Full text of "The evolution of the steel ball industry"

See other formats

8 I •

HOOVER STEEL BALL Co

ANN ARBOR, MICHIGAN

MANUFACTURERS OF

B-A-L-L-S

STEEL, BRONZE, BRASS
COPPER, ALUMINUM

AND OTHER MATERIALS IN BOTH
HIGH AND COMMERCIAL GRADES

MADE IN AMERICA

TO THOSE WHO ARE INTERESTED IN OUR PRODUCT WE WILL,
UPON APPLICATION, GLADLY FORWARD OUR REGULAR CATALOG
AND PRICE LIST, GIVING THE TRADE NAME OF OUR DIFFERENT
GRADES OF BALLS, GUARANTEED ACCURACY AND QUALITY,
PRICES AND TERMS

TJ/07/

FOREWORD

WE TAKE PLEASURE IN PRESENTING
THIS TREATISE ON THE MANUFAC-
TURE OF STEEL BALLS AND WISH
TO THANK THOSE WHO HAVE SUPPORTED
US IN SUCCESSFULLY OVERCOMING THE
PREJUDICE AT ONE TIME EXISTING
AGAINST AMERICAN MADE BALLS.

THIS SUCCESS HAS BEEN DUE TO THE
QUALITY OF OUR PRODUCT COMBINED
WITH EXPERT KNOWLEDGE OF THE
REQUIREMENTS OF OUR CUSTOMERS.

IN THE FOLLOWING PAGES WE HAVE
ENDEAVORED TO SHOW STEP BY STEP
THE VARIOUS STAGES THROUGH WHICH
A BALL PASSES FROM THE ROUGH STEEL
BLANK TO THE MIRROR -LIKE FINISHED
SPHERE.

THE OBJECT OF THIS TREATISE IS TO
LAY BARE FACTS WHICH HAVE HITHERTO
BEEN GENERALLY UNKNOWN AND IF WE
SUCCEED IN STIMULATING FURTHER
INTEREST IN THE BALL INDUSTRY, THIS
WORK WILL NOT HAVE BEEN IN VAIN.

HOOVER STEEL BALL Co.

ANN ARBOR, MICHIGAN

372827

L. J. HOOVER

PRESIDENT AND GENERAL MANAGER

COMMITTEE OF THREE HAVING ENTIRE CHARGE OF FACTORY
MANAGEMENT AND MANUFACTURING

• »» *•» *

•M »• » •

hi 0}

\- "

(fl ,_

cr z

u <

> -i

O a-

0 H

1 Z

Manufacture of Steel Balls

DURING recent years the application of ball
bearings in machine design has increased rapidly,
and this type of bearing is now used in many
machines where plain bearings were formerly
considered good enough. Until German export
facilities were shut off by the war, the majority of the
steel balls used in these bearings were made by the Deutsche
Waffen und Munitions Fabriken of Berlin, Germany, and
the product of this firm has become so celebrated that many
persons think the steel ball industry was developed by the
Germans. As a matter of fact, the art of ball making goes back
to a very early date, and the development of original methods
for doing this work is attributed to the Chinese. To those who
have credited the Germans with the development of commercial
methods of ball manufacture, it will doubtless be of interest to
learn that the first commercial steel balls were made in this
country under basic patents granted to Richardson of the
Waltham Emery Wheel Co., Waltham, Mass., and that the
original ball making machinery for the plant of the Deutsche
Waffen und Munitions Fabriken was designed and built in the
United States and shipped to Germany ready for use. This
will be explained in detail in connection with the following
historical outline of important epochs in the steel ball industry.

HOW THE STEEL BALL INDUSTRY
CAME INTO EXISTENCE

IT HAS been stated that basic patents for dry grinders used
in roughing out ball blanks to a spherical form were granted
to Richardson of the Waltham Emery Wheel Co., in 1887.
These patent rights were subsequently sold to the Cleveland
Machine Screw Co., Cleveland, Ohio, which had control of
patents on ball making machinery taken out by John J. Grant.
One of the first firms to manufacture steel balls on a commer-
cial basis was the Simonds Rolling Machine Co., of Fitchburg,
Mass., and the Fitchburg Steel Ball Co. was subsquently formed
by employes who left the Simonds firm. After a brief career,

the latter firm was taken over by the Cleveland Machine Screw
Co., and with facilities acquired through its own development
work and purchase from other companies, it was in a position
to manufacture the majority of balls used in the bicycle trade.
In this connection it will be of interest to note that up to the
year 1899 balls one-half inch in diameter were the largest size
that were manufactured in quantities.

About 1890 the Cleveland Machine Screw Co. designed and
built for the Deutsche Waffen und Munitions Fabriken, of
Berlin, Germany, equipment used in its original steel ball plant
and this marked a most important step in the trade, owing
to the reputation for making high-grade balls that was later
acquired by this firm. The machines built and shipped to
Germany had no reference to American manufacturing rights,
and the Cleveland Machine Screw Co. continued to operate its
plant in the usual way.

In 1894 when a consolidation of bicycle manufacturers was
effected, the Cleveland Machine Screw Co. was sold to the Pope
Mfg. Co. of Hartford, Conn., which at that time started to
manufacture its own balls for use in bicycle bearings. The
requirements of balls for the bicycle trade were not nearly as
severe as the standards which must be met by balls used in
high-grade annular bearings at the present time. This was
largely due to the fact that the cup and cone form of races was
employed, allowing compensation to be made, and while this
form of race did not enable ball bearings to be operated under
the most efficient conditions, it was the means of overcoming
discrepancies due to inaccuracies in the size of the balls. Up
to this time there had been six or seven firms engaged in the
manufacture of steel balls, but with the decline of the bicycle
industry a number failed.

In 1901 the Standard Roller Bearing Co., Philadelphia,
Pa., acquired all obsolete and existing plants engaged in the
manufacture of steel balls. L. J. Hoover, who was formerly
in the employ of the Standard Roller Bearing Co., left that firm
in 1906 and formed the Grant & Hoover Co. at Merchantville,
N. J. The name of this firm was later changed to Atlas Ball
Co., and the plant transferred to Philadelphia, Pa., where it is
still in operation. On March 1, 1913, the Hoover Steel Ball Co.
of Ann Arbor, Mich., was organized by Mr. Hoover for the

purpose of engaging in the manufacture of high-grade steel balls
to take the place of those imported from Germany. When the
European war started in 1914, the blockade of German ports by
the British Navy shut off the supply of steel balls formerly
exported by that country to the United States, and the insistent
demand of consumers for balls made in this country imposed
a heavy strain upon the facilities of domestic producers. Some-
what similar conditions existed in all branches of the machinery
trade, making it difficult for the ball manufacturers to increase
the capacity of their plants ; but the management of the Hoover
Steel Ball Co. showed commendable initiative by contracting
for the entire output of machine building firms with which orders
were placed for special machinery required in ball manufacture;
and these firms were given financial assistance to enable them
to handle work with the greatest possible rapidity. As a result,
the Hoover Steel Ball Co. has greatly increased its capacity, the
growrth being well illustrated by Fig. 1 and the illustration in the
center of the book, that show, respectively, the original factory in
which the firm started manufacturing in March, 1913, and the plant
as it appears at present. An idea of the magnitude of the business
will be gathered from the fact that the consumption of steel
runs in excess of 500 tons a month, and calculated on the basis
of J^-inch balls, the daily production is between 25,000,000
and 30,000,000 balls per day.

Fig. 1. Original Plant in which Hoover Steel Ball Co. started Manufacturing
Operation in March, 1913.

RAW MATERIAL OF THE STEEL
BALL INDUSTRY

THE steel from which balls are made comes to the factory
in coils or straight rods, according to its size. Stock
less than 11/16 inch in diameter comes in coils and is
known as "wire," while all stock exceeding 5/g-inch in diameter
comes in straight bars. The size of the stock is referred to in
thousandths, i. e., stock ^g-inch in diameter is known as 0.375-
inch stock. The following is a specification of steel wire
used for making balls: carbon, 0.95 to 1.05 per cent; silicon,
0.20 to 0.35 per cent ; manganese, 0.30 to 0.45 per cent ; chromium,
0.35 to 0.45 per cent; sulphur and phosphorus, not to exceed
0.025 per cent. The following analysis is typical for the larger
sizes of stock which comes in straight bars: carbon, 1.02 per
cent; manganese, 0.40; silicon, 0.21; chromium, 0.65; sulphur,
0.026; and phosphorus, 0.014 per cent. A well equipped
laboratory is maintained in which chemical and physical tests
are conducted on each shipment of steel to determine its suit-
ability for manufacture into balls, and an unloading ticket must
be signed by the head of the laboratory before the steel is taken
from the cars into the plant. Some very interesting conditions
have been brought to light by the laboratory work, and a later
section of this article will be devoted to a discussion of tests
conducted on the raw material and product, data obtained from
these tests, and a description of methods and apparatus used
in the laboratory.

PRODUCTION OF BALL BLANKS
BY COLD-HEADING

BALL blanks made from stock ranging from 1/16 up to and
including j^-inch m diameter are formed on special
cold-headers designed for the production of ball blanks
by the E. J. .Manville Machine Co., Waterbury, Conn. A
battery of these machines is shown in operation in Fig. 2, and
in this connection it may be mentioned that the Hoover Steel
Ball Co. is equipped with machines of the following sizes: 00, 0,
1, 2, 3, and 5. Production of ball blanks by the cold-heading
process has several advantages in its favor. In the first place,
there is practically no waste, with the exception of about 0.040

Fig. 2. General View in Cold-header Department; Blanks for All Sizes of Balls
up to Y^-inch Diameter are made on Cold-Heading Machines.

inch of metal left on the blank to provide for finishing. Blanks
can be held to this close limit because the steel is worked
cold and there is no tendency for it to become decarbonized.
One man can look after three or four machines, so that the
cost of labor is almost negligible. Cold-headers used in the
production of ball blanks are of the type commonly known as
single-blow solid-die machines, and the way in which they
operate can best be explained in connection with Fig. 3. These
machines consist of a heavy framed which completely surrounds
the working parts of the machine, thus insuring a high degree
of rigidity. At one end of the machine there is a driving shaft
B ; and. at the opposite end of the frame is die-block C. Between
the sides of the frame is a movable ram D that actuates the heading
punch E. Wire F to be made into ball blanks enters the machine
through feed rolls G and then passes through cut-off quill H.
At the side of the machine is supported a bracket / in which
slide / may be reciprocated by a crank motion from the main
driving shaft. Slide J has a cam groove cut in it in which roll K
is fitted ; this roll is mounted on cross-slide L, so that a lateral

Fig 3. Plan View of Cold-header Mechanism Illustrating Method of Operation.

motion is imparted to cut-off knife M located on the end of
cutter-bar L.

A ratchet feed advances the wire through the cut-off quill
until it comes into contact with a stop, which is not shown
in the illustration. This stop checks forward motion of the stock
when a sufficient length has passed the cut-off knife to produce
a ball blank of the proper size. Cut-off knife M is advanced in
the manner just described, severing the wire, but retaining it on
the cut-off blade by means of a spring finger. Advance of the
cut-off knife and wire slug is continued until the slug reaches a
position directly in front of the opening in die TV. Here it is
held stationary long enough for punch E to begin to push the
slug of metal into the die, at which time cut-off knife M retreats

Table I. Capacities of Cold-headers in Ball Blanks per Hour

Size of
Cold-
header

Capacity for
Ball Blanks
Diameter in
Inches.
Max. Size.

Production
of Blanks
per Hour

Size of
Cold-header

Capacity for
Ball Blanks
Diameter in
Inches.
Max. Size.

Production
of Blanks
per Hour

3/16

7800

7/16

6300

9/32

7200

1/2

6000

3/8

6900

9/16

4800

Note — Due to time loss in setting up, trouble with stock and breakdowns, the actual
average rate of production is from 80% to 90% of above values.

Table II. Size of Stock Used for Making Balls on Cold-headers

Diameter of
Ball — Inches

Diameter of
Stock — Inches

Diameter of
Ball — Inches

Diameter of
Stock — Inches

1/8

.100

5/16

.235

5/32

.120

3/8

.275

3/16

.145

7/16

.320

7/32

.170

1/2

.365

1/4

.190

9/16

.395

9/32

.220

5/8

.440

and allows punch E to continue its work by pushing the blank
to the bottom of the die cavity. After the slug F has been
headed it is ejected by the knock-out pin O which is
advanced by the mechanism operated by lever P, which
also receives its motion from a crank at the side of the machine
connected to the main driving shaft. In this way the ball
blank is knocked out of the die and dropped through an opening
into a receptacle placed to receive it, this being clearly shown in
Fig. 2. Table II gives the diameter of stock used in making
blanks for several different sizes of balls, and is presented to
show the enlargement that takes place during the heading
operation. Various grades of steel* have been used for making
dies employed on the cold-headers, but the most satisfactory
results have been obtained with the following grades ^'Sander-
son" or "Viking Special" made by the Crucible Steel Co. of
America; "Intra" made by the Hermann Boker Co.; "Gyro"
made by Braeburn Steel Co.; and tool steel made by William
Jessop & Sons.

HOT-FORGING BALL
BLANKS

IT HAS previously been stated that blanks for balls exceeding
^g-inch in diameter are hot-forged from straight bars, and
in handling this work multiple dies are employed which
produce strings of balls containing up to ten balls, according
to the size. The stock is heated in "Frankfort" furnaces made
by the Strong, Carlisle & Hammond Co. of Cleveland, Ohio;

Fig. 4. View of Stock Racks in Hot-forging Department where Ball Blanks
exceeding Y^-inch diameter are made.

these are oil furnaces which are operated with oil at a pressure
of 8 pounds per square inch, and air at a pressure of 2 pounds
per square inch. Twelve bars are arranged in the furnace as
shown in Fig. 5. The hammer-man takes out the bar at the
left-hand side of the furnace, and after forging a string of balls
at the end of this bar and cutting it up into individual ball
blanks, returns the bar to the furnace at a point at the extreme
right. In this way, the bars are used in rotation, which prevents
any bar from becoming overheated. This is a matter of con-
siderable importance, because the furnaces are maintained at a
temperature somewhat in excess of 1800 degrees F. in order to
provide for heating the stock as rapidly as may be necessary;
but should it happen that steel was left in the furnace for an
undue length of time, there would be danger of burning the steel.

The multiple forging dies are shown in detail in Fig. 6,
in which it will be seen that each die opening is elliptical; the
purpose of this is to provide a clearance space at each side into
which excess metal will flow. It must be borne in mind however,
that while this illustration only shows four die openings, the
number of openings runs up to ten, according to the size of ball
blanks that are being forged. In the cross-sectional views,
the dimensions of the die are indicated by letters, and in Table
III are given diameter A of cherrying cutter, distanced between

Fig. 6. Type of Die used for Hot-forging Ball Blanks for Balls exceeding
%-inch Diameter.

Table HI. Dimensions of Hot- forging Dies for Ball Blanks

Diameter
of Ball,
Inch

Diameter
A of Die,
Inch

Distance
B between
Centers,
Inch

Depth C of
Die, Inch

Depth D, of
Bridge, Inch

Diameter E
of Stock,
Inch

3/4

0.775

0.910

0.387

0.065

0.625

7/8

0.905

1.060

0.452

0.065

0.729

1.035

1.210

0.517

0.075

0.823

centers, and depth C to which the cherrying cutter is sunk in
making the dies for three sizes of balls, and these data are
presented to indicate how dimensions of the dies vary for differ-
ent sizes of balls. The depth D of the gate between adjacent
diesis a matter of considerable importance, because, it determines
the size of the neck between adjacent balls, which is depended
upon to hold the string of balls together until they are sheared.
Also this depth must be regulated so that there is no tendency
to draw the stock adjacent to the neck and form a pipe in the
ball blank, which would have a highly detrimental effect on its
structure. A land of approximately one-third the diameter of
the ball is provided for clearance at the bottom of the die and
the upper die member. The dies are made from a special die
steel made by the Ludlum Steel Co. of Watervliet, N. Y., or from

Fig. 5. "Frankfort" Oil-heated Furnaces made by Strong, Carlisle &
Hammond Co., in which Bars are Heated for Hot-forging Operation.

"Firth-Sterling Special," made by the Firth-Sterling Steel Co.,
McKeesport, Pa. This is not an alloy steel, but a regular tool
steel adapted for making hot-forging dies. In order to produce
round balls in such dies, the bar is turned between each stroke
of the hammer, which results in bringing the balls to a close
approximation of the spherical form. Along one side of each
die is a pipe with a number of holes drilled in it through which
water flows onto the dies and work.

In purchasing stock for the production of ball blanks for
the hot-forging method, it is matter of considerable importance
to have all bars of the same length. This is due to the fact that
when there is considerable variation in length, some bars will

be used up before others, with the result that it is necessary to
finish up a number of short pieces in the furnace before putting
in an entire new charge. At the end of each bar there is left
what is known as a "short end," and experience has shown
that these short ends cannot be forged into ball blanks of the
regular size, as they fail to fill out the dies properly. On this
account, short ends are collected and forged into ball blanks of
the next smaller size. By ordering stock in bars of a specified
length, ' 'short-ends" are eliminated.

After being forged, the hot string of balls is taken to punch
presses made by the Ferracute Machine Co., Bridgeton, N. J.,
which are placed beside the Bradley helve hammers on which
the forging operation is performed, the arrangement being
clearly shown in Fig. 7. The punch presses are equipped with
multiple shearing dies, which consist of a lower die member
with holes of the same size as the balls and a multiple punch
carried in the ram, one punch being in line with each opening
in the die. The string of balls is dropped into place and the

Fig. 7.

C. C. Bradley Hammer and Ferracute Power Press in which a String of
Ball Blanks is Forged and Cut up into Individual Balls.

press tripped, resulting in pushing the balls through the holes
in the die and leaving the scrap metal which is brushed off
before the next operation is performed. The bar is then returned
to the right-hand side of the heating furnace, as previously
mentioned, and is moved to the left each time a heated bar is
removed, until it reaches the extreme left ready for another string
of balls to be forged from the heated metal at its end. Three
sizes of helve hammers made by C. C. Bradley & Son, Inc.,
Syracuse, N. Y., are used for forging ball blanks, which have
capacities for striking blows of 125, 150 and 300 pounds.

ELEVATION OF DIE AND PUNCHES

PUNCH. HOLDER

FLAN OF DIE

Figure 8. Type of Die used for Shearing String Forgings into Individual Ball Blanks.

Fig. 8. shows the construction of shearing punches used
for cutting up the string forgings into individual ball blanks.
At A is shown the form of punch-holder used, which will be seen
to consist of a cast-iron shoe with four set-screws for holding
the punches. These are secured in a clamp B which is made
by drilling holes of the proper size for the punch shanks in a
block of the desired form and then sawing this block in half;
the punches are then put in place and the entire clamp secured
in punch-holder A . The diameter C of these punches is usually
made about J^-inch less than the diameter of the balls in the
string forging that is to be cut up. A plan view of the die is
shown at D, and it will be evident that the spacing £ between
holes in this die is the same as the center distance between the

die cavities in the forging die. Also a bridge is provided in the
shearing die of sufficient depth to retain the neck left between
adjacent ball blanks on the string forging while the balls are
pushed through the die. After the shearing operation has been
completed, the scrap metal is brushed off the shearing die before
the next set of ball blanks is cut up.

In has been mentioned that balls ranging in size from ^-inch
up to about 2j^-inches in diameter are made by forging strings
of blanks according to the process which has just been described.
In the case of the larger sizes of balls — from &/% to 4 inches in
diameter — single blanks are usually forged under a steam
hammer, making one blank at a time at the end of the bar.
Slugs of the proper size are first cut off to the required length
and both ends chamfered, the length of stock being determined
by the weight of the finished balls after making a proper allow-
ance for the material removed in finishing. These blanks are
placed in the oil furnace and heated to a forging temperature;
and each time a blank is removed to be forged a new slug of
metal is put into the furnace in its place. Dies used for this
kind of forging are of an entirely different form from those used
in string forging; they are cupped out to the desired diameter,
but are only turned to a depth of one-quarter the diameter of
the ball to be forged and are not relieved. When the blank has
been heated, the hammer-man places it in the die and the hammer
is worked very slowly until the blank begins to take a spherical
shape, when quicker and heavier blows are struck. Owing to
the shallowness of the die, the operator has ample room to turn
the ball in all directions, and he can therefore produce an almost
perfect sphere. Blanks up to 8 inches in diameter are forged
without varying more than 0.005 inch from a true spherical form.

ROUGH DRY-GRINDING

THE method of making ball blanks varies according to their
size, small blanks being made on cold-headers and large
blanks forged from hot metal according to the methods
which have just been described. After this preliminary
work, all sizes of balls go through essentially the same treatment
certain minor modifications being made according to the quality
of the balls; and the method of treatment may also vary some-

what in the case of balls of extremely large size. These modifica-
tions from standard practice will be taken up in detail.

Blanks made by either the cold-heading or hot-forging
process are first sent to the dry-grinding room, where they

Fig. 9. Side View of Dry-grinder, showing wheel dropped away from work, a

Charge of Balls ready to be dropped into Grinding Position, and Ball

being measured for Size in Test Indicator.

are subjected to a rough-grinding operation § before going to
the heat-treating department. This rough-grinding results in
removing a considerable part of the surplus metal and bringing
each ball to a much closer approximation of a truly spherical
form than it is possible to obtain in forgings made by either of
the methods that have been described. In the case of hot-forged

Fig. 10. Front View of Grinding Machine, showing Grinding Wheel raised to

Operating Position and Tray of Ground Balls just removed from Machine;

Balls seen in Ring are not in Grinding Position

ball blanks, this rough-grinding also removes the decarbonized
steel from the surface of the blanks produced in forging.

An exception to the general method of procedure is made
in the case of balls from 1/16 to 3/16 inch in diameter. Such
balls are not dry-ground before being heat-treated, but they
get a rough and a finish dry-grinding after being hardened.
Figs. 9 to 11, inclusive, show the type of machine on which
the dry-grinding operation is performed, and the best idea
of its construction and method of operation will be obtained
by reference to the two views shown in Fig. 11. The main

Fig. 11.

Front and Side Views of Dry-grinding Machine, illustrating
Principle of Operation.

parts of this machine consist of a carborundum grinding wheel
A and an iron ring B which are driven in opposite directions.
Two rings C and D are supported by spiders in such a way
that there is a space between the beveled edges of the inner
and outer rings sufficient to allow ball blanks that are to be
ground to project through this space. In the side view of the
machine illustrated in Fig. 11, these rings are shown with the
wheel lowered, but when the machine is in operation the balls
held between rings C and D are in contact with grinding wheel
A ; and ring B presses down and holds them against the grinding

wheel. In order to provide for grinding the balls uniformly,
the spindles on which grinding wheel A and driving ring B are
carried are placed eccentric to each other, which results in giving
the balls an oscillating motion in addition to their motion of
rotation ; and as a result of this combined movement all surfaces
of the ball blanks are exposed to the action of the grinding wheel,
which results in bringing them to a close approximation of the
spherical form. The way in which the upper and lower spindles
of the machine are driven is best illustrated in Fig. 9, which
shows how open and crossed belts are brought to the machine
pulleys from an overhead countershaft.

Probably the best way to describe the operation of one of
these dry-grinders is to start at the point where a charge of
ball blanks has been ground down to the required size and is
to be removed from the machine. To provide for doing this,
the head which supports grinding wheel A is carried on a slide
on the base of the machine. Secured to the bottom of this slide
is a rack E that meshes with a pinion at the end of cross-shaft F.
Keyed to the opposite end of shaft F is a worm-wheel G that
meshes with a worm actuated by hand-wheel H that provides
fine adjustment. Secured to the bed of the machine is a disk /,
and in order to drop grinding wheel A out of contact with the
work held between rings C and D, the spring latch carried by
lever J is withdrawn from a notch in disk / and the lever is
moved to the left until the latch engages a stop notch in disk /,
which limits the downward motion of the grinding wheel. It
will be seen that sufficient clearance is, now provided between
grinding wheel A and rings C and D to enable tray K to be
swung into position to catch the balls when they are discharged
from the holding rings.

It will be seen that inner ring D is supported by a spider
secured to the lower end of rod L, and in order to discharge
the ground balls, ring D is dropped by pushing down lever M.
This drops the inner ring and allows the ground balls to fall into
tray K. When lever M is released, ring D is returned to its
original position by means of a compression spring N. During
the time that the charge of balls in the machine is being ground,
a fresh charge of blanks is placed in the space between driving
ring ,5 and outer ring C; a few of these balls will be seen in position
in Fig. 9. After the ground balls have been removed and inner

ring D has been returned to the position shown in Fig. 11, it is
necessary to place the charge of new blanks in position to be
ground. This is done by dropping both rings C and D sufficiently
so that the balls held between outer ring C and driving ring B
may drop into position, after which the two rings are returned
to the location shown in Fig. 11. This result is accomplished
by means of lever O that is carried at the end of a cross-shaft
which has a pinion at its right-hand end meshing with the rack
P cut in the sleeve that supports the spider on which outer
ring C is carried.

In order to drop a charge of balls into place, the spring
latch carried by lever 0 is released and this lever is pulled forward
which results in dropping both rings C and D, due to the fact
that rod Z/, supporting inner ring D, is pinned to the upper end
of sleeve P, to which the outer ring is connected by means of the
spider. When the balls have been dropped into position as
indicated, grinding wheel A is raised into contact with the work

Fig. 12. Special Grinding Machines for Grinding Rings shown at
C and D in Fig. 11.

by raising lever /. Rings C and D are ground to a smooth surface
and fine edge in order that the balls may run freely and reach
through the space to come into contact with the grinding wheel
A. This is done on special grinding machines, the method of
grinding the inner and outer rings being clearly illustrated in
Fig. 12. Lever Q at the front of the grinding machine operates
a clutch that provides for starting or stopping the machine.
It will be seen from Figs. 9 and 10 that the grinders are provided
with an exhaust system to carry away the dust of the wheel.

HEAT TREATMENT

DURING the process of making the steel for the balls and
in forging and rough-grinding the ball blanks made
from this steel, severe internal strains are likely to be
set up in the metal that would often be of sufficient magnitude
to cause the balls to be broken when subjected to only a small

Fig. 13. Charging End of American Rotary Gas Furnaces in which Balls
up to One Inch Diameter are Heat-treated.

part of their rated load carrying capacity. Trouble from this
source must be eliminated, and this is done by subjecting the
balls to a preliminary annealing operation in rotary gas furnaces
made by the American Gas Furnace Co. of Elizabethport, N. J.,
before the final hardening operation. The same type of furnace

Fig. 14. Discharge End of American Rotary Gas Furnaces, showing Quenching

Tanks and Deflector through which Balls are delivered to

Baskets at Bottom of Tanks.

is used for the annealing and hardening operations, but for the
former the delivery chute on the furnace is arranged to discharge
the balls into pans, as shown at A in Fig. 13, while for the latter
the balls are discharged into a quenching tank, as indicated in
Fig. 14. The form of retort used in these American gas furnaces
is shown in Fig. 15, and it will be seen to have a spiral path

Fig. 15. Cross-Sectional View of "Nichrome" Retort used in Rotary Gas Furnaces.

through which the balls pass as the retort is revolved. At the
loading end of each furnace there is a hopper that is kept filled
with ball blanks, and the retort draws blanks from this hopper
and passes them through the furnace at such a rate that the
steel is heated to the desired temperature when the balls are
discharged. For annealing, a temperature of 1300 degrees F.
is employed, and for hardening the balls are raised to a tempera-
ture of from 1425 to 1475 degrees F. according to the size and
the composition of the steel. Pyrometers made by the Hoskins
Mfg. Co. of Detroit, Mich., are used to determine the tempera-
ture of each furnace.

QUENCHING THE STEEL
BALLS

IT HAS been mentioned that the same type of furnace is
used for both the annealing and hardening operations, the
only change being to place the tube so that the ball blanks
are discharged into a pan in the case of annealing, and into
the quenching tank in the case of the hardening operation.
The retorts used in the furnaces were formerly made of cast
iron, and great trouble was experienced through their destruction
after they had been in service a short time. This trouble has

been over-come by substituting "Nichrome" in place of cast iron,
and retorts made of this material last indefinitely.

In hardening, there is a difference of practice according to
the size of the balls, those of 5/16-inch diameter and less being
quenched in oil while balls of larger size are quenched in water.
Balls made of some grades of steel are quenched in pure water
and others are quenched in brine. In all cases the quenching
tanks are provided with a device of the form shown in Fig. 14,
which consists of a series of conical sheet metal deflectors through
which the balls pass before reaching the wire mesh basket at
the bottom of the tank. The purpose of these sheet metal cones
is to deflect the course of the balls so that they follow a winding
path and are completely cooled before reaching the bottom of
the tank. One complete furnace charge can be run into one
of these wire baskets and when this is filled, the entire outfit
is lifted out of the tank by means of an electric hoist as shown,
and the balls are then removed from the basket. The depth of

Fig. 16. ''Frankfort" Oil Furnaces for use in Heat-treating Balls over One inch

Diameter, and Quenching Tank in which these Balls are Hardened. Note

Hoskins Pyrometer for showing Temperature of Furnaces.

the quenching tank is about 14 feet. Rotary furnaces are used
for annealing and hardening the smaller sizes of balls, and in
the case of balls one inch in diameter and over, ''Frankfort"
oil furnaces are employed, into which the balls are introduced
on trays as shown in Fig. 16. When the balls are heated to the
proper temperature, these trays are withdrawn and the balls
are dumped into the quenching tanks provided with the sheet
metal cones described. The reason for quenching small balls
in oil and large balls in water is that the oil does not absorb the
heat as rapidly as the water, and in the case of very small balls,
the shock of dropping them into water would result in strains
so great that many balls would either be cracked or broken, and
the strength of those balls in which there were no visible defects
would be seriously impaired. In the case of large balls, there
is sufficient heat to prevent trouble from this cause. From
time to time sample balls are tested by breaking them on an
anvil and examining the structure of the steel to make sure that
the heat-treatment is producing the desired results. Provision
must be made for preventing over-heating of the oil or water
in the quenching baths, and this is done by having a circulating
system through which the oil or water passes into a reservoir
outside the building and then through a coil in this reservoir
and back to the tank. In this way the contents of the quenching
tank are kept in continual circulation, preventing overheating.

SPECIAL TREATMENT TO RELIEVE
INTERNAL STRAINS

DURING the process of hardening, internal strains are
set up in the balls, and it is necessary, of course, to
relieve the strains without effecting the surface hard-
ness of the balls.

This is done by immersing the balls which are carried in
wire baskets, in a tank of boiling water for two hours. The
equipment used for this purpose is shown in Fig. 17.

This practice is only followed in the case of balls that are
hardened by quenching in water or brine.

Besides relieving the internal strains, the hot water prevents
the balls from rusting after their removal, as the hot balls dry
off very rapidly.

Fig. 17. Water Bath in which Severe Strains are Removed from Balls Quenched
in Water by subjecting them to Temperature of Boiling Water for Two Hours.
This Treatment also enables Balls to Dry Rapidly and Prevents Rusting.

FINISH
DRY-GRINDING

AFTER being hardened, the balls are sent back to the dry-
grinding room, where they are subjected to what is
known as a finish dry-grinding operation. This is the
same as the rough dry-grinding that the balls receive before harden-
ing, except that it is done with a finer wheel which results
in removing the scale produced in hardening and also
reducing their diameters a little closer to the finished size.
For the rough-grinding operation, wheels of No. 40 grit are
employed. On the finish-grinding, the grit of the wheel varies
according to the size of the balls. Wheels of No. 60 grit are
used for all balls exceeding 5/16-inch in diameter, while for
smaller balls wheels of 90 or 100 grit are employed. In all cases
the machines are driven at the required number of revolutions
per minute to give a surface speed of 4500 to 5000 feet per minute
at the point where the ring wheel engages the balls.

A VISITOR who is conducted through the plant of the
Hoover Steel Ball Co. finds it exceptionally easy to
become acquainted with what is going on in each shop,
because, although the plant is large, it is engaged in
making a single product, manufacturing operations on different
sizes of balls being conducted in essentially the same way through
out. This condition stands out in marked contrast to that found-
in plants engaged in the production of a variety of different
parts, as the manufacturing operations necessarily vary, making
it more difficult to see just what is being done.

c • oce *Dm ©E© c>Fo

Fig. 18. (A) String of Hot- forged Bail Blanks. (B}Ball Blanks made by Cold-
heading Process. (C) Rough Dry-ground Balls. (D) Rough Dry-ground Balls after
Hardening. (E) Finish Dry-ground Balls. (F) Oil-rolled Balls. (G) Oil-ground
Balls. (H) Polished Balls ready for Inspection.

Fig. 18 shows the condition of the product at each step in
the process of manufacture, and it will be of interest to study
this illustration carefully, as it shows just what is done to the
balls by each operation through which they pass before comple-
tion. At A is shown a string of hot-forged ball blanks before
they have been sheared apart, and at B are illustrated two ball
blanks made by the cold-heading process. Blanks produced
by either of these methods are first subjected to a rough dry-
grinding operation which reduces them to an approximately
spherical form, as shown at C, although the surface is covered
with a multitude of small flats and scratches left by the grinding
wheel. At D are shown two rough-ground blanks after they
have been subjected to the process of heat-treatment, and it
will be noticed that their appearance is essentially the same as
that of the rough-ground blanks shown at C except that the
surface is darkened as a result of the heat treatment. Two
blanks are shown at E, which have received the finish dry-
grinding after being hardened, and it will be noticed that the
appearance of these blanks is the same as that of the rough-

ground blanks C except that the flats and scratches are not so
pronounced. At F and G are shown two blanks that have gone
through a process known as ' 'oil-rolling" and two blanks that
have been through the oil-grinding process. The appearance of
both these balls is practically the' same except that the oilground
balls have been reduced to exactly the desired size. At H are
shown two finished balls after being polished, ready to be sent
on to the inspection department, where they will be subjected
to a series of rigid tests.

OIL-ROLLING BALLS IN
TUMBLING BARRELS

AFTER receiving the finish dry-grinding, the balls are of
approximately spherical form, but the surface is covered
with flat spots and scratches left by the grinding wheel
and there is still a considerable amount of excess metal on the
balls to be removed. The first step is to subject them to a process
known as oil-rolling which consists of tumbling a charge of balls
in an iron barrel containing oil and abrasive. This oil and
abrasive is refuse from machines on which a subsequent opera-
tion known as "oil-grinding" is performed; this operation will be

Fig. 19. View in Oil-rolling Department, showing Special Tumbling Barrels
of Large Capacity.

described in detail later, and the nature of the abrasive will be
explained at that time. Most of the tumbling barrels used in
this department have capacity for a charge of 1500 pounds of
balls, and these were built especially for the Hoover Steel Ball
Co. ; but some 800-pound barrels made by the Baird Machine Co.
of Bridgeport, Conn., are also employed. Some of these barrels
are shown in operation in Fig. 19. The purpose of oil-rolling
is to smooth off the flats and scratches left by the dry-grinders
and to remove excess stock, about 0.004 inch being allowed
for removal in the oil-grinding operation. Balls up to 1^2-inch
in diameter are given this oil-rolling treatment.

It is necessary to leave the balls in these tumbling barrels
from twenty to thirty-six hours, according to the amount of
stock that must be removed, and as each ball rotates in such a
way that its entire surface is uniformly exposed to the action

Fig. 20. Oil-grinding Machine on which Final Grinding Operation is performed

— Attention is called to Dials showing Approximate Time when Grinding

will be Finished, and Indicator for Testing Size of Balls.

of the abrasive and of the balls adjacent to it, this treatment
results in the production of perfect spheres. When the time has
almost arrived at which the balls should be removed, a
number are selected at random from the contents of
each barrel, taken out and measured with a micrometer
in order to see how closely they approach the required size.
The oil-rolling is then continued with successive gaugings until
the balls have been reduced to the required dimension plus 0,004
inch, after which they are removed from the barrels, cleaned,
and then taken to the oil-grinding department. In reducing
balls by the process of oil-rolling, it occasionally becomes neces-
sary to add more abrasive to the supply of oil and abrasive ob-
tained from the oil-grinders. When this is done, No. 36 carborun-
dum is used, as this coarse-grain abrasive increases the speed at
which the balls are reduced to the required size.

HOW THE PROCESS OF OIL-GRINDING
IS CONDUCTED

THERE are two main grades of balls made in the Hoover
factory, known as "Micro-chrome" and "Commercial"
balls, the former being the better quality. Both grades are
reduced to the final size by the process known as "oil-grinding"
that is conducted on machines of the form shown in Figs. 20

Fig. 21.

Side and Front Views of Oil-Grinding Machine,
Illustrating Method of Operation.

and 21. The construction and operation of the oil-grinding
machines will be best understood from Fig. 21, which shows
details of its construction. These machines are provided with
two iron rings A and J3, each of which has an annular groove cut in
it of a suitable size to accommodate the balls C to be ground.
It will be noted that there is a small groove at the bottom of the
annular groove in the lower ring A , which provides for holding a
supply of oil and abrasive. Ring A has the annular groove for the
balls cut at the bottom of a larger groove, and ring 5 has a flange
in which the ball groove is cut that drops into this large groove
in ring A ; the arrangement will be readily understood from the
illustration. It will, of course, be understood that the grinding
ring is rilled with balls, the number that constitutes a complete
charge varying according to the size of balls that are being
ground.

To provide for loading and unloading the machine, lower
ring A is drawn out onto a table D which is provided for that
purpose, and after a fresh charge of balls has been put in place
this ring is pushed back into position under the upper ring B
that is secured to the spindle of the machine. A sheet metal
shield is then pushed into place in front of the rings in order to
prevent splashing of the oil. Ring A is located in approximately
the desired position by means of a hole in the machine bed
into which an extension on the under side of ring A drops, but
the extension on this ring is a loose fit in the hole to allow ring
A to align itself properly with ring B.

The upper ring is secured to the spindle, and in order to
start the grinding operation it must be lowered into contact
with the balls carried in the annular groove of ring A. This is
accomplished by a rack on the spindle sleeve that meshes with
pinion E secured to lever F.

In order to raise ring B out of contact with the work so that
ring A may be drawn out onto turntable D, lever F is pulled down
into the horizontal position shown in the illustration. In this
position spring latch G drops into a notch on ring H that is
secured to the frame of the machine, thus holding ring B in the
suspended position. After the machine has been reloaded and
it is desired to drop ring B into contact with the work preparatory
to starting the grinding operation, spring latch G is withdrawn

from the notch in ring H by pulling back grip / that is connected
to the end of the rod on which latch G is carried. Then the wheel
is lowered by gravity, care being taken to hold tight to the
crank at the end of lever F so that it is slowly raised to a vertical
position instead of flying up and allowing ring B to drop heavily
onto the balls carried in the lower ring.

It will be seen that there are three grinding heads provided
on each machine, and these are furnished with independent
tight and loose pulley drives, so that any head may be stopped
without interfering with the operation of the other two. This
is done by throwing the belt from the tight to the loose pulley
by means of lever /, which actuates the belt shifter. The oil-
grinders are provided with a dial similar to that of a clock,
so that the time for grinding can be observed; the grinding
operation usually takes from twenty to forty-five minutes, ac-
cording to the size of the balls and the amount of stock that
must be removed. When the machine is set up ready to start
the grinding operation, this dial is set to the approximate time at
which the grinding operation will be completed, and a little while
before this time is reached several balls are selected at random
from different points around the ring, and are measured with an
indicator to see how near they come to the required size. The
dials on the machine and the test indicator are shown in Fig. 20.

Fig. 22. Small Tumbling Barrels for Cleaning Balls in Sawdust, and Riddles foi
Separating Sawdust from Balls.

CLEANING AND POLISHING
OIL-GROUND BALLS

AS SOON as the balls have been ground down to the desired
diameter, they are removed from the machine and taken
to tumbling barrels containing hardwood sawdust, in
which they are rolled for a sufficient length of time to clean off
all oil and abrasive. The charge in each tumbling barrel is then
taken out and put into riddles through which the sawdust is
sifted, as shown in Fig. 22, to separate it from the balls; the
balls next go to the tumbling barrels containing a mixture of oil

Fig. 23. Kegs in which Balls are Polished by Rolling in Leather.

and Vienna lime. They are rolled in this mixture for a sufficient
length of time to give them a preliminary polish, after which
they are removed and again cleaned in tumbling barrels filled
with hardwood sawdust. The sawdust is sifted from the balls
in riddles, after which they are rolled for from twenty to twenty five-
minutes in kegs containing strips of kid similar to that from which
gloves are made, the arrangement of this polishing equipment
being shown in Fig. 23. Rolling the balls in this way gives them
a high polish, which is the final step in the process ; and the finished
balls are then ready to be taken to the inspection department.

The following data concerning conditions under which
oil-grinders are operated and abrasives and oils used on these
machines will prove of interest. It has been mentioned that
two main grades of balls are made, which are known as ' 'Micro-
chrome" and ' 'Commercial" the former being the better quality.
On the "Micro-chrome" balls the grinders are run at 195 revolutions
per minute and the abrasive used is a mixture of No. 3-F car-
borundum and "Atlantic Red" machine oil made by the Standard
Oil Co. On "Commercial" balls, the grinders are run at a speed of
325 revolutions per minute and the abrasive is an equal mixture
of Nos. 180 and 150 carborundum to which No. 4 "Road Oil"
is added, this oil also being the product of the Standard Oil Co.
Used oil and abrasive from the grinding machines is collected
and used in the tumbling barrels.

SPECIAL TREATMENT FOR
LARGE BALLS

CERTAIN variations from the practice described in the
preceding paragraphs are necessary in the case of large
sized balls which would be too heavy to handle in tumbling
barrels. For instance, "Commercial" balls over 1^-inch in diameter
and "Micro-chrome" balls over 5/g-inch in diameter are burnished
on oil-grinders running at high speed and in which very fine
abrasive and light oil are used instead of being subjected to a
tumbling operation in barrels containing a mixture of oil and
lime, as previously described. If large balls of this kind were
put in a tumbling barrel, there would be too much shock from
the balls striking one another; hence the variation in practice.

PRODUCTION OF OIL-
ROLLED BALLS

IT HAS been explained that in the regular process of manu-
facture the balls go from the tumbling barrels to the oil-
grinders on which they are reduced to the required size
ready for polishing. There are some cheaper grades of balls,
however, that do not go to the oil-grinders; these balls are reduced
to size by oil-rolling in the tumbling barrels, after which they
are polished and sent to the inspection department. The method
of polishing is the same as that to which the better grades are
subjected, which was previously described. In oil-rolling the
balls, a mixture of No. 36 carborundum and No. 4 "Road Oil"
is used in the tumbling barrels.

MANUFACTURE OF BRASS, BRONZE
AND COPPER BALLS

IN ADDITION to its regular product, the Hoover Steel Ball
Co. does quite an extensive business in the manufacture of
brass, bronze and copper balls of various sizes. One important
use of these balls is for various forms of valves, although they find
a number of other applications. The general features of the
methods used in producing these balls are the same as those
employed in making steel balls, but there are certain modifications
which will prove of interest. Brass, bronze and copper ball blanks
up to 1 ^-inch in diameter are produced on Manville cold-headers,
and blanks for balls exceeding this size are cast. In the case
of very large balls the practice is often adopted of making the
blanks hollow, which is done by casting them with a sand core
that is subsequently removed. Then in order to prepare the
blank for finishing, the holes left by the core prints are drilled,
reamed and tapped so that threaded plugs may be screwed in.
These hollow ball blanks are then subjected to the regular process
of manufacture, and it is a difficult matter to detect the place
where the plugs have been screwed in.

As in the case of steel balls, these blanks are first subjected
to a process of dry-grinding to make them approximately spheri-
cal. Brass, bronze and copper balls are too soft to stand treatment
in tumbling barrels, as they would be covered with bruises from
impact with each other. After being dry-ground, they receive

the regular process of oil-grinding and are then polished in
machines of the same design as those used for oil-grinding; but
in polishing, the balls are rolled in oil without any abrasive,
which results in giving them quite a high polish, although the
surface produced is not as highly finished as in the case of steel
balls which are subjected to burnishing and polishing operations
after being oil-ground. In treating brass, bronze and copper balls
in the oil-grinding machine, care must be taken not to subject
them to too great pressure, and in order to guard against this
the rings on the machine are filled with brass and steel balls
arranged alternately; the steel balls support the pressure of the
upper ring and the head on which it is carried, and allow the
balls to be ground and polished without being subjected to
sufficient pressure to flatten them.

INSPECTION OF FINISHED
BALLS

AFTER each step in the process of manufacture, the balls
receive a general inspection to make sure that nothing
is wrong with the adjustment of the machines or with the
material from which the balls are made that will prevent the
production of balls that come up to the standard. After receiv-
ing their final polish, the finished balls go to the inspection
department, where they are subjected to a number of searching
tests in order that all defective balls may be eliminated and
that those balls which pass inspection may be divided into
various grades according to the accuracy of their dimensions.

The first step is to clean the balls thoroughly, which is done
by placing them in metal baskets provided with long handles
so that the load of balls may be dipped into gasoline to remove
grease and particles of leather carried over from the polishing
department. After this washing, the balls are put into canvas
bags and rolled on a table so that the bags will absorb the
gasoline and wipe off the dirt. The balls are given a preliminary
wiping in one of these bags, after which they are placed in a
second bag that is cleaner and insures the removal of the last
traces of gasoline and dirt.

MAKING PLATE
INSPECTION

AFTER cleaning, the first actual examination is conducted
on what are known as "inspection plates, "one of which is
shown in Fig. 24. These plates are used on benches that
run all the way around the two inspection rooms, so that ad-
vantage may be taken of the liberal amount of daylight provided
by the windows which extend from below the bench up to the
ceiling. The plates are made of glass and painted black. A
reflector is set up at the back of each inspection plate which
throws light on the balls; and a strip of thin flexible cardboard is
drawn back and forth beneath the balls to rotate them and
bring all surfaces into view. Several times while making this
inspection all the balls on the plate are rubbed with a cloth
to change their axes of rotation and insure exposing the whole
surface. The first step is to pick out balls having cracks, flats,
etc., and these are sold as seconds or scraps.

Fig. 24. Type of Glass Plate on which Preliminary Inspection is Conducted.

During the next step in the process of inspection, attention
is paid to a white spot on each ball that is thrown from the
reflector at the back of the inspection plate. As previously
mentioned, a card is drawn back and forth under the plates
to make them revolve, and the inspectors first pick out what
are known as "wigglers," which is the name given to balls
that are out of round and go through a series of contortions
while being rolled. After this has been done, the balls on the
plate are gone over carefully and all those that show any defect
are picked out. During this process of inspection, the balls
are sorted into eight grades, as follows: (1) "Cracked," balls
that have received their cracks from any cause, (2) "Junk,"
balls which have flats, holes, etc.; (3) "Rubbish," same
defects as (2) but not so bad; (4) "Dead soft," balls that
are covered with small pits caused by impact with hard balls
during the process of tumbling; (5) "Out of round," balls known
as "wigglers" by the inspectors; (6) "Fifth grade," balls with
small cuts and scratches on them; (7) "Fourth grade," balls
showing same defects as "Fifth grade," but not of so serious a
character; (8) Balls having no defects sufficiently serious to
be visible to the eye. The inspectors engaged in making the
plate inspection are provided with small magnets somewhat the
shape of a pencil with which they handle the balls with amazing
dexterity.

Disposal of the defective balls varies somewhat according to
their size. Many of the small balls with defects of the kind
referred to are sold to various manufacturers, according to the
class of service required of them. For instance, very poor balls
are sold to novelty makers. Other balls that are not good enough
for use in high-grade ball bearings are plenty good enough for
the use of certain manufacturers of hardware specialties, such
as roller bearing castors for furniture, roller bearing roller skates,
etc. Large balls that are found defective are returned to the
manufacturing department, where they are ground down to a
smaller size in order to remove the defects from the surface of
the metal ; and these balls are again carried through the regular
process of manufacture.

GAUGING BALLS FOR
SIZE

BALLS that are used in annular bearings must be of abso-
lutely the same size in order to give satisfactory results.
If this is not the case, the large balls will support all the
load, and the undue amount of service to which they will be
subjected will cause them to be destroyed more rapidly than
would otherwise be the case. In order to fit properly in" the
races, it is desirable for the balls to be of exactly the specified
size, but provided all the balls are of the same size, they are
capable of giving very satisfactory results even though they
are either slightly over or under the specified size. In the
final process of inspection, the balls are gauged and sorted out
into different grades, according to whether they are of exactly
the specified size or somewhat under or over this size. Attention
is called to the fact that this variation in high-grade steel balls
does not exceed- a few ten- thousandths inch. As balls of the
different grades are all of the same size, they are capable of
giving perfectly satisfactory results. Some users of balls gauge
them at their own plants and make this sub-division, while others
buy gauged balls ready for assembly.

In gauging those balls which show no defects in conducting
the plate inspection, practice varies according to the size of
the balls, but in all cases the object is the same, namely, to
sort the balls out into those which are of absolutely the desired
size and those which vary by different degrees either above or
below the standard. Balls up to and including ^-inch in diameter
are gauged on automatic machines which sort them into seven
different grades, as follows: balls exceeding 0.0002 inch over size;
balls 0.0002 inch over size; balls 0.0001 inch over size; balls of the
specified size; balls 0.0001 inch under size; balls, 0.0002 inch
under size; and balls more than 0.0002 inch under size. Auto-
matic gauging machines are used for this grading, two batteries
of such machines being shown in Figs. 25 and 26. The balls
are placed in hoppers A, at the bottom of each of which there
is a plate in which a number of holes are drilled in a ring, these
holes being of slightly larger size than the balls to be gauged.
The plates are revolved, and as each hole comes into line with
the delivery tube, the ball carried in this hole drops into the

Fig. 25. Close View of Battery of Automatic Gauging Machines
with Inclined Blades.

tube and runs down over gauge blades B which are set at a slight
angle to each other so that balls of the different sizes referred to
will drop between the gauge blades and enter tubes that carry
them to the proper drawers in the cabinets beneath.

It will be seen that two types of machines are shown in
Figs.. 25 and 26. In Fig. 25 the gauge blades are placed on an
incline so that the balls run over them by gravity, and as the
balls are always in contact with the gauge blades, the tubes lead-
ing to the drawers of the cabinet can be placed much closer
together than on the type of machine shown in Fig. 26, where

Fig. 26. Close View of Battery of Automatic Gauging Machines
with Horizontal Blades.

the gauging blades are in a horizontal position. On the latter
type of machine an agitator is necessary to keep the balls moving
over the gauge blades. This agitator consists of a crank C and
connecting-rod D that actuates a link mechanism which causes
a horizontal bar to rise in the space between the gauging blades.
This bar rises slightly and then moves forward, carrying the
balls with it, after which the agitator bar slowly drops and
leaves the balls once more supported on the gauging blades. In

this way the balls are moved along over successive tubes and
finally drop through between the gauging blades — the position
being determined by the size of the balls — so that different sizes
of balls are sorted out as previously described. A stop checks
the progress of the ball as it passes onto the gauging blades, and
prevents it from rolling too fast. The gauging blades are set by
master balls, in order to have the desired angle between them;
and before the balls are packed, the accuracy of the blade setting
is tested.

SPECIAL INDICATOR FOR
TESTING BALLS

FOR gauging balls larger than 5/s-inch in diameter use is
made of. an instrument of the form shown in Fig. 27.
This will be seen to consist of an ordinary Brown &
Sharpe dial test indicator accurate to 0.0001 inch, that is set

Fig. 27.

Dial Indicator with 10 to 1 Leverage Ratio, for Testing
Accuracy of Balls to 0.0001 Inch.

up on the table on which is also carried a holder for the ball to
be tested. Connection between the ball and the dial test indicator
is made by a lever, the fulcrum of which is so placed as to give
a ratio of 1 to 10, and in this way readings obtained are accurate
to 0.0001 inch. The girls who conduct this inspection handle
the balls very rapidly and sort them out into different sizes
according to the amount of deviation from the normal size.

COUNTING AND PACKING
BALLS

IT IS necessary to use great care in handling finished balls to
prevent them from becoming rusty. On this account it
would not do to have the balls touched by the fingers.
For these reasons, several methods of mechanical counting have
been developed which give extremely satisfactory results. The
apparatus used for this mechanical counting is shown in
Fig. 28. The balls are placed in hopper A and dropped
down in holes in sliding plate B, which is pushed forward
so that the holes are under the hopper during the "loading"

Fig. 28. Methods used for Counting Balls Preparatory to Packing.

period. The plate is then drawn forward to allow the balls to
drop out into a box placed to receive them. Each stroke of the
plate counts out one hundred balls, and plates for counting balls
of various sizes are made interchangeable so that all of them
may be used on a given machine. Balls up to J^-inch in diameter
are counted by the machine, and balls from 9/16 to J/g-inch in
diameter are counted mechanically by means of board C, into
the grooves of which the balls are loaded up to an index line.
Plates of this kind are made for various sizes of balls, and each
plate holds 500 balls. Large balls are counted by hand, care
being taken not to touch the balls with the bare fingers. After
counting, the balls are packed in cartons lined with waxed paper,
and these are packed in substantial wooden boxes for shipment.

RESEARCH
DEPARTMENT

IT IS obvious that in the tonnage manufacture of a product
that must meet such exact requirements as balls for use
in high-grade annular bearings, the greatest care must be
taken in the selection of raw material and in conducting each
step in the process of manufacture in order to produce balls
that will pass the inspection department. In addition to the
requirements of high-grade balls that were referred to in the
description of various examinations that are conducted by the
inspectors, it is absolutely necessary for the balls to be of uniform
hardness and strength because this is the only way of being
sure that all balls will possess the necessary durability and
elasticity.

Assurance must be obtained that the steel received at the
factory is of a suitable grade to produce balls that will fulfill
the specifications before manufacturing operations are started,
because if the balls were finished before it was found that they
were defective, the raw material and the labor involved in
converting this material into finished balls would be lost. Data
showing that the steel fulfills these specifications 'are obtained
from the results of tests conducted in the testing department
which is equipped with all the necessary apparatus for making
physical and chemical tests upon the raw material. In addition,
this department is referred to by heads of the various manufac-

turing departments when any case of trouble arises, such as failure
of the balls to harden properly, the production of more than the
usual number of balls with cracks, and other troubles of this
kind. Some exceptionally interesting facts have been brought to
light as the result of work conducted in the metallurgical
department and chemical laboratory.

TESTING

RAW MATERIAL

THERE are sidings from the Ann Arbor Railroad entering
the plant so that cars may be run directly to the building
in which the raw material is received and to the building
where the finished balls are packed for shipment. The method
of procedure in testing raw material is the same for both bar
stock and coil, and consists of taking at random a number
of each kind in proportion to the quantity received and from
the end of each of which is cut a sample. One end of this
sample is etched in dilute hydrochloric acid for fifteen minutes.
After this has been done, the surface of the metal is carefully
examined to see that it is free from seams. The acid tends to
accentuate any surface defects that may be present, so that
those that might be invisible in the bar as it comes to the
plant can be quite easily seen after the treatment. In ball
manufacture it is highly important for the stock to have a
flawless surface, because any slight defects are carried right
through the process of manufacture and are likely to become
accentuated, with the result that balls produced from this stock
will be rejected by the inspectors.

The regular routine tests of the raw material inspected in
the laboratory also include a Brinell hardness test. This is
especially important in the case of "wire" under 11/16 inch in
diameter that is converted into ball blanks by the cold-heading
process, because excessive hardness of this material is likely
to give trouble through the breakage of the cut-off knives or
the dies used on the cold-headers. In order to give the best
possible results, stock for the cold-heading machine should have
a Brinell hardness of not over 170. A sufficient number of
samples to represent the average uniformity of the shipment
are examined for pipes, segration or decarbonization, and when

necessary microphotographs are made, which together with
their accompaning reports, put definitely on record the condi-
tion of each shipment. Samples are also taken for chemical
analysis from each shipment and the percentage of the most
important elements determined, this being influenced by the
kind of material received and the effect of these elements on the
finished product. In cases where laboratory tests do not show
that the stock is defective, an "unloading ticket" is made out
and sent to the stock- room, authorizing the material to be
taken from the cars and placed in storage, ready to be drawn
out on requisition by the manufacturing department.

On the following pages are given our specifications for
coil and bar stock, and a consideration of these will show the
care taken in the selection of raw material used in the manu-
facture of Hoover steel balls.

HOOVER STEEL BALL CO.

SPECIFICATION NO. 1.
Chrome-Carbon Steel Wire — Cold Drawn.

ANNULMENTS:

1. This specification supercedes all previous specifications, or letters of instruction,
covering this material.

MANUFACTURE:

2. The material must be made by the Electric or Crucible process.

QUALITY:

3. The material must be of highest quality in every respect, of uniform composition,
and free from slag or other segregation.

The wire must be free from imperfections, such as pipes, seams, checks or lamina-
tions either on the surface or in the section of the wire.

WORKMANSHIP AND FINISH:

4. The wire must be of good workmanship, must have a good surface finish, and
must be true to diameter ordered within the limits of plus .002" and minus .002".
If the wire is out-of-round, the mean of the largest and smallest measured diameter
must be equal to the size ordered, but in no case can they exceed the limits of
plus .002" and minus .002".

COMPOSITION:

5. Upon receipt of the material at destination, drilling may be taken from the
several coils, selected at random, for analysis, and must show the composition
of the material to be uniform and within the following requirements.

Carbon .95 % to 1.05 %

Chromium .35% to .45%

Manganese .80% to .45%

Silicon .20% to .35%

Phosphorus under .025 %

Sulphur under .025%

CONDITIONS:

6. The material must be thoroughly and uniformly annealed and the fracture
must be close grained.

The Brinnell hardness (5 m/m Ball under 1000 Kg. pressure) must not exceed
170 at any point in the length or any point in the cross section of the wire, so
that when blanks made therefrom are cold upset into the form of a Ball, no
defects will open up in the outside surface of the Ball.

The wire must be free from any decarbonized surface and after hardening must
show a close grained velvety fracture.

COIL SIZE, WEIGHT AND CONDITION:

7. Coils must be reeled uniformly and the layers must be bound together securely
with separate tie wires to keep them in good shape during transportation so that
they can be unwound properly without tangling. If the ends of the coil are
tapered down or imperfect in any way, they must be "cropped" off.

Coils may be covered with a coating of oil or grease to protect them from excessive

rusting during transportation, but the coils must be free from any hard or gritty

foreign matter that would interfere with their proper operation in the heading

machine.

The coils must not be less than 18" inside diameter or greater than 34" outside

diameter. Wire of heavy cross section should be wound in as large a coil as

possible, but within the outside diameter limit given above.

The coils should weigh not less than 90 pounds or more than 110 pounds for wire

above .235" diameter. Coils of wire below .235" diameter may weigh as low as

70 pounds.

REMARKS:

8. Material which fails to meet the above requirements will be rejected and returned.
The manufacturers must pay all transportation charges on rejected material.
Ann Arbor, Mich., January 1st, 1917.

HOOVER STEEL BALL CO.

SPECIFICATION NO. 5.
Chrome-Carbon Steel Bars— Hot Rolled.

ANNULMENTS:

1 This specification supercedes all previous specifications or letters of instruction
covering this material.

MANUFACTURE:

2. The material must be made by the Electric or Crucible process.

QUALITY:

3. The material must be of highest quality in every respect, of uniform composition,
and free from slag or other segregation.

The bars must be free from imperfections such as pipes, seams, checks or lamina-
tions either on the surface or in the section of the bar.

WORKMANSHIP AND FINISH:

4. The bars must have as good a surface finish as is consistent with good hot rolling
practice. They must be free from excessive scale, and must be true to diameter
ordered within the following limits.

Minus 0 and plus .005" for sizes under 13/16" diameter.

Minus 0 and plus .010" for sizes over 13/16" diameter.

If the bar is slightly out-of-round, the mean of the largest and smallest measured
diameter must be within the minus and plus limits given above.
Appended to this specification is a table giving the prevailing sizes (diameter)
of stock which we use, and the corresponding decimal sizes. We reserve the
right to change this list from time to time when necessary but the order or contract
calling for the material will specify the size wanted.

As an example, if our order calls for 13/16" plus .010" (Decimal .823"), the
manufacturer may supply this as large as .833" diameter but no smaller than
.823" Diameter.

COMPOSITION:

5. Upon the receipt of material at destination, drillings may be taken from the
several bars, selected at random for analysis, and must show the composition
of the material to be uniform and within the following requirements.

Carbon .90% to 1.00%

Chromium .60% to .70%

Manganese .30 % to .45 %

Silicon .20% to .35%

Phosphorus under .025 %

Sulphur under .025 %
CONDITIONS:

6. The material must be thoroughly hotvworked to produce a fine grain and must
not, subsequent to this hot working, be subjected to a high temperature such
as would produce a coarse grain.

The surface of the bars must be free from decarbonization to the extent that upon
removing .005" from the diameter of the bar, the remaining section will retain
its full quota of carbon as called for under composition.

The bars must be cut to uniform lengths as ordered. A preferred length will be
specified on the order, also a minimum length and a maximum length, but in no
case may intermediate lengths be supplied. For example, a 13/16" plus .010"
diameter bar will be ordered cut to lengths 64", 73" and 82" with 73" as the
preferred length.

SHIPPING:

7. When two or more different sizes are shipped together in the same car, they must
be so arranged and located in the car that they will not become mixed during
transportation .

REMARKS:

8. Material which fails to meet the above requirements will be rejected and returned.
The manufacturers must pay all transportation charges on rejected material.
Ann Arbor, Mich., January 1st, 1917.

TESTS OF SEAMY COLD-
DRAWN WIRE

IN DESCRIBING the inspecting of balls, reference was made
to the rejection of those in which cracks are found.
These exist almost entirely in balls up to and including
5/8-inch in diameter, the blanks for which are made by the cold-
heading process; it seldom happens that cracked balls are
found in sizes over j^-inch, blanks for which are made by the
process of hot-forging. A study of this subject reveals the
fact that after cold-heading, ball blanks very often had some
sort of crack, and in a great many cases these were quite
deep. At first it was thought that this was due to faulty
annealing or to some element in the steel, which had a tendency
to make the metal brittle, but subsequent investigation showed
that this was not the case.

STUDY OF SEAMS IN STEEL
BARS AND WIRE

DEFECTS revealed by etching the metal in dilute hydro-
chloric acid run lengthwise of the bar; sometimes these
extend for the entire length of the coil, while in other cases
only one end is found to be defective. For want of a better
name, the laboratory has called these defects ' 'seams," and it
has been proved that wire with seams will in all cases be split
to some extent during the process of cold-heading, while that
without seams will produce perfect balls in the cold-heading
machines. In some cases the cracks opened up in the balls while
cold-heading are not so deep that they cannot be eliminated dur-
ing the subsequent treatment to which the blanks are subjected ;
but in other cases it may happen that these splits in the blanks
are so deep that they reach below the surface of the finished
balls, in which case the balls will be rejected by the inspectors.
The investigation conducted in the laboratory relative to
troubles resulting from stock having seams or scratches have
developed the following information: (1) Cold-drawn wire
on which the surface is apparently quite smooth, and on which
no seams are visible, is found in many cases to possess minute
laps or seams which are made visible by etching with dilute

hydrochloric acid. (2) Although these seams may not be deep
on the original wire, they are accentuated by the stretch which
the surface of the wire undergoes during the cold-heading
operation. (3) Such cracks are likely to be still further ac-
centuated in hardening, and in many cases they will cause the
ball to split in half.

In making a study of the effect of seams on the steel,
it is the practice, as previously mentioned, to etch the stock
with dilute hydrochloric acid for fifteen minutes.

The action of the acid first lays open any surface defects
which may be closed so tightly by the pressure of the cold-
drawing operation that they will be invisible to the eye
unless subjected to the acid treatment. The acid also makes
the cracks black, and subsequent grinding exposes the white
surface of the adjacent metal so that the crack is brought into
as great prominence as possible.

TESTING FOR SEAMS IN STOCK BY
APPLICATION OF PRESSURE

RECENTLY another test for revealing these seams has
been developed, which consists of upsetting short blanks
cut from the bars. These test blanks for wire having a
diameter of .275", are 7/16-inch high and are ordinarily

Fig. 33. (A) Samples cut from Steel with Seam in Surface, and Same Samples

partially and fully upset, indicating how Seam opens up through Application

of Pressure; (B) Similar Samples from Steel without Seam, which

show No Tendency to Split.

subjected to a pressure of 20,000 pounds, which results in
flattening them out to a height of 3/16-inch, or to a
pressure of 50,000 pounds, which flattens them out to a
height of 3/32-inch. In all cases where there are seams in the
wire, these test samples are split open by this pressure, while a
perfect wire without any seams is not damaged by the treat-
ment. At A in Fig. 33 is shown a sample cut from wire con-
taining a seam and the same blank partially and fully upset;
it will be noticed that, although the seam in the wire is small,
it has been widened out considerably by the upsetting. At B
in the same illustration is shown a similar set from perfect wire,
comprising a blank and partially and fully upset samples, and
it will be seen that the upset sample does not show any tendency
to split.

In order to give some idea of the extent to which the seam
at A was deepened by the upsetting treatment, section a-b
through the blank and section c-d through the flat disk were
polished, and photomicrographs of these are shown in Fig. 34.
At A in Fig. 34 the seam in the original wire was about 0.010
inch in depth, while at B the depth of the seam after the blank
has been upset has been increased to approximately 0.050 inch.
From this it will be apparent that seams in the wire that do
not appear to be of sufficient depth to give trouble may become
very objectionable because of the tendency to deepen during
the conversion of the stock into ball blanks. Upset disk B
is of about the same diameter as a ball blank made from this
wire by the cold-heading process, so that it has been subjected
to about the same amount of stretch in upsetting that would
ordinarily take place in making a ball blank by the cold-heading
process. To show how trouble may develop in this way, a
ball 0.375-inch in diameter is produced from a blank 0.400
inch in diameter, so that the blank is reduced 0.025-inch on
the diameter, or approximately 0.013-inch on the radius. This
leaves 0.050 minus 0.013, or 0.037-inch of the split extending
below the surface of the finished ball, which will certainly lead
to its rejection by the inspectors.

It appears that hardness of the wire does not cause splitting
of the upset blank. Tests conducted with a view to establish-
ing this fact have shown that blanks made from seamless steel

Fig. 34. Photomicrographs of Sections on Lines a-b and c-d in Fig. 33,
indicating Increase in Size of Seam through stretching of
Metal Surface in Upsetting.

with a high Brinell hardness number did not split under the
most severe conditions of upsetting, while blanks of metal
with a low Brinell hardness number, but with seams on their
surfaces, were frequently split during the process of cold-heading.
Specifications under which steel is purchased for the production
of ball blanks in cold-heading machines call for metal with
a hardness number not exceeding 170 as determined by the
Brinell method, but slightly harder stock is capable of being
worked with fairly satisfactory results.

HOW SEAMY STOCK ACTS IN
COLD-HEADING MACHINE

IN ORDER to confirm the accuracy of the conclusions reached
in regard to the action of seamy stock when worked up into
ball blanks in the cold-heading machines, tests were
conducted by placing coils that had bad seams in them on the
cold-headers and observing the kind of ball blanks that were
produced. In every case it was found that the blanks produced
from such stock showed bad cracks, as shown at A in Fig. 35.
In the inspection department, cracks found in finished balls
were at one time commonly referred to as "fire cracks" on the
assumption that they were developed during the process of heat-
treatment, but they are now designated as ' 'header cracks." In

this illustration attention is called to the fact that at the top
and bottom of each ball blank there is a small projection formed
by pressing the metal into the knock-out pin hole in the header
dies. These have been termed ' "poles," and it will be noted
that the poles lie on the axis of the wire. Midway between
the two poles there is a band or "fin" caused by the metal being
forced out between the two header dies; and this fin has been
termed the "equator" of the ball.

Fig. 35. (A) Cold- header Ball Blanks, showing Splits running from Pole

to Pole, (B) Finished Balls produced from Blanks Split during

Cold- heading Operation.

It will be noted at A in Fig. 35 that the header cracks run
from pole to pole. At B in the same illustration are shown
some finished balls with the same kind of cracks, and it has
always been found that cracks in the finished balls have been
lengthened to a considerable extent, the ends of these cracks
terminating in very fine lines. This can be readily understood
when we consider that a small crack or fine sharp tool mark on a
piece to be hardened causes a weak spot which in many cases
will result in splitting the piece during the process of heat-
treatment. At A in Fig. 36 are shown some balls that were
picked out in the inspection department because they had fire
cracks; these were sent to the laboratory and fractured to
reveal the grain of the metal. It will be noticed — particularly
in the third ball of the third line — that at the extreme left of
the fracture there is a dark spot near the surface, which is

the mark left by the original crack produced during the cold-
heading operation. Then to the extreme right there is a
fresh fracture which represents all the metal that the ball had
to hold it together after being hardened.

Attention is called to the fact that the middle of the ball is
black and oily; this is the hardening crack into which the oil
and abrasive have found their way during the oil-rolling and

rveo

Fig. 36. (A) Fractures of Balls shown at (B) in Fig. 35, showing Original

Header Crack, Fire Crack and Fracture of Uncracked Metal, (B) Etched

Balls, showing Crack from Pole to Pole and Crack on Equator.

grinding operations. The crack produced in cold-heading was
the cause of a further cracking of the ball during the process
of heat-treatment. At B in Fig. 36 are shown some finished
balls that were rejected by the inspectors because of cracks.
Before being photographed these balls were etched with dilute
hydrochloric acid, and it will be noticed that the cracks run
from pole to pole, and in some cases there are also secondary
cracks following the line of the equator. The way in which
these equatorial cracks are produced can best be explained by
reference to Fig. 37. At A is shown a longitudinal section of
the wire which has been etched with hydrochloric acid to
reveal the structure of the metal. Attention is called to the
lamellar structure, which is characteristic of any steel and is no
reflection upon its quality. These laminations run lengthwise of
the coil. At B is shown a section of a headed ball blank made
from a piece of this wire and etched with acid to bring up the

Fig. 37. (A) Section of Steel Stock, showing Lamellar Structure; (B) Cross-
section of Cold-header Ball Blank, showing Distortion of Steel Structure; (C)
Cross-section of Header Cracked Ball Blank; (D) Ball Blank shown in Cross-
Section at (C); (E) Cold-header Ball Blank with Large Fin; (F) Perfect Cold-
header Ball Blank; (G) Etched Ball, showing End Grain of Steel at Equator;
(H) Etched Ball, showing End Grain of Steel at Pole.

structure of the metal. Here it will be seen that the lamina-
tions have arranged themselves in a manner similar to magnetic
lines of force running from pole to pole.

At D and E are shown header-cracked ball blanks, and it
will be noticed that blank E shows an unusually large fin on
one side. Blank D shows the split on one side and also a por-
tion of the split extending into the fin. Blank F is properly
headed and shows no crack or excessively large fins. Referring
to the view shown at C, which is a cross-section of blank D,
it will be seen that the split extends into the fin, and it will
also be noted that the crack extends below the surface of the ball,
although it comes to the surface at each end at points near
the poles. This is due to the fact that the split does not penetrate
the ball at right angles to the surface, but runs on a slant. Instead
of compressing and filling up the open space in the ball, material
has been pressed outward and made a large fin; when this fin

is ground away, the crack is quite evident. At G and H are
shown finished balls that have been etched with acid to show the
grain at the equator and at the poles, respectively.

HEADED BLANK

FINISHED BALL

Fig. 38. Diagram illustrating Distortion of Steel Structure in Cold-header
Ball Blank similar to that shown at (B) in Fig. 37.

Referring again to the sectional view of the wire shown
at A in Fig. 37, and also to the cross-section of a ball blank
made from this wire shown at B, it will be seen that the structure
of the steel has been greatly disturbed during the process of
cold-heading to produce the ball blank. In Fig. 38 is shown
diagrammatically the way in which this disturbance takes
place. It will be seen that the ends of the fibers come to the
surface at the poles and at both sides of the equatorial fin;
and when the ball is etched the steel is attacked more rapidly
at these points. The peculiar marks shown at G and H in Fig. 37
are the result of this disturbance of structure. The conclusion has
been reached that when a ball with so-called ' 'header cracks" is
etched with acid and shows two end poles and two equatorial marks
with a wide crack running from pole to pole or possibly a secondary
crack running between the two equators, this crack is a header
crack which is caused by a seam or lap in the steel from which
the ball was made. The internal stress due to the structural
distortion illustrated in Fig. 38 is completely normalized by the

annealing treatment to which all Hoover balls are subjected.
A number of these headed balls with header cracks were
heated in an electric furnace in the laboratory and quenched
in water at 1500 degrees F. ; every ball was further cracked by
this treatment, and several of them fell in half or were easily
broken by a light hammer blow. Another lot of headed balls
with no header cracks was heated in the electric furnace and
quenched in water at 1600 degrees F., and not a ball was cracked
in hardening. Balls quenched in water at 1500 degrees F. that
broke during the process of heat-treatment are shown at A in Fig.
39, while the balls quenched in water at 1600 degrees F., without
damage are shown at B in the same illustration. At this excessive
temperature the grain of the metal was coarsened, but no hard-
ening cracks were produced and it required considerable force
to break the balls. Several finished balls were next selected
in the inspection department that showed very slight header
cracks. These balls were hardened at 1500 degrees F. and cracked
in the process of hardening exactly as before. The characteristic
black mark left by the original header crack is shown at one side
of the balls at C in Fig. 39. Another lot of finished balls showing
no header cracks was hardened at 1600 degrees F. and none
of the balls was cracked, views of the fractured surfaces of these
balls being shown at D in Fig. 39. This confirmed the accuracy
of previous tests, and from these data the following conclusions
were drawn: (1) The header crack forms a weak spot, so that
when the ball is hardened, even at the proper temperature,

••••

?•• • ••

•••

Fig. 39. (A) Fractures of Header Cracked Balls that Split when re-heat-treated
in Laboratory at 1500 Degrees F., (B) Fractures of Perfect Balls that did not
Split when re-heat-treated at 1600 Degrees F.; (C) Fractures of Balls with
Slight Cracks which broke when re- heat-treated at 1500 Degrees F., (D)
Fractures of Perfect Balls that did not break when re- heat-treated at 1600

Degrees F.

what the inspectors call a "fire crack" is likely to be produced.
(2) A ball with no header cracks can be hardened at an excessively
high temperature without producing a fire crack.

Another hardening test was made with four samples of wire,
two pieces of which showed seams, and two pieces that did not.
The seamy pieces of wire were quenched at a temperature of
about 1500 degrees F. in water and hardening cracks developed
along the seams. The two pieces without seams were quenched
in water at a temperature of 1600 degrees F. and no cracks
developed. All these tests show that with small blanks with-
out any header cracks, it is practically impossible to produce
fire cracks in the automatic hardening furnaces; when cracks
are produced they are started in cold-heading and not through
the process of heat-treatment. The shape of the ball is in its
favor, as it insures uniform quenching and a minimum of internal
strain. Application of too high a temperature would tend to
increase the size of the grain in the steel and make it brittle
and unfit for use, but it would not produce hardening cracks.

EFFECT OF HARDNESS
OF WIRE

WHEN the wire used in making ball blanks on cold-headers
is too hard, there is a tendency for it to break off instead
of shearing as it should. When trouble of this sort is
encountered, it is likely to be accentuated by the fact that the
blank is often carried to the heading die in a sidewise position,
which results in the development of abnormal pressure in the
die. Working hard stock of this kind results in breaking the
cut-off knife or the dies on the cold-heading machine. This
condition of excessive hardness does not usually exist for
the entire length of the coil ; wire may shear off and head nicely
for some time, when suddenly a hard spot will be reached and
then the dies or the cut-off knife is likely to suffer. After this
hard spot has been passed, the wire may be all right for another
period of considerable duration. With the view of showing
the relative condition of hard and soft spots in the wire, slugs
of metal were selected at a point where trouble was encountered
from this cause, and again at a point where the operation of
the cold-header was entirely satisfactory. These were tested

Fig. 40. (A) Fracture of Hard Metal Slug; (B) Fracture of Normal Metal Slug;

called to Decarbonization at Circumference; (D) Etched Surface of Normal

Steel with No Decarbonization at Circumference.

by the Brinell method and it was found that the hard slugs
had a Brinell hardness number of 215, while the soft slugs only
showed a Brinell hardness number of 190. The latter is really
higher than it should be, as 170 is specified for steel to be used in
cold-heading machines.

"Fig. 41. (A) Decarbonized Surface shown at (C) in Fig. 40 magnified to Sixty-
two Diameters; (B) Same Magnification as at (A), showing Condition of
Practically No Decarbonization.

At A in Fig. 40 is shown the fresh fracture of a slug of hard
metal and attention is called to the coarse grain as compared
with the finer grain of the normal steel shown at B. The hard
specimen was very brittle and easy to break, while the normal
steel was tough and capable of bending considerably before
being broken. Specimens of these two steels were next polished
and etched, with the result shown at C and D, respectively.
These are transverse sections cut through the wire, and attention
is called to the coarse grain of the steel shown at C\ the ring at the
surface is a band of decarbonized steel apparently produced by
the application of too high an annealing temperature. The
normal steel shown at D has a fine grain and there is no
indication of decarbonization.

At A in Fig. 41 is shown the decarbonized band of steel sur-
rounding section C in Fig. 40, which is magnified to 62 diameters,
instead of 5.25 diameters, as in the case of the previous illustra-
tion. It will be noted that the extreme edge of this photomicro-
graph is somewhat indistinct, owing to the slightly rounded
edge formed while polishing the specimen. The decarbonized
surface of this stock would not be entirely removed in the process
of grinding, and would result 'in the production of either soft
balls or balls with soft spots. AtB, Fig. 41, we have the condition
where there is practically no loss of carbon at the surface.
At A and B in Fig. 42 is seen a decided contrast between the

Fig. 42. (A) Pronounced Pearlitic Structure with large cells and Boundaries
of excess Cementite, indicating Application of too High an Annealing Temper-
ature; (B) Fine-grained Structure, showing Condition obtained with Proper
Annealing Temperature. Both Samples magnified to 225 Diameters.

structure of the slug of hard metal and that taken from the
normal wire. At A there is a pronounced pearlitic structure
with large cells and distinct boundaries of excess cementite,
which also indicates the application of too high an annealing
temperature. At B the structure is fine grained, which is the
condition produced by employing the proper annealing
temperature. Where lack of uniformity is discovered in
the hardness of the wire, it is probably due to application of too
high an annealing temperature.

CAUSE OF SOFT SPOTS
ON BALLS

SOME valuable discoveries have been made in the laboratory
as a result of work that was started with some other object
in view. For instance, an investigation that was started
with the view of determining the effect of slight seams found in
a certain shipment of steel at the time of the preliminary tests.
These seams were not considered serious enough to justify rejec-
tion of the steel, but after the first lot of blanks had been finish
dry ground, tests were made. This was done by etching a number
of balls in dilute hydrochloric acid, to see if the seams had been
removed in grinding. The balls were immersed in the solution,
and after being etched for fifteen or twenty minutes they were
removed, and cleaned.

When treated in this way, the balls are usually a light gray
color over their entire surface, but the particular lot of balls
referred to could not be uniformly etched. At first it was thought
that a film of grease or some other foreign matter was interfering
with the action of the acid, but a second trial resulted in the
same mottled appearance of the etched balls. Part of the surface
was light gray, while other parts were dark gray and almost black.
Balls with these spots are shown in Fig. 43 and no matter how
often they were re-etched, the same spots always appeared and
they were of the same outline as those developed by the pre-
vious etching. Some of the unetched samples were examined,
and it was found that a considerable quantity of black scale
was left on the balls, i. e., the forging had not been cleaned
up properly after the finish dry-grinding. At this stage the ball

Fig. 43. Finish Dry-ground Balls after being etched with Hydrochloric Acid,
showing Mottled Appearance due to Soft Spots produced by Decarbonization

of Steel.

consistently measured 1.135 inch, i. e., within 0.010 inch of the
finished size — -\Y% inch.

Thus far results seemed to indicate that the forging blanks
were under size, so five samples were selected at random and
measured. The measurements of these five blanks are given
in Table 4, reference to which will show that dimensions A across
the poles and dimension B near the poles were of ample size;
and the surfaces at or close to the poles were also smooth and
well filled out. However, these conditions did not exist around
the equator, where it will be seen that dimension C was scant
in many balls, and additional trouble was caused by the fact
that the surface was very rough and covered with "hills" and
' Valleys." In making these equatorial measurements with a
micrometer, the distance is taken across the tops of the ' 'hills,"
while the dimensions in the 'Valleys" will obviously be consider-
ably less. It is doubtful, therefore, whether three out of five of

these samples would clean up in the rough dry-grinding. A
re-examination of the etched dry-ground balls showed that the
peculiar black spots did not appear at the poles as frequently
as they did at the equator; and when a new file was applied to
the black spots shown in Fig. 43, it was found that they were
dead soft, while the light gray spots were very hard. The
sclerescope hardness of ten of these balls was taken and averaged
as follows: black spots, 48; gray spots, 70.

Table IV. Measurements of Balls across Poles, near
Poles and at Equator.

1.168
1.169
1.170

1.166
1.161
1.151
1.175
1.152

1.161
1.167
1.145
1.172
1.170

1.160
1.163
1.145
1.170
1.158

1.166
.162
.150
.159

.167

The reason for these spots will be understood from the
photomicrographs presented at A and B in Fig. 44, which are
taken from polished surfaces at the extreme outer surface of
the black and white spots on the balls. These surfaces were
prepared and photographed in exactly the same way; instead
of polishing a flat on the ball, the spherical surface was polished,
because a flat surface having any width whatever would also
be at a considerable depth below the surface of the ball, and
would not reveal conditions that it was desired to investigate.
Difficulty was experienced in polishing this spherical surface,
and so the photographs reproduced in Fig. 44 show polish marks
rather too distinctly, but these have no bearing upon the accuracy
of the results obtained in the investigation. At A is shown a
large percentage of free ferrite, indicating a hypo-eutectoid
structure of about 0.30 to 0.40 per cent carbon; in other words,
the metal is similar to a mild steel. On the other hand, the
condition revealed at B is practically a pure eutectoid structure
of pearlite, this steel having from 0.85 to 0.90 per cent carbon.
Specifications under which the steel is purchased call for from
0.95 to 1.05 per cent of carbon, so that in this regard it fulfills
requirements.

Fig. 44. (A) Photomicrograph of Black Soft Spots on Balls shown in Fig. 43,

showing Large Percentage of Free Ferrite or Hypo-eutectoid Structure; (B)

Photomicrograph of Hard White Spots on Balls shown in Fig, 43, indicating

the Desired Eutectoid Structure.

A further test was conducted by preparing flat surfaces of
considerable depth on the balls and examining these under the
microscope; and in both cases it was found that photomicro-
graphs obtained in this way indicated metal containing its full
percentage of carbon. Hardness tests show that the metal
directly under the decarbonized spot is soft and indicate
not only that the decarbonized surface fails to harden, but that
it also forms a sort of insulator and retards the proper
hardening of the eutectoid steel beneath it. Therefore, the
decarbonization plus its effects means a soft area of decided
depth, so deep, in fact, that when the ball is finished the
soft spot still appears. Having reached this conclusion, specimens
of the raw material were prepared by cutting sections trans-
versely from the bar, and these were prepared and photographed
Fig. 45 illustrating the conditions that were revealed in this way.
It will be noted that the steel shown at A is decarbonized to a
depth of 0.010 inch — 0.020 inch on the diameter of the ball-
while in the sample shown at B there is no decarbonization. It
was this steel with the decarbonized surface that produced balls
showing soft spots in the tests.

Fifty of these balls showing soft spots were taken to the
laboratory, where they were again heat-treated, and the result
was that the balls came out hard. It was not considered, how-
ever, that this indicated defective heat-treatment in the process

Fig. 45. (A) Photomicrograph of Transverse Section of Decarbonized Edge of
Steel — Magnification, 125 Diameters; (B) Photomicrograph of Transverse
Section of Steel showing No Decarbonization — Magnification, 125 Diameters.

of manufacture, because it might have happened that the
operation of finish dry-grinding removed enough metal from
the surface so that the balls would harden properly, although
they were prevented from doing so at the time of the original
treatment by the decarbonized steel that covered the surface
of the balls. Because of the oval shape of the forgings, the depth
of decarbonization varies at different spots on the rough-ground
surface of the balls; for example, at the poles there is little or no
decarbonization, while around the equator the decarbonization
is quite deep. When a ball is reduced to the finished size, the
following conditions will be found : (1) decarbonized areas where
the original decarbonization on the rough ball was deep; (2)
soft areas where the original decarbonization on the rough ball
was shallow; (3) hard areas where there was little or no de-
carbonization on the rough ball. In cases (2) and (3) the steel
has its full percentage of carbon, and when the balls are rehard-
ened some of the soft spots disappear, while the spots devoid
of carbon still remain soft. It would be possible to reduce these
balls to a smaller size and reclaim them by rehardening, but this
subsequent heat-treatment has a tendency to roughen their
surface slightly, which necessitates subsequent grinding opera-
tions jhat would probably reduce the diameter from 0.015
to 0.020 inch, so that allowance must be made for this
reduction in size.

To overcome trouble from the use of stock that is decarbon-
ized at the surface, special forging dies were made which produce
oversize ball blanks, so that the diameter at the equator measures
from 0.060 to 0.080 inch more than that of the standard finished
balls. The same stock forged in a regular die would make a
blank 0.025 inch to 0.035 inch larger than the finished size. In
the present case it is found that these would not clean up, but
left soft and decarbonized spots on the surface of the finished
ball. For this reason, the special forging dies were produced.
This practice was adopted because, owing to the slow deliveries
made by the steel mills, it was desired not to reject any steel of
this size that could possibly be used.

DEVELOPMENT OF A DEVICE FOR
SEPARATING HARD AND SOFT BALLS

OWING to shipment to the factory of a large quantity of
low carbon steel through an error made at the steel mills,
and which escaped the rigid sampling to which every
car of steel received at the Hoover plant is subject, about
seven tons of this material was converted into ball blanks
before it was attempted to harden them. This was due to
the fact that a large supply of blanks of the same sizes
had accumulated, and these were naturally sent through the
heat-treating department ahead of blanks made from this
shipment of steel. When the blanks had been heat-treated, they
were tested in order to determine the nature of the results
obtained, and while a number of balls broke with a fine-grained
fracture and showed a hardness that was all that could be desired,
almost 10 per cent of the balls were found to be dead soft. When
these balls were subjected to pressure they flattened out instead
of breaking in the usual way. A peculiar mottled effect was
noted on the balls found to be file hard, while the soft balls
were a dull black color ; but this difference in appearance was not
sufficiently marked to enable the balls to be separated, and even
had this been possible, the length of time required to eliminate
defective balls by this method would have been prohibitive.
With a view to overcoming this difficulty, a device was de-
veloped which is shown in diagrammatic form in Fig. 46. Its
principle of operation is based on the fact that when balls are

dropped on a hardened steel anvil there is considerable difference
in the height of the rebound of hard and soft balls. The balls
to be tested roll down an incline plane and drop upon a hardened
steel block, from which they rebound; the hard balls rise high
enough to pass over a "hurdle" into a box, while the soft balls
do not reach this height and are deposited in a second box. To
test the efficiency of this device, 119 balls taken from one of the
tote pans in the shop were run through the drop test; 79 dropped
into the "hard bin" and 40 into the "soft bin." These balls were
once more thoroughly mixed and again run through the ap-
paratus with the same result as in the previous case. Additional
trials confirmed the accuracy of the apparatus. This method of
separation proved so satisfactory that a regular equipment
has been built for use in the dry-grinding room, where it is
used for separating hard and soft balls.

/ !

HARD O
BALLS ,f:

"ji^lr

ZT\ 47" i—

_,__47 .4 >,

-t— -„ \

i S& / <**'
3lW x \ t

SOFT-
BALLS

\\ /

y§n&sj

HARD

A ANVIL" |

Fig. 46. Diagram illustrating Principle of Apparatus developed for
Automatic Separation of %-inch Hard and Soft Balls.

CONCLUSION

MANY of the cases of trouble to which reference has been
made are of rare occurrence, but it is obvious that they
exert a powerful influence on the quality of the product
turned out in the factory. Also, the conditions brought to light
by these investigations are exceptionally interesting. It was on

this account that they were selected for discussion in the present
treatise, in connection with the regular work of the laboratory,
and not because they really belong to a description of routine
work of testing the raw material and product of a factory en-
gaged in the manufacture of steel balls.

CRUSHING AND DEFORMATION
TESTS

THE old method of determining the crushing load of a
ball was to test a single ball between two hardened steel
plates. It is obvious that if the plates were not of uniform
hardness the crushing loads would also lack uniformity, because
the plates would be indented during the test and the softer plate,
being indented the greater, would present more supporting area
to the ball, and thereby increase its resistance to crushing.
Inability to produce plates of absolute uniformity puts this
method out of the question as a standard test.

The Hoover Steel Ball Co. has developed the Three-Ball
test as a standard. Three balls super-imposed, as shown in
the illustration, are subjected to a gradually increasing pressure
until rupture occurs, and the amount of pressure is recorded at
this point.

We wish to emphasize that testing by the Three-Ball method
will yield results somewhat lower than by the plate test by
reason of the fact that the contact points are very minute and
therefore the pressure per unit of area is tremendous.

The plate test is very often used by some ball manufacturers
to deceive the buyer by making him believe he is getting a better
ball by reason of the high crushing load.

Believing that a table of crushing loads would be of very
little value to our customers, as a guide to determine the safe
working load of the ball, and that such a table might be mislead-
ing, we refrain from publishing same.

It is evident that the safe working load that a ball will
carry depends not only upon the quality of the ball, but also
upon the type of bearing in which it is to run, the shape, material
and finish of the ball race, etc.

We stand ready at all times, however, to give our customers
information as to crushing strength and elastic strength, and
to give our opinion as to the most suitable size and type of
ball for any particular work, after we have received full particulars
of the bearing and the nature of the work for which it is required,
load, speed, etc.

Ball Crushing Apparatus

I
I

Cft;
Sfe

Fracture of a hard surface tough center ball. Note the flattening and cone of rupture at the points
of contact, formed when the balls were crushed.

HOOVER STEEL BALLS HAVE A HARD
SURFACE AND A TOUGH CENTRE

CO-OPERATION with our customers and extensive service
tests of our balls have developed a method of heat
treatment which while simple in its theory is difficult of
practical control, and this control is only made possible by
automatic hardening machines which eliminate the personal
element.

It is not a difficult matter to harden a ball clear through
to the centre, as it is merely a question of quenching at a tempera-
ture sufficiently high to harden the interior, but this method is
without due regard to the exterior. Hardening a ball under
these conditions produces an over-heated exterior which is
necessarily brittle, and strength cannot be restored by tempering.

Hoover balls are heat-treated to produce a sufficiently
hard exterior and a tough semi-hard interior, producing the
qualities most needed in ball bearings. The surface is sufficiently
hard to withstand wear, without being so brittle as to flake or
peel. The interior is sufficiently tough and elastic to stand the
strain of heavy loads.

This type of ball must not be confused with a low grade
steel ball "case hardened" on the surface and with a soft core.

When we speak of hard surface and tough centre we refer
to a high grade alloy steel in which there is a gradual merging
of hardness at the surface to semi-hardness at the core, without
a distinct line of demarkation as in the "case hardened" ball.

The above photograph shows the fracture of a hard surface*
tough centre ball, of which the Hoover Steel Ball Co. is the
exponent.

SMOOTH AND MIRROR-LIKE surface finish must be maintained in every ball
leaving our plant and to this end extensive microscopic examinations are regularly
made. The constancy and effect of the many abrasive materials used are kept under
rigid control.

Microphotographs 1, 2, 3 and 4 show the highly magnified surfaces of several
makes of balls for comparison.

No. 1 shows the Hoover standard.

At the top of this page is shown the apparatus on which microscopic examinations
and photographs are made.

CHEMICAL
LABORATORY

WE HAVE an up-to-date Chemical Laboratory which co-
operates with the Metallurgical Department in the
control of the raw material which is used in the pro-
duction of the Hoover Steel Balls, as well as the solution of the
different problems which are constantly arising in a plant that is
aiming to produce a product as near perfect as scientific methods
and human efficiency can make it.

Drillings, and in some cases millings, are taken from the
samples which are brought to the Metallurgical Department
from each shipment of steel which is received at the plant,
whether for the production of balls or to be used for the produc-
tion of machine parts that may be required in the plant. These
drillings, or millings, as the case may be, are analyzed in the
Laboratory. The percentage of carbon, manganese, phosphorus,
sulphur and chromium is determined in all steel used for the
production of balls. In the case of Header Die Steel, which is
used to make the dies that forge the balls, the percentage of
carbon is determined on each bar, and if it should be an alloy
steel other elements are determined, and a complete analysis
is made on one sample taken from the shipment.

Samples are also taken from each shipment of Brass and
Bronze Wire or Rod, which is received at the plant, and from
which are made our Brass and Bronze Balls. These are analyzed
to determine the quantities of tin, lead, copper, iron, zinc, and
also any elements that might have an injurious effect on the
service rendered by the finished balls. As the composition in
a great measure controls the hardness, resistance to abrasion,
resistance to corrosion, and therefore the life of the finished balls,
it can be seen how very important it is that a careful analysis
should be made of all raw material from which these balls are
produced.

A great many oils and greases are used in the plant for
various purposes, and these must be all carefully tested and
graded. For instance, the finished balls are packed in a mixture
of an oil and a grease and it is very important that these should
be absolutely free from any element such as acids, sulphides or
water, as any of these would etch and oxidize the surface of the

balls in a short time so that they would be rendered useless for
our customers. The same thing applies to the paper used in
lining the boxes in which the balls are packed for shipment. The
leather used to put the final polish on the balls must also be free
from acid and moisture, or they would be rejected by the Inspec-
tion Department on account of rust spots. A surface so highly
finished is very sensitive and must be carefully protected not
only during production but in the packing, and this is the reason
the surface of the balls is so carefully covered with oil to prevent
even the atmospheric moisture affecting them.

Fuels in the form of coal, oil and gas are also graded and
combustion problems investigated in this department. The effect
on the finished product of the different modes of handling the
balls during production must be considered, and even the humi-
dity of the room in which they are inspected has to be reckoned
with. It can, therefore, be seen that besides the routine control
of raw material, etc., a number of interesting questions arise
from time to time which the Chemical Laboratory must assist
in solving.

* £«

« gS

S-2
•c e,sc

g °°
s l:

a •£»

S B te

•s •?§
S g!

^ C o,

c o^S e e

•*•»

•si

•5!
a

<»> c

11:

"S.B

I Is

•

s
si

• «e 0 fc

a v. ^~

ill!

5 h *S

«.

1111!

5-8

cc «> 0

5*1

o S o

8is&

lol

« c

§.'=
£§

c-o

S «>
8?

=- o

TUMBLING BARREL ROOM— This room is equipped with a variety
of tumbling barrels, cleaning barrels and rotary kegs, all of which serve some
special purpose, depending upon the size of ball or the grade of finish desired.

*!!
Ill

OQ S-«

^ c c

•$•2

""* a>

•?*

•i"B

i c

> c

100

•a h

•80

I!
ft

=»
= i

101

- *

102

103

104

105

106

107

108

109

WEIGHTS OF STEEL BALLS

Diameter of Decimal
Ball (Inches) (Inches)

WEIGHT PER BALL

Grammes
(Metric)

Ounces

(Avoir)

Pounds

(Avoir)

1-16 .0625

.0166

.00096

.00006

3-32 .09375

.0547

.00193

.00012

1-8 .125

.1302

.00457

.00029

5-32

.15625

.2552

.00898

.00056

3-16 .1875 .4408

.01552

.00097

7-32 .21875

.6993

.02461

.00154

1-4 .25

1.0463

.03680

.00230

9-32 .28125

1.4865

.05231

.00327

5-16 .3125

2.0415

.07184

.00449

11-32 .34375 2.7141

.09550

.00597

3-8

.375 3.5226

.12400

.00775

7-16

.4375 5.5871

.19722

.01229

1-2

.50 8.3498

.29392

.01837

9-16 .5625

11.8923

.41856

.02616

5-8 . 625

16.2947

.57520

.03585

11-16 .6875

21.6873

.76336

.04771

3-4 . 75

28.1872

.99200

.06200

13-16 .8125

35.7585

1.25872

.07867

7-8

.875

44.7872

1.57648

.09853

15-16

.9375 55.0169

1.93664

.12104

66.8257

2.35232

.14702

-1/16

.0625 80.1379

2.8199

.17626

-1/8

.125

95.1271

3.3473

.20923

-3/16

.1875

111.8809

3.9369

.24608

-1/4

.25

130.4965

4.5919

.28702

-5/16

.3125 151.0656

5.3157

.33226

-3/8

.375

173.6764

6.1113

.38199

-7/16

.4375

198.4563

6.9833

.43649

-1/2

.50

225.4820

7.9343

.49594

-9/16

.5625

254.8682

8.9683

.56057

-5/8

.625

286.6917

10.0881

.63056

-11/16 .6875

321.0544

11.2973

.70614

-3/4 .75

358.0711

12.5998

.78756

-13/16 .8125

397.8185

13.9985

.87498

-7/8 .875

440.398

15.4968

.96864

1-15/16 .9375

485.939

17.0993

.06880

534.491

18.8077

.17559

2-1/8

2.125

641.101

22.5591

.41007

2-1/4

2.25

761.019

26.7788

.67382

2-3/8

2.375

895.037

31.4947

.96859

2-1/2

2.50

1043.924

36.7340

2.29608

2-5/8

2.625

1208.474

42.5239

2.65798

2-3/4

2.75 •• «'

1389.436

48.8916

3.05600

2-7/8

2.875

1587.727

55.8691

3.49213

1803.881

63.4751

3.96755

3-1/8

3.125

2038.920

71.7457

4.48451

3-1/4

3.25

2293.482

80.7033

5.04440

3-3/8

3.375

2568.460

90.3792

5.64920

3-1/2

3.50

2864.492

100.7960

6.30031

3-5/8

3.625

3183.875

112.0345

6.99997

3-3/4

3.75

3523.164

123.9734

7.74903

3-7/8

3.875

3887.462

136.7923

8.55028

4275.876

150.4599

9.40458

4-U4

4.25

5128.882

180.4756

11.28073

4-1/2

4.50

6088.179

214.2314

13.39065

4-3/4

4.75

7160.402

251.9608

15.74896

8351.420

253.4605

18.36854

C = Contents in Cubic Inches.
= 4/3 TT R3 = 4.1888 R3 = .5236 D3

W = Weight of Steel Balls in pounds.

= R3 (.28065 X 4.1888) = 1.17558 R3 = .14695 D

110

FORMULA FOR DETERMINING PITCH DIA.

OF BALL CIRCLE AND CLEARANCE

BETWEEN BALLS

Notation:

Di = Pitch Dia. of Ball Circle.

D2 = Dia. of Circumscribed Circle.

Ds = Dia. of Inscribed Circle.

d = Dia. of Balls.

N = Number of Balls in the Ring.

S = Clearance Between Each Pair of Balls.

Di = (d+S)XCSC.

180 °\
N /

/ISO

)3 = Di-d

/180
= DiXSIN. f

180 °\

IT)-'

The following table gives the value of the CSC. and SIN. for "N" Balls.

No. of Balls

"N"

Angle a
180°

CSC.
180°

SIN.
180°

30°

2.00000

.50000

25°— 42'— 51.43"

2.30476

.43388

22°— 30'

2.61313

.38268

20°

2.92381

.34202

18°

3.23607

. 30902

16°— 21'— 49.09"

3.54947

.28173

15°

3.86370

.25882

13°— 50'— 46.16"

4.17858

.23932

14 12° — 51'— 25.72"

4.49396

.22252

15 12°

4.80973

.20791

11°— 15'

5.12583

. 19509

17 10°— 35'— 17.65"

5.44219

. 18375

18 10°

5.75877

. 17365

19 9°— 28'— 25.26"

6.07554

. 16459

20 9°

6.39247

.15643

21 8°— 34'— 17.14"

6.70950

. 14904

22 8° — 10' — 54.55"

7.02667

.14231

23 7° — 49' — 33.91"

7.34394

.13617

24 7°— 30'

7.66130

.13053

25 7°— 12'

7.97873

.12533

6°— 55'— 23.08"

8.29623

.12054

6°— 40'

8.61380

.11609

6°— 25'— 42.86"

8.93140

.11196

6°— 12'— 24.82"

9.24907

.10812

6°

9.56677

. 10453

111

THE CIRCLE

d = Diameter of Circle.

C = Circumference of Circle.

C=7rd =3.141593 d

A = Area of Plane Surface.
7r = 3.141593

Trd2

A = = .785398 d2

Areas of Circles are to Each other as the Squares of their Diameters.
THE SPHERE

V = Volume of Sphere.

d = Diameter of Sphere.

S = Area of Convex Surface.

.523599 d3

Surfaces of Spheres are to each other as the Squires of their Diameters.
The Volume of a Shpere = 2/3 the Volume of its Circumscribing Cylinder.
Volumes of Spheres are to each other as the Cubes of their Diameters.

BALL DIA.
IN INCHES

C RCUM.
N INCHES

AREA

VOLUME
CU.- INCHES

SECTION
SQ. INCHES

CONVEX SURFACE
SQ. INCHES

/SZ

.09818

.00077

.00307

.00002

/16

. 19635

.00307

.01227

.00013

/SZ

.29452

.00690

.02761

. 00043

. 39270

01227

.04909

.00102

/SZ

.49087

.01917

. 07670

.00200

/16

. 58905

.02761

.11045

.00345

/32

.68722

. 03758

. 15033

. 00548

.78540

.04909

.19635

.00818

/32

.88357

.06213

.24851

.01165

.98175

.07670

. 30680

.01598

11 32

1.0799

.09281

.37123

.02127

3 8

.1781

.11045

.44179

. 02761

IS SZ

.2763

. 12962

.51848

.03511

7 16

.3744

.15033

.60132

.04385

15 32

.4726

. 17257

. 05393

1 Z

.5708

.19635

.78540

. 06545

9 16

.7671

.24850

. 99403

.09319

5 8

.9635

.30680

.2272

.12783

11 16

.1598

37122

.4849

.17014

S 4

.3562

.44179

.7671

. 22089

.5525

.51849

.0739

. 28084

7/8

.7489

.60132

.4053

.35077

15/16

.9452

.69029

.7611

.43143

.1416

.7854

.1416

.52360

1/16

.3379

.8866

.5466

. 62804

1/8

.5343

.9940

.9761

.74551

3/16

.7306

.107J

.4301

.87681

I/*

.9270

.2272

.9088

.0227

.1233

.3530

.4119

.1839

3/8

.3197

.4849

.9396

.3611

7/16

.5160

.6230

.4919

.5553

1/Z

.7124

.7671

.0686

.7671

/16.

.9087

.9175

.6699

.9974

.1051

.0739

.2957

.2468

1 /16

.3014

.2365

.9461

.4978

.4053

.6211

.8062

1 /16

.6941

.5802

.321

.1177

.8905

.7612

.044

.4514

1 /16

.0868

.9483

.793

.8083

.2832

.1416

.566

.1888

/16

.4795

.3410

.364

.5939

.6759

.5466

.186

.0243

/16

.8722

7583

.033

.4809

.0686

.9761

.904

.9641

/16

.2649

.2000

.800

.4751

.4613

.4301

.7Z1

.0144

/16

.6576

.6664

.666

.5829

.8540

.9087

.635

.1813

/16

.0503

.1572

20 629

.8103

.2467

.4119 21.648

.4708

1 /16

.4430

.6727 22.691

.164

.6394

.9396

23.758

.889

I /16

.8357

.2126

24.850

.649

.0321

.4918

25.967

.443

1 /16

.2484

.7771

27.109

.272

.4248

.0686

28.274

.137

1/16

6211

3662

29.465

1/8

.8175

.6699

30.680

'979

3/16

.014

.9798

31.919

.957

1/4

.210

.2958

33.183

]

974

i/l«

.407

.6179

34.472

.031

S/8

.60S

.9462

35.784

20.129

7/16

.799

.2806

37.122

21 . 268

1/Z

.996

.6211

38.484

22.449

.192

.9678

S9.872

23.674

.388

.321

41.283

24.942

11/16

.585

.680

42.719

26.254

3/4

.781

44.179

27.611

13/16

977

.tit

45 . 664

29.016

7/8

174

.798

47.173

SO 466

15/16

.370

48.708

3 .965

12.566

12 566

50.465

33.510

112

DECIMAL EQUIVALENTS OF FRACTIONS
OF AN INCH

Fract.

Dec.

Fract.

Dec.

Fract.

Dec.

Fract.

Dec.

—

.015625

—

. 265625

—

.515625

—

.765625

—

.03125

—

.28125

—

.53125

—

.78125

—

.046875

—

.296875

—

.546875

—

.796875

.0625

.3125

.5625 —

.8125

1 37

i 53

.078125

.328125 —

.578125 ! —

.828125

! 64

.09375

•H

.34375

.59375 —

.84375

—

. 109575

—

.359375

.609375

—

.859375

.125

.375

.625

.875

.•

. 140625

. 390625

.640625

.890625

. 15625

.40625

.65625

. 90625

—

.171875

—

421875

.671875

—

.921875

—

.1875

—

.4375

.6875

—

.9375

.203125

.453125

.703125

.953125

.21875

.46875

.71875

.96875

—

.234375

.484375

.734375

.984375

—

.25

.75

TABLE OF DECIMAL EQUIVALENTS OF MILLI-
METERS AND FRACTIONS OF MILLIMETERS

1/100 mm. = .0003937".

mm. Inches mm. Inches mm. Inches | mm. Inches mm. Inches

1/50 = .00079

"/50 = .00866

21/50= .01654

31/50= .02441

41/50= .03228

2/50= .00157

i«/50 = .00945

22/50 =.01732

3*/50 = . 02520

42/50 =.03307

3/50= .00236

13/50 =.01024

23/50=. 01811

33/so =.02598

43/50 = . 03386

4/50 =.003 15

14/50 = .01102

24/5o= .01890

34/50 = . 02677

44/so = . 03465

5/50=. 00394

15/50=. 01181

25/50= .01969

35/50 = . 02756

45/50= .03543

6/50 =.00472

16/50 = .01260

26/50= .02047

36/50 =.02835

46/5o =.03622

, 7/50= . 00551

17/50= .01339

27/50= .02126

37/50= .02913

47/50= .03701

8/50 = . 00630

18/50 =.01417

28/50 = . 02205

38/M>=. 02992

48/so =.03780

9/50= .00709

19/50= .01496

29/50= .02283

29/50= .03071

49/50 = . 03858

I0/5o= .00787

S°/50= .01575

so/so =.02362

40/50= .03150

10 mm. = l Centimeter = 0.3937 inches

10 cm. =1 Decimeter = 3.937 inches

10 dm. =1 Meter =39.37 inches

25. 4 mm. = 1 English Inch.

113

CONVERSION TABLE

DECIMAL EQUIVALENTS OF MILLIMETERS IN INCHES

1 m/m to 500 m/m.
1 m/m = .03937027"

nun.

Inches

m.m.

Inches

m.m.

Inches

m.m

Inches

m.m

Inches

man

Inches

m.m

Inches

]

.03937027

7ft

2.87402971

145

5.70868915

eii

8.50397832

287

11 .29926749

358

14.09455666

429

16.88954583

.07874054

2.91339998

14(i

5.74805942

217

8.54334859

•288

11 .33863776

359

14.13392693

430

16.92921610

.11811081

2.95277025

147

5.78742969

-218

8.58271886

289

11 37800803

300

14.17329720

431

16.96858637

.15748108

7(i

2.99214052

14S

5.82679996

-219

8.62208913

291

11 41737830

301

14.21266747

432

17 00795664

.19685135

3 03151079

149

5.86617023

221

8.66145940

-291

11.45674857

302

14.25203774

433

17.04732691

.23622162

3.07088106

150

5.90554050

-2-21

8 70082967

292

11.49611884

563

14 29140801

434

17.08669718

.27559189

3.11025133

151

5.94491077

•2-2-2

8.74019994

•2!)3

11 53548911

304

14.33077828

435

17 12606745

.31496216

3.14962160

152

5.98428104

•2-23

8.77957021

•2!) 4

1 1 . 57485938

505

14.37014855

430

17.16543772

.35433243

3 18899187

153

6.02365131

-2-24

8.81894048

•2!)5

11 61422965

500

14 40951882

437

17 20480799

.39370270

3 22836214

154

6.06302158

-2-2.-,

8 85831075

296

1 1 65359992

507

14.44888909

438

17.24417826

. 43307297

3.26773241

1 55

6 10239185

22 (i

8.89768102

•297

11.69297019

568

14.48825936

439

17.28354853

.47244324

3.307102C8

156

6.14176212

-227

8 93705129

298

11 73234046

309

14.52762963

440

17.32291880

.51181351

3.34647295

157

0.18113239

•2-2S

8.97642156

2!)!)

11 77171073

370

14.56699990

441

7.36228907

.55118378

3.38584322

15S

6.22050266

22!)

9.01579183

Kin

11.81108100

571

14.60637017

44-2

17 40165934

1.5

.59055405

3 42521349

159

6.25987293

230

9.05516210

301

11 85045127

37-2

14.64574044

443

7.44102961

. 62992432

3.46458376

160

6.29924320

-231

9 09453237

i()-2

11.88982154

373

14.68511071

444

17.48039988

.66929459

3.50395403

K.I

6 33861347

-2:! 2

9.13390264

503

11 92919181

574

14.72448098

145

17.51977015

.70866486

.54332430

162

6.37798374

2:!:!

9 17327291

504

1 1 . 96856208

575

14.76385125

440

7.55914042

1!)

.74803513

.58269457

1 63

6 41735401

234

9 21264318

505

12.00793235

570

14.80322152

447

7.59851069

.78740540

. C2206484

K!4

6.45672428

235

9 25201345

500

12.04730262

377

14.84259179

44S

7.63788096

.82677567

.66143511

Ki5

6.49609455

230

9.29138372

507

12.08667289

!78

14.88196206

149

7.67725123

2-2

.86614594

. 70080538

100

6.53546482

-237

9 . 33075399

ios

12 12604316

379

14.92133233

150

7.71662150

.90551621

.74017565

107

6.57483509

238

9 37012426

!()!)

12 16541343

580

14.96070200

451

7.75599177

•24

.94488648

!>o

. 77954592

168

6.61420536

23!)

9 40949453

510

12.20478370

581

15.00007287

452

7.79536204

«5

.98425675

.81891619

169

6 65357563

240

9.44886480

ill

12.24415397

iS -2

15.03944314

153

7.83473231

.02362702

. 85828646

170

6 69294590

241

9 . 48823507

312

12 28352424

383

15.07881341

454

7.87410258

. 06299729

!»!»

.89765673

171

6 73231617

21-2

9 52760534

313[12 32289451

384

15.11818368

155

7.91347285

. 10236756

100

93702700

17-2

6.77168644

243

9.56697561

314 12.36226478

585

15.15755398

450

7.95284312

.14173783

101

.97639727

173

6 81105671

244

9 . 60634588

3 la! 12 40163505

580

15.19692422

457

7.99221339

.18110810

102

01576754

174

6 85042698

245| 9.64571615

316|l2. 44100532

387

15.23629449

158

8 03158366

.22047837

103

05513781

175

6.88979725

246

9.68508642

517 12.48037559

5S8

15.27566476

15!)

8.07095393

3-2

.25984864

104

. 09450808

176

6.92916752

247

9.72445669

31812.51974586

18!)

15 31503503

400

8.11032420

.29921891

10.-,

.13387835

177

6 96853779

248

9.76382696

319 12.55911613

.390

15.35440530

401

8.14969447

.33858918

100

.17324862

17S

7 00790806

249

'9.80319723

320 12.59848640

5!)!

15.39377557

40-2

8.18906474

.37795945

107

.21261889

17!)

7.04727833

250

9.84256750

321 12.63785067

392

15.43314584

403

8.22843501

.41732972

108

.25198916

ISO

7 . 08664860

251

9.88193777

322

12.67722694

393J 15. 4725 1611

404

8 . 26780528

. 45669999

109

29135943

181

7.12601887

252

9 92130804

i23

12 71659721

39415.51188638

165

8 30717555

. 49607026

110

.33072970

18-2

7.16538914

253i 9.96067831

i-24

12 75596748

595

15.55125665

400

8.34654582

.53544053

111

. 37009997

183

7.20475941

254'lO. 00004858

525

12 79533775

590

15.59062692

407

8 38591609

.57481080

11-2

40947024

184

7.24412968

255

10.03941885

520

12.83470802

597

15.62999719

108

8 42528636

.61418107

11.'!

44884051

185

7 . 28349995

256

10.07878912

i-27

12 87407829

598

15.66936746

469

8.46465663

4-2

.65355134

114

.48821078

186

7.32287022

-257

10.11815939

•528

12 91344856

i!)!)

15 70873773

470

18 50402690

.69292161

115

52758105

187

7.36224049

258

10 15752966

529

12.95281883

400

15 74810800

471

18.54339717

.7?229188

116

56695132

188

7.40161076

2;,i)

10 19689993

;:!(i

12 99218910

401

15.78747827

47-2

8.58276744

4.)

.77166215

117

.60632159

18!)

7.44098103

-2(10

10 23627020

!31

13 03155937

40-2

15.82684854

473

18.62213771

4(i

.81103242

.64569186

190

7.48035130

-261

10 27564047

532

1 3 07092964

403

15.86621881

474

18.66150798

.85040269

11!)

68506213

11)1

7.51972157

•262

10 31501074

133

13.11029991

404

15.90558908

475

18.70087825

.88977296

120

72443240

192

7.55909184

203

10.35438101

•534

13 14967018

105

15 94495935

470

8 74024852

4!)
50

.92914323
.96851350

121
122

. 76380267
.80317294

193
1!)4

7 59846211
7.63783238

201
-265

10 39375128
10 43312155

i:i.-,

530

13 18904045
13.22841072

407

15 98432962
16.02369989

477
478

8 77961879
8 81898906

00788377

123

.84254321

195

7 67720265

200

10.47249182

537

13 26778099

408

16.06307016

479

8 85835933

.04725404

124

.88191348

196

7 71657292

•207

10 51186209

i:is

13 30715126

40!)

16. 10244043

480

8 89772960

.08662431

1-25

92128375

197

7.75594319

20S

10 55123236

i:59

13.34652153

410

16 14181070

481

8.93709987

.12599458

126

. 96065402

IDS

7.79531346

•26!)

10.59060203

540

13 38589180

411

16.18118097

48-2

8.97647014

.16536485

127

. 00002429

HI!)

7 . 83468373

•270

10.62997290

341

13 42526207

41'2

16 22055124

483

9 01584041

.20473512

1-2S

. 03939456

-200

7.87405400

•271

10.66934317

54-2

13 46463234

413

16.25992151

484

9 05521068

24410539

129

.07876483

-201

7.91342427

•27-2

10.70871344

343

13 50400261

414

16.29929178

485

9 09458095

.28347566

130

.11813510

-20-2

7.95279454

273

10.74808371

•544

13 54337288

415

16.33866205

480

9 13395122

.32284593

131

. 15750537

•20:!

7.99216481

•274

10.78745398

545

13.58274315

410

16.37803232

487

9 17332149

(i()

.36221620

132

.19687561

-204

8.03153508

•275

10.82682425

340

13 62211342

417

16 41740259

488

9 21269176

.40158647

133

23624591

205

8 . 07090535

•27 (i

10 86619452

547

13 66148369

4 IS

16.45677286

48!)

9 25206203

44095674 .134

.27561618

206

8 11027562277

10 90556479

348

13.70085396

419

16 49614313

490

9 29143230

.48032701 135

31498645

•207

8 14964589

•278

10.94493506

549

13 74022423

420

16 53551340

491

19 33080257

.51969728

136

35435672 208

8.18901616

279

10.98430533

550

13 77959450

421 16 57488367

492

19.37017^54

55906755

137

.39372699209

8.22838643

•280

11.02367560

351

13.81896477

422 16 61425394

493

19 40954311

6(i

.59843782

138

43309726 210

8.26775670

281

11.06304587

552

13.85833504

42316.65362421

494

19.44891338

.63780809 139
.67717836:140

47246753211
.51183780212

8.30712697
8.34649724

•282
2 S3

11 10241614
11 14178641

553
•554

13 89770531
13 93707558

424'ie. 69299448
425 16 73236475

495
496

19 48828365
19.52765392

71654863!l41

.55120807213

8 38586751,284

11.18115668

555

13 97644585

426

16.77173502

497

19.56702419

.75591890142

59057834214

8.42523778 285

11.22052695

550

14.01581612

4-27

16.81110529

498

19 60639446

79528917 143

62994861

•215

8.46460805286

11.25989722

357

14 05518639

428

16.85047556

491)

19.64576473

.83465914 144

66931888

500

19.68513500

114

CONVERSION TABLE

MILLIMETER EQUIVALENTS OF FRACTIONAL INCHES
& inch to 12% Inches

6" 7'

10' 11'

12'

25 3995

50.7990

76.1986

101.598

126.998

152.397ll77.797

203.196

228.596

253.995279.394

304.794

1/64 0.3968

25.7964

51 1959

76.5954

101.995

127.394

152.794178.193

203.593

228.992

254.392279.791

305.191

1/32 0.7937

26.1932

51.5928

76.9923

102.391

127.791

153.190178.590

203.990

229.389

254. 7891280.188

305.588

3/64J .1906

26 5901

51.9896

77.3892

102.788

128.188

153.588178.987

204.386

229 . 786

255s 1861280. 585

306.985

1/161 .5874

26 9870

52.3865

77.7860

103.185

128.585

153.984J179.384

204 . 783

230.183

255. 5821280. 982

306.381

5/64 .9843

27 3838

52.7834

78.1829

103.582

128.982

154.381179.781

205.180

230.580

255. 9791281. 379

306.778

3/32' 3812

27.7807

53 1802

78.5798

103.979

129.378

154.778180.177

205.577

230.977

256.376^281.776

307.175

7 /64 7780

28.1776

53.5771

78.9766

104.376

129.775

155.175180.574

205.974

231.373

256.773i282.173

307.572

1/8 .1749

28.5744

53.9740

79.3735

104.773

130.172

155.572 180.971

206.370

231.770

257.170;282.569

307.969

9/64 .5718

28 9713

54.3708

79.7704

105.169

130.569

155.969181.368

206.768

232.167

257.567,282.966

308.366

5/32 9686

29 3682

54.7677

80.1672

105.566

130.966

156. 365181. 765

207.164

232.564

257. 964(283. 363

308.763

11/64 .3655

29.7650

55.1646

80.5641

105.963

131.363

156. 762182. 162

207.561

232.961

258. 360^283. 760

309.160

3/16 .7624

30.1619

55.5614

80.9610

106.360

131.760

157.159 182.559

207.958

233.358

258.757J284.157

309.556

13/64 1592

30.5588

55.9583

81.3579

106.757

132.156

157.556182.956

208.355

233.755

259. 154(284. 554

309.953

7/32 .5561
15/64 .9530
1/4 ! .3498
17/64 .7467

30.9556
31.3525
31.7494
32.1462

56.3552
56.7520
57.1489
57.5458

81.7547
82.1516
82.5485
82.9453

107.154
107.551
107.948
108.344

132.553
182.950
133.347
133.744

157. 953 183. 3521208. 752
158. 350183. 749 209. 149
158.747184.146209.546
159.143 184.543209.943

234.152259.551284.951
234. 5481259. 948'285. 347
234. 9451260. 3451285. 744
235 . 342!260 . 742 286 . 141

310.350
310.747
311.144
311.541

9/32J .1436

32.5431

57.9426

83.3422

108.741

134.141

159. 540(184. 940

210.339

235.739

261.139:286.538

311.938

19/64 .5404

32 9400

58.3395

83.7391

109.138

134.538

159.937185.337

210.736

236.136

261.535'286.935

312.334

5/16 7.9373

33.3368

58.7364

84.1359

109.535

134.935

160.334 185.734

211.133

236.532

261.9321287.332

312.731

21/641 8 3342

33.7337

59.1333

84.5328

109.932

135.331

160.731 186.131

211.530

236.930

262.329287.729

313.128

11/32; 8.7310

34.1306

59.5301

84.9297

110.329

135.728

161.128186.527

211.927

237.326

262.726!288.126

313.525

23/64- 9. 1279

34 . 5274

59.9270

85.3265

110.726

136.125

161.525186.924

212.324

237.723

263. 123^288. 522

313.922

3/8 9 5248

34 9243

60 3239

85.7234

111.122

136.522

161.922187.321

212.721

238.120

263.520288.919

314.319

25/641 9.9216
13/3210 3185

35.3212
35.7180

60.7207
61.1176

86.1203
86.5171

111.529
111.916

136.919
137.316

162.318 187.718213.118
162.715188.115|213.514

238.517263.9171289.316
238. 914 264. 3131289. 713

314.716
315.113

27/6410.7154

36.1149

61.5145

86.9140

112.313

137.713

163.112;188.512

213.911

239.311

264.710290.110

315.509

7/1611.1122

36.5118

61.9113

87.3109

112.710

138.109

163.509188.909

214.308

239.708

265.107290.507

315.906

29/6411.5091

36.9087

62.3082

87.7077

113.107

138.506

163.906189.305

214.705

240.105

265.504'290.903

316.303

15/32H1.9060

37.3055

62.7051

88.1046

113.504

138.903

164.303189.702

215.102

240.501

265. 901 j29 1.300

316.700

31/6412.3029

37.7024

63.1019

88.5015

113.901

139.300

164.700190.099

215.499

240.898

266.298i291.697

317.097

1/2 12.6997

38.0993

63.4988

88.8983

114.297

139.697

165.097190.496

215.896

241.295

266.695,292.094

317.494

33/6413.0966

38.4551

63.8957

89.2952

114.694

140.094

165.493190.893

216.292

241.692

267.092292.491

317.891

17/3213.4934

38.8930

64.2925

89.6921

115.091

140.491

165.890191.290

216.689

242.089

267.488j292.888

318.287

35/6413.8903

39.2899

64.6894

90.0989

115.489

140.888

166.287191.687

217.086

242.486

267.885293.285

318.684

9/1614.2872

39.6867

65.0863

90.4858

115.885

141.284

166.684192.084

217.483

2-I2.SS3

268. 282:293. 682

319.081

37/6414.6841

40.0836

65.4831

90.8827

116.282

141.681

167.081192.480

217.880

243.279

268.679294.079

319.478

19/3215.0809

40.4805

65 . 8800

91.2795

116.679

142.078

167.478192.877

218.277

243.676

269.076294.475

319.875

39/6415.4778

40.8773

66.2769

91.6764

117.075

142.475

167.875il93.274

218.674

244.073

269.473,294.872

320.272

5/8 115.8747

41.2742

66.6737

92.0733

117.472

142.872

168.271193.671

219.071

244.470

269.870295.269

320.669

41/6416.2715

41.6711

• 17.070(1

92.4701

117.869

143.269

168.668194.068

219.467

244.867

270.266295.666

321.066

21/3216.6684

42.0679

67.4675

92.8670

118.266

143.666

169.065194.465

219.864

245.263

270.663296.063

321.462

43/6417.0653

42.4648

67.8643

93.2639

118.663

144.063

169.462194.862

220.261

245.661

271.060|296.460

321.859

11/1617.4621

42.8617

68.2612

93.6608

119.060

144.459

169.859195.258

220.658

246.058

271.457'296.857

322.256

45/6417.8590

43.2585

68.6581

94.0576

119.457

144.856

170. 256I195. 655

221.055

246.454

271. 8541297. 253

322.653

23/32 18. 2559

43.6554

69.0549

94.4545

119.854

145.253

170.653196.052

221.452

246.851

272.251297.650

323 . 050

47/6418.6527

44.0523

69.4518

94.8513

120.250

145.650

171.050196.449

221.849

247.248

272.648298.047

323.447

3/4 19.0496

44.4491

69.8487

95.2482

120.647

146.047

171.446196.846

222.245

247.645

273. 0451298.444

323.844

49/6419.4465

44 . 8460

70.2455

95.6451

121.044

146.444

171.843197.243

222.642

248.042

273.441:298.841

324.241

25/3219.8433

45.2429

70.6424

96.0419

121.441

146.841

172.240197.640

223.039

248.439

273. 8381299. 238

324.638

.51/5420.2402

45.6397

71.0393

96.4398

121.838

147.237

172.637198.037

223.436

248.836

274.235i299.635

325.035

13/1620.6371

46.0366

71.4362

96.8357

122.235

147.634

173.034198.433

223.883

249.232

274.632300.032

325.431

53/64 21.0339

46.4335

71.8330

97.2326

122.632

148.031

173.431 198.830

224.230

249.629

275.029300.428

325.828

27/3221.4308
55/6421.8277
7/8 22.2245

46.8303
47.2272
47.6241

72.2299
72.6267
73.0236

97.6294
98.0263
98.4232

123.029
123.425
123.822

148.428
148.825
149.222

173 . 828 199 . 227 224 . 627
174. 224 199. 624i225. 024
174. 6211200. 021 225. 420

250:026!275.426;366.825
250.423,275.823,301.222
250.820276.220:301.619

326.225
326.622
327.019

57/6422.6214
29/3223.0183

48.0209
48.4178

73.4205
73.8173

98.8200
99.2169

124.219
124.616

149.619
150.016

175. 0181200. 418J225. 817
175. 415200. 815!226. 214

251 .217)276 . 616;302 . 016
251. 614 277.013;302. 413

327.415
327.812

59/6423.4151

48.8147

74.2142

99.6137

125.013

150.412

175.8121201.211

226.611

252.011

277.410!302.810

328.209

lD/16'23.8120

49.2116

74.6111

100.011

125.410

150.809

176.209201.608

227.008

252.407

277.807i303.207

328.606

61/6424.2089

49 . 6084

75.0080

100.408

125.807

151.206

176.606202.005

227.405

252.804

278.204303.603

329.003

31/3224.6057

50 . 0053

75.4048

100.804

126.203

151.603

177.003202.402

227.802

253.201

278.601304.000

329.400

63/6425.0026

50.4021

75.8017

101.201

126.600

152.000

177.399202.799

228.198

253.598,278.998304.397

329.797

115

DICKINSON BROS GRAND

UNIVERSITY OF CALIFORNIA LIBRARY
BERKELEY

Return to desk from which borrowed.
This book is DUE on the last date stamped below.

3 1972*

LD 21-100m-9,'48(B399sl6)476

72 -1 PM 4

Internet Archive Audio

Featured

Top

Images

Featured

Top

Software

Featured

Top

Books

Featured

Top

Video

Featured

Top

Mobile Apps

Browser Extensions

Archive-It Subscription

Save Page Now

Full text of "The evolution of the steel ball industry"

See other formats