

e.8

ft.

8t f

I'-
J

U^BXJUfbjW.A. No. 1.
Babso SOOO jds.

^ M.V.D. 144 Ft.

.1.68 "

«J5 -

I M.V.D. 1
< M.U.D. 1
( M.D. J

kTB-L. Rine. W. A. No.
Rmafe 3000 yda.

M.V.D. 1.44 Pi.
M.H.D. 841 "
MJ>. .8.98 '•

Fig. 111.— Accuracy Targets, 10- and 12-inch B. L. Rifles.

10-inch Crozier wire wound gun and round 283 in the 10-ineh
forged steel gun. Before these firings the guns had become so
eroded that the original bands on the projectiles were too small to
seat the projectile or take the rifling properly. New and enlarged
bands were used in each case with the result of restoring the
accuracy of the guns practically to the original standard.

The Disappearing Carriage.

130. The utility of the disappearing carriage as a system was
seriously attacked in 1899 by the Major General commanding the
army, followed by the Board of Ordnance and Fortification over
which he presided at that time. On October 8, 1900, the Board,
by a majority vote of the members present at the time, recom-
mended that no additional disappearing carriages be manufactured
for use on high or medium sites, and that no more should be manu-
factured for use on low sites until the proportion of those to be
placed on such sites should have been reduced to one-third of the

410 ORDNANCE FOR THE LAND SERVICE.

t6tal number to be so located. The objections were directed gen-
erally: first to the use of disappearing guns, and second to criti-
cisms of the then adopted type of disappearing carriage. The
objections and criticisms on both points have been very fully and
completely answered by papers published in the reports of the
Chief of Ordnance, U. S. Army, for 1900, appendix 32, 1901,
appendix 58 and 1902, appendix 1. These references can be con-
sulted by any desiring to acquaint themselves with all the bear-
ings of the subject. The question was finally settled (appendix
1, Keport of 1902) by the report of a board of seven members
appointed by the President pursuant to act of Congress. The
membership of this board comprised Colonel Wallace F. Kan-
dolph. Chief of Artillery; Captain Eugene H. C. Leutze, U. S.
Xavy; Major John G. D. Knight, Engineers Corps; Major
Charles Shaler, Ordnance Department; Major Albert S. Cum-
mins, Artillery Corps; Captain William II. Coffin, Artillery
Corps; Mr. John K. Freeman, of Providence, K. I., with Captain
Richmond P. Davis, Artillery Corps, as recorder. The board
was directed to test the system by firing not only the thirty
rounds from a 10-inch disappearing gun prescribed by Congress,
but also thirty rounds from a 10-inch barbette and ten rounds
each from 6, 8 and 10 inch guns mounted on both disappearing
and barbette carriages. The board visited five artillery posts
and fired over 150 rounds with full service charges, and in con-
clusion recorded its opinion that the general mechanical prin-
ciples involved in" the chief elements and movements of the Buf-
fington-Crozier disappearing carriage are admirably adapted to
their purpose. In these trials 30 continuous rounds were fired
from the 10-inch disappearing carriage in 27 minutes 10 seconds.
The average interval between rounds being 54.3 seconds, the
shortest interval 47 seconds.

The more this question is studied in connection with the in-
creased accuracy and power of gun fire from shipboard the more
reason there appears for affording the best possible protection
for the gun and its mount, and the gun crew as well. This re-
quires guns to be mounted either on disappearing 'carriages or
else to be thoroughly protected by shields. One gun that can
be kept in action at a critical time is evidently better than a
number that are so exposed as to be silenced or seriously dam-
aged in action. The straight front shield usually employed on
barbette carriages affords a very limited amount of cover, and a

ORDNANCE FOR THE LAND SERVICE.

411

D
<

412 ORDNANCE FOR THE LAND SERVICE.

shield thin enough to be penetrated by shell is a source of danger
rather than protection.

131. The illustration, Fig. 112, showing an Armstrong gun in
the Takau forts, China, in 1900, is well in point. They were
attacked by gunboats carrying rapid fire guns only at ranges
between 2,000 and 3,000 yards. The inner sides of the shields
were covered with blood when examined after the capture of the
forts, and around one gun were found the bodies of 35 dead
Chinamen. One gun was put out of action by a projectile that
entered under the chase and injured the mechanism of the car-
riage. This is the sarile pattern of shield regarding which the
testing board of 1902 remarked, " With the non-disappearing
Armstrong carriage with shield the gunner has better protection
than he has on any other mount examined by the board." The
4.7-iiich Armstrong gun shields have a thickness of 4 inches at
the front and one or two inches on the flanks. The 6-inch Arm-
strong carriage shield is 4^ inches thick at the front. The qual-
ity of the steel is such, however, that even the fronts can be per-
forated by 5 and 6 inch rapid fire guns at battle ranges.

132. The thickness of shields prescribed for our barbette car-
riages is l^-inch for 6-pounder, 2 inches for 15-poundcr and
4^ inches for all larger calibers, including the 5 and 6 inch rapid
fire and the sea-coast barbette carriages. This steel is required
to be face-hardened of the best quality. The curved shield
recently tested on a dummy 6-inch barbette pedestal mount,
model 1900, is made of face-hardened, Harvey ized steel 4^ inches
thick. It was able to keep out projectiles of 5-inch and smaller
calibers, but was perforated by a 6-inch projectile fired with a
reduced velocity to simulate a battle range of about 3,000 yards.
The cost of this shield, about $10,000, scarcely compensates for
the partial protection it affords.*

133. In firing from a disappearing carriage the sighting of a
gun in direction and elevation is given before the gun rises into
the firing position, when it can be immediately fired and the gun.
remains exposed to hostile fire for a few seconds only.

* Fig. 118 sbows the 6-inch barbette carriage shield with the axis of the gun
turned 60 degrees from the line of observation and after 9 rounds had been fired
against it, namely, one 8.2-inch; five 5-inch; and three 6-inch projectiles. The
perforation seen in the side exposed to view was made by a 6- inch capped A. P.
shot fired with a striking velocity corresponding to about 8,100 yards range and at
an angle of 50.5 degrees between the line of fire and a tangent plane to the plate
at the point of impact.

ORDNANCE FOR THE LAND SERVICE.

413

4:14 ORDNANCE FOR THE LAND SERVICE.

134. A very important attribute of the disappearing carriage
besides the protection of the gun mount and gun crew from
hostile fire, is the concealment it affords on the approach of an
enemy as compared with a barbette carriage and shield. When
the vicinity of the disappearing battery is masked by trees or
shrubbery, which should be planted if the natural features are
bare, the disappearing battery becomes most difficult to identify
at a distance, while a barbette gun is always in evidence and
when furnished with a shield becomes very conspicuous.

135. The model 1901 12-inch disappearing carriage. Fig. 114,
is controlled by electric motors in manceuvring, and combines
other improvements. The time for raising into battery is
reduced to about six seconds, or less than one-half the time re-
quired with the previous models of 12-inch carriage. The
traversing and elevating are controlled by the gunner on the
sighting platform through electric motors, and the telescopic
sight is so supported in a bracket geared with the elevating ap-
paratus of the gun that the gunner can himself give the neces-
sary elevation for range with the same facility as with gims on
pedestal mounts. The telescopic sight has a 3-inch triple objec-
tive, 15-inch focal length, Brashear-IIastings erecting prisms
and two eye pieces giving powers of 12 and 20 with a field of 4
and 3 degrees respectively. The eye end of the telescope wdll be
provided w-ith a rack and pinion for focusing and with adjustable
cross wares. Small electric lights are provided for night use to
illuminate the scale diaphragm in the telescope and the deflection
and elevation scales of the instrument. The sight includes a
range drum graduated in yards. These sights and the telescope
are furnished by the firm of Warner & Swasey, Cleveland, Ohio.

Methods of Range Finding and Fire Control.

136. The methods of fire direction in the sea-coast service
include three cases. One, where the gunner aims directly at the
target without predicted range and direction, and continues to
fire as rapidly as possible by observing as far as practicable the
errors of previous shots. This method is used generally w'ith
rapid fire guns. Two, where the gunner gives direction to the
gun by observing the target through the telescopic sight, and the
gun is elevated to give a predicted range obtained from the
range-finding instrmnents. Three, when the range and direction

ORDXAXCE FOR THE LAND SERVICE.

415

Q
O

-<
O

o
z

K
§

c
z

416 ORDNANCE FOR THE LAND SERVICE.

are both obtained by the range-finding appliances and transmitted
to the gun, which is aimed accordingly. In this case it is not
necessary for the gunner to see the target. The case applies to
all mortar fire and to the special case of salvo points, the range
and direction of which are known beforehand, and the guns can
be pointed to be discharged when the moving target reaches a
salvo point. Cases two and three are of most general applica-
tion, especially for distant fire.

137. The ranges are first obtained by either the depression
or horizontal base range-finding instruments, and by means of
a plotting board and accessories in the battery commander's
station are reduced to the position of the gun that is to be fired
and then transmitted by telephone and telautograph to the bat-
tery officer at the gun. When the gun is reported ready for
firing the battery commander discharges it by pressing an electric
button at his station connected by wire with the firing mechanism
and electric primer in the gun. Should the electric firing fail
a cannoneer stands ready to pull a lanyard and fire the primer by
friction. In case of a moving target the predicted range and
direction sent to the gim are calculated to set the gun sufficiently
ahead of the target so that the projectile will reach the target
when it arrives at the predicted point. The time of flight of the
projectile is necessarily involved in this prediction.

138. This is a very brief outline of the methods employed
which involve in practice a number of instruments and acces-
sories, and a very complete system of governing rules for fire
direction and for fire control by the commander of a group of
batteries.

139. Two methods of measuring ranges are employed. One,
by the depression range finder and one by a horizontal base with
azimuth instruments at each end. The depression' range finder
being complete in itself affords much greater convenience than
the horizontal base in operation. The difficulties with the hori-
zontal base are especially great in recognizing the target; that is,
to insure that both observers are looking at the same vessel. The
limitations in the accuracy of the depression instrument, how-
ever, which must be mounted at a height of at least 60 or bettor
100 feet above sea level, coupled with the difficulty of securing
this elevation at many sea-coast forts — imless by building expen-
sive and very conspicuous towers — necessarily lead to the use of
the horizontal system in many cases. The system now being

ORDNANCE FOR THE LAND SERVICE.

417

worked out is based upon a judicious combination of the two
classes of instruments.

140. The Warner & Swasey depression range finder, Fig. 115,
has been constructed with full knowledge of the requirements of
an instrument of this kind and fulfills these requirements better
than any other known instrument for the purpose. It has been
tested at the Proving Ground as well as by artillery officers at
various posts in target practice and is now being supplied in
quantity for the service. In effect, when the instrument in any
given station has been adjusted for height of tide and for refrac-
tion it enables the distance to a target to be read directly from
the yard scale by bringing the horizontal cross-wire on the water-
line of the object. The same firm has also designed a self-con-
tained horizontal base measuring instrument. A model of one-
fourth size with a base six feet three inches long has been tested
with satisfactory results.

Fig. 115. — Warner & Swasey Depression Range Finder

PAPERS

OV THE

CHICAGO MEETING

(XLIXth)

OF THE

AMERICAN SOCIETY OF MECHANICAL ENGINEERS.
MAT 3l8t TO JUNE 3rd, 1904.

No. lOM.

PROCEEDINGS

OP THB

CHICAGO MEETING

(XLIXth)

OF THB

AMERICAN SOCIETY OF MECHANICAL ENGINEERS.

Maj 8l8t to June 8rd, 1901

LOCAL EXECUTIVE COMMITTEE.

RoBBRT W. Hunt, Chairman.

Louis Mohr, Chairman Finance Committee.

George M. Brill, Chairman Entertainment and Excursion Committee.

Wm. L. Abbott, Chairman Hotel and Reception Committee.

Francis W. Lane, Chairman Printing Committee.

W. H. V. RosiNO, Chairman Transportation Committee.

Paul M. Chamberlain, Chairman Meetings Committee.

J. H. Warder, Secretary Local Committee. *

The forty-ninth meeting of the American Society of Mechanical
Engineers was made noteworthy by the fact that it was a joint
meeting of the American Society with the Institution of Mecfhan-
ical Engineers of Great Britain. The Council of the American
Society had invited the British Institution, by a formal vote,
suggesting that the latter body make their spring meeting of 1904
an American Meeting, with a view of visiting the great exposition
in St. Louis which would be in progress in commemoration of
the centennial of the purchase of Louisiana from the French in
1803.

This invitation was cordially accepted by the Council of the
Institution of Mechanical Engineers and arrangements were at
once begun by the executive officers of the two institutions to
carry out the common purpose of such a joint meeting.

422 PROCEEDINGS OF THE

It was recognized that a very significant feature of the visit of
the British Engineers to the United States would be the oppor-
tunity and -privilege of visiting the American installations and
study American practice in the line of their particular interests. '

To this end the American Society had organized headquarters
in the various cities to which the visiting engineers were directed
by accrediting letters, and through which channels the arrange-
ments for the local visits were provided. The visiting engineers
had come to America sufficiently far in advance of the date of the
meeting to accomplish their purposes in part before the meeting,
and were arranging to visit other cities after the meeting.

The date for the joint meeting was set for May 31st to June
3rd, and in the city of Chicago, rather than in St. Louis, in order
that the advantages might be reaped from the ample and con-
venient hotel accommodations in Chicago, and so that the real
business of the meeting might be completed before the distrac-
tions and interests of the exposition should compete for available
time of the visitors.

The headquarters of the Society were opened on the morning
of May 31st,*in a room in the Auditorium Hotel, and it became
at once obvious that the numbers in attendance from American
and British sources would be unusually large. The total registra-
tion during the four days exceeded the highest figures reached
in any previous gathering, and were approximately as follows:

American Society 350

British " ^ 75

Ladies and guests 518

Total 943

The registration system was carried out by a card principle in
multiple, and this meeting was the first at which a satisfactory
method was used for the carriage by each member of his name as
a feature of the lapel badge of distinction of member and guest.

The name was used only by members of the two societies and a
different color for the card was used for the American and Brit-
is^h members. The plan worked smoothly and acceptably.

The opening day was given over to incidental excursions to
the underground subways of the city and to miscellaneous and
individual interests.

The first session was arranged to be held in the Music Hall of
the Auditorium for Tuesday evening.

OHIGAOO MEETING. 493

Opening Session. Mat SIst, 8.30 p. m.

The first session was held in the Music Hall of what is desig-
nated as the Fine Arts Building, and was called to order by Mr.
Robert W. Hunt, of Chicago, Chairman of the Local Committee.
On the platform with Mr. Himt were Mr. Ambrose Swasey,
president of the Society; Mr. J. Hartley Wicksteed, president
of the English Institution; Mr. Edgar Worthington, secretary
of the English Institution, and Prof. F. R. Hutton, secretary of
the American Society.

Mr. Hunt introduced the Comptroller of Chicago, the Hon.
Lawrence E. McGann, who had been delegated by the corporation
to represent it in welcoming the engineers as guests of the city.

Mr. McGann spoke in detail of the obligation which the city
recognized to the work of the engineer, and referred to the finan-
cial obstacle which had prevented progress in these directions at
the rate which the city would have been glad to follow.

President Swasey of the American Society spoke of the Society
meeting once again in Chicago, referring to the meetings in 1886
and in 1893, and commenting on the growth of the Society during
these intervals He spoke of Chicago being the only city where
the Society has ever held three of its mid-year meetings so far.
He made his welcome at this point include particularly the pleas-
ure of the American Society in welcoming the British Institution
to a joint session under such favorable circumstances.

President J. Hartley Wicksteed of the Institution of Mechani-
cal Engineers of Great Britain expressed his pleasure in taking
part in the welcome of the engineers, and gave reference to the
pleasure of the visitors in coming to the various cities of the
United States, and noting the differences which were characteris-
tic of them.

At the close of the responses of the respective presidents, a
letter was read from Mr. James Forrest, past-secretary of the
Institution of Civil Engineers of Great Britain, in which he ex-
pressed the wish that the joint meeting might strengthen the bonds
of international fellowship and social union between the two great
branches of the English-speaking race.

At the close of the formal exercises the meeting adjourned to
an informal reception, tendered by the local residents, in the par-
lors of the hotel, at which a light collation was served, and which
gave opportimity for the members to renew old acquaintances

424 PBOOEEDINGS OF THE

and form new ones, and for the members of the two societies to
come into social contact.

Second Session. Wednesday, June 1st, 10 o'clock a. m.

The second session was called to order in the large ball-room
of the Auditorium Hotel on the ninth floor, overlooking the lake.
The meeting was called to order by President Swasey of the
American Society, who took occasion to report to the meeting
the action of the Council under the provisions of the Constitution,
on the previous afternoon, whereby, after nomination in due
form, Past-President John E Sweet, one of the founders of the
Society, had been elected to honorary membership.

He spoke in fitting words of Professor Sweet as a capable man,
an able engineer, and that if any living member of the Society was
entitled to the name of " Father of the Society of Mechanical
Engineers" it was the member whom the Society. had honored,
and in honoring whom it honored itself. The announcement of
this election was received with applause.

The Secretary reported for record the Report of the Tellers
of Election as follows:

REPORT OF TELLERS OF ELECTION.

The undersigned were appointed a committee of the Council
to act as tellers under By-Laws 6, 7, and 8, to scrutinize and count
the ballots cast for and against the candidates proposed for mem-
bership in their several grades in the American Society of
Mechanical Engineers, and seeking election before the Forty-ninth
Meeting, Chicago, 1904.

They have met upon the designated days in the office of the
Society, and have proceeded to the discharge of their duty. They
would certify for formal insertion in the records of the Society,
to the election of the following persons, whose names appear on
the appended list in their several grades.

There were 661 votes cast on the ballot ending May 14th, 1904,
of which 50 were thrown out on account of informalities. There
were 596 votes cast on the ballot ending May 21, 1904, of which
32 were thrown out because of informalities. The tellers have
considered a ballot as informal which was not endorsed, or
where the endorsement was made by a facsimile or other stamp.

^T:raltri\r-u^''/^

Henri G. Chatain, )

ton.

CHICAGO MEETING.
May 14, 1904.

425

To BE Voted for as Members.

ADen, W. M.
Andrew, J. D.
Angus, J. J.
Arnold, George
Baker, J. C. W.
Ba\i8ch, F. E.
Bidle, W. S.
Coster, M.
Darlington, P. J.
Davol, G. K.
Diemer, H.
Dunbar, W. O.
Gerhard, W. P.
Goddard,D.

Harris, W. A.
Hood, O. P.
Howells, C.
Jenness, C. H.
Kelman, J. H.
Kenyon, A. L.
Lafore, J. A.
Larkin, W. H., Jr.
Lenfest, B. A.
Ix)rd, J. E.
MacMurray, J. T.
Meier, K.
Met<'alf, L.
Muckle, J. S.
Powell, J. E.

Randall, D. T.
Saflford, A. T.
Sage, H. M.
Sanborn, F. E.
Smith, A. L.
Smith, P. F., Jr.
Stillman, G. F.
Teele, F. W.
Toltz, M. E. R.
Trampe, J. A. C. L. de
Walker, G. S.
Wells, E. C.
Wells, J. H.
Wolf, F. W.

For Promotion to Full Membership.

Bump, B. N. Hannah, F. A. Logan, J. W.

Marks, L. S. W^allace, D. A.

To be Voted for as Associates.

Gillette, J. W.
Johnson, W.
Judd, Horace
Logan, J.

Ludeman, E. H.
Merrill, J. L.
Norton, F. L.
Noyes, H.T., Jr.
O'Neil, J. G.

Parkhurst, F. A.
Hippey, S. H.
WiUiams, G. W.
Woods, F. F.

Libbey, J. H.

Buckingham, H. H.
Canby, H. B.
Diederichs, H.
Dudley, S. W.
Gamon, E.
Gillett,W.L.
Glasgow, C. L.
Jaok.son, P.
Kelley.G.C.
Kenney, L. H.
King, R. S.

For Promotion to Associates.

Robbins, C. C.
To be Voted for as Juniors.

Knowlton, F. K.
Locket t, K.
Massie, J. H.
Matthews, F. E.
Morgan, L.
Murphy, J. K.
Nate, E. H.
Newbury, G. K.
Norris, H. L.
Powell, E. R.
Rantenstrauch, W,
Richards, C. D

Schaeffer, S.
Schlemmer, E.
Scott, E. F.
Slawson, H. H.
Snodgrass, J. McB.
Sponsler, C. F.
Stacy, H. W.
Townscnd, H. P.
Waclialofsky, C. J.
Woodward, R S., Ji.
young, C. H.

426

PROCEEDINGS OF THE

BiAY 2l8T, 1904.
To BE Voted for as BIembbrs.

Bennett, C. W.
Clemens, A. B.
Corbett, R. H.
Elmes, C. W.
Haven, H. M.
Holdsworth, F. ]

Belsley, Clay.
Bouton, G. I.
Collier, W. H.

Beebe, M. C.
Garred, U. A.
Hill, H. H.

Jones, B. N.
Loomis, B., Jr.
Lucas, H. M.
Massa, R. F.
Monteagle, R. C.
Nicholson, S. T.
Reid, Joseph.

Rivett, Edward.
Schulte,G.H.
Slade, A. J.
Smith, E. J.
Stivers, W. D.
Taylor, C. L.

FoR^ Promotion to Full BIembersuip.

Doty, Paul.
Hutchison, R.
Jacobs, Ward S.
King, J. H.

Price, A. M.
Watts, Geo. W.
Whitaker, H. E

To BE Voted for as Associates.

Nelson, J. W.
Obert,C.W.

Pond, H. O.
Stevens, H. L.
Van Winkle, E.

For Promotion to Associate Membership.

Langoltz, Robt.

Barnard, E. E.
Chasteney, C. D.
Cobleigh, H. R.
Gardner, Henry

Morrison, H. S.

To BE Voted for as Juniors.

Heaton, H. C.
Hodge, Geo. O.
Horton, W. H.

Knoop, T. M.
Midgley, F. W.
Sibley, M. M.
Sibson, H. E.

Under the provisions of the Constitution and By-Laws, the
Chairman announced these gentlemen in the foregoing list to be
duly elected to membership in their respective grades.

Mr. Wilfred Lewis, Chairman of the Commitee on Standard
Forms of Machine Screws, reported in a letter that very satisfac-
tory progress had been made, and the committee confidently ex-
pected to complete its work in ample time for the presentation of
a report at the annual meeting of the Society in December.

The Chair then called upon the Committee representing the
Society in the planning and labor concerning the proposed en-
gineering building which Mr. Andrevf Carnegie had projected for
the use of the societies who should take part in its advantages.
This report, in the absence of other representatives of the Society,

CHICAGO MEETING. 427

was presented by the Secretary, and is made a special appendix to
the Proceedings of this meeting.

At this point the President called for new business, motions
and resolutions.

Mr. Fred J. Miller, of New York, presented at this point a pro-
posed amendment to the method of amending the Constitution of
the Society. Th^ debate on this amendment was as follows:

Mr, Fred J. Miller. — ^In accordance with paragraph 59 of
the Constitution, I wish to offer an amendment to that para-
graph. Perhaps I had better read the paragraph as it is in the
Year Book, so that if any discussion takes place we may discuss
it intelligently. May I do that?

President Swasey, — As you please.

Mr. Fred J. Miller (reading). — " At any semi-annual meeting
of the Society any member may propose in writing an amendment
to this Constitution. Such proposed amendment shall not be
voted on at that meeting, but shall be open to discussion and to
such modification as may be accepted by the proposer. The pro-
posed amendment shall be mailed in printed form by the Secretary
to each member of the Society entitled to vote^ at least sixty
days previous to the next annual meeting, accompanied by com-
ment by the Council, if it so elects. At that annual meeting such
proposed amendment shall be presented for discussion and final
amendment, and shall subsequently be submitted to all members
entitled to vote, provided that twenty votes are cast in favor of
such submission." That means twenty votes at that meeting.
" The final vote on adoption shall be by sealed letter-ballot, clos-
ing at twelve o'clock noon on the first Monday of March fol-
lowing."

Paragraph 58 reads:

" The letter-ballot, accompanied by the text of the proposed
amendment, shall be mailed by the Secretary to each member
of the Society entitled to vote at least thirty days previous to the
closure of the voting. The ballots shall be voted, canvassed and
announced as provided in the By-Laws. The adoption of the
amendment shall be decided by a majority of the votes cast. An
amendment shall take effect on the announcement of its adop-
tion by the presiding officer of the semi-annual meeting next fol-
lowing the closure of the vote."

I call attention to the fact that according to this paragraph of
the Constitution if a member propose any amendment to the Con-
stitution he need not secure the approval of any other member

428 PROCEEDINGS OF THE

of the Society. It does not even require that the proposed amend-
ment shall be seconded. If it is merely proposed at a semi-annual
meeting, in writing, though it may then be discussed, a vote upon
it is prohibited at that time; it may only be discussed and the
proposer may accept amendments if he choose, but no matter if
he is unable to secure the approval of a single other member of
the Society for the proposed amendment, the Secretary is, never-
theless, compelled to print it, at the expense of the Society, and
to send it to every member of the Society to be voted upon. Then
it must come back to the Secretary, must be again discussed, and
if it then receive twenty votes in the meeting it must be again
sent out to every member to be voted upon a second time, and,
if it then receives their approval, it is adopted. Now, in case
no amendment is made at the second presentation of the amend-
ment, that is, if no modification of it is made, the Society is
placed in the position of sending out to the full membership
identically the same thing to be voted upon a second time. That
is to say, though it has been sent out once and approved, it must
be again sent out for a second vote, even though no alteration
whatever may have been made.

In place of those two paragraphs I propose the following:

Amsndmbnt to the Constitution.

"At any semi-annual meeting of the Society any member may propose in
writing an amendment to this Constitution. Such proposed amendment shall
then be open to discussion and to such modification as may be accepted by the
proposer, and if it then receives in its favor 20 votes from among the members
present it shall, at least 60 days previous to the next annual meeting, be mailed
in printed form by the Secretary to each member of the Society entitled to vote,
accompanied by commments by the council if it so elects. The vote upon the
amendment shall be by sealed letter ballot closing at 10 a.m. on the first day of
the annual meeting first succeeding the semi-annual meeting at which the amend-
ment wa3 proposed. The ballots shall be voted, canvassed and announced as
provided in the by-laws, and if a majority of the votes cast are in favor of the
amendment it shall, after declaration of that fact by the presiding officer there-
upon take effect and be in force."

It seems to me that this amendment obviates the difficulty into
which this clause of the Constitution as it now stands may lead us.
I have, I may say, some other amendments which I think should
be made, but I do not present them now, because I do not wish to
propose them under this law as it now stands, as I would be
compelled to propose them without their having received the ap-
proval of any other member, and I wish simply to get this under

CHICAGO MESTINO. 429

way in order that neither myself nor any other member who
wishes to propose an amendment may be compelled to propose
it in the manner in which the Constitution now requires it to be
done.

President Swasey. — Gentlemen, you have heard the amendment
proposed by Mr. Miller, which is now open for discussion. What
have you to say in regard to it?

Mr. R. H. Soide. — ^Having been a member of the Committee on
the Revision of the Constitution I am familiar with the circum-
stances and considerations which led up to the adoption of the
particular clause which it is now sought to amend; there was not
one article of the entire Constitution and By-laws for which so
many alternative arrangements were offered as this one; the com-
mittee spent a great deal of time in discussing the best method of
amending the Constitution, and finally decided on the form which
has since been adopted; and the imderlying principles in reaching
that conclusion were these:

First of all it was considered wise to give every member of
this Society a right to stand up on this floor and to suggest an
amendment to the Constitution, and it was purposely made to read
so that it was not necessary for him even to secure a second; it
was considered that it ought to be the fundamental right of every
man to submit an amendment to the Constitution, and at least
claim a hearing for it; the other idea was that it would be a
sound principle that the main debate on that proposed amendment
should be precipitated into the arena of discussion at the annual
meeting of the Society; the fact is recognized that the annual
meeting is always the largest, and therefore it is the one at which
it may be assumed that an amendment would receive the most
careful consideration and at the hands of the largest proportion
of the membership.

That is the logic which pervaded the minds of the members
of the Conmiittee on Revision of the Constitution, and which led
up to their suggestion of that particular clause which it is now
proposed to amend.

Mr. Fred J. Miller. — ^In regard to what Mr. Soule says as to the
right of every member to propose an amendment, I would be
the last one to deny that right, for I most thoroughly believe in it;
but I call attention to the fact that my amendment toes not really
abridge the right of a member to propose an amendment; he still
has the right to propose, but there is no use in putting the Society
to the expense of printing his amendment and sending it out to

430 PROCEEDINGS OF THE

the membership if he cannot get twenty members to agree with
him that the amendment is a desirable one; because it is then
morally certain that his amendment cannot be adopted, and the
bother and expense of the general vote will be to no purpose.

I have studied the Swiss Initiative and Referendum to some
extent. Any citizen of the Swiss Republic has a right to pro-
pose a law, but it is there clearly seen that it would be folly to
arrange so that any Swiss citizen could simply write out a proposed
law, and that it should then be immediately submitted to every
member of the Oovernment to be voted upon. The Swiss Consti-
tution requires that the proposer of the law shall have back of it
the support of a certain nimiber of citizens first secured by the
proposer. That is what we want here. Another point which occurs
to me is tfiis: That an amendment may be sent out to the member-
ship to be voted upon, and it may be approved by them; the mem-
bers expressing their desire to have the Constitution so amended,
and yet that amendment comes up in the next annual meeting of
the Society, and is discussed if at that time it fails to receive
twenty votes, the will of the Society is set aside, i.e., although
the membership at large has voted in favor of the amendment
it cannot be adopted, simply because it cannot be sent out for a
second vote by the membership at large. I do not think that is
right.

President Swasey. — ^Is there any further discussion? If not,
the amendment will take the regular course laid down by the
Constitution.

No other new business being presented the meeting took up
the professional papers as follows:

The paper by Mr. Harrington Emerson on " A Rational Basis
for Wages " was discussed by Messrs. Emerson Bainbridge, E. J.
Chambers, and J. Hartley Wicksteed of the British Institution,
and Mr. H. H. Suplee of the American Society.

Two papers presented by British authors were then read by
the Secretary of the British Institution.

The paper by Mr. George Watson was entitled "The Burning
of Town Refuse," and the paper by Mr. C. Newton Russell was
entitled "Refuse Destruction by Burning, and the Utilization
of Heat Generated." President Wicksteed opened the discus-
sion, which was participated in by Mr. Alfred Saxon, Mr. Charles
Wicksteed, and Mr. G. R. Dunell of the British Institution.

At the close of these papers the meeting took a recess until the
evening of the same day.

OHIOAOO MEETINO. 431

Third Session. Wednesday Evening, June Ist, 8.30 p. m.

The third session was called to order in the Music Hall of the
Fine Arts Building by President Ambrose Swasey. The papers,
of this evening were specifically devoted to the Power Plant
Problem and Practice, especially on the development of the steam-
turbine. In the discussion of the three papers, Messrs. Storm
Bull, Greorge I. Eockwood, D. S. Jacobus, H. H. Suplee, George
W. Colles, H. L. Doherty, and Alex. Dow took part. Messrs.
Hodgkinson, Rearick, Rice, and Kerr spoke in the closures. The
fidl titles of the papers were as follows: " Some Theoretical and
Practical Considerations in Steam-Turbine Work," by F. Hodg-
kinson; " The De Laval Steam-Turbine,'' by E. S. Lea and E.
Meden; "The Curtis Steam-Turbine," by W. L. R. Emmet;
" Different Application of Steam-Turbines," by A. Rateau; " The
Potential Efficiency of Prime Movers," by C. V. Kerr.

FoiTRTH Session. Thursday Morning, June 2nd.

The session was opened by* the presentation of two papers on
the Tall Office Building Problem, by Messrs. Bolton and J. H.
"Wells. These papers were respectively entitled " The Power
Plant of the Tall Office Building," and were discussed by Messrs.
Bryan, Colles, Rockwood, Suplee, Bunnell, Gifford and Mr. Nistle
of the American Society.

The next paper was from the English Society, and was en-
titled " The Middlesbrough Dock Electric and Hydraulic
Power Plant," discussed by Messrs. Barr, Alfred Saxon, and
John Etherington ; the paper entitled ^' Use of Superheated Steam
and of Reheaters in Compound Engines of Large Size," by Mr.
Lionel S. Marks, was discussed by Messrs. Barrus, Kerr, and Rock-
wood. The paper by Mr. William P. Flint, entitled " Commercial
Gas Engine Testing and Proposed Standard of Comparison," was
discussed by Messrs. Mathot and Chambers of the English Society,
and Professor Jacobus of the American.

At the close of the discussion the meeting took a recess until the
final session on Friday morning.

Final Session. Friday Morning, June 8rd.

The Joint Meeting was the guest for this session of the Lewis
Institute of Chicago, West Madison and Robey Streets. The ses-
sion was introduced by the presentation by Secretary Worthing-
ton, on behalf of Mr. William Campbell, of the report of the

482 PROOEEDINOS OF THB

Institution's committee on "Effects of Strain and Annealing,"
prepared by Dr. William Campbell. This report was presented
in ab&tract by the use of lantern slides projected upon a screen.

Following the presentation of this report, which was com-
mented on by President Wicksteed, the paper by Mr. William
J. Keep was read on " Cast-Iron, Composition, Strength and
Specifications," which was discussed by Messrs. West and Emer-
son of the American Society and Mr. Chambers and Mr. Crosta
of the British Institution.

At this point. President Wicksteed took the chair for the pres-
entation of the paper by Prof. J. T. Nicholson on " Experi-
ments with a Lathe-Tool Dynamometer," which was presented by
Mr. Daniel Adamson of Manchester, England, with whom he had
been associated. In the discussion, Messrs. Wicksteed, Emerson,
McQeorge, Benjamin, and Pilton took part.

President Swasey of the American Society resumed the chair
for the presentation of the 4)apers on locomotive testing, by
Messrs. Hitchcock and Qoss, respectively, entitled: "Locomotive
Testing Plants," by Prof. W. F. M. Goss; " Road Tests of Con-
solidation Freight Locomotives," by Prof. E. A. Hitchcock.

Messrs. Bement and Worthington took part in the discussion.

With this group of papers the presentation of technical mat-
ter was completed.

Director Carmen of the Lewis Institute introduced at this
point the Hon. C. C. Kohlsaat, Judge of the United States Court
in Chicago and president of the Board of Trustees of the Lewis
Institute. Judge Kohlsaat expressed the pleasure of the Trustees
of the Institute in having the body its guests for the morning,
and referred in brief to the desire of the Institute to make itself
useful in the spread of engineering education and in fitting young
men to carry on the work of the world in industrial lines.

Mr. Robert W. Hunt spoke in feeling words of the death
during the progress of the convention of Mr. David R. Eraser,
a veteran member of the Society resident in Chicago, who had
been a member of the Local Committee of Arrangement, and
who was a foremost representative of the industrial interests of
the city. Due memorial action was to be taken by the Society
in its published volume of Transactions.

President Swasey flt this point read from the chair the fol-
lowing minute:

The American Society of Mechanical Engineers, speaking for its own members,
desires at the outset to record its pleasure in the fortunate outcome of its plans

CHICAGO MEETING. 438

which have brought the Institution of Mechanical Engineers from England to
be its guests and co-workers in the pleasures and labors of its Chicago Convention.
It was a happy thought of Mr. James Rowan of Glasgow to present to the Ameri-
cans the possibility of holding a meeting of the Institution in the United States.
It is with great pleasure that the Americans have been permitted to receive so
many members of the Institution and to take them into all the elements, both pro-
fessional and social, of such a successful convention. The Americans hope that
their guests will carry home with them a still more earnest recognition of the fact
that the two branches spring from a common stock by race and language, and
that the laws and practice of engineering must necessarily be one on both sides
of the Atlantic.

At the close, President Wicksteed of the British Institution
pose and took the chair, and presented the following minute on
behalf of the British Society:

The Institution of Mechanical Engineers, as represented by its President,
Council, visiting members and Secretary, desires to take occasion on this last
session of the joint meeting to pass a resolution which shall convey to the Ameri-
can Society the warm thanks and recognition which they desire to express for all
that has been done for their pleasure and profit during the American visit of
the Institution. They desire to thank the Council of the American Society and
the executive officers who represent it for coiu*tesies, official and personal, and
for the arrangements which have been made for them, the visiting engineers, to
have opportunities of studying American industry in its home field.

The Institution desires further to record the pleasure its members have de-
rived from the attentions and opportunities which have been extended to them
during their stay in Chicago, and in their joint participation in the Chicago
Meeting. The visitors ask that, in the resolutions of thanks for specific things
which are about to be presented to the joint meeting, they may be particularly
iDchided, and that they as well as the society may be identified with these resolu-
tions.

At the close of this action, the Secretary of the American So-
ciety presented on behalf of both organization the following reso-
lutions of thanks and recognition, which were unanimously
adopted:

1. Resolved, That the American Society of Mechanical Engineers and the
Institution of Mechanical Engineers desire to express to the City of Chicago and
its city Government the sincere thanks of the joint meeting for their courtesy,
extended by a representative of the City in the person of its Comptroller, in
addressing a welcome to the visitors at their opening session on Tuesday evening.
They would express their pleasure in having brought to their notice the recogni-
tion which the Corporations was kind enough to express of the debt which
C^cago owes to the engineering talent which has helped to make the City what
it is.

2. The American Society and the Institution of Mechanical Engineers have
■ received from the Illinois Steel Company at South Chicago a notable courtesy

in the permission and arrangements for the joint visit through the Illinois Steel

434 PR00EEDINQ8 OF THE

Works on Wednesday afternoon. The Mechanical En^eers recognize what
is meant by a permission of this sort an4 the difficulties in handling so large a
number through the complicated and dangerous departments of a productive
establishment of this magnitude. They ask that the Illinois Steel Company and
Mr. W. A. Field, the General Superintendent, and other officers of the Company
will accept the sincere thanks of the party for the success attending the visit,
and for the admirable way in which the visit and entertainment were planned.

3. The joint session of the American Society of Mechanical Engineers and
the Institution of Mechanical Engineers have receievd from the Illinois Central
Railway most notable courtesies, entertainment and provision during their
Chicago visit. They would ask that that Company and in particular Mr . W . H . V.
Rosing, Assistant Superintendent of machinery, will accept the sincere thanks
of the Engineers for their tender of free transportation over their lines during
the Chicago Convention, and for the special and admirable arrangements wherdOy
the Company and its agents have served the convenience of the visitors on their
excursions and visits about the City.

4. The joint meeting of the Mechanical Ekigineers of {England and America
ask that Marshall Field and Company will accept the sincere thanks of the societies
for the opportunity given to the ladies and members to visit the wonderful
establishment which b known by their name, and for the courteous arrangement
for the luncheon of the visitors on the afternoon of Thursday at that great mer-
cantile establishment. The character of the hosts on this occasion is a factor in
economic and commercial development which all engineers are watching with
the keenest interest in its bearing upon the problem of economic handling be-
tween producer and consumer.

5. The joint meeting in Chicago has been made memorable for the Mechanical
Engineers by the particular importance which has been given at this meeting to
the subject of the Steam Turbine. The Engineers, therefore, ask that the
Commonwealth Electric Company and the Chicago Edison Company will accept
the sincere thanks of the visitors for the privilege of the visit to the Fisk Street
Station and the chance to see the woriung of a power station of this magnitude.
They would couple this vote of thanks in particular with the name of Mr. Loub
A. Ferguson, Second Vice-President of the Company, and with those of Mr.
Abbott and members of the local committee who have been instrumental in
bringing about this pleasant result.

6. Those members of the joint meeting who have taken occasion to visit the
Stock Yards and the Packing House of Swift and Company, ask that that Com-
pany and Mr. EJdward S. Swift in particular and other officers of their corporation
will accept sincere thanks for the invitation to visit their great undertaking, and
to study their admirable, humane and careful methods which have been followed
in the prosecution of their business and in the attainment of their wonderful
results.

7. To the Trustees of the Art Institute and to Mr. H. N. Carpenter, Secretary,
the Engineers beg to extend their most sincere appreciation for the distinguished
courtesy accorded by these gentlemen, in permitting the reception of Thursday
evening to be held in the midst of the attractive collections of that Institute.
The relation between engineering and art is a close one, even if by many its exist-
ence b not detected . Art b a child of economic progress and wealth . Engineering
is the foundation of real productive wealth. It has given the Engineers great
pleasure that their reception should be accorded to them in the midst of such
congenial and appropriate surroundings. *

CHICAGO MEETING. 435

8. The Engineers ask that the Trustees and Faculty of the Lewis Institute
of Chicago will accept the sincere thanks of the joint meeting for the invitation
to hold its closing session in the Hall of the Lewis Institute. The Engineers
recognize what the profe^ion owes to the work of the technical educators and
are glad to record by this resolution their hearty sympathy in the work of the
Institute and their recognition of the far-reaching efifect of such education, both
in industry, in civilization and in culture. They would express in particular
their thanks to Professor P. M. Chamberlain for his instrumentality in bringing
about this pleasant result, and heartily appreciate the courtesy for the enter-
tainment and luncheon which the Trustees have provided.

9. It is to the Trustees of the Sanitary District that the Mechanical Engi-
neers owe the courtesy of an invitation to inspect the Drainage Canal, a great
work in the interests of Chicago, which has been prosecuted under their direction.
The thanks of the Societies and the Institution are also due to the Atchison,
Topeka and Santa Fe Railroad Company for the courtesy of the special tcain
provided by that Company through Mr. Kendrick, third Vice-President of the
Railway system. The visiting Engineers ask also that the Western Society of
Engineers, by whose active co-operation this excursion has been originated, will
accept the sincere appreciation of the visitors for their share in the successful
visit.

10. Beginning with the very first day of the attendance of visitors and
continuing beyond the official end of the meeting have been the courtesies of the
Illinois Tunnel Company. The Engineers desire to return thanks to the Com-
pany and to Mr. George W. Jackson in particular for the courteous invitation to
^'isit the tunnels and to study under such favorable circumstances their possi-
bilities and their working. The Mechanical Engineer is entirely wonted to the
experience that some of his most successful achievements are where the eye of
the superficial observer knows nothing of their real excellence. That we should
have had the opportunity to go under ground and under competent guidance
study the hidden detail is an experience for which we desire to record our thanks*

11. It must be known to any Mechanical Engineer that the success which
comes to an acceptable product of a machine depends upon the skill, the fore-
thought and the careful administrative attention of the brain of the Engineer
that has originated the machine. In the successful meeting of a Convention,
such as our Chicago meeting has proved itself to be, the Engineer behind the
product is the local committee of resident members in whose hands we have been
but raw material, and which have succeeded in turning out so successful a finished
product as the Chicago meeting. The visiting Engineers ask that Mr. Robert
W. Hunt, with a membership common to both societies, who has acted as chair-
man of the local committee, will accept for himself and for his associates a sincere
recognition for his and their share in bringing this delightful result to pass.
They ask also that Mr. George M. Brill, efficient and capable Chairman of the
EIntertainment and Excursion Committee, who has abandoned his liome tlmt he
might be near his responsibilities, will allow himself to he included in our wann
expression of tlmnks and recognition. The meeting would not have ]>een the
success which we have enjoyed, had less gifted, interested or assiduous hands
been at the throttle.

12. While the ladies are important and appreciated factors as captains of
industry in the control of the individuals who paas resolutions, it is unfortimate
that they do not have a speaking voice in the proceedings of the official part of
this meeting, and that from "seconds in command" and not from the captains

486 PROCEEDINGS OP THE CHICAGO MEETING.

themselves must come the voicing of the thanks which the ladies of our party
must tender to the efficient and delightful group of ladies organized imder the
local committee for the pleasure of the visitors. This resolution is to give voice
to the official thanks of the visiting ladies for the drive and the luncheon on
Wednesday, for the organization and direction of the charming visit on Thursday
and for the crowning delight of the drive on Friday. The ladies ask also that
Mrs. Robert W. Hunt will accept this unsatisfactory effort of mere men to ex-
press for them the thanks of the ladies for the delightful reception on Thursday
afternoon.

The men and the women leave for home with delightful memories of new
friendships formed, of old friendships strengthened and a memory of a week
full of pleasure, satisfaction and profit.

These resolutions being put by the chair they were unani-
mously adopted.

At the close of the sessions the Society and its guests were
entertained by the ladies of Lewis Institute at a most satisfactory
luncheon on the top floor in the department of the Institute
devoted to domestic science, and afterwards took a train on the
Atchison, Topeka and Santa Fe Railroad for the controlling
works of the drainage canal.

On the way the party visited the distributing yards of the rail-
way companies. The party were escorted by float a short dis-
tance on the canal itself, .and after visiting the work of the con-
trolling outflow, re-embarked on the train and returned to Chi-
cago.

In the evening, the Society and its guests were again enter-
tained by the Local Committee, the entertainment taking the most
attractive form of an orchestral concert in the Auditorium Thea-
tre by the Thomas Orchestra of Chicago.

This concert was made noteworthy by its ending. A precentor
came forward on the stage, and led the audience in two verses
each of the American national hymn, " My Country 'Tis of Thee,*'
and the English form of the same air, " God Save the King."

With this incident the meeting officially closed.

The next meeting of the Society will be its regular annual
meeting, to be expected in the city of New York the first week
in December.

No. lOM.

APPENDIX.
CARNEGIE GIFT TO ENGINEERING.

THIRD CIRCULAR.

In former circulars the members of the Society have been ad-
vised that at the Saratoga Meeting in June, 190t3, there was re-
ported in official form the proposed gift to engineering by Mr.
Andrew Carnegie, a member of the Society, and that at the New
York Meeting in December, 1903, a supplementary report was
made of the work of the Society's Committee up to the beginning
of that year. The first report will be found on page 870 of
Volume XXIV., as paper No. 976, and the second on page 34 of
Volume XXV., as No. 1008.

At the Chicago Meeting in May, 1904, a further report of
progress was presented for the information of the members as
a feature of the business session of that meeting. The report was
presented by the Secretary of the Society, who is one of the
representatives of the Society on the Committee, and was as
follows:

It will be recalled by those who have followed the history of
the undertaking that in February, 1903, Mr. Andrew Carnegie
made his first offer of a million dollars in order that the four
national engineering societies might be housed in a proper build-
ing, which should give necessary accommodation for the executive
offices of the engineering societies, and should contain auditoriums
for meetings and give adequate space for the libraries of the
societies, the plan also to include accommodations under a separate
roof, but as a part of the general scheme, for the Engineers'
Club, which is the social organization having no necessarily pro-
fessional outlook.

The action which was taken on the receipt of Mr. Carnegie's
proposition was to appoint a Conference Committee, consisting
of three members from each of the four national societies and
from the Engineers' Club. This Conference Conmiittee held

438 APPENDIX.

frequent meetings for the consideration of its problem, and dis-
charged its duties under a provisional organization until March,
1 904, when by the letter-ballot of the American Society of Civil
Engineers — for reasons satisfactory to themselves — it was de-
cided by that body that they would not partake in the benefit of
Mr. Carnegie's gift.

The Conference Committee was then at once faced with the
problem: Would the donor give his proposed gift to the three
engineering societies and the Engineers' Club, and would the
three societies be able to make themselves responsible for the
maintenance and conduct of the building, or would the problem
become greater than these three bodies could cope with?

The matter was immediately laid before the Councils of the
three societies and on careful consideration of the resources of
each body, the three organizations, the American Society of
Mechanical Engineers, the American Institute of Electrical En-
gineers, and the American Institute of Mining Engineers, decided
that it would be within their power to resume this responsibility.

Careful estimates of the cost of operating the building were
prepared, and by special meetings the decision was reached that
each society was prepared to go forward in co-operation with the
other two. When this action was communicated to Mr. Carnegie,
with characteristic promptness of decision, he* at once wrote an-
other deed of gift, stating his willingness to present to the three
societies and the Engineers' Club a building which should cost
a million and a half dollars, in recognition of the larger scope of
the proposition as submitted to him by the three societies.

It is the purpose of the Trustees that this building should
provide for all organizations which have engineering as their
basis in the matter of suitable auditoriums, executive offices, and
most of all, for their libraries.

Mr. Carnegie's letter of gift has, therefore, constituted the rep-
resentatives of the three engineering societies and the Engineers'
Club, an organized committee of twelve, who for the moment are
to administer what has been called the " Carnegie Trust for En-
gineering." This committee of twelve has the responsibility for
the expenditure required for adequate buildings. While the gift
is as one, there are in reality two buildings. The three societies
are to occupy a frontage of 125 square feet on the north side of
Thirty-ninth Street in New York City, with a depth of 100 feet.
The Engineers' Club is to occupy a site 50 feet wide by 100 feet
on the south side of the Fortieth Street frontal of the same eity

APPENDIX. 430

block. The buildings do not quite oome to each other in the rela-
tion of the upright and the arm of a capital letter " T/' because the
Engineers' Club stands nearer one end than the other of the long
frontage of the Engineering Building.

The Conference Committee, in its capacity as Trustees, at once
made arrangements to have the design of the building competed
for and the architects selected by the competitive method.

A programme of competition was prepared with some elaborate
ness, deciding a number of questions in advance in order that
the competition might be on the conmion basis. Among these
questions were, first, that the auditorium which is a central feature
of such a proposed engineering building should have such a capa-
city that the best voice of an author or speaker could be distinctly
heard in every part of it. This brought to the front at once an
interesting question as to the floor space and height within which
an untrained speaker taking part in the meetings could be heard
and understood. It was plainly absurd to make the auditorium
80 lai^ that discussions were a mere dumb show of moving lip
and gesture. It seemed on conference with theatrical people,
ministers, architects, and public speakers, that such ordinary
speaker could be heard if the audience did not exceed one thou-
sand persons on the floor, with a possible additional five hundred
seated in a commodious gallery.

The committee would be very glad to receive additional light
on this question from any members who may have experience to
oflFer.

There are also to be smaller auditoriums having a capacity of
two hundred and fifty persons and downward, so that smaller
meetings shall not be lost in a large auditorium, and so that meet-
ings of different societies of different sizes can be held at the same
time. Such smaller auditoriums will also be cosy for monthly
reunions.

The auditorium must, necessarily, by ordinance of the Building
Department of New York City and as a matter of common sense,
be located near the level of the street.

Above the auditoriums will be the office floors, one for each of
the societies, giving accommodation for its reception rooms, offices
for secretary, accountants, stenographers, and similar adminis-
trative departments, and the meeting room for coimcils and- com-
mittees.

At the most liberal calculations for the present needs of the
societies, a floor area of 5,000 square feet would seem to be ample.

440 AP^NDIX.

Each floor will furnish over one-and-a-half times this space, so as
to leave room for growth and future needs. It has been quite
interesting to note that the three societies, calculating indepen-.
dently, ask the committee for practically the same amount of
floor space.

Above the three floors allotted to the founders^ societies can be
certain additional floors, if the funds available shall make this
possible, which shall be assigned to the use of the other engineer-
ing organizations which will naturally be the beneficiaries of Mr.
Carnegie's gift. Such bodies are the Electro-Chemical Society,
American Gas Light Association, Electric Light Association, and
others. These accommodations will not be rented, but will be
furnished to the users, on a basis of a pro rata of operating ex-
penses. It is plainly unfitting that there should be any profit-
making in the conduct of the enterprise from a commercial point
of view.

On the top floor or floors, which will probably be about tho
twelfth floor from the street, will be located the consolidated
libraries and the reference and reading rooms.

It is an interesting fact that when the three societies met to-
gether by appointed meetings to consider the library problem in
an engineering building of this character, it was at once the sense
of these representatives of the three societies in their conferences,
that there was before them the opportunity of a lifetime to create
a great engineering library, with the strength and interest and the
knowledge and personnel of the three societies behind it. The
representatives of the New York Public Library, which will be
located in a building just across the street, were called in at this
conference, and their advice asked on the possibilities of a scheme
of federation, whereby the New York Public Library and the
Engineering Library could be mutually serviceable.

By locating the reading-room on the top floor it will not only be
more quiet, but it will be in the better air, above the fly-line, and
it will obtain light from the roof.

The Committee feel emphatically that this should be a work-
ing library, in which workers and students should be able to have
access to the shelves and the books themselves.

To secure this greater development of the library idea will make
a demand upon the societies themselves, but it is the belief of the
Committee that the societies will be glad to enlist increasing
energy and effort in return for what the library development will
bring.

APPENDIX, 441

If the plans of the Committee are realized the library of this
building will be The Engineering Library of the* Atlantic Sea-
board.

In carrying out the plans of the competition, the Committee
have secured Professor William R. Ware as Professional Adviser
on architectural questions. He is perhaps of all American archi-
tects one who has had the widest experience in judging competi-
tions of the character which the Committee has had in mind.

With the advice of Professor Ware, and what could be gotten
from other sources, the printed Programme of Competition in
pamphlet form was issued on the 5 th of May.

In view of advice given us from many sources it seemed plain
that six weeks was a long enough time to allow architects for the
consideration of their preliminary plans. In this view, counting
forward from the 5th of May, the competition was announced
closed on June 20th.

The competition is of the character which is known techni-
cally among architects as " a mixed competition," in that sLk
firms have been invited to compete, and will be compensated,
whether successful or not by a fee of $1,000. Besides this the
competition is open to any architect in good professional standing,
who has been in practice under his own name for over two years,
to send in competitive drawings, and he will receive consideration
exactly the same as the salaried or invited firms. All are com-
pelled to compete under an absolute incognito. Four prizes of
$400 each will be the award to the four most successful plans
in the " open class."

Of course, the competition means nothing for the present but
the selection of the architect, who will prepare the plans finally
approved by the Committee. It does not mean in any sense that
the plan of the successful architect is to be the plan approved by
the Committee in final form. The Committee reserves the right
to change and modify and remake the drawings of the successful
competitor until its own ideals shall be realized.

It may be of interest to state that seventy-five applications have
been received, and while we know that some of these writers do
not intend to compete, but simply ask to recieve the competition
because it was a cleverly prepared document, yet the Committee
feels sure that there will be a sufficient number of competitors to
make an interesting variety when the Committee comes to decide.
Nothing can be done in the way of actual construction until after
the first of July of the current year, since the site of both build-

442 APPENDIX.

ings is now occupied by residences, the leases on which will only
expire on t^hat day.

The buildings must be removed and the work of final settling
on details of plans must follow before the actual construction
can be begun.

Th^ Committee has also had under consideration the problem of
conducting the building after it should be completed and the
construction problems have been solved. The present Conmiittee
are the Building Committee, and their functions will end with
the completion of the building.

This has been met by appointing a Committee on Organization,
through whom a charter has been secured from the State of New
York, and signed by the Governor, May 11th, 1904.

The general scheme of organization provides for the election of
trustees by the three societies, who enter the undertaking as
founders, under the deed of gift.

The Engineers' Club will manage its building independently
as soon as construction is completed.

These trustees are to be elected by the Councils of the individ-
ual societies, and will be three in number from each body, the term
of one expiring each year. These trustees will have the respon-
sibility of financing and managing the administrative detail of
the building as a whole for the benefit of the founders' societies
themselves, and the other engineering societies who will be par-
ticipants in the privileges of the building.

The by-laws of the body provided by the charter will have to
be very carefully prepared with advice of counsel, to provide for
many questions in advance which are incident to the creation of
a trust of this character.

The Committee proposes to proceed slowly with the considera-
tion of these questions of organization, but hope that before the
annual meeting they will be provided with a full and carefully
thought-out scheme which will be reported to the Society at that
time.

F. K. HuTTON, Secretary,

SUPERHEATED STEAM AND REHEATERS IN COMPOUND ENGINES. 443

^o. 1030.*

THE USE OF SUPERHEATED STEAM AND OF RE-
HEATERS IN COMPOUND ENGINES OF LARGE
SIZE,

BY LIONEL 9. MARKS, rAMBKIDOK, MASS.

(Associate Member of the Society.)

1. The object of this paper is to collect and present to the
Society the results of a number of unpublished tests made during
the past five years on several high-speed, two-cylinder compound
engines, all built by the same makers, and all of 'the same type.
The engines tested differ from one another only in size, in cylinder
proportions, and in their working conditions. The investiga-
tions were made to determine the performance of the engines
under different loads, both with and without jacketing and reheat-
ing. A comparison of the results for the different tests throws
some light upon the influence on the thermal efficiency of large
sized four-valve compound engines of the following factors :

(a) The use of a reheater.

(6) The use of moderately superheated admission steam.

(d) The size of the engine,
(c) The .cylinder proportions.

2. The results recorded here are for tests made on nine separate
engines and for twenty-eight different tests. The engines are
parts of three electric lighting plants situated in or near Boston.

Engine A is at the L Street station. South Boston, of the
Edison Electric Illuminating Company — a plant which at the
time of the test was the property of the Boston Electric Light
Company. The tests at this station were conducted by the \ATiter.

Engines B, C, D, E & F are at the Atlantic Avenue station
of the Edison Electric Illuminafting Co., and they v:ere tested

* Presented at the Chicago meeting, May and June, 1904, of the American
Society of Mechanical Engineers, and forming part of Volume XXV. of the
TranMrtions.

444 SUPERHEATED STEAM AND REHEATERS IN COMPOUND ENGINES.

SUPERHEATED STEAM AND REHSATEBS IN COMPOUND ENGINES. 445

under the joint supervision of an employe of the company and
a representative of the engine builders. Messrs. C. H. Parker
and H. Cook, both members of this Society, were in charge of
most of the tests, but in some tests their places were taken by
Messrs. C. R. Brown and S. G. Colt respectively.

Engines Oy H & K are at the new plant of the Cambridge
Electric light Company, and were tested by the writer.

Description of the Enginea.

3. The engines tested vary from 750 to 2,500 rated horse
power, and were all built by Mcintosh, Seymour & Co., of
Auburn, N. Y. They are all vertical, high-speed, two-cylinder,
cross-compound, direct-connected units with overhanging cranks.
Each cylinder is supported on a heavy, hollow, cast-iron frame
at the back and on two inclined steel standards in front. Each
II. P. cylinders is jacketed on the barrel, and both heads and the
jackets are piped in series ; the steam enters the jacket on the top
head, passes into the barrel jacket, goes to the jacket on the lower
head and then to the reheater coils. In this way a very active
circulation in the jackets is ensured. As there is no separate
steam supply to the reheater coils, nor any separate drain from
the H. P. jackets, it is not possible to use either jackets or
reheater alone. The receiver is a large cylindrical drum at the
back of the engine and close to the cylinders. The reheater
consists of one or more coils of pipe in the receiver. The L. P.
cylinder is unjacketed.

4. The valves are of the flat, gridiron type, unbalanced and
of short stroke. The steam valves on both H. P. and L. P.
cylinders consist of a main valve cutting off at about .8 stroke,
and a Kider cut-off valve, the movement of which can be varied
so as to give any desired cut-off. The main steam valves and the
exhaust valves on each cylinder are driven from an eccentric on
the main shaft through a system of links and levers. The cut-off
valves are driven by auxiliary eccentrics which are controlled by
a fly-wheel governor. The action of the valves is rapid; the
openings for admission and exhaust of steam are large.

The fly-wheel governors are designed to control the speed
within two per cent, variation between zero load and full load.
At the Cambridge electric light station the position of the gov-
ernor weights can be regulated when the engine is running by

446 SnPEBHEATED BIEAH AND BEHEAIEB8 IK COUfOUND ENQINES.

e-W

SUPERHEATED STEAM AND EEHEATERS IN COMPOUND ENGINES. 447

means of a small electric motor fastened to the fly wheel and
controlled from the switch-board. This device is valuable fot
synchronising in parallel running.

The Arrcmgementa for Testhig.

6. All the engines, with one exception to be noted later, are
fitted with jet condensei:s so that their steam consimiptions had

Pig. 117a.— Sectional Plan of 60" x 56" L. P. CYLmDBB Showing Adhisbion
AND Exhaust Valves.

to be determined by measuring the boiler feed. This neces-
sitated the adoption of adequate precautions to prevent leakage
from the feed mains, from the boilers and from the steam pipes.

448 SUPERHEATED STEAM AND REHEATERS IN COMPOUND ENGINES.

The steam was obtained in every case from Babcock & Wilcox
boilers, fed from a pump which handled only the water going
to the test boilers. These boilers supplied steam only to the
engine under test; they were examined for general tightness;

!<■

{<■

Fig. 117b.— SroB Elevation op 60" x 56" U P.
Cylinder Showing Valve Gear.

were shut off from the Holly return systems and had their
blow-offs blanked. The steam on its way to the engine
under test went first to a section of the main steam header,
which was isolated by gate valves from the rest of the header.
These gate valves were tested for tightness with full steam press-
ure on one side and atmospheric pressure on the other, and were

SUPERHEATED STEAM AND REHEATERS IN COMPOUND ENGINES. 449

found practically tight in all cases. In order, however, to pre-
vent even slight leakage the steam pressures were kept the same
on both sides of the valves during the tests. In most of the
tests the steam supply to the engines was superheated so that the
steam pipe drips could be closed; in the cases where wet steam
was supplied the condensation in the pipes was returned by
gravity to the test boilers.

6. The total leakage from the boilers and steam pipes was
determined by leakage tests after nearly every run. To make
the leakage test, the fires were allowed to bum down at the end
of the run and were kept in such condition as just to be able to
maintain the steam at the test pressure when the supply to the
engine was shut off completely. The observation of the rate of
lowering of the water level in the boilers under these conditions
gave the necessary information for determining the rate of leak-
age loss from the boilers and steam pipe.

7. The feed heaters were all tested and found to be tight.
The reheater coils were also found to be tight except in one
engine and in that case the only run made was without the
reheater in use.

8. The weight of steam condensed in the jackets and reheater,
when these were in use, was determined by collecting the con-
densed steam in a vessel of known capacity provided with a
gauge glass. The drainage from the receiver was determined in
a similar way. The arrangement of jackets and reheater pre-
vented the separate determination of the amounts of condensa-
tion occurring in each.

9. The diameters of all the cylinders were gauged when hot.
The clearances were not measured directly; the values used for
the calculations were given by the engine builders and were
determined by them from the working drawings.

10. The weighing scales were examined by the local sealer of
weights, and their accuracy in the writer's tests was ensured by
further proving with special test weights. All gauges, ther-
mometers and indicators were calibrated. The indicators used
in the tests of engine A were calibrated under steam pressure by
comparison with a mercury column at the Crosby Steam Gage
Company's works, and later were tested at the Engineering Lab-
oratory of Harvard University under steam pressure by compar-
ison with known rotating weights. The results of these two cal-
ibrations are given below.

450 SUPERHEATED STEAM AND REHEATERS IX COMPOUND ENGINES.

SUPERHEATED BTEAM AND REHEATERS IN COMPOUND ENGINES. 451

Number of
Instrument.

Nominal
Scale of Spring.

t .

Actual Scale of SrniNo.

Makxr'8 Naxb.

By Mercnry
Column.

ByRotatlng
WelghU.

Tabor

788
1926
4908
4909

100

98.9
99.7
19.48
19.6

98.8

Tabor

99.4

Crosby

19.4

Crosby

19.57

11. The load on the engine was entirely electrical, and con-
sisted of part of the station load supplemented when necessary
by an adjustable water rheostat load. There was no difficulty
in any of the tests in keeping the total load constant.

Tests of Engine A.

12. In June, 1899, the writer, assisted by students in engi-
neering of Harvard University, made the acceptance test of the
engines then installed at the L Street station of the Boston
Electric Light Company. A description of the plant may be
found in the " Electrical Worid and Engineer," for May 27, 1899.
Of the three similar units then installed, engine A (number 1 of
the plant) was selected for test because it had been running
Icmger than the others.

13. This engine is rated to develop 2,400 indicated horse-
power at .23 cut-off, 4,128 indicated horse-power at .6 cut-off
and has a maximum cut-off at .8 stroke. Its piston speed is 960
feet per minute; its cylinder dimensions 28 inches and 58 inches
by 48 inches. It is direct connected to a 1,500 kilowatt, three-
phase General Eleetric generator delivering current at 2,250 volts.
The fly wheel is 16 feet diameter and weighs 100,000 pounds.
The main shaft is 26 inches diameter; the bearings 21 inches
by 42 inches. The steam pipe is 10 inches, the exhaust 24 inches
diameter. The ratio of low pressure to high pressure piston dis-
placements is 4.3 to 1. The reheater coil has 777 square feet of
heating surface, but the steam supply to it was too small.

14. The engine was apparently in fii*st class condition; the
valves and pistons were tight when at rest. The steam was ob-
tained from two 650 H. P. boilers, which supplied saturated steam
to the engine. The amount of feed water was determined by
direct weighing.

15. Two full load tests were made: one with the jackets and

452 SUPERHEATED STEAK AND REHEATERS IN COMPOUND ENGINES.

reheater in use, the other with both out of use. The principal
results of the test are given in Table I. The combined indi-

TABLE I.
Qenrral Results of Tests on Engine A.

4
6
8
•7
8
9
10
11
12
13
14
IS
16
17
18
19
20
21
22
28
24
25
26
27
28
29
30
81
St
33
34
85
36
37
38
89
40

Namber of teit

Nominal load

Wilborwitboot jackets and reheateni

Darationof test, nonra

H. P. cylinder diameter, Inches

L.P. " ** "

Pinton rod diameter, inches

Clearance H. P. cylinder, per cent

L.P. ** "

Ratio L. P. to H. P. displacement

Revolutions per minote

Gauge pressure at throttle

steam qaulity at throttle, per cent

Initial steam pressure in H. P. cylinder^ by cards

Steam pressure in receiver, ganee

Snperlieat of steam entering I^. P. cylinder, Fahr

Effective vacuum in L. P. cylinder, by cards, inches

Vacuum in exhaust pipe, Inchei*

Temperatore of exhaust steam, Fahr

Commercial cut-off H. P. cylinder, hexd end

** ** '* •* crank end

" " L.P. " headend

** ** ** " crank end

I. H. P., H. P. cylinder

" L.P. "

TotalLH.P

Percentage power developed by H. F. cylinder

Total dry steam per I. H. P. per hour, pounds

Wet steam in jackets and reheater per l. H. P. per hour

Percentage of total steam used in j ickets and reheater

Percentage of cylinder steam drained from receiver

Quality at cut-off, H. P. cylinder

Quality at release, H. P. cylinder

Quality at cut off, L. P. cylinder

QualitV at release, L. P. cylinder

B. T. U. per I.H.P j)er minute, mensnred above ideal feed temperature.

Thermodynamic efficiency, per cent

B. T. U. per I. H. P. per minute, for Rankine ideal cycle

Thermodynamic efflcienoy compared with ideal cycle, per cent

Percentage saving by jacketing and reheating

Full

With

28.14
58.12
5.5
8.6
6.85
4.8
118.46
159
99.46
156
21

85.5
24.06
25.0
127
.20
.80
.29
.81
859
1,136
1,994
48

18.59
.92
6.8
1.6
82
86
92.5
95.5
24^
17.8
160
65.4
1.8

Pull

Without

28.14
68.12
5.5
8.6
6.8S
4.8
118.40
161
99.5
155
20

28.05
24.5
182
.895

.au6

.33
.8:i5
961.1
1,022.6
1.988.7
48.5
18.68

6.0
76
80
87
98
248
17.1
161

cator cards for the two tests are given in Figs. 119 and 120. A com-
parison of the two tests shows the conditions to have been very
similar during the two runs with some advantage in the matter
of vacuum for the first test. The reheater was not very effective
owing to the deficient steam supply — it superheated the receiver
steam 35 degrees Fahr.

16. The economic results as indicated by the steam consump-
tions in line 28 show practically no saving at all by the use of
the high pressure jackets and the reheater. The steam con-
sumption, however, is not a fair quantity by which to judge the
performance of the engine, since in Test 1 about 7 per cent, of
the total steam is rejected as condensation at about the boiler
temperature, whereas in Test 2 all is rejected at the condenser

8UFERHXATSD STEAM AND REH£ATERS IN COMPOUND ENGINES. 453

temperatiire. Since, in these tests, engine performances alone
are being studied, and the efficiencies of the feed heaters are not
considered at all, the performance of the engine is best measured
by the amount of heat that must be given to the feed water per
I. H. P. per minute, the feed being supplied at what may be
known as the ideal feed temperature. The ideal feed temper-

Engine A. - Test U
Full Load, with Jackets and
Reheater.
1994-I.U.P.

150^

- -

L.P.

100-

— tfO

\ '

^f»

H.P.

y''

LH.

8tr<

«*

1185|I.H.F
4

^
<*

,^_^

■--^

—

— ^-

17*^1 11 niAQ r<ii Vi

\ \ 1 1 \ \ 1 \ 1

80 100 130 140 l(

Jfar»«.L^

Am.3mnk JUI4 Co.,.\. T.

Fig. 119.

ature is obtained by mixing the condensations in the reheater
and receiver, each at its own proper temperature, with the con-
densation water in the condenser supposed to be at the exhaust
pipe temperature. The heat units per I. H. P., measured above
the ideal feed temperature, necessary to form steam such as
exists at the throttle, is an accurate measure of the performance
of the engine, and it is this quantity which is given in line 36,
and from which the calculation in line 40 of the saving by jacket-

454: SUPERHEATED STEAM AND EEHEATERS IN COMPOUND ENGINES.

ing and reheating has been made. An examination of the com-
bined cards in Figs. 119 and 120, or lines 32 and 33 of Table 1,
show that the H. P. jackets have been moderately effective, raising
the quality of the steam in the H. P. cylinder 6 per cent,
throughout the expansion. The small total gain by both jackets
and reheater indicate then that the reheater is of small value.

150

Engine A* - Test 2*

Full Load, without Jackets and

Reheater.

1983.7-I.II.P.

120 140 IGO

Fig. 120.

or may even be a source of loss when it superheats only 35
degrees.

Teats of Engine B.

17. The engines B, C, 2), E and F are all part of the Atlantic
Avenue plant of the Edison Electric Illuminating Company, the
station numbers of these engines being 7, 8, 9, 10 and 11 respec-
tively. This plant has been described in the " Electrical World

SUPERHEATED STEAM AND EEHEATEBS IN OOICPOUND ENGINES. 455

and Engineer," of May 18, 1901, and also in the paper by Messrs.
Moultrop & Curtis, published in Volume XXIII. of the Trans-
actions of this Society. The engines 0, D, E and F are aU of
the same size and are rated at about 2,400 indicated horse power.
Engine B develops only about one-half the power of the other
engines; it is also the only one of these engines fitted with a
surface condenser.

18. Owing to the comparatively short duration of test neces-
sary when the steam consumption of an engine is measured by
weighing the discharge from a surface condenser, it was practic-
able to carry out a much more complete series of tests with
this engine than with any of the others. The engine B is much
more distant from the boilers than the other engines tested, and
in consequence the superheat at the throttle is very low, and drop
of pressure in the steam pipe is considerable.

19. The engine is intended to run at 100 revolutions per
minute and to develop 1,200 indicated horse-power at .22 cut-off
with 160 pounds steam pressure and 26 inches vacuum. At .6
cut-off it develops 2,200 indicated horse-power. The cylinder
dimensions are 23 inches and 48 inches by 48 inches, and the en-
gine drives an 800 kilowatt General Electric direct-current gener-
ator. The reheater consiats of coils of brass pipe aggregating 440
square feet total heating surface. The high pressure clearance
is 2 per cent.; the low pressure is given as 3 per cent. The
diameter of the fly wheel is 15 feet and its weight 65,000 pounds.
The main bearings are 17 inches diameter, 35 inches long; the
diameter of the shaft between the bearings 20 inches. The
diameters of the steam and exhaust pipes are 9 inches and 20
inches respectively.

20. Ten separate tests were made on this engine in order to
determine its economy at different loads, both with and without
the jackets ajid reheater in use. Of these tests, five (viz.: tests
3, 4, 6, 8 and 10) were with the jackets in use, and were with
loads increasing from one-quarter load to an overload of one-
quarter. Four tests (viz.: tests 5, 7, 9 and 11) were without the
use of jackets and reheater and were with loads varying from
one-half to an overload of one-quarter. Another test (number
12) was made at full load with jackets and reheater in use and
with a greater boiler pressure than in the previous runs so as to
test the engine more nearly under its rated conditions.

21. The principal results of these ten tests are given in Table

456 SUPEBHEATED STEAM AND REHEATEB8 IN OOMPOUND ENGINES.

M
O

QQ O

u
O

*- 1 *ss558?asi5|ifeg

t^tiSSc

SS8 'OOOO « M

15.16

"'4!97'
74.0
78.0
860
89.0

267

265

257

61.6

0># ^ MCDOM 'ofttOOTOO^CDaO

14.57
0.906
8.9

88.0

04.0

98.0
261
255

250

16.7
168
66 1
4.0

14.41

"i'.in

60.0
75.0
76.0
88.0

267

SS8 ;

o ^ Sioe«oio«8nooc>ioo

SS9 ioooo

M iD«0

88s ^oooo
«-^' :8888|g g

17.0
152
60.6
7.6

xi«* ixeoooooo ^ao8ooot-<o

t>oooo

»8S8S|5 5 2|S

jS -i«iioao8ao-<«'OOOOOaoQO

?l!8

8 S'

oa?^ o.gg 'si-gssssiss

■".6

£11

0-7 ^

la a

B o r*

n i s = e^ s^ 8

S8S8gg g S88«'

5888|g g SSS

t;:^!

sis.

ill

111

c c S

5 • • :.2jK

a ^

g 8 be

i-tO* CO ■«

««,t.oo«o-««;2;o25:; • 2858l85aSSiS 8 8S8S

SUPERHEATED STEAM AND REHEATERS IN COMPOUND ENGINES, 457

II. The combined indicator cards are shown in Figs. 6 to 15.
On examining the conditions under which these tests were made
it will be seen that the superheat was slight in all cases, but was

§ S 8 B 8 S

Engine K - Test 3.

One-quarter Load with Jackets

and Reheater

309-I.II.P.

•

— -

■-

__)*_

s
3

•

^\'iS

8 S 8 8

a 8

8 9 8 8 S
-80JUSS9JJ a)n[osqy

greater during the series of tests without jackets and reheater.
The vacuum was not constant, but fell at the higher loads; it
averaged higher than in the tests of any of the other engines.
In order to eliminate the eflFect of variation in the vacuum on

458 SUPERHEATED STEAM AND REHEATEBS IK COMPOUND ENGINES.

the economy of the engine a correction has been applied reduc-
ing all the results to an eflFective vacuum of 26 inches. This
correction is obtained by adding to the mean effective pressure

§ 8 S ^ 8 S

Engine B. - Test 4.
Half-Load with Jackets and
Relieater.
552-I.H.P.

~j_

•2-S-

- I
4

§ -

8 S 8 8 8

S 8 8 it 8 8 S
< eajnssajd a^njoeqv

of the low pressure cylinder the difference between the absolute
pressure corresponding to 26 inches vacuum and that correspond-
ing to the actual effective vacuum shown by the cards and
recorded in line 11. The corrected results are given in line 29,

SUPERHEATED STEAM AND REHEATER8 IN COMPOUND ENGINES. 459

and for the purpose of comparison of results the figures in this *
line are the most satisfactory to use, since they not only eliminate
the efficiency of the feed water heaters but also reduce all results
to a common vacuum.

Engine B. - Test 5.
Half-Load without Jackets
and Reheater.
548-I.H.P.

i <

\ i

t i

5 i

p;^
J

^ \

^^^

■

— '

g|8§§8SSS

8 S il S S S
— 8a.in888Jj o:^nio8qy

22. The corrected heat consumptions per indicated horde-
power per minute, and the method of variation with the engine
load, are shoA\Ti graphically in Fig. 131; the same figure shows also
the i>ercentage of the total steam used in the jackets and re-

460 SUPERHEATED STEAM AND BBHBATEKS IN COMPOUND ENGINES.

'heater. The most strikiDg feature of the curves is the small
variation of the efficiency of the engine when jackets and reheater
are used throughout the range of load from one-half load to one-
quarter overload. Within this range the total variation of econ-

•

Ensinc B. - Test 6.
Three-quarters Load with Jackets
and Reheater.
845-I.H.P.

- 1 —

■ "4

S«

1-4

—

i^ ^

$ S

i ^

^ I

I i

S 1

I %

I \

2 X

I s

\ %

I t

% %

3 •=

% I

a
:3

-99.tn888jj* dmiosqy

omy did not exceed 2 per cent., and the economy was greatest at
about the rated load. It should also be noted that the percentage
of the total steam used in the jacket and reheaters is lowest when
the efficiency is greatest.

SUPERHSATED 8TEAH AND REHEATEBS IN COHPOUND ENGINES. 461

- 23. Without the jackets and reheater in use the economy in-
creases with increase of load, and in such way as to indicate that
there would be but little advantage in the use of jackets and
reheater when the engine is much overloaded. With light loads

8 !

S S 1

S 8

^ -

Engine B. - Test 7.

Three-quarters Load- without

Jackets and Reheater.

831-I.H.P.

•

*«

■ c
s

—

— ■

)\.

8 !

\ %

i i

■4 »■

i s

5 t

i S

•

-80.11

1S88J

\ !

5 %
in[08

d o

§1

up to full load the saving by jackets and reheater ranges from 9
to 4 per cent, and is obtained by using from 10 to 7 per cent, of
the total steam in the jackets and reheater. The increase in econ-
omy with the load and the reheater in use would have been

462 SUPERHEATED STEAM AKD REHEATERS IN COMPOUND ENGINES.

greater had the reheater been of greater capacity. As is seen in
line 10, the superheat of the steam entering the low pressure cyl-
inder decreased from 100 degrees to 46 degrees as the amount of

8 8

Engine B. • Test 8.

Full Load with Jackets and

Reheater.

1047-I.II.P.

«f4

_Pk

Vyt

•^

*^^

—

— '^

§ I

-8

g|8gS§88

^ 8 S 9 8 8 S
•* sainssdJj e^nioaqy

steam used increased. This is probably the reason for the close
approach of the two heat consumption curves in Fig. 130. An
examination of the indicator cards, Figs. 120 to 129, and of the
steam qualities tabulated in lines 23 to 26 of Table 11., shows
that the qualities during expansion in the high pressure cylinder

SUPERHEATED STEAM AND REHEATEBS IN COMPOUND ENGINES. 463

are but slightly affected by the steam jackets. On the other
hand, the qualities during expansion in the low pressure cylinder
show very clearly the beneficial results of reheating, especially

S 3 § S S §

^ 8 s

in the low load tests where the superheat in the receiver was
greatest. The rise of the end of the expansion line of the high pres-
sure cards at light loads, Figs. 121, 122 and 123 results from the
return of steam from the receiver to the high pressure cylinder

464 SUPERHEATED 8TEAH AND REHEATEBS IN COMPOUND ENGINES.

due to the lifting of the exhaust valve from its seat when the
pressure in the cylinder becomes less than that in the receiver.

Tests of Engines C, D, Band F.

24. The engines C, 2?, E and F are all located in the new
engine room of the Atlantic Avenue station of the Edison

Engine B. • Ted tO.

Ono and one-qnarter Load "with

Jackets and Reheater.

1498-I.H.P.

<
1

i^l!^

SJ-

■J

p.*

'Wl

S?SS388£

S S 9 S 8 3
-soanssojj ajn[osqy

8 ^

Electric Illuminating Company, and are almost identicaDy sim-
ilar units, each rated at about 2,400 indicated horse-power and

SU7EBHEATED STEAK AND REHEATEBS IN COMPOUND ENGINES. 465

direct connected to a 1,600 kilowatt direct current General
Electric generator. The cylinders are all 29 inches and 60 inches
by 56 inches; the fly wheels 18 feet diameter and 130,000 pounds

2 8 8 g

One and one-qaarter Load without

Jackets and Beheater.

1516-I.H.P.

{

,/j

///

9
(7

)l 1

^ /

f »-

I-

_^J

% %

\ %

! 1

\ \

I S

\ %

i i

\ s

\ %

s s

\ s

s I

"o ^

o>|

-so.msso.ij 0)n]O8qy

in weight; the main bearings 24 inches diameter and 48 inches
long; the shafts between the bearings 27 inches diameter; the
steam and exhaust pipes 10 inches and 24 inches diameter respec-
tively. The high pressure clearances are 2.75 per cent.; the
low pressure clearances 4 per cent

466 8UP£RH£AT£D STEAM AND REHEATERS IN COMPOUND ENGINES.

25. Engines C and D have each 740 square feet of reheating
surface in the receiver; the other two engines have each 800
square feet. The only other diflFerences between these four

100
90 >

oo«

can

iliti

" ^

— "

IM)

4RA

1*0

14M

Engine B. - Test 12.

Fall-Load with Jackets and

Iteheater.

U04-I.H.P.

IIU

100

jw 70

|50

1415

I.H.P

CD *^

P^ 80

87.6^

s..

0801

.H.P.

•*>-

"^::

:r -^

===

^''

Volumes-

100

190 140

Fig. 130.

engines are minor diflFerences in the valve gearing. Engines
C and D were tested in 1901; engines E and F in the following
year. All these four engines exhaust into a central condensing
plant consisting of jet condensers and Blake twin vertical air
pumps.

St7PERHEAT£D STEAM AND REHEATERS IN COMPOUND ENGINES. 467

The measurement of the feed-water was accomplished by
the use of 2-inch Worthington hot water meters, one at each
boiler employed during the list. These meters were calibrated
by sending known weights of water through them at each of

'^ So OD »• O

/I ■

^ ^ !P «;

8 »

'^ CO

three different speeds. The middle speed in each case was ap-
proximately the average speed of working of the meters during
the test. The meters showed errors which were practically con-
stant throughout the range of speeds employed, so that their
readings could be used with complete confidence. Boiler leak-
age tests were made as usual after each run.
31

468

SUPERHEATED STEAM AND UEHEATERS IN COMPOUND ENGINES.

•J

go
2 o

2
o

SUPERHEATED STEAM AND REHEATERS IN COMPOUND ENGINES. 46^

26. The following tests were made on these engines :

Engine C was tested at full, three-quarters and half loads \vith

jackets and reheater in use, and at full load without steam in the

jackets and reheater.

Engine D was tested at half load and at full load, both w^th

and without steam in the jacket and receiver, but as a drip

Fig. 133.— Engine C.

connection, which should have been closed^ was found to be
partly opened after the two half-load tests, the results of these
tests are not given here, and the table of results contains only
those obtained at full load.

Engine E was tested at full and at half load with jackets
and reheater in use.

Engine F was tested only at full load, with steam in the
jackets and reheater.

470 SUPERHEATED STEAM AND REHEATERS IN COMPOUND ENGINES.

27. The principal results of these nine tests are collected in
Table III. The combined indicator cards for the same tests are
shown in Figs. 134 to 142.

28. The tests on engine C show the efficiency of the engine to
change very slightly with change of load, — ^the better economy
at half load being principally due to the better vacuum, though
this is offset in part by the lower superheat of the admission
steam.

The saving by jacketing and reheating at full load is only
3 per cent. The jacketing is shown by the steam qualities in
the high pressure cylinder to be of no value with 80 degrees
admission superheat, and the reheater is of but small value when
it superheats only 60 degrees.

29. The economic results of engine D are better than on
engine as a result of higher superheat of the admission steam
and a better vacuum. The effect of the higher superheat is most
plainly shown by the very high steam qualities during expansion
in the high pressure cylinder. The saving in this engine by the
use of jackets and reheater is 4.5 per cent. — a little greater than
in engine C. The full load tests on engines E and F gives
results agreeing very closely with those obtained on engine D —
the running conditions were very similar in all three cas^s.
The half load test on engine E gives the same economic result as
the full load test on the same engine.

Tests of Engines O^ S and K.

30. These engines are part of the new plant of the Cambridge
Electric Light Company, and correspond to station numbers 1, 2
and 3 respectively. A description of the plant is to be found in
the " Engineering Record " for November 1, 1902.

31. Engines G and H are exactly similar units, 18 inches and
38 inches by 42 inches, developing 760 indicated horse-power at
.24 cut-off with 135 pounds initial steam pressure and 26 inches
effective vacuum, and direct connected to 600 kilowatt, 60-cycle
alternating generators built by the General Electric Company.
Engine K has cylinder dimensions 31 inches and 64 inches by 48
inches, develops 2,320 indicated horse-power at .24 cut-off with
135 pounds initial steam pressure and 26 inches effective
vacuum and is direct connected to a 1,500 kilowatt General Elec-
tric generator.

SUPERHEATED STEAM AND REHEATEBS IN COMPOUND ENGINES. 471

fr4

< §

J3 St-t-«iO*-S8'*l-'-'iO

o <d9S«o

22 ^ *0

^^9

•0 le o -« '^ » o e» 9

- eg*

B98:j^«88es'-«S8

8 S'

S«»'«OOOk^2*'

>o<^e« «

w*-5*a|es:sgs«25^2

&s^

8 ;:

23 -^ I'

82^

«o<«iiOio«aDH«*t>-o«»et ao

o2:f^^8|?g'«S358g||5?

00 HSm

S S '^

09 t«

SSSS^SS?

ssssgsse

rift.

8a88gs|e^

S£SSgSS(2

risssgssii

g St'-lOiO*-!

gt^OO 0090

8 S

'SS3S|r:SS

* '^ M sf 'i - s *; a sj cLi^ ^
H ^ [1- £u CKJ'O'o^ t- a &- a*

^•t«^««t-Q0<»oj;Q«j;-5Ot5 S SiSS^SISISSSiSg^SS

472 SUPERHEATED STEAM AND REHEATERS IN COMPOUND ENGINES.

32. A slight leak was found in the reheater coils of engine JJ,
so that this engine was run only without steam in the reheater.
Engine K had been used but very little, and had not worn down

iUL

00
80
TO

Stei

im (

Jual

ties

180

170

140

180
190

Engine C - Test \Z.

Pull Load with Jackets and

Reheater

2267-I.H.P.

110

100

60
50

•

)98.4I

H.P.

48.

g 80

O 20
'S in

1168.(
51

I.H.I

====^

^ n

Voh

lfar*t. X

imee

•4

a/13

to such good condition as the other two engines. Its governor wae
found to stick at very light loads, and the consequent hunting
made it impracticable to obtain satisfactory friction cards from
the engine when running without load. In all other respects

SUPEBHEATED 8TEAH AKD REHEATEB8 IN COKPOUKD ENGINES. 473

the conditions were favorable for satisfactory tests of all three
engines.

33. During the tests the whole of the steam generated in the

100
90
80
TO

tear

iQi

alit

170

lao

Engine C - Test 14.

Full Load without Jackets

and Reheater.

2239.8-I.H.P.

120

100

■70

C4»

1S17J

I.H.P

»%

S M

■

-^^

1
■^10

lOStSt
45

.H.P.

:s=-

^^^^

Volumee-

GO 80

Fio. 185.

100

120

140

plant was used for the engine under test and its auxiliaries.
The main engine was supplied from one or more main boilers;
the auxiliaries from an auxiliary boiler. The feed to the main
boilers was determined by direct weighing. The engine under

474 SUPEEHEAT^ STEAM AND REHEATEBS IN COMPOUND ENGINES.

test carried the total station load, and to this was added the
power absorbed by a water rheostat, continuously adjusted so as
to keep the sum of the loads constant. The tests were made in

■"

Sic

Qua

itie«

ITO

100

J80

Engine C - Test 15.

Three-quarters Load with Jackets

and Eeheater.

1831-I.H.P.

ino

w
80
TO
60

« 40

Ph 80

^4I.H
47.8^

P.\

—

)

S 9n

■^

"" -J

9671
52

U.P.

-^^=^

=^^

100

Volumes-

Jfarft*, £.&

Pig. 136.

100

120

140

April and May, 1903, and were conducted by the writer, assisted
by students in engineering of Harvard University. On engine
G two full load tests were made, one with, the other without
steam admission to the jackets and reheater. Only one test

SUPERHEATED STEAM AKD REHEATEBS IN GOMPOUKD ENGINES. 475

was made on engine H; a test mth about 20 per cent, overload
and without steam in the jackets and reheater. Engine K was
tested at full, half and quarter loads with the jackets and re-
heater in use, and at half load without them.

■^^

100
90

Qua

litiei

^70-

IflO -
lao-
1^ -

180 -
120 -

Ei^

grille

C-

Test

t6.

HO -

Half Load with Jacketa and
Beheater

100 -
90 -
80-

1238

.9 I.

H.P.

iw ■ -
80-

80--

-k

O An \

L>»71.^

\ «J*

s"-

66(L2

.H.P.

■■* ,

2 *"

"■ ^

S-

"0 20 40 60 80 100 120 146

ATol 11 TTl Af

Fro. 137.

MvU^XJI,

34. The principal results of the tests on engines G, B. and K
are collected in Table IV. The combined indicator cards for
the seven tests are given in Figs. 143 to 149.

476 SUPERHEATED 8TEAH AND REHEATERS IN COMPOUND ENGINES.

Sao ^o^Sok«oS8aoee«HieSSS

8«^

^aoj5j-5«iodW2j8ggo5g«gg • • • 'sa«»2|ggs *« *'- *SS8Sg2|SSJ

O9oeot«<o et

^^oi

Saaj3o»aog-5^«Vc5gjrggjj:

2 ooeo9oc

'SSS8Sgd|U:gf

8S OkS27-4iO OkOO^V^^M

g S8 o»SS'«>aDoiMo»«e

5 -- -

^zs^Z^Sti^^^^^^SSiM^^^Bm^^^^nk^^^:;^^^^^^^^

SUPERHEATED STEAM AND REHEATEBS IN COMPOUND ENGINES. 477

35. The engines O and H are the smallest of the whole series
tested. The results of the two full load tests of agree very
closely with the results of the tests 9 and 8 of engine B under

lu)
90
80

' St

litic

earn

Qui

4M £^.

170 -^

lao

150 —

MU —

Engine D* - Test M.

Full Load with Jackets and

Reheater.

2201.6-I.H.P.

ISO -

110 ■ -

100 - -

00 --

80 - -

70 -

tOCULl

lpN;

£ 40 —

2 80 —

^55^

« 20 —

UW.I

LH.P
9%

10 —

::>

Volui

nes-

Fie. 188.

full load, although this latter engine is 50 per cent, larger than
6. The reheaters gave practically the .same superheat in both
cases; the steam pressures in the high pressure cylinders were
practically the same, so that the disadvantage of smaller size and
poorer vacuum in engine Q is just about offset by the greater

478 SUPERHEATED STEAM AND REHEATER8 IN COMPOUND ENGINES.

superheat of its admission steam. The gain by the use of jackets
and reheater was about 7 per cent, in both cases. It will be
seen by an examination of the steam qualities in the high pres&-

Stea

m ^

uali

tieT

180
I'm

ftn

15m

Ensine D«-Test )8.

Full Xx)ad without Jackets and

Eeheater.

2215.1-I.H.P.

14tf)

inn

A fm

.) RA

a
C

89.01

■■>

g40

6S.8

/■

Q 20

< in

(

1086.1
46

I.H.P

:::::

Mtrt

.^^

^olu

mes

9
Fi<

8
3. 18t

Ml* 2Mt

ure cylinder that the jacket is effective in this comparatively
small cylinder even with moderate superheat at the throttle.
The performance of engine H at 20 per cent, overload is con-
siderably better than that of engine B under approximately sim-
ilar conditions. That the superheat of the admission steam is

SUPERHEATED STEAM AND REHEATEBS IN COMPOUND ENGINES. 479

probably responsible for this is suggested by the higher quality
of steam in the high pressure cylinder. This particular
engine, JET, was known to be in better condition than G, its

CO
100
150

Steam (Qualities

100
00
80

Engine E. - Test 19.
Full Load with Jackets and
Reheater.
}.1-I.H.P.

120 140

Fig. 140.

pistons were tighter and it had run with unusual smoothness
from the beginning. The distinctly better performance of this
engine, as compared with the similar engine 0, is to be accounted
for only by the superior condition of the engine.

36. The tests on K again emphasize the fact brought out on

480 SUFEBHEATED STEAM AND REHEATERS IN COMPOUND ENGINES.

tests of C and E that the efficiency throughout the ordinary
range of loads is practically constant in these engines, though it
naturally falls off when the load is reduced to one-quarter. The

170

leo

150
140
180
ISO

100
00
80
70
GO

(» 60

S40

C-iao

^ 80

-510

100

Ste

am J

iual

ities

80
<7n

Engine E.. Test 2a

Half-Load with Jackets and

Beheater.

1258.9-I.n.P.

]

\ I.H.P.

\50^^

626

I.H.F

49.7jr

Volumes-

Fig. 141.

100

120 140

advantage of jacketing and reheating is again found to be about
7 per cent

Temperature-Entropy Diagraans.

37. Temperature-entropy diagrams for the tests on engines
0^ H and K are given in Figs. 150 to 155. There is one omission,

SUFEBSEATED STEAM AND REHEATEBS IN COMPOUND ENGINES. 481

that of test 27. The combined indicator card, Fig. 148, for this
test shows that considerable steam returns from the receiver to
the high pressure cylinder after release^ and the assumption of

100

earn

iliti.

170

lao

tRA

140

lao

190

100
90

Engine F« - Test 2\.

FuU Load with Jackets and

Eeheater.

2171.1-I.H.P.

(b50

|»

T' rm

lioo

UI.I

4&A%

— ^

l.fl

182.7
SS.

.n.p

^--

^ 10

— ■

Vohinics-

100

ISO 140^

Fig. 142.

constant weight of acting substance, necessary for the construc-
tion of the temperature-entropy diagram, would give superheated
steam in the exhaust from the high pressure cylinder. It is
then impossible to draw a temperature-entropy diagram without
^ossly distorting the facts.

482 SUPERHEATED STEAM AND BEHEATEBS IN COMPOUND ENGINES.

38. In each of the Figs. 150 to 155 the high pressure and low
pressure cards have a common pair of reference curves. As
drawn they are valuable mainly in indicating when the heat
losses occur. They assume that the total steam acting in each

150
140
180
190
110
100

m «
2
g SO

£40

-2 80

- —

teai

lali

;ie8

—

Engine G* - Test 22.

Full Load Test, without

Beheater.

714.7-I.H.P.

8.91.

a.p.

^5;

•

{

S»

1.8 I.I
46.3^

i.p.

i::^

t; -^

■

Markt, LA

8 16

Volumes, Cu.Ft.-

48 60

Fro. 143.

cylinder (the sum of the steam admitted and of the clearance
steam) weighs just one pound. The relative amounts of work done
in the two cylinders will consequently be shown correctly in such
a temperature-entropy diagram only in the case when the actual
total weights of steam acting during expansion are the same in
both cylinders.

39. In the engines under consideration the weights of steam

SUPERHEATED STEAM AND REHEATEBS IK COMPOUND ENGINES. 483

acting in the two cylinders were different in every case; (a)
because the weights of clearance steam remaining over in the
two cylinders is different in every case, and (6) because in many
of the tests part of the steam admitted to the high pressure

150
140

lao

leo

110
100
90
80
70
00
50

® 80
< 10

Ste

Qua

litic

—

Engine G. - Test 23.

Full Load Test, with Keheater.

715.5-l.H.R

7.11.
48.6

H.P.

«H.4
51

I.H.I

4)r

oof
sojT

"0 8 10

Volumes, Cu.Ft.

48 56

Fia. 144.

cylinder does not go to the low pressure cylinder, but is drained
from the receiver. In order, then, to get a correct relation of
the temperature-entropy diagrams of the two cylinders, each
diagram should be drawn with respect to its own pair of refer-
ence curves, and each pair of reference curves should preferably
be drawn for the actual weight of steam present during expan-
sion per revolution.

40, Fig. 156 shows the temperature-entropy diagram for the

484 SUPEBHEATED STEAM AND REHEATERS IN COMPOUND ENGINES.

full load test (No. 25) on engine Ky drawn to give correct quan-
titative results for one revolution. The curves oi, cdy are the
reference curves (i.e., the entropy curves for water and for dry
and saturated steam) for the high pressure card; the low press-
ure card has ef and gh for its reference curves. The position

90^

160

150

140

180

120

110

100

g 50

^ 80
^ 10

^am

Qui

ilitii

Engine H. - Test 24.

Overload without Relieater.

857.6-I.H.P.

462

.8 I.I

.p.

bl%

V —

391

8I.B

.p.

:::::;

irz

— .

46^

96%
80^

Jtfarfta,£jK

Volumes, Cu.Ft.

48 56

Fig. 145.

of any point on these cards with respect to the proper pair of
reference curves gives the quality of the steam at that point on
the assiimption that all the admission and clearance steam are
present in the cylinder. The areas of the cards, when multiplied
by the product of the scales of ordinates and abscissae, give the
heat equivalent of the actual work done per revolution.

41. To find the thermodynamic efiiciency of each of the cyl-
inders and of the whole engine some additional construction is
necessary. The diagram must show the heat added; that is,

SUPERHEATED STEAM AND EEHEATERS IN COMPOUND ENGINES. 485

the heat of formation of the admission steam measured ahove
its proper starting temperature. It is necessary, then, to ehm-
inate the clearance steam from the diagram. To accomplish
this, take the points k and I on the two cards corresponding to
the beginnings of compression. At this point of the cycle experi-

100
iSO
1-10

lao

120
110

aoo

00
J^ 80

S 00

S. 50

^__

■^

;am

alit

'eV

Engine K* - Test 25.

Full Load with Reheater

2184-I.H.P.

|im

L8T.

4S.8J

\—

1190

2I.B
61.85

.p.

~-_

96%
90%

e&%

85 50 75

Volumes, Cu.Ft. >

100

1« 150 175 300

Fia. 146.

ment has indicated, and it is customary to assume, that the clear-
ance steam, which alone remains in the cylinder, is dry and satu-
rated. The point Je may be taken as a point on a new reference
line giving the entropy of the admission substance as water at that
temperature and the new reference curve Jco may be drawn for
the weight of substance admitted to the high pressure cylinder per
revolution. The dry and saturated steam reference curve np
for the same weight of substance is constructed as usual by tak-
ing horizontal ordinates such as op, fcn, of length equal to the

486 SUPERHEATED STEAK AND REHEATEBS IK GOMPOUKD ENGINES.

increase in entropy in converting the weight of water substance
admitted per revolution from water at a certain temperature into
dry and saturated steam at the same temperature. The point n
of the new reference curve np will fall on the curve cd for the
total steam acting, because the point h has its entropy equal to

100^

100
150
140
180

190
110
100
00
80
TO
S 00

^90

St^am

Qualities

Engine K. - Test 26.
Half Load with Reheater.
1195.9-LH.P.

9)%
70%

'0 25 80 75

Volumes, Cu.Ft. —

Mmrk^LJB.

Fig. 147.

the sum of the entropy of the dry and saturated clearance steam
plus the entropy of the admission steam as water at the same
temperature.

42. In a similar way a new pair of reference curves, Zj, rs,
can be drawn for the low pressure cylinder, eliminating the clear-
ance steam.

The area under Bop (measured down to absolute zero of tem-
perature) gives the heat supplied to the high pressure cylinder
if the steam admitted is dry and saturated; as it is actuaUy

SUFEBHEATED STEAM AND REHEATERS IN COMPOUND ENGINES. 487

superheated, the area imder Bopt is the correct measure of the
total heat going to the cyUnder per revolution. In finding the
efficiency of the engine there must be added to this the h^at given
up in the high pressure jackets and the reheater. The heat given
up in these two places cannot be separated in these tests, so it

100 ji
00^

' —

teai

lalit

les

11 A

Engine K. - Test 27.
Quarter Load Test with
Beheater. y
71&-T.H.P.

IIM -

IW ~

w -

, 80 "

70-

« w- -

S HA

l„

J »

10
0-

885.3

.H.P

■ — -

25 50 75

Volumes, Cu.Ft.

100

125

150

175 200

Fio. 148.

will all be assumed to be given to the high pressure cylinder.
The steam going to the jackets and reheater gives up only its
latent heat and superheat ; consequently, if pu is the increase of
entropy during vaporization of the steam used in the jackets and
reheater per revolution, and if uv is the entropy line for the super-
heating of all the steam going to the engine per revolution, the
area under Bouv gives the total heat going to the engine per revo-

488 SUPERHEATED STEAM AND REHEATERS IN COMPOUND ENGINES.

lution. The heat in that part of the steam going through the
cylinders is measured above the low pressure exhaust temper-
ature, sfid that of the steam used in the jackets is measured above
the saturation temperature of steam at the admission pressure.
The area under Bauv gives then the total heat supply per revolu-
tion measured above the ideal feed temperature. The work done
by the ideal (Rankine) engine working under the same conditions
is represented by the area BouvD. The efficiency of the actual
engine compared with the ideal

__ H. P. area 4- L. P. area
"" B ouv JJ

The efficiency of the high pressure cylinder, considered as a
separate engine, compared with the ideal •

_ H. P. area
~" X uvw'

The efficiency of the low pressure cylinder compared with the
ideal

— ^' ^' ^^"^^
"" Aqsy C*

43. It is to be noted that Rankine's unjacketed cycle has been
used as the ideal although the high pressure cylinder is jacketed.
It is also to be observed that the efficiency of the high pressure
cylinder as defined above is less than its actual efficiency because
of the assumption that the heat of the reheater steam goes to
the high pressure cylinder.

The Value of Sigh-pressure Jackets amd of IieheaiM%g.

44. There are certain facts which the writer believes may be
postulated with reference to the effectiveness of jacketing and
reheating, and which are supported by the results of these tests.

45. The saving by jacketing varies with the following factors :
(a) It increases as the cutroff becomes earlier.

(h) It decreases as the superheat of the entering steam in-
creases.

46. There is no saving by reheating when the reheater does

SUPERHEATED STEAM AND REIIEATEBS IN COMPOUND ENGINES. 489

no more than to dry the steam. If there is any advantage in
using dry and saturated steam in the low pressure cylinder over
the use of wet steam, it can be obtained by the use of a separator
between the cylinders. In tests 2, 7, 9 and 24, when the re-
heater was not in use^ the receiver acted as an almost perfect

25 50 75

Volumes, Cu.Ft.

Mmrk»,L.iL

150 175 aoo

Fig, 149.

separator, taking away all the condensatiou which calculations
showed to have occurred in the high pressure cylinder. If the
reheater merely vaporizes the condensed steam at the expense of
a practically equal quantity of high pressure steam it is not only
non-effective but is also, probably, a source of actual loss, since
more work could have been obtained from the total steam used
if it had all gone into the high pressure cylinder. The reheater

490 SnP£BHEA.TED STEAM AND REHEATEBS IN 0OMPOT7ND ENGINES.

then should be regarded merely as a superheating device for the
steam entering the low pressure cylinder, and it may be expected
to be more effective the greater the amount of superheat it gives
the receiver steam. It is probable that the reheater would be
more effective if the only work were superheating, as it might be
if the wet steam exhausting from the high pressure cylinder
passed through a good separator before reaching the reheater.

Engine G« - Test 22*
Full Load without Reheater.

800

S750

,^^

oTOO

—

S660

)

H
5

- —

[

— -

O a^

CO 600

.2 .4

Entropy

Jlbrla, £.&

1.0

Fig. 150.

1.4 1.6 1.8 «.0

Am.Btaik JM> <W^ T.

As the reheater wiU have less to do when the engine is running
at low loads it may be expected to give a higher superheat at low
loads, and consequently to be more effective. With the above
in mind, an examination of the tests already described (the prin-
cipal results of which are collected in Table V. for more easy
inspection) throws some light on the conditions under which the
jacketing of the high pressure cylinder and the practice of i*e-
heating are desirable, and shows the saving to be expected from
them in large size engines.

SUPERHEATED STEAM AND REHEATERS IN COMPOUND ENGINES. 491

*'SS3S|^3

J- to e< eo Oft f!! iC

oeoo m

ssssgss^ss

ssssg^s

-«^§a*5|s^a|s j-'seffisiss

« «e« tooi

2«? I a55a*s8|s*

e8a5$|2S^

a»«o««o <<«•

»«-|§a5g$gs|gj§2 j-iggggggsg

•<

a '

-- n

»2^§a95S5:|8;gS

'*8?e8|SS

s»«»!| S9?S28»|S''

*«^§§5?S55^S5|

»8S8SgJSS

SSSS^SS^"

ssssgss

••^•-|lss5Ks||s|2 :«eeteg|Ss

— <-r

9aD«ococo-<(fio eo^eo f-i

g sa*s^s8|s»'-ssas|£:8-'ss

111
111

•J3

Si"

3 a r* o 3 r '■■ "" -"

:^2

M S %• '"^ ^ u c

— -i"o

^^2

eSl-

'V 10(000

^ ooi-»"^S8

SK«^ I SS^SJggSSBSS'-* *88a88|Sf2»^

SwS|ls2Sggs|8|S :«ag8S8|5:S

ja <-i30 M^S iO ettOC059>0 ^

go- I S83g{2^8SS« *«aB88|5;fe*-'SS

-r-« Sec

8«:« g 88S8g8a|d»

2«- I 8gSS?f8«|d^

g§8SSg8^^Sn

oo^oeo ko-^

^^^ % 8S8|$8a|;:^

2«^g 88S|^SS|«<*'

i3 lo «> [^e»

2«»| 8SSfeSfe|g|2'^

j '-'5J03 ^ io®t»aco»o»-

Eli
I!!
sll

£.

d a o r; o ^ ■ " :.-^ -

8S88gs'^

$:2e

ssssgsrf

sSoS;2|?52i

*o-^ta t-

8S8SgSj2'|8

S888§SSs

. C2 S 9

£ :«

'^%%':L

iass

i:^::^:^ ' C t «-"-d
I fl a ^ ^ ;i T I, ^D y

^oteo "V *^®^®^S^^*'ZH!SS^S^Sa«SSSSeS(3

4:92 SUPERHEATED STEAM AND REHEATERS IN COMPOUND ENGINES.

47. In engine A, where saturated steam is used, the jackets,
at full load, keep the steam about 6 per cent, dryer in the high
pressure cylinder and the reheater superheats 35 degrees. As
the greater part of the 6.8 per cent, of the total steam which is
condensed in the jackets and reheater must have been condensed
in the latter place, it is obvious that the reheater cannot have
contributed anything to the small total saving of 1.3 per cent.

820

•

ifCtl

740

^ -m

[ (

-"^

O nan

Engine G* - Test 23.
Full Load with Reheater.

S MO

A
\

500

.2 .4

Entropy

1.0 1.2 1.4 1.C 1.8 2.0 2.2

Fig. 161.

48. The conditions in engine B are much more satisfactory for
effective action of jackets and reheater, especially at the lower
loads. The engine is smaller, and its reheater is more effective.
At half load, with 75 degrees superheat of the receiver steam,
the saving was 9 per cent. ; at three-fourths load with 60 degrees
superheat there was 7 per cent, saving; at full load with 46 degrees
superheat there was still 7 per cent, saving, and even at one-

SUPERHEATED STEAH AND REHEATERS IN COMPOUND ENGINES. 493

"ID

1 ^Mk

1 T«

'-m

)

;;n9

)

^^^

t3»

Engine H* - Test 24.
Overload without Reheater.

.2 .4 .6

Entropy ^

1.0 1.8

Fig. 152.

1.4

1.0

1.8 8.0 2.2

quarter overload with 26 degrees superheat there was 4 per cent,
saving.

49. The larger engines C and D with 80 degrees and 98
degrees initial superheat and 60 degrees superheat by the re-
heater show but 3 per cent, and 4.5 per cent, saving respectively.

60. The engine is only one-third the power of engines C and
P, consequently the jackets are much more effective (raising
the steam quality 10 per cent, in the high pressure cylinder), so
that with 49 degrees superheat going to the low pressure cyl-
inder the saving is 7.5 per cent.

51. The tests on engine K at half load, with 59 degrees super-
heat by the reheater, show 7.2 per cent, saving. The engines
C, D and K have sufficient initial superheat and are of such
size as to make the high pressure jackets of but little value, so
that the savings shown are due principally to the action of the
reheater.

494 SUPERHEATED STEAM AND REHEATERS IN COMPOUND ENGINES.

52. A study of the above results appears to indicate that the
reheater will not justify its use (except as a separator) unless it
superheats the low pressure admission steam at least 30 degrees.
An examination of the qualities at release in the low pressure
cylinders indicates that 100 degrees superheat of the receiver
steam will probably be enough to make the steam dry and satur-

820

710

)

^700

i AAA

(

J COO

— '

(

^^^

Snan

Q> 580

Engine K. - Test 25.
Full Load with Keheater.

3 540

600

Marl

k.. Lit.

utr<

spy

.•1

1.
Fio

. 15,

2.2

!r.r.

ated at release. As it is not desirable to have superheated
steam at release this suggests the probable desirable limit to
the amount of superheat to be given by the reheater.

The Value of Moderate Superheating,

53. The engines (7, 2?, jF, Fy 0, 11 and K were all supplied
with steam from Babcock and Wilcox boilers, fitted with super-
heaters giving from 100 degrees to 125 degrees superheat at the
boiler when running at the rated power. The amount of super-

SUPERHEATED STEAM AKD REHEATEBS IK COMPOUND ENGINES. 495

heat at the engine depends on the load at which the engine is
nmning; (a) because the superheat at the boiler decreases as its
load is decreased, and (b) because the fall of temperature in the
steam pipe increases as the weight of steam passing through it
diminishes. For these two reasons the superheat was less at low
loads than at higher loads except in some cases where the number

(

)

inn

"^^

H in

—

S 20

E-am

— =

Engine K. - Test 26.
Half Load with Reheater.

soo

Entropy

.8 1.0 1.2

Fig. 154.

1.4 l.G 1.8 2.0 2.2

of boilers used could be decreased as the load decreased. As the
superheat going to the low pressure cylinder when the reheaters
M'ere in use varied in the opposite way; that is, increased with
decrease of load, these two variations tended to offset one
another in their influence on the engine efficiency. The tests
without the reheaters in use will then be the most valuable for
showing the influence of the superheat of the high pressure
steam. The tests at full load show that engine A uses 248
British thermal units per indicated horse-power per minute with

496 SUPBEHEATED STEAM AND REHEATERS IN COMPOUND ENGINES.

no superheat; engine C of about the same size with 78 degrees
superheat uses 239 British thermal imits, and engine D with
98 degrees superheat uses 226 British thermal units, a saving of
about 9 per cent. ; some of which, however, is also due to a better
vacuum and better cylinder proportions. Engine B, a smaller
engine, uses 267 British thermal units with 15 degrees super-

T80

740

■• 700

O AAA

■^

«""

^ G20

o 580

Engine K. - .Test 28. .
Half Load without Reheater.

BOO
(

) .2

.6 .8 1.0 1.2 1.4 1.0 1.8 2.0

M0^,LA ^

Fio. 155.

heat — which is practically the same result as that obtained from
G, a still smaller engine, with a poorer vacuxmi but with 72
degrees superheat. In the test of D with 98 degrees superheat
the quality of the steam at release is 87 per cent., so that it is
evident that when the jackets are not used a much' greater super-
heat is desirable in order to prevent condensation in the high
pressure cylinder — ^probably at least 150 degrees mil be neces-
sary. An even greater superheat will be necessary to keep the
steam dry in both cylinders. With the jackets in use and with

SUFEBHEATED STEAM AND BEHEATERS IN COMPOUND ENGINES. 497

98 degrees superheat the quality in the high pressure cylinder at
cut-oflF is 99 per cent., and at release is 94 per cent. The ad-
vantage gained by superheating is, of course, greater in the
smaller engines.

The Variaiion of Economy with Engvns Load.

54. The engines 5, C, E and K were all tested at several loads
so as to determine the effect of variation of engine load on the

.8 1.6

^ Entropy—

6.4 7.2 8.0

Um. Ana iVbte OW...V. T.

Fig. l.r).

efficiency of the engine. In all the cases (engines C, E and K)
where moderately superheated steam was admitted to both cyl-
inders the important fact was brought out that the heat con-
sumption per indicated horse-power is practically constant
through a range of load varying from one-half load to full load,
and probably even to a considerable overload. The apparent
exception in the better performance of engine C at half load is

496 SUPERHEATED STEAM AND REHEATEBS IN COMPOUND ENGINES.

no superheat; engine C of about the same size with 78 degrees
superheat uses 239 British thermal units, and engine D with
98 degrees superheat uses 226 British thermal units, a saving of
about 9 per cent. ; some of which, however, is also due to a better
vacuum and better cylinder proportions. Engine B, a smaller
engine, uses 267 British thermal units with 15 degrees super-

820

780
740

.£3

^ 700

, Degre

3 MWt

^ Gao
2

—

■^

O 580

Engine K. - .Test 28. .
Half Load without Reheater.

o wo

-3

500

9 .i

.6 .8 1.0 1.2 1.4 1.6 1.8 2.0 ^

Fig. 155.

heat — ^which is practically the same result as that obtained from
G, a still smaller engine, with a poorer vacuum but with 72
degrees superheat. In the test of D with 98 degrees superheat
the quality of the steam at release is 87 per cent., so that it is
evident that when the jackets are not used a much* greater super-
heat is desirable in order to prevent condensation in the high
pressure cylinder — probably at least 150 degrees will be neces-
sary. An even greater superheat will be necessary to keep the
steam dry in both cylinders. With the jackets in use and with

SUPERHEATED STEAM AKD REHEATERS IN COMPOUND ENGINES. 497

98 degrees superheat the quality in the high pressure cylinder at
cut-off is 99 per cent., and at release is 94 per cent The ad-
vantage gained by superheating is, of course, greater in the
smaller engines.

The Variation of Economy with Engine Load.

54. The engines B, C, E and K were all tested at several loads
so as to determine the effect of variation of engine load on the

,^^ Entropy

Fm.

eflBciency of the engine. In all tLe cas^s ■''frniriii^^ C, E and K)
where moderately superheat^ iteam was zAiuwjA to both cyl-
inders the important fact was broight out that the b^at con-
sumption per indicated hor^i^-power is prar:»;<ra^y ojfTxrX^zX
through a range of load Tarrine frorij on^-half l^/ad t/> f *'! loa-i,
and probaMy even to a c>>n-;ier2ih*e over>/ad Ir.e ap;/arer*
exception in the better fi^rf'rrr.ir.':^ of 'rE^Ine C at bi^f >/a/i U

498 SUPERHEATED STEAM AND REHEATERS IN COMPOUND ENGINES.

probably due to a better vacuum at that load. The general
result was to be expected because the effect of superheat is to
reduce the amount of heat disappearing during admission^ and,
consequently, to permit the increased expansion at low loads to
occur without excessive cylinder condensation. In engine B,
with very small superheat of the admission steam, there is a
slight decrease in economy as the load decreases from full load,
and in the tests without reheater, and consequently with no
superheat going to the low pressure cylinder this decrease in
economy is very marked.

55. It is, perhaps, hardly necessary to emphasize the fact that
the constancy of heat consumption referred to above is in terms
of the indicated horse-power. In all these engines the friction
horse-power is low, and the mechanical efficiencies at the rated
loads are high, varying (see Table V., line 25) from 91.4 per
cent, to 94 per cent, and averaging 93.2 per cent. The friction
horse-powers were determined by taking cards with no load on
the engine, but they really represent more than the friction of
the engine proper, since they include some losses properly charge-
able to the generator — such as the brush friction, the armature
windage, the bearing friction of the armature and, in some cases,
a low excitation of the field. Consequently, the real mechanical
efficiencies of the engines are somewhat higher than the quan-
tities given in the table.

56. The low friction of the engine causes the heat consump-
tion per electrical horse-power to change a comparatively small
amount as the load decreases; the heat consumption per kilowatt
per hour will be some 6 or 7 per cent, greater at half load than at
full load.

The Radiation amd Conduction Heat Losses,

57. In those tests where the reheater was in use, the knowl-
edge of the exact condition of the steam entering the low press-
ure cylinder permitted the determination of the heat lost by
radiation and conduction from the high pressure cylinder and
the receiver. The total heat of the steam coming to the engine
is the sum of the four following quantities :

(1) The heat going to the low pressure cylinder.

(2) The heat equivalent of the work done in the high press-
ure cylinder.

SUPERHEATED STEAM AND BEHEATERS IN COMPOUND ENGINES. 499

(3) The heat escaping with the reheater and receiver drainage.

(4) The heat lost by radiation and conduction from the high
pressure cylinder and the receiver.

The first three quantities can be calculated from the observa-
tions made on the test, and consequently, the last quantity
can be determined by the heat balance stated above. The radia-
tion and conduction loss was calculated for most of the tests,
and was found to vary from ^ to 1 per cent, of the total heat
supply to the engine at full load. As the low pressure cylinder
is as carefully lagged as the high pressure cylinder it appears that
from 1 to li per cent, of the total heat supply to the engine at
full load will be lost by external radiation and conduction. The
larger percentage appliea to the smaller engines.

Piston Leakage.

58. The steam qualities at cut-off and release, given in lines 16
to 19 of Table V., show considerable variation in different en-
gines running under practically the same conditions. The
reason for this variation is apparently not far to seek, and de-
pends on a phenomenon to which but small consideration is gen-
erally given in the discussion of steam engine performances.
This phenomenon is piston leakage.

59. Most of the engines were tested when at rest for piston
leakage, before the runs were made, and in no case was there
any but slight leakage. It is probable, however, that a piston
which is quite tight when at rest will leak when running. The
static leakage tests were made for a small number of piston posi-
tions and did not insure static steam tightness in every position.
There is evidence moreover to show that even if the piston is
tight in every position when at rest, it may leak when in motion
owing to the breaking up of the oil film on the cylinder walk. If
to the results of the leakage test of the piston is added the knowl-
edge that an intelligent engineer has of the condition of the cyl-
inders* of which he has had charge, it is probable that a more
accurate statement can be made as to the tightness of the pistons.
From such data the following statements may be made as to the
condition of the engines tested.

Engine B, neither piston perfectly tight, but both in good
condition.

Engine C, no appreciable leakage.

500 SUPEBHEATED STEAM AND BEHEATEBS IK COMPOUND ENGINES.

Engine 2?, both cylinders in very good condition.

Engine Ey high pressure cylinder very good; low pressure
piston had not worn down to maximum tightness.

Engine F^ high pressure cylinder had been scored a few weeks
before test and had not worn quite tight; low pressure cylinder
unusually good.

Engine Hy high pressure cylinder very good; both cylinders
better than engine 0.

60. The above conditions, as known before the tests, will be
found to explain most of the variations in steam quality to which
reference has been made. For example, of the two similar en-
gines E and F^ the latter shows lower quality at cut-off in the
high pressure cylinder, notwithstanding a greater initial super-
heat — ^and this quality is seen to decrease throughout expansion.
Leakage past the high pressure piston readily accounts for this.
In the low pressure cylinders of these two engines, the phenom-
enon is reversed, and the remarkably high quality of the steam
in engine F is presumably due to the unusually good condition
of the cylinder.

61. Similarly comparing tests 22 and 24 on the exactly sim-
ilar engines and H under practically similar conditions, a
marked advantage is seen in the quality of the steam during expan-
sion in the high pressure cylinder during the latter test — a result
to be expected from the known better conditions of that cylinder.

62. These examples could be multiplied were it desirable.
The effect of the piston leakages, when moderate, on the engine
economy is not very great since steam leaking by the high press-
are piston will be available for doing work in the low pressure
cylinder.

ConclvMona.

63. In summing up the general results of the tests the follow-
ing conclusions appear to the writer to be justified when applied
to large size, high-speed, compound, four-valve engines of com-
mon proportions.

The jacketing of the high pressure cylinder is of but little
value when moderately superheated steam (100 degrees Fahr.)
is used.

Reheating is probably a source of loss unless it superheats the
receiver steam at least 30 degrees Fahr., and is not fully effective
unless it superheats about 100 degrees Fahr. In the latter case

SUPERHEATED STEAM AND BEHEATEBS IK OOKPOUND ENGINES. 501

it may be expected to effect a saving of 6 to 8 per cent, of the^
total heat used per indicated horse-power.

Jacketing the low pressure cylinder is shown by the steam
qualities during expansion in the low pressure cylinder to be un-
necessary and therefore undesirable when the reheating is effec-
tive. The effect of admitting moderately superheated steam to
both the high pressure and low pressure cylinders is to keep
the heat consumption per indicated horse-power practically con-
stant throughout a considerable range of loads— from half load
to about one-quarter overload.

The variation within the ordinary limits of the ratio of stroke
to diameter in large size engines of the same power when using
moderately superheated steam^ does not have any marked effect
upon the economy of the engine. The size of the engine is an
important factor in determining its efficiency. The engine
has about 10 per cent, greater heat consumption per indicated
horse-power than K, which is three times larger.

DISCUSSIOlf.

Mr. W. R. A. Harris. — As the discussion on above was closed
rather early, I should like to pass a few remarks on the subject.

First. I think the tests to have been satisfactory should have
been all of one duration, and should have gone ^arther than steam
economy, which, for superheating, is not sufficient, coal con-
sumption being the essential point; for if the steam economy
gains 10 per cent, at a cost of 15 per cent, more fuel owing to
superheating there is a loss.

Second. It would be well to know the type of superheater and
distance from engine, i. e., length of pipe, etc. — as it is a small
amount of superheat.

There are many places on the Continent and in England where
superheat is used as high as 650 degrees Fahr. to 700 degrees
Fahr., with an economy in coal consumption of 7^ per cent.

Another point to bear in mind in favor of superheating, for
new plants, is the increjised velocity of superheated steam over
saturated, enabling smaller pipes to be used and a greater piston
speed maintained.

Mr, George Barrus. — T have not had an opportunity to
study this paper very carefully, but I have noticed one
or two omissions in it. For example, the author states

502 SUPERHEATED STEAM AND BEHEATERS IN COMPOUND ENGINES.

that the tests on engine A were conducted by himself. Also the
tests on engines G, // and K; but tests B, C, Z>, E and F were
conducted by Mr. Parker and Mr. Cook, members of the Society,
and by two other parties. It would be interesting if he stated why
the results of the tests made by these gentlemen are incorporated
in his paper.

I notice in all the tables what is termed " quality of steam at cut-
off and release." I do not think that expresses what the author
really means. I suppose he means what i« usually called the
" percentage of steam accounted for by the indicator." This per-
centage shows how much steam appears on the indicator card, but
it does not show the exact condition of the steam as to the amount
of moisture it contains, as would be understood from the use
of the term " quality." I think the tables ought to be revised
to meet this objection.

Then it seems to me the tables ought to be arranged in accord-
ance with the forms established by our conmiittee on engine testing,
and if they are not arranged in that way there should be some ex-
planation about it. If those forms are of no use let us know it.
If they are all right the author, as a member of the Society, ought
to follow them.

There are no copies of the original indicator diagrams in this
paper. The author should supply them for each of the tests. The
reason I make this staJtement- is, that I saw the report of the test
on engine A, as well as some of the original diagrams which were
taken. I was struck with the fact that the diagrams were
poor. I do not mean that they showed poor engine performance,
but they showed poor indicator work. You all know that if a
vertical engine is not solid on ite foundation and the indicator
outfit is not properly connected, the indicator is very unsteady,
and the diagrams obtained will likewise be shaky and un-
reliable. These diagrams were very poor in that respect, and if
it had been my test I should have placed no reliance on their in-
dications.

Mr, C. F. Kerr, — An investigation that I had the privilege
of making about a year and a half ago on the theory and practice
of superheated steam, convinced me that the reheater would come
again into solid favor with the general use of superheated steam.
When the reheater receiver is used vnth saturated steam the
trouble is that we cannot get a sufficient amount of reheating
surface; you cannot put enough reheating surface into the

SUPERHEATED STEAM AND REHEATER8 IN COMi»OUND ENGINES. 503

reheater to secure as much superheat as you will need, and tests
have shown this quite conclusively. In the 5,000 horse-power
engine at the Waterside station, in New York, which was re-
cently tested, it was shown that both jackets and reheater, either
together or alone, were practically useless throughout the work-
ing range of load. How much heat can you expect to get into
a cylinder full of steam in a fraction of a second through cylinder
walls that may be five or six feet in diameter? That difficulty,
coupled with the fact that in the reheater with saturated steam
you cannot get much superheat, is the reason that tests show both
jacket and reheater to be nearly useless. I think that with 200
degrees superheat and a receiver with an adequate amount of
superheating surface, it is possible to get enough superheat to
keep down condensation to the point, at least, of cut-off, and
that will solve the problem of initial condensation.

If you will note the table on page 49, in which the principal
results of all the tests are given, and look over the various tests,
you will find this to be true, I believe, that with the highest super-
heat at the throttle and with 60 degrees superheat at the low
pressure admission, and perhaps less than the average vacuum,
we find the lowest steam consumption in the 28 tests, which is
11.57 pounds per indicated horse-power hour. We find 100 per
cent, as the quality of the steam at cut-off in the low pressure
cylinder, the lowest heat consumption in British thermal units
per minute, and the highest dynamic efficiency compared with the
ideal cycle. This seems to me to ffhow that the superheating
receiver is bound to come into favor with the general use of
superheated steam.

Mr. Oeorge I, Rockwood, — ^The author gives the conclusions he
derives from these tests at the end of his paper, which is com-
mendable; but after all they hardly seem to justify so much labor
or to be especially valuable or new. On the contrary, the paper
seems to be merely a threshing over of old straw.

The performances of all of these engines have been beaten
many times by other engines of the same make and type; they
were also beaten many years ago by an engine of only 200 horse-
power having a higher cylinder ratio. I maintained twelve years
ago that compounds having a higher cylinder ratio would beat
those with the ordinary ratio by 10 or 15 per cent. To see how
that view would apply in the present case, these tests speak of 13.5
pounds of steam per indicated horse-power per hour on an engine

504 SUPEBHEATED STEAM AND REHEATEBS IN COMPOUND ENGINES.

of 2,400 horse-power (which is really a very poor performance).
Mr. Barms tested an engine a year ago of 500 horse-power, in
which test the performance was 11.2 pounds. This engine was a
high-ratio compound, and beat the 2,400 horse-power engine by 17
per cent., both engines operating " under similar external working
conditions." This would seem to call in question the conclusion
that the smaller engine is about 10 per cent, less economical
than the larger. This conclusion of the author appears to be
general and not confined to these engines alone.

I am inclined to agree that the supposition that the larger the
engine of a given type, the greater its economy will be is true;
but hardly think these tests prove it. Even if the test of the 500
horse-power high-ratio compound be compared with that of the
2,500 horse-power high-ratio compound engines at the Waterside
station, to whidh Mr. Kerr has just referred, it wdll be seen that the
smaller engine is still the most economical of all. This, however,
comes from the fact that the Waterside engines, being designed
for use with superheated steam, had poppet valves, which involved
15 per cent, clearance in the first cylinder, with a resulting total
loss of economy of 6 per cent.

With regard to the value of reheaters, the author does not note
two or three important considerations. The principml one is that
a reheater increases the economical power of a given engine to an
extent out of all proportion to its cost. Even, therefore, if it does
not increase the economy of the engine — and I feci very sure
that if properly designed it cannot decrease the economy — its use
is more than justified. Nothing is said, I believe, about draining
the water of condensation from the high pressure exhaust pipe
before the steam meets the reheating surf-aoe. This is necessary if
economy is aimed at.

Some people still like to test steam-engines. To me the subject
is of only passing interest, as I believe the steam-turbine has
routed the reciprocating machine completely for all work which
can be done by electrical transmission of the power; and there is
now hardly to be found any place where the motor is not of greater
advantage than shafting and belting.

Prof. L, S. Marks* — The writer regrets that Mr. Barrus did
not have an opportunity to study tliis paper very carefully
before he criticised it. He vciU. find his criticism in part answered

* Author's closure, uuder the Rules.

SUPERHEATED STEAM AND REHEATERS IN COMPOUND ENGINES. 505

in the paper itself, but as the criticism includes also other points,
the writer will endeavor to reply fully.

In paragraph one is given the reason why the writer included
with his own tests all the available unpublished tests on engines of
the same make. It may also be pointed out that these other tests
were not published by those employees of the company who
had conducted the tests, because of a special regulation of the com-
pany forbidding such action on their part; and that their inclusion
in this paper was with the express permission of the president of
the company.

The terra " quality of steam at cut-off " was used because of its
greater brevity^ and in the belief that the members of this Society
understand what is meant hereby; and further, the meaning of the
term is made abundantly clear by the discussion under the heading
of " piston leakage." The tests were presented with a special ob-
ject in view, and the writer, while appreciating fully the excellent
work done by the Committee on Engine Testing, in arranging
forms for the presentation of results, preferred to use such forms
for his own tests as should exhibit most clearly the results which
he wished to emphasize. With respect to the publication of
copies of the original indicator cards, the writer believes he had
done' as much as is desirable in publishing the combined cards.
The publication of copies of the original indicator diagrams for
each of the tests would have increased largely the bulk of the
paper without at all increasing its value.

The author cannot but express his extreme surprise at the state-
ment of Mr. Barrus, that he had seen some of the original
diagrams for the test on engine A. These diagrams have not been
out of the possession of the writer from the time they were taken,
and cannot have been seen by Mr. Barrus. His statement that
these diagrams were poor must also be denied. They were seen
at the time of the tests, and were found satisfactory both by the
Consulting Engineer of the Boston Electric Light Co. and by the
engineer representing the engine builders. The engine founda-
tions were very solid, the engine itself very stiff, the indicating
gear was a simple photograph device, designed by the engine
builders, with but few joints, strong, new, and wdth no discernable
back-lash. The indicators were gone over and oiled every hour.
There was no suggestion of shakiness in the cards. The writer
considers the indicator cards for this test to be as reliable as can be
obtained from engines of this size and epeed, and cannot but con-

606 SUPERHEATED STEAM AND REHEATERS IN COMPOUND ENGINES.

sider that Mr. Barrus' remarks on these cards, which he has never
seen, are either the product of a faulty memory or of a vivid
imagination. Copies of the actual cards for the tests on engine A^
on July 20, 1899, are added to this paper, Figs. 157-160.

The writer finds himself in complete agreement with the re-

Enginc A. H.P. Head End.

July 90, 1800. 12.80 P.M.
Engine pressure by gauge- 155

Area- 1.60 Lengths 3.72

Jlwi.Bk.lM0 Ck.Jl.r.

Fio. 157.

marks of Mr. C. V. Kerr. The results of the test of the 5,000
horse-power engine, at the Waterside station in New York, are
such as the present tests would show to be probable, unless the
reheater were of unusual size and efficiency. The larger the
engine the less is the gain to be anticipated by the adoption of the
usual devices for securing economy.

Engine A. H.P. Crank End.
July 20, 1809. 12.30 P.M.
Area->2.10 Length- 3.80

Awi.1u.S9f c^jf.r.

Fig. 158.

Mr. Rockwood's numerous criticisms, although they do not
touch the substance of the paper, had perhaps better be answered
individually. So complete a series of tests on large size engines of
one type has not, the writer believes, been recorded before. They
give certain information for which the owners of the engines were
willing to undertake the not inconsiderable cost and inconvenience

SUPEEHEATED STEAM AND REHEATERS IN COMPOUND ENGINES. 507

involved in the making of these tests. The results so obtained
have value to all those who have charge of such engines — to
all who want to know whether engines of 1,000 to 3,000 indicated
horse-power should preferably be run with or without jackets,
vnth or without reheaters, and at what range of loads they may

Engine A. L.P. Head End.

July 20, 1899. 12.80 P.M.

Vacuum pressure by gauge « 12.2

Area«.2.84 Length»8.89

Am.Bk.y9t0a>..s.r.

Fig. 159.

be economically run. Such information the writer does not con-
sider to be the threshing of old straw. The information on these
points, obtained on small laboratory engines up to 100 indicated
horse-power, is of no value for general industrial practice; and
it is principally on such small engines that information of the
nature indicated has been obtained hitherto.

Engine A. L.P. Crank End.
July 20, 1890. 12.30 P.M.
Area«8.19 Length .8.92

Am.Bi.SuU Cu.,Xy.

Fig. 160.

Moreover, this paper does not protend to give " record " results,
but only the results of good modern practice. The writer has a
high opinion of the possibilities of the high-ratio compound asso-
ciated TOth Mr. Rockwood's name, but he would point out that his
conclusion with reference to the effect of the size of an engine on

508 SUPERHEATED STEAM AND REHEATERS IN COMPOUND ENGINES.

its economy is expressly restricted to engines of ordinary propor-
tions, and intentionally omits a comparison between different types.
The writer has no doubt, however, that his general conclusion will
be found to apply satisfactorily to a comparison of two of Mr.
Rockwood's engines. The comparison instituted by Mr. Rockwood
between his own 500 horse-power high-ratio compound and the
very different Waterside station engine referred to by Mr. Kerr,
can hardly be expected to prove anything.

The writer cannot agree with Mr. Rockwood as to the desirability
of having a reheater in the case, where it does not increase the
economy of the engine. A separator in the high-pressure exhaust
and a readjustment of the cut-off valves will produce a similar
result with less trouble and first cost.

With reference to Mr. Rockwood's suggestion that these tests
were carried out to gratify the writer's passion for testing steam
engines, it may again be noted that these tests were all either
acceptance tests, or tests designed by those in charge of the plants
to yield certain information as to their performance, without any
special reference to the partialities of the writer. The pleasure of
working up such tests is perhaps hardly so great as Mr. Rockwood
seems to imagine.

GAS ENGINE TESTING AND STANDARD OF COMPARISON. 509

No. 1081.*

COMMERCIAL GAS ENGINE TESTING AND PROPOSED
STANDARD OF COMPARISON.^

BY WILLIAM P. FUKT, BA8T PITTSBURG, PA.

(Member of the Society.)

1. The value of accurate and comprehensive testing of all prime
movers, and the determination of a basis of comparison upon which
to judge the relative economy of those of any particular class can-
not be overestimated.

2. To the manufacturer of heat motors, especially, it is un-
questionably necessary, and at the same time most expedient, to
incorporate into the scheme of engineering organization an effi-
cient testing department, which shall at once be capable of secur-
ing accurate data of the operation of the finished machines with-
out being burdened with any unnecessary refinements. The sim-
plest reliable methods and apparatus are therefore obviously best
suited to accomplish the desired results, and it is then left only to
determine the most feasible basis upon which to make compari-
sons of the characteristic data obtained.

3. The friction brake is by long odds the most satisfactory

* Presented at the Cliicago meeting (May and June, 1904) of the American
aociety of Mechanical Engineers, and fonning part of Volume XXV. of the
Transactions.

f For further discussion on the same topic consult Transaciiona as follows :
No. 843, vol. xxi., p. 396 : '* An Efficiency Test of a 125 Horse-Power Gas Engine."

C. H. Robertson.
No. 875, vol. xxii., p. 152 : "Efficiency of a Gas Engine as Modified by Point of

Ignition." C. V. Kerr.
No. 895, vol. xxii., p. 612 : *' Efficiency Tests of a 125 Horse-Power Gas Engine."

C. H. Robertson.
No. 949, vol. xxiii., p. 686 : ** Temperature of Exhaust Gases.*' R. H. Fernald.
No. 950, vol. xxiii., p. 705: "Working Details of a Gas Engine Test." R. H.

Fernald.
No. 989, vol. xxiv., p. 1048 : "Method of Testing Gas Engines." E. C. Oliver.
No. 990, vol. xxiv., p. 1063; "Performance of an Internal Combustion Engine

Using Kerosene and Gasolene as Fuel." Halladay and Hodge.
No. 991, vol. xxiv., p. 1095: "Test of a 12 Horse-Power Gas Engine." C. H.

Robertson.

510 GAS ENGINE TESTING AND STANDARD OF COMPARISON.

GAS ENGINE TESTING AND STANDARD OP COMPARISON. 511

means of measuring the power developed. Its accuracy depends
on the following factors :

First, The accuracy of standard platform scales. This is easily
verified by standard weights.

Second. Allowance for the weight of the blocking on the scales
and the unbalanced weight of the brake band and arm. These are
both easily determined by actual weighing.

Third, The brake radius, or distance from center of the shaft
to the Jcnif e edge supporting the brake arm.

Fourth, The revolutions per minute.

Fifth. The steadiness with which the scales are kept balanced.

4. If we compare the above simply verified facts with those re-
quired by an indicator test, it will be at once apparent how muc6
superior the brake is to the indicator as a means of quickly and
reliably measuring power. Then also the available or brake horse-
power, and not the indicated horse-power, is the thing for which
engines are run.

77ie Friction Brake:

5. The brake is applicable to large as well as small engines, as
may be seen by reference to Figs. 161 and 162, showing respec-
tively a 300 brake horse-power double-acting tandem and a 25
brake horse-power single-acting vertical gas engine undergoing test.
Fig. 163 shows several engines set up in readiness for shop test.

6. In practice it is necessary to lubricate the brake wheel for the
purpose of making the friction uniform, and to prevent the wooden
cleats becoming locally heated and taking fire. It is found that
strips of fat salt pork make the best lubricant, for they slowly fry
and thus keep the wheel uniformly and continuously greased. The
wheel must, of course, be cooled, and this is usually provided for
by casting re-entrant lips on it, which will retain a layer of water
by virtue of centrifugal force when the brake wheel is in motion.

7. A stream of water is fed into the brake wheel at one point
and scooped out by a stationary pipe at another point. By this
means the temperature of the brake wheel is controlled.

8. When the temperature and lubrication of the brake wheel
are kept constant, it is not difficult for a man ^vith a wrench to
keep the brake-band tension screw so adjusted that the scale beam
of the platform scales is always substantially balanced. If the load
becomes too great, he slackens the tension, and if too small, he
increases it

512 GAS ENGINE TESTING AND STANDARD OP COMPARISON.

GAS ENGINE TESTING AND STANDARD OF OOMPARISON. 513

9. Occasionally considerable annoyance may be experienced by
sudden changes of friction occurring which are due to a little
water getting on to the outside of the brake wheel. This seems
to chill the lubricant and have a disturbing effect out of all pro-
portion to the cause. It is a small practical point well worth at-
tention, for with irregular variations in the friction, the man at
the brake cannot keep the load constant. The remedy is very
simple — ^keep water off the outside of the brake wheel.

The Indicator Card:

10. The indicator, while not as useful for power determina-
tions, is of great value for showing what is going on in the cyl-
inder, and cards should always be taken in connection with the
brake test. It will often be superfluous to calculate the indicated
horse-power from thenu If engineers in charge of gas engines
made more frequent use of the indicator, they would frequently
be able to obtain increased satisfaction by detecting faulty adjust-
ments, particularly of the point of ignition.

Eng^ine Speed:

11. Where an engine is being tested on a constant load and
quality of gas, the governor should hold the speed constant enough
to warrant the use of a good make of hand-speed counter.

Where the above condition is not met a continuous counter must
be employed.

Caa Measurements:

12. A good meter, whose accuracy is checked occasionally, by
means of prover tests, over the range for which it is used, is the
best instrument for this purpose. The temperature and pressure
of the gas at the meter and the barometer reading (uncorrected
for sea level) are needed for correcting the meter readings to
standard conditions.

Calorimeter determinations or chemical analyses will give the
heat value of the gas.

Ca7nparison of Tests:

13. After tests are made we still desire to know what they mean.
We must have a standard with which to compare the results ob-
tained on a given engine and as a basis for predicting what that
engine ought to do.

514 GAS ENGINE TESTING AND STANDARD OF COMPARISON.

0A8 BKOINB TS8TINO AND 8TANDABD OP 00HPABI80N. 515

TABLE I.

Data and Calculatiokb for Fioubb (5).

1^ X 24 Gas Engine— Test No. 904.

Type— Horiiontal, doable-acting, tandem, single crank, foor cycle throttling
engine.

Diameter of cylinders 16^ inches.

Stroke 24

Diameter of piston rod (in three explosion chambers) 5f "

Diameter of tail rod (in one explosion chamber). 2f "

Full load test speed 181 B.P.M.

Number of charges per two revolutions 4

Suction displacement per minute 981 cuUc feet.

« J ^i * ^ 8^ cubic feet . ^^^

Reduction factor = -— ; — . . , ^ = 0.862

981 cubic feet

Actaal Readings.

822 B. H. P.

8,430,000 B. T. U.

118-8 B. H. P.
1,207,000 B. T. U.

281 B. H. P.
8,080.000 B. T. U.

99.0 B. H. P.
1,066,000 B. T. D.

148 B. H. P.
1,840,000 B. T. U.

50.8 B. H. P.
646,000 B. T. U.

17.1 B. H. P.
905,000 B. T. U.

6.0 B. H. P.
818.000 B. T. U.

NoTK.— The eecond column of flgnres are obtained by moltlplying those in the flnt colomn by
the redaction factor 0.86S.

14. The most natural standard is the cubic feet of gas used
per brake horse-power hour. When, however, we attempt to use
this, we find that for each engine it is a function of the load car-
ried. Its value varies considerably at full load, and very rapidly
at light loads, until at no load it reaches infinity. We must make
several tests at different loads in order to get the law of this varia-
tion. When we calculate the gas consumption per brake horse-
power hour at light loads, we find that a very small variation in
the total amount of the gas used, or of the load at which the test
is made, makes a very large and unmeaning variation in the cubic
feet per brake horse-power hour.

15. This unmeaning characteristic of the gas per brake horse-
power curve at li^t loads leads one to prefer the use of a curve
plotted with the brake horse-power as abscissae, and the cubic
feet of gas, or better yet the British thermal units of gas, per hour
as ordinates. Fig. 164 shows two such curves, one for an 8 by 10,

516

GAS ENQINB TB8TINO AND STANDARD OF COMPARISON.

and the other for a 16^ by 24 gas engine. These two curves, how-
ever, are so different in size that they are incomparable with each
other, even though the engines require about the same number of
British thermal imits per brake horse-power hour at corresponding

—

■""

-n

I6K X 24 GAS CNQINC TEST
SHOI>TCST*004

leHxa^

y*!

— /

CNG

— ^— 1

B.M.F.

OAS

B.T.O.

R.F.M.

C.F.
per hoar

Effective
per hour

3,0

00,

NX)

823

8940

WWtt

178

14S

2110

1.840.000

184

n.i

IMO

906,000

186

-fe

2j(

00,

)00