Skip to main content

Full text of "Grinding and grinding machines"

See other formats




B 3 016 77E 

A /rflTTTl r?r\ Tr"Tr^ 


No. 38 

A Doll^fc Worth of Condensed Information 

rinding and 
Gifidmg Machines 


Price 25 Cents 


Types of Grinding Machines 

Principles of Grinding, by OSKAR KYLTN 8 

Economy in Grinding, by JOHN J. THACHER 21 

The Disk Grinder 29 

Grinding Kinks and Examples of Grinding - 33 

Cost of Grinding, by H. F. No YES - 44 

The Bursting of Emery Wheels - - 46 

The Industrial Press, 49-55 Lafayette Street, New York 
Publishers of MACHINERY 











Types of Grinding Machines 3 

Principles of Grinding, by OSKAR KYLIN 8 
Economy in Grinding, by JOHN J. THACHER - 21 

The Disk Grinder - 29 

Grinding Kinks and Examples of Grinding 33 

Cost of Grinding, by H. F. NOYES 44 

The Bursting of Eme*y:^sels. .s ~ - 46 

Copyright. 1910, The Industrial Tress, Publishers of MACHINERY, 
49-55 Lafayette Street, New York City 

In the present the second edition of this Reference Series Book, 
two additional chapters on "Types of Grinding Machines" and "Econ- 
omy in Grinding" have been included. In order to provide space for 
this additional material, the chapters on "Lapping Flat Gages" and 
"The Rotary Lap," included in the first edition of this treatise, have 
been eliminated. These two chapters, together with additional material 
relating to gage making and lapping, are included in MACHINERY'S 
Reference Series No. 64, "Gage Making and Lapping." 



History of the Universal Grinding- Machine 

The universal grinding machine has had so great an influence on 
modern machine shop methods, and has done so much to raise the 
standard of workmanship and to increase the economy of production 
that a few words relating to the history of the development of this 
machine may be of interest. 

The origin of the modern universal grinding machine is found in 
the crude grinding lathes of the early sixties. Mr. Joseph R. Brown, 
senior member of the firm now known as the Brown & Sharpe Mfg. Co., 
was intimately connected with the development of these grinding 


Fig. 1. Universal Grinding Machine made by the Brown & Sharpe Mfg. Company 

lathes into the universal grinding machine. The grinding lathe, as 
first built at the Brown & Sharpe Mfg. Co.'s works, was intended for 
the accurate and economical manufacture of the company's own prod- 
ucts, and there was no idea of putting the machines on the market. 
In this respect the origin and development of the grinding machine 
was very much like the origin and development of the universal milling 
machine. The first work for which the grinding machines were de- 
signed was for grinding needle bars, foot bars and shafts of the Wilcox 
& Gibbs sewing machines. The first machine was built in 1864 and 


4 _ - ;\V 38- GRINDING 

1865, and one of these early machines is still in use in the Brown & 
Sharpe works. Cylindrical grinding, however, was done at the Brown 
& Sharpe works as early as 1862, this being indicated by the existence 
of drawings of a back-rest, dated September 22, 1862, which contains 
the essential features of a solid grinding machine back-rest of to-day. 

These early machines were not grinding machines in the present 
meaning of the word, but were grinding lathes using, to a considerable 
extent, the parts of a 14-inch Putnam lathe. A great number of these 
were sold both in this country and abroad. Mr. Brown, however, real- 
ized the need of building a new machine designed especially for grind- 
ing, and iJi 1868 the design for such a machine was made. This design 

Fig. 2. Plain Grinding Machine made by the Landis Tooi Company 

shows a machine containing most of the essential elements of the 
universal grinding machine of to-day. None of these machines were 
built at this time, however, on account of the pressure of other matters, 
and it was first in 1874 that working drawings were made for a com- 
plete machine containing practically all the features of the modern 
universal grinding machine. The first of these machines was exhibited 
at the Centennial Exposition in Philadelphia, 1876. The judges at the 
Centennial Exposition were especially impressed with this universal 

The development of the grinding art since the Centennial Exposition 
has brought out many improvements and refinements, but the essential 



S o> 

45 OJ .S ~ 

J! O 

S S 

a ,_; 



+J O ro 

en &., -2 
3 PQ 5 

i? *> 

5 S S 

6 No. 38 GRINDING 

and is distinctly a manufacturing machine. It is adapted for grind- 
ing plain straight and taper work, and all work that can be revolved 
on two dead centers'. These machines are made especially heavy 
and represent the highest development of machines of their class. 

In Fig. 3 is shown a grinding machine of a heavier type than the 
two previously shown. This machine is built by the Norton Grinding 
Co., Worcester, Mass., and is known as a 20- and 32- by 168-inch self- 
contained grinding machine, the name being derived from the fact 
that the machine swings 20 inches', except in the gap indicated, where 
it will swing 32 inches. "This particular machine is motor-driven, the 
motor being of 20 horsepower capacity for large and heavy work. The 
foot-stock is arranged for sensitive adjustment in order to secure 
straight work. The grinding wheel used in this machine is 24 inches 

Fig. 4. Internal Grinding Machine made by the Bath Grinder Company 

in diameter by 4 inches face. One of the principal features- of the 
machine is the gap arrangement, the gap being 44 inches long. The 
plates shown under the grinder bed are intended to be carefully em- 
bedded to straight-edge and level in the foundation of the machine. 
Wedges are used under the machine resting on these plates. These 
wedges are adjusted with two nuts, one on either side of the projecting 
end of the wedge, the nuts being operated on threaded studs fixed in 
the base of the machine. Both the plates and the wedges are machined 
to insure broad permanent contact. The total length of this machine 
is 22 feet, and it weighs 25,000 pounds. 

In Fig. 4 is shown the Bath duplex internal grinding machine. 
When this machine was placed on the market in 1909 it represented 
an entirely new departure in internal grinding machines. This machine 
places internal grinding on as practical a basis as that which external 


grinding has achieved during the past decade. The principal advantage 
of the Bath duplex grinder is that the arrangement of the grinding 
spindles and the work-holding head, or heads, makes it possible to 
gain considerable time in the grinding and gaging of internal work. 
Two pieces can be ground on the machine simultaneously, and it is 
not necessary to shift the reciprocating slide in order to gage, insert 
or remove the work. It is possible to use two grinding wheels at 
once, one operating from each end of the work. It is also possible to 
use a number of grinding wheels mounted on a supported spindle 
between the two grinding heads and to quickly grind the inside of a 
sleeve or bushing by having one wheel after the other enter the work, 
the previous wheel, of course, leaving the work before the next one 
enters. This saves considerable time, as it makes it unnecessary to 
reverse the reciprocating table for each cut. 

Fig. 5. Face Grinder made by the Diamond Machine Company 

The novel feature which, in particular, distinguishes this machine 
from older designs is that the grinding wheels and spindles pass in 
through the back end of the head-stock spindle as shown in Fig. 4 r 
instead of running into the head-stock spindle from the front. 

In Fig. 5 is shown a large face grinder designed and built by the 
Diamond Machine Co., of Providence, R. I. This machine is designed 
for the general run of surface grinding operations in the ordinary 
machine shop. It consists, as seen from the illustration, of a recipro- 
cating table sliding on a long bed. The table carries the work back 
and forth in front of the face of a large ring emery wheel. The longi- 
tudinal table movement is obtained by an open and crossed belt 
reversing mechanism in the back of the machine, which is connected 
to the table rack by heavy gearing. The mechanism used is similar 
to that of a planer. Machines of this type can be profitably applied 
to such operations as the finishing of machine columns, pipe flanges 
and a great number of similar pieces requiring a plane surface, but 
not a high degree of accuracy. 



The development of the grinding machine has made rapid progress 
during the last few years, and the process, of grinding is more and 
more recognized as having both economical and technical advantages, 
as compared with the old methods of obtaining finish. This is espe- 
cially true regarding plain cylindrical grinding, and this is due chiefly 
to the fact that the machines for this kind of grinding are easier to 
build, and in general more efficient, than machines for other kinds of 
grinding. It is probably true, however, that there is more misunder- 
standing among engineers and workmen in regard to cylindrical grind- 
ing than in the case of any of the other mechanical arts. Nearly 
every operator has a different theory, and each maker of grinding 
machines has his own method of grinding. 

Relative Time Required for Finishing 1 by 
Turning- and Grinding- 

It has often been claimed, by people who have had long and thorough 
experience in regard to this subject, and whose testimony, therefore, 
must be considered as having weight, that time can be saved in finish- 
ing a cylindrical piece of work by taking a roughing cut with an ordi- 
nary cutting tool, leaving about from 0.008 to 0.010 inch of metal, and 
grinding off this amount instead of taking a second cut in the lathe 
and finishing the piece by filing. One great advantage of the grinding 
machine is the closer finish that can be obtained. Mr. C. H. Norton, 
President of the Norton Grinding Co., Worcester, Mass., in a lecture 
before an engineering class at Columbia University, dealt with the 
question of relative time and cost of finishing by grinding and turning 
as follows: 

"A shaft Qy 2 inches diameter, 10 feet long, rough turned cheaply to 
within about 1/32 inch of the required finished size, can be finished 
straight, round a'nd to size by turning with a tool that cuts off a chip 
removing the 1/32 inch from the diameter and leaving a good surface 
that requires some filing and polishing with emery cloth, and the time 
required to thus finish the shaft to a limit of 0.0005, plus or minus, is 
from seven to eight hours, according to the quality of the cutting tool 
and of the material in the shaft, and to the skill and ambition of the 
workman. To remove the 1/32 inch from the diameter, and finish the 
"shaft to a limit of 0.0005 inch, plus or minus, by the grinding method, 
with a modern grinding machine, requires from one to two hours 
according to the ambition of the operator, and the finish is superior 
to that obtained by the other method. Here is a case where the grind- 
ing wheel, cutting the material to a powder with microscopic cutting 

* MACHINERY, April, 1908, and May, 1900. 


points, is more economical than the steel tool cutting a chip large 
enough to be seen and handled. 

"The statement that it requires enormous power to grind steel to 
a powder, while cutting it into larger chips does not, should be given 
careful consideration. In the case of the 10-foot bar we use an average 
of approximately eight horsepower from one and one-half to two hours 
when grinding. When turning to finish we use about two horsepower 
from seven to eight hours. In the case of the 10-foot bar we require 
a nice surface and accurate measurements to a thousandth of an inch 
or less, and in both the lathe and the grinding machine we remove 
1/32 inch, more or less, from the diameter. The production of such 
a grade of v.ork when removing 1/32 inch, more or less, shows greatest 
economy by the grinding method. When, however, the surface can 
be very rough and the diameter may vary within much larger limits, 
a steel tool cutting deeply will remove the same amount of metal in 
shorter time. If the object is simply to remove a certain number of 
pounds of metal, turning it off with a steel tool is cheapest. But, as 
we know, nearly all round work must have accurate, or approxi- 
mately accurate, diameter, and from approximately smooth to 
very smooth surface. The great majority of round work must finally 
have a better surface and more accurate dimensions than can be ob- 
tained with a steel tool when it is cutting at sufficient speed and depth 
to enable it to remove material faster than by grinding. Therefore, 
the limited amount of material that is removed by the finishing opera- 
tion on accurate, and approximately accurate, work can be more 
economically removed by grinding than by turning. 

"There are also cases when work not requiring a smooth surface 
may be more cheaply ground than turned. A case that illustrates this 
is that of some bridge pins, which were from 42 inches to 8 feet long 
and from 3 inches to 18 inches diameter. These pins were roughly 
turned from the billet to within 1/32 or 1/16 inch of the required 
diameter, and then completed by another, or sizing, cut. The limit 
of variation from size is 0.010 inch; if the turning tool is made to cut 
the right diameter at the starting end, the cutting edge of the tool 
must not wear off quite 0.005 inch when doing the work of removing the 
steel from the entire length of the pin. Now, in order to ensure that 
the tool will not wear away enough to cause an error beyond these 
limits, it becomes necessary to revolve the work so slowly that the 
same results can be obtained more quickly by grinding, although the 
surface may be far from smooth. Grinding is accomplished by a 
number or rapid cuts, and during the final or light cuts the grinding 
wheel does not wear at all, so that we are enabled to produce work 
of uniform diameter regardless of the length. 

"While practically all round work is turned before grinding, there 
is a portion of such work that- is most economically ground without 
turning. Owing to certain shapes or structural weakness it sometimes 
becomes difficult to turn; in such cases grinding is more economical 
than turning. An extreme case of this kind is that of a shaft, or bar 
of steel, 9/16 inch diameter and 10 feet long, with 1/16 inch to be 

10 No. 38 GRINDING 

removed from the diameter to produce an accurate %-inch bar within 
a limit of 0.0005 inch, plus or minus. It is easy to understand how 
difficult it would be to turn this bar. It is, however, very easy to 
grind such a bar to the limits, and in short time. The roughing cuts 
that take the place of the turning easily remove the stock to within 
a few thousandths of an inch in about ten minutes, while hours would 
be consumed in turning such a bar to even coarse limits. 

"In the case of slender work that springs badly when it is turned, 
the work can, many times, be ground more quickly than it can be 
turned and ground ; because, when grinding off the material, the spring 
is ground out as it occurs, owing to the many cuts, or passes, of the 
grinding wheel; while when it is turned, with one cut over the piece, 
it must be straightened before the finishing cut is taken. It is true, 
however, that the majority of work should be turned before grinding." 

Another fact in connection with cylindrical grinding to which Mr. 
Norton calls attention is that in order to secure the greatest economy 
by the use of grinding machines, less should be paid for the turning 
than when the work is to be finished in the lathe. With well con- 
structed grinding machines, the coarser the turning, the quicker the 
grinding can be done. It is no longer necessary to turn either smoothly 
or correctly to size. A variation of 1/32 inch, more or less, on large 
work is of no moment, and on small work a variation of 1/64 inch 
is permissible, and the surface may be very rough in all cases. A 
large part of the economy is secured by cheap turning. 

Grinding Hard Alloy Steels 

In some special cases, when the steel to be finished is so hard that 
it cannot be cut by means of a cutting tool, the grinding machine has 
to take the place of the lathe entirely. Of course, the work in this 
case cannot be done so cheaply as in the case of ordinary kinds of 
steel, but still it can be done with fair economy. As the piece is taken 
entirely rough and put up in the grinding machine, there is, consider- 
ing the errors in casting, about 1/8 inch up to 3/lt> inch on the 
diameter that has to be ground off. When so large an amount of 
metal has to be removed by grinding, another problem than that dealt 
with when only 0.008 to 0.010 inch has to be ground off, presents itself. 
The writer at one time designed three special grinding machines, two 
for external and one for internal work, all for very heavy duty. Herein 
are given a few of the conclusions arrived at while designing these 
machines. Being, as mentioned, mostly used for heavy grinding, the 
machines may differ some from the common light grinding machines, 
but the principles remain, in general, the same. 

Any machine tool must, of course, be designed heavy enough not 
only to take all the strains produced by the action of the cutting tool 
or wheel, but to prevent all, or nearly all, vibration and chattering 
of the machine itself. This is true of the grinding machine more 
than of any other machine tool. Rigidity is a very important factor 
in the efficiency of the machine, both in regard to heavy grinding and 
grinding for very exact sizes and high finish. 


Influence of Vibrations on Action of Grinding Machines 
The grinding wheel rotating at a high speed tends to jar its bearings 
and supports. Vibrations of this kind would result in an oscillating 
motion of the grinding wheel perpendicular to its own axis of rotation 
and along the line connecting the center of work with the center of 
the wheel. The frequency of these vibrations depends entirely upon 
the weight of the oscillating parts. The cause of the vibrations is that 
the center of gravity of the rotating parts, grinding wheel, shaft, 
pulley, etc., is not entirely the same as the center line of rotation. 
This is partly due to the uneven structure of the material. It is 
very plain to everybody that the oscillating grinding wheel cannot 
cut to its lull capacity. The length of the oscillations might not be- 


of Wheel, 








large, perhaps only one-thousandth of an inch or a fraction thereof, 
but the cut will be just so much deeper one moment than the next 
following. Only at one moment, when the wheel is furthest in, will 
it cut to its full capacity. 

It is very important, in order to secure nice running of the wheel, 
to have the belts in good order, and to have the boxes closely adjusted, 
even though they run a trifle warm. Because of the high speed of 
the shaft, the boxes ought to be made with ring-oiling devices. This 
would allow a closer adjustment, and secure a better running of the 
shaft. However, as far as the writer knows, there are no grinding 
machines on the market equipped with ring-oiling boxes. The slides 
should, for the same reason as the boxes, be adjusted closely, even 
though they slide hard. 

Speed of Grinding Wheels 

The peripheral speed of the grinding wheel should be approximately 
from 5,000 to 5,500 feet per minute. There are occasionally cases when 
higher speed is desirable, but with higher speed there is danger of 
the wheel creaking. The wheel should, however, never be run slower 
than 5,000 feet per minute, because it becomes less efficient at slower 

Above will be found a table which gives the number of revolutions 
per minute for specified diameters of wheels to cause them to run 
at the respective periphery rates of 5,000 and 6,000 feet per minute. 



R. P. M. for 

R. P. M. for 

Surface Speed 

Surface Speed 

of 5,000 feet 

of 6,000 feet 



























12 No. 38 GRINDING 

Experience has shown that for grinding work with fairly large diam- 
eter, better results are obtained by using a comparatively small wheel 
than by using one with too large a diameter. The explanation of this 
fact is that the wheel of smaller diameter clears itself faster from the 
work, while the larger one has a larger contact surface, and, there- 
fore, the specific pressure between wheel and work becomes reduced, 
and the metal removed by the wheel stays too long a time between 
the wheel and work, and prevents the particles of the wheel from 
cutting properly into the work. The peripheral speed must, however, 
be the same for the smaller wheel as for the larger one. 

Surface Speed of Work 

The proper surface speed of the work varies somewhat with the 
material and kind of work to be done. The grinding machine builders 
recommend 15 to 30 feet as a good average speed range for ordinary 
kind of work. For cast iron this can be slightly increased. The writer 
has had experience in grinding a very tough and hard steel (man- 
ganese steel), and has found the right surface speed in this special 
case to be as low as 6 to 8 feet a minute for rough grinding. For 
the finishing grinding, the speed should be somewhat higher than for 
the rough grinding. For delicate work the speed should be slow, 
because the work could easily be damaged by forced grinding. 

As a general rule, for determining the surface speed for a certain 
kind of material, one can say that a brittle material, as cast iron, 
takes a high speed, while a tough and hard material, as the best tool 
steel, takes a slow speed. For grinding close to size and for high 
finish, the depth of the cut must be small, and higher surface speed 
can consequently be used. 

Many of the grinding machines on the market are built so as to 
have the work revolving on two dead centers. This is done more for 
the sake of being able to obtain accuracy than for the sake of increas- 
ing the cutting efficiency of the machine. 

Traverse Speed of Grinding Wheel 

The traverse speed of the grinding wheel should for ordinary grind- 
ing be three-fourths of the width of the wheel, that is, for one revo- 
lution of the work the wheel should travel three-fourths of the width 
of the face. If the wheel be traversed slower, the new cut is over- 
lapping the old one more than necessary, and too large a part of the 
wheel is idle. It is, however, necessary that the new cut overlap the 
old one with about one-fourth of the width of the face, because the 
edges of the face easily become rounded off, and, if the travel be too 
rapid, the result is an uneven surface. 

The capacity of the wheel, within certain limits, of course, is pro- 
portional to the width of the face. A certain specific pressure between 
wheel and work is required for the highest cutting capacity. A wider 
wheel requires consequently a larger total pressure. But many of 
the machines now on the market are not rigid nor heavy enough to 
stand the pressure needed for a fairly wide wheel, cutting at full 
load, without vibration and chatter. The grinding machines on the 


market have not, in the writer's opinion, yet reached their full capac- 
ity. Wider wheels should be used, and the machines should be designed 
and built heavier in order to take the load of the cutting wheel, with- 
out perceptible vibration of the machine. 

For the final smooth finish, a slower traverse speed should be used, 
especially if the face of the wheel is not kept a perfectly straight line. 
A smoother surface is obtained by using a slower traverse speed. 
The part of the wheel which is overlapping, while theoretically it does 
not cut, still wears away the unevenness left from the first cut, and 
thus to some extent polishes the surfaces. 

While grinding a plain cylindrical piece of work, the grinding wheel 
should not be allowed to travel too far past the ends of the piece 
before reversing; it is only waste of time. The wheel should be 
reversed when three-fourths of its width is past the end of the work. 

Depth of Cut 

Tne depth of the cut to be taken depends upon the material, kind 
of wheel, and the work done. It should be deep enough to permit the 
wheel to do its utmost. This is, of course, true only about pieces that 
are rigid enough to stand a heavy cut. The grinding operator himself 
will have to determine the depth of cut for each individual case, 
judging it by the prevailing conditions of work, machine, and wheel. 

When the piece to be ground, owing to the hardness of the material, 
cannot be roughly finished by a cutting tool before being placed in 
the grinding machine for the final finish, there is often up to 3/16 inch 
on the diameter to be removed by grinding. Employing the same 
principle as when the piece is roughly turned in a lathe, previous to 
the grinding operation, the work should first be put up in a machine 
equipped with a coarse and wide grinding wheel. A wheel of this 
kind is capable of removing stock rapidly. The piece should be fin- 
ished to within 0.005 inch of the finished diameter in this machine, 
and then moved to a machine equipped with a finer grain wheel, and 
the final finish given to it. 

The Grinding Wheel 

For heavy grinding, the alundum wheel is the best for removing 
stock rapidly. The carborundum wheel will give a smoother finish, 
and is to be recommended for the large majority of other classes of 
grinding. Emery is less abrasive, but gives a higher polish. Most 
grinding wheel manufacturers recommend their medium grade, M. 

The question as to what is the very best wheel for finishing any 
particular piece cannot be definitely answered. On the next page is 
given a table of wheels which can, with advantage, be used in the 
cases mentioned. This table is recommended by one of the largest 
grinding machine manufacturers. 

Grit No. 24 may be too coarse for any but rough classes of work, 
but if mixed with No. 36 it gives a fair result. No. 30 used separately 
is capable of a very fair commercial finish, but if mixed with No. 46 
will give as fine a finish as is desired by the majority of the grinding 
machine users, and at the same time it retains the rapid cutting 

14 No. 38 GRINDING 

capacity. Nos. 46 and 60 are as fine as is necessary for almost any 
manufacture, although finer than these are used by some concerns 
who require a very high gloss finish. 

A satisfactory grinding wheel is an important factor in the produc- 
tion of good work. In machine grinding, it is desirable, in order that 
the cut may be constant, and give the least possible pressure and heat, 
to break away the particles of the wheel after they have become dulled 
by the act of grinding. It is the capacity of yielding to or resisting 
the breaking out of the particles which is called grade. The wheel 
from which the particles can be easily broken out is called soft, and 
the one that retains its particles longer is called hard. It is evident 
that the longer the particles are retained, the duller they will become, 
and the more pressure will be required to make the wheel cut. Retain- 
ing the particles too long causes what is familiarly known as glazing. 
A wheel should cut with the least possible pressure, and must therefore 
be sharp. This sharpness is maintained by the breaking out of 
particles. Therefore, a wheel of proper grade, cutting at a given 
speed of the work, possesses "sizing power," or ability to reduce its 
size uniformly without breaking away its own particles too rapidly; 



Grit No. 


a -, ( Ordinary shafts . . 
a?"J Steel tubing or very 
bteelj Hght shalts 

24 to 60 
24 to 60 

Two or three grades softer 
than medium. 

Tool steel or Cast iron '. 
Internal grinding 

24 to 60 
30 to 36 

Medium or one grade softer. 
Medium or several grades 

obviously if the work is revolved at a higher speed, the particles will 
be torn away too fast, and the wheel will lose its sizing power. 

The properties of toughness and hardness of the material to be 
ground have a retarding influence on the grinding because they make 
the material stick to or clog the wheel. The ground-off material, in- 
stead of being thrown away from the wheel by the centrifugal force, 
gets in between the particles of the grinding wheel. It is self-evident 
that this has a greatly retarding effect on the cutting quality of the 
wheel. A brittle material, on the contrary, does not have the tendency 
to clog the wheel, but the stock ground off is immediately thrown 
away from the wheel, leaving the particles free to cut without the 
retarding action of undue friction, and the generation of more than 
the due amount of heat. If we take into consideration only these 
properties of the material to be ground, the tough or leady material 
requires a soft wheel, because the particles must break away fast 
enough to prevent the wheel from being clogged. In this case, the 
particles do not wear enough to become dull, but must break away 
before this. When grinding a brittle or hard material, on the con- 
trary, the wheel is less liable to be clogged, the particles do not need 
to break away so soon, and, therefore, a harder wheel should be used. 



However, the wheel jnust not be so hard that the particles get too 
dull and become inefficient as cutting agents before they break away. 

Importance of Wheel Running- True 

In order to obtain the full efficiency of the grinding wheel, it must 
be run perfectly true; that is, cut evenly all the way around. The 
grinding wheel detects its own errors. A slight difference in the 
sparks indicates that the wheel is out of true. The eccentric wheel has 
about the same kind of action as the one which is vibrating because of 
too weak supports. Furthermore, the edge of the grinding wheel 
should be kept perfectly straight. If the edge be curved, however 
slightly, a curved cut will be the consequence. Many grinding machines 

Machiner U ,X.Y. 
Fig. 6. Fixture for Truing- Emery Wheel with Diamond 

give inefficient results because the edge of the wheel is not kept in 
a true straight line. The operator seldom appreciates the great im- 
portance of this, and, therefore, the foreman should watch the men 
closely in regard to this point. 

The best tool for truing the wheel is the diamond, but, this being 
rather expensive for shops where not very much grinding is done, the 
usual emery wheel dresser can be used to good advantage. In truing 
the wheel, the dressing tool should be kept stationary and rigidly 
supported, and the wheel should be traversed back and forth, until a 
true edge is obtained. Fig. 6 shows a fixture and arrangement for 
wheel truing with a diamond. 

Wet and Dry External Grinding 

Nearly all plain cylindrical grinding is now r done wet. There are 
many reasons why the w T et method is to be preferred to the dry. Be- 

16 No. 38 GRINDING 

cause of the friction between the grinding wheel particles and the 
work, as well as between the cut-off material before it leaves the wheel 
and the work, more or less heat is generated. If this heat is not 
carried away, the work will be burned. Besides, the edge of the grind- 
ing wheel would be highly heated, but the center would still remain 
comparatively cool, and the outside would expand and there would be 
danger of the wheel breaking. It is found that the water has a soft- 
ening effect upon the wheel; therefore, a harder wheel is required for 
wet grinding than for dry. 

Machines with Two Grinding- Wheels 

The grinding machines on the market are equipped with only one 
grinding wheel, but there is no reason why two grinding wheels can- 
not be employed to advantage. In this case one wheel is to operate 
on each side of the work. As both of the wheels are to throw the 
sparks and the water down, one of the wheels has to cut with the 
revolving of the work, that is, the peripheries of the wheel and the 
work are going downwards. This is, of course, not the ideal condi- 
tion, but, when, the work is revolving at a slow peripheral speed, there 
is not much difference in the cutting capacity of the two wheels. 

It is self-evident that, when employing two wheels, one at each side 
of the work and just opposite each other, the traverse speed of the 
wheels must be twice as fast as in the case of only one wheel, or three- 
fourths of the width of the wheel for one-half revolution of the work. 
Otherwise one wheel will overlap the cut of the other. 

The two machines for external grinding which the writer designed 
have two wheels working according to the principle previously de- 
scribed. Fig. 7 gives an idea of the arrangement used on one* of these 
machines. The principal features of the design can be studied direct 
from the illustration without any further comments. 

One new feature of these machines is that each grinding wheel is 
driven independently by a motor. This motor is mounted above the 
wheel spindle, and is belted directly to it. Special attention has 
been paid to designing the support of the motor in order to prevent 
the vibrations of the motor from being transferred to the grinding 

Internal Grinding- 

The development of internal grinding machines has not advanced 
as fast as that of machines for external grinding. It has even gone so 
far that one man holding a prominent position with one of the largest 
grinding machine manufacturing concerns in the country has said 
that in his opinion, the internal grinding machine is a mistake from 
start to finish, and that it will never be made a success. This, how- 
ever, is rather too broad a statement in face of some recent develop- 
ments in this direction. 

The statement just quoted, nevertheless, was not made without good 
reason. As we have already seen, the rigidity of the arrangement for 
supporting the grinding wheel is a very important factor for all 
efficient grinding. The internal grinding machine does not very well 



lend itself to the employment of any rigid and heavy fixtures, and the 
grinding wheel must necessarily be small, and therefore lacks the 
strength to stand a heavy cut. The designer, when designing the fix- 
tures for internal grinding, has an entirely different problem to solve 
than when designing those for external grinding, where it is com- 
paratively easy to obtain ample rigidity. The internal grinding wheel 

Fig. 7. Grinding Head for External Grinding Machine 

must be mounted at the end of a small spindle which projects past 
the bearing far enough to enable the wheel to reach past the end of 
the hole to be ground. Such a spindle rotating at a high speed is 
liable to vibrate, especially if pressure be applied at the end of it, 
as is here the case. 

Sometimes, however, it becomes absolutely necessary to grind, in- 
ternally, even a comparatively large amount of stock. This is the 
case when finishing manganese steel, this material being so hard that 



it cannot be cut by any kind of tool steel. Take the case of bores of 
manganese steel car wheels. As the grinding of the bores must be 
done without any stock having previously been removed from the 
rough casting, on the average about one-eighth inch of metal must 
be ground off from the hole. All the errors in the cored hole, as eccen- 
tricity in reference to the circumference of the wheel, etc., must be 
corrected by grinding. A hole cored in a manganese steel casting is 
always comparatively much rougher than a hole cored in cast iron, 
and all this must be taken into consideration, when determining the 
amount of stock to leave for the grinding process. 

Desig-n of Heads for Internal Grinding 

The fixture used in the internal grinding machine designed for grind- 
ing these wheels is shown in Fig. 8. Internal grinding fixtures gen- 
erally have a long extension bearing, as shown in Fig. 9. This serves 
to support the spindle as near to the grinding wheel as possible; but 

Machinery, N.T. 

Fig. 8. Grinding Head for Internal Grinding 

the diameter at the root of this extension, that is, nearest to the box, 
cannot exceed the diameter of the grinding wheel. 

The spindle, shown in Fig. 8, is made solid, and has the largest 
diameter possible for the size of the grinding wheel. An increased 
amount of rigidity and a greatly increased simplicity is gained by this 

When working, the grinding wheel produces, especially in dry grind- 
ing, very much dust. When inside a hole the dust cannot very easily 
get away, but whirls about in the hole. If the spindle has a bearing 
near to the grinding wheel, the dust will find its way into the journal. 
This drawback is entirely eliminated by having a large solid spindle 
without a bearing near to the grinding wheel. 

As to the relation between the overhanging part of the spindle and 
the distance between centers of the boxes, there are many factors that 
come into consideration in regard to this relation, such as the design 
of the boxes, the diameter of the spindle, how close the spindle can be 
allowed to run in the boxes, etc. However, the distance between the 
centers of the boxes should be made as large as the general design 
conveniently permits. 

Fig. 8 shows at A the support for the motor. This support is 


placed on the top of the top rest. The driving pulley is placed between 
the bearings, so that the support could be made as rigid as possible. 

It was found by actual experience with these fixtures, that when 
the grinding wheel was taking a fairly heavy cut, the spindle did not 
vibrate nearly so much as when the wheel was running idle. The 
springing quality of the spindle, and the pressure between work and 
wheel made the wheel cut without any chattering worth mentioning. 

Regarding the peripheral speed of the grinding wheel, what has 
already been said with reference to external grinding is equally 
applicable to internal grinding. 

Because of the lighter fixtures, the speed of the work should be 
slower than for external grinding. The writer has found the right 
cutting speed for hard and tough steel to be, for heavy grinding, about 
seven feet a minute. For the finishing, the speed can, with advantage, 
be somewhat higher. The wheel should travel three-fourths of its 
width for one revolution of the work, the same as for external grinding. 

Wet and Dry Internal Grinding 

One point that has been much discussed in regard to internal grind- 
ing is whether it shall be conducted wet or dry. Some grinding 



Fig;. 9- Common Construction of Grinding: Heads for Internal Grinding: 

machine designers have advanced the opinion that it, by all means, 
must be done dry, but others claim the wet method to be superior. 
For light finishing grinding one method might be considered as good 
as the other, because so small an amount of heat is generated that 
there is no danger of burning the material or breaking the wheel. 
But, for heavier grinding, a considerable amount of heat is generated, 
and it becomes necessary to carry it off by water. At least, such is the 
writer's own experience on this subject. At a test recently conducted 
to find out the actual difference between dry and wet internal grinding, 
it was found that the cutting quality of the grinding wheel was about 
the same in both cases, but, with a heavy feed and dry grinding, the 
work was highly heated, and the wheel broke after about half an 
hour's run, while, with wet grinding, the wheel stood the heavy cut 
continuously without breaking. 

The water can be injected into the hole in a stream about 1/16 inch 
in diameter. In addition to carrying away the heat, the water serves 
to wash away the removed stock from the hole. 

Tests have been undertaken on the above mentioned internal grind- 
ing machines, in order to find out the time required to grind the 
bores of a certain kind of manganese steel car wheels. Two different 

20 No. 38 GRINDING 

kinds of wheels were tested. The first one, a 20-inch diameter wheel, 
had a bore 2% inches in diameter and 5% inches long, and it was to 
be ground for a press fit. The second one, an 18-inch diameter wheel, 
had a bore 3^4 inches in diameter and 4y 2 inches long, and was also 
to be ground for a press fit. Four wheels of each kind were ground 
during the course of the test, and it was found that the actual time 
for the grinding operation, not including the time required for put- 
ting up the work in the machine, was, for the first kind of wheels, 1 
hour and 23 minutes for all four, and for the second kind, 1 hour and 
9 minutes for four wheels. Considering that the bores of the wheels 
were not previously turned, but entirely rough, as the wheels were 
taken directly from the foundry, and considering the hardness and 
toughness of the steel, the results obtained were considered good. 
The time of putting the work in the machine was about 6 to 8 
minutes for each wheel. As the machines work automatically, one 
man is able to run three machines. Counting 8 minutes for the put- 
ting up of each wheel, the man is able to grind one wheel of the first 
kind in 30 minutes, and one wheel of the second kind in 26 minutes. 

The work was revolved at a speed of 7.7 revolutions per minute. 
This makes a peripheral speed, for the first case, of 5.8 feet per minute, 
and for the second case, of 6.6 feet per minute. The grinding wheel 
used was a 2-inch diameter, 1-inch face, No. 46 grit, O grade alundum 
wheel. It was run at a speed of 4,750 feet per minute. 

The traverse speed of the work was as high as 0.84 inch per revolu- 
tion of work. This allowed the wheel to overlap the old cut by only 
0.16 inch, but, as the grinding wheel was trued very carefully, this was 
found to be all that was required for obtaining a nice smooth surface. 
The traverse feed was not slowed down, but remained the same while 
doing the final finishing, and a very satisfactory finished hole was 
obtained. The test was made throughout with wet grinding. 

For heavy cylindrical grinding, which has especially been referred 
to, the width of the wheel used varies between 1% and 2% inches, 
regardless of the diameter. In some special cases narrower wheels 
than iy 2 inch are used, but these special cases are exceptions to the 
general practice, and must be recognized as such by the machine build- 
ers and users. Although larger wheels are used, there is no doubt 
that the best range of diameters of wheels is between 12 and 18 inches. 
For how wide a wheel the grinding machine of the future can be 
designed, has yet to be decided; but, wider wheels and heavier ma- 
chines point the direction of the road which the designer and machine 
builder should follow for the development of the grinding machine. 



It is very often the case that a grinding machine falls short of its 
highest possible output by reason of the inattention of the operator 
to some of the short cuts and time-saving methods that have been 
highly developed in the use of other machine tools. It is the case 
with grinding machines as with other machine tools, that the develop- 
ment of short cuts and kinks of various sorts greatly increases the 
aggregate output of the machines. The lack of such time-saving 
methods is the reason for the unfavorable attitude of some firms to 
the grinding process. It is almost always the experience of a demon- 
strator sent out by the manufacturer of grinding machines that his 
results are not maintained by the operator. First one little kink is 
lost sight of, then another, and the time increases very slightly on 
each individual piece ground; consequently the aggregate of the 
day's output is soon considerably below what it should be. 

A grinding machine will size work to a commercial degree of 
accuracy with remarkably little attention on the part of the operator, 
but the quantity output of a machine is very largely dependent on 
the ability and willingness of the operator to hustle; hence the reason 
for the almost universal prevalence of the "piece system" in the grind- 
ing department. 

Spotting 1 "Work for the Back-rests 

Spotting work for the back-rests is a great help to the operator in 
several ways. When a piece of work is placed in the machine, assum- 
ing that the automatic cross-feed stop shield shown at A, Fig. 10, has 
been adjusted from a previous piece ground, the grinding wheel should 
be run back from the work about 1/32 of an inch. This is accom- 
plished by turning the cross-feed handwheel B about one revolution 
in the opposite direction to that in which it is automatically revolved. 
The table carrying the work can now be moved by the handwheel 
marked C, provided for the purpose, bringing the various back-rests 
successively in front of the grinding wheel, as shown in Fig. 11. The 
wheel is then fed into the work by hand without reciprocating the 
table; it should be fed into the work until the diameter where the 
shoes of the steady-rests bear is within one-thousandth of an inch 
of the finished size. The guard A, Fig. 10, as it approaches the pawl 
D. serves as a gage to determine the extent to feed the wheel in. 
This operation takes a very short time and provides a smooth surface 
for the bronze shoes. This surface is so near the finished diameter 
that the shoes are not worn large before the work is reduced to the 
finished size, therefore the work is accurately and steadily supported 

* MACHINERY, May, 1910. 



during the finishing cuts. This is found particularly advantageous on 
hardened work where a large amount of stock is left for finishing and 
where chatter marks are more apparent if the supporting shoes of 
the back-rest do not fit the work very closely. It is also found to be 
a great saving on the wear of the shoes themselves. 

Grinding to a Shoulder 

A time-consuming error among grinder operators is made on account 
of the prevalence of the idea that the reversing dogs on a grinding 
machine should be set to reverse the table traverse when the grinding 
wheel is within a very few thousandths of an inch of a shoulder on 

Fig. 1O. Controlling Mechanism of the Brown & Sharpe Grinder 

the work as at E, Fig. 11. It is a fact that most grinding machines 
will reverse within one or two thousandths of an inch of the same 
place each time, provided the rate of table travel is not changed. 
The variation in the depth of center holes in the work makes it neces- 
sary for the operator to try the reversing of the machine by hand 
after placing each piece in position to make sure that the wheel will 
not gouge the shoulder, as it would surely do if the center holes 
were a little smaller than in the piece previously ground and a close 
limit for reversing were used. There are two ways to prevent the 
wheel gouging the shoulder on the work; one is the adjusting of the 
reversing dog by the screw F, Fig. 10, for each piece ground, which 
is a time-consuming operation; the other, and better method, is to 


bring the shoulder on the work up to the wheel by hand (using hand- 
wheel C after spotting for the back-rests as described), then feed 
the wheel straight into the work, reducing the diameter next to the 
shoulder for a distance equal, of course, to the width of the wheel, 
to the finished size. This is quickly and easily done by using the 
knock-off shield A against the pawl D for a gage as described. The 
table may then reverse for the subsequent complete grinding of the 
piece when the wheel has advanced not closer than % inch or more 
from the shoulder, and the edge of the wheel next to the shoulder 
does not become worn away or rounded because it runs off the work, 
or nearly off, at each end of the piece. A wasteful truing off of the 
wheel is thus easily avoided. 

This operation of necking the work at a shoulder with the full 
width of the wheel, obviates the necessity of a dwell of the table at 
that reversing point. If a machine dwells when reversing at the 
shoulder end of the traverse, it must of necessity dwell at the other 
end where the wheel usually runs nearly off the work; here the dwell 
is not only of no va'ue but it is very likely to cause the wheel to 
grind the end of the work undersize. While this dwell is only 
momentary, it is quite a factor in a day's output that can be readily 

If for any reason the necking of the work with the wheel is not 
deemed expedient and a dwell is required, this dwell is needful only 
once or twice during the grinding of a given piece and can be pro- 
duced at will by the operator pressing the knob H, Fig. 10. Pressing 
this knob stops the power traverse of the table, which may then be fed 
over by hand, using hand-wheel G, to face up the shoulder, and the 
dwell may be prolonged to allow one or more revolutions of the work 
as the particular quality of work and wheel may require, instead of 
the length of dwell being dependent entirely on the speed of the 
reciprocating table, as is the case when the dwell is automatically 
supplied by the gearing of the table. The table traverse is started 
after the dwell by pulling the knob h to the position shown in Fig. 
10. A dwell so produced is not duplicated at the other end of the 
work except at the will of the operator. 

It should be clearly understood that a grinding machine can be 
reversed with the shoulder on a piece of work % inch or more from 
a grinding wheel, then stopped at the reversing point and "forced 
over" this % inch or more beyond the normal reversing point without 
disturbing the reversing dogs and without subjecting the reversing 
mechanism to any strain. 

In Fig. 13 are shown the elements of the reversing mechanism of 
the Brown & Sharpe plain grinders, which shows quite clearly how 
this traverse beyond the reversing point is accomplished. When the 
reversing dog J or K, Fig. 10, strikes and reciprocates the reversing 
lever L, the motion is transferred by its fulcrum stud and lever M 
to the arm :v which compresses the reversing spring 0. When this 
arm, which compresses the spring, has moved far enough to give con- 
siderable tension to the spring, the taper lug on the arm A' raises 





the latch P. thus releasing the yoke R which is connected to the re- 
versing clutch 8, this deriving its power from gear T. The spring, 
which is under compression, throws the reversing clutch as soon as 
the latch releases the yoke R. This movement of the yoke is sufficient 
to relieve the compressed reversing spring so that the reversing lever 
can traverse farther than the position where reversing takes place 
without unduly compressing the reversing springs. Thus the facing 
up of a shoulder on the work slightly beyond the reversing point 
without disturbing the reversing dogs on the sliding table can be 



Machinery, X.T. 
Fig. 13. Reversing Mechanism of the Brown & Sharpe Grinder 

readily accomplished without undue strain on any part of the re- 
versing mechanism. 

Truing- the Wheel 

When truing the periphery of a grinding wheel for all regular 
cylindrical work, a bort diamond, mounted in a suitable holder, is 
used, as shown in Fig. 14. This holder should be so mounted in its 
support that the distance from the diamond to the support V is as 
short as possible, thus avoiding spring or vibration in the holder 
which produces an irregular surface on the grinding wheel, appearing 
on the work in the form of a mottled effect or chatter. When work 
of large diameter is being ground, the wheel should be brought for- 
ward to the position shown in Fig. 14 when truing it off. The time 
consumed in moving the wheel forward with the cross-feed handwheel 



is more than offset by the more rapid cutting of the wheel by the 
diamond, and, furthermore, the surface of the wheel is in better 
shape, as stated. 

For internal grinding, it has been proved economical by practice 
to true off the grinding wheel with a piece of a large wheel that has 
been worn down to such a small diameter as to render it useless for 
grinding; this piece of wheel must be harder than the wheel being 
trued off. This is a much quicker process than using a diamond, as 
the piece can be held in the hand and pressed to the grinding wheel 
as often as occasion requires. For this same purpose, bricks of various 

Fig. 14. The Way the Diamond Tool should be mounted for Truing the Wheel 

abrasive materials have been made and are generally used when pieces 
of a worn-out wheel are not available. 

Desig-n of Pootstock 

The footstock of the grinding machine is of enough importance 
to warrant more attention than is generally allotted to it. The design 
that is in common use among grinding machine manufacturers, in- 
cludes among other elements what might be called a spring-actuated 
spindle. The spring, which forces the center against the work, is 
primarily for the purpose of allowing the work to expand from the 
effect of the heat- developed in grinding. As heat is very largely 
dissipated by the use of water, the more practical value of the spring- 


actuated footstock is to apply a firm pressure of the center against 
the work without sufficient force to distort it. This is very difficult 
to do with a screw and handwheel and is accomplished on a lathe 
by setting the center solid against the work, and withdrawing it 
until the work can be easily turned by hand. This represents a loose- 
ness between centers intolerable on a grinder and also causes a great 
waste of time. When grinding heavy work, there are two reasons 
why it is often necessary to clamp the footstock spindle solid after 
inserting the work in the machine. The weight of the piece of work 
tends to crowd the footstock center back on account of the angle of 
the center; also the momentum of the piece when the table reverses 
at the footstock end of its traverse, tends to pound the center away, 
and any looseness thus developed will render futile any attempt to 
produce round work. 

Time Required for Grinding 

In Fig. 12 are shown several samples of work that have been fin- 
ished on the grinding machine. Without exception every piece has 
been machined complete before it is sent to the grinding department; 
all threads have been cut, keyways and slots milled, holes drilled, etc., 
therefore ail external and internal strains have been equalized in the 
pieces before they are ground. The grinding process is the most free- 
cutting process known to metal workers and should be the last cutting 
process, as it distorts the work the least. The piece marked A is an 
overhanging arm for the milling machine, made of machinery steel 
4y 2 inches diameter and 69 inches long. These pieces require an 
exceptionally good finish and are ground complete in thirty minutes 
for each arm. They are revolved in the grinding machine by a pin 
temporarily driven into one end near the periphery, this pin engaging 
the driving arm on the headstock pulley; with this arrangement pieces 
of sufficient size can be ground from one end to the other complete, 
while such small pieces as those marked C and D must be turned 
end for end to complete the grinding. These last pieces are about % 
and y inch diameter, respectively, and 10 inches long. They are 
ground at the rate of fifteen per hour and have a limit of 0.00025 inch 
either side of the dimension given. The shaft marked E is about 40 
inches long and 1 inch in diameter where it is ground; this piece 
can be readily dogged at one end. These are ground at the rate of 
twenty minutes each with a tolerable variation of 0.00025 inch larger 
or smaller. The tapered collet shown in the center of the engraving 
is ground in four minutes. The milling machine spindle marked F 
is ground complete in "one hour. The limits are very close, viz., 
0.00025 inch total variation, and the taper behind the collar is ground 
to a gage. The smaller spindle marked G has the same close limits 
as the larger one, and is ground complete at the rate of seventy in 
sixty-four hours. The spindle marked J, which is 34 inches long, has 
a threaded guard over the end and on this guard the dog is clamped. 
Thirty-nine of these spindles are ground complete in thirty hours. 
The screw machine spindle marked H is a very difficult piece to 

28 No. 38 GRINDING 

finish owing to the fact that it is bored out its entire length so that 
it is practically a hardened steel shell which is cut away at the smaller 
end. When grinding these spindles in large lots, they -are roughed 
out all over in large quantities, then finish-ground later. It requires 
thirty minutes to completely grind one spindle. 

This last example illustrates very clearly how difficult it is to esti- 
mate the time required for grinding a piece of work, as every feature 
of the piece enters into the problem, and if it were not for the two 
slots which so cut away and weaken the small end that the grinding 
wheel cannot be forced into the work, these spindles could be ground 
in about ten minutes less time for each one. Placing the work on an 
arbor for grinding very seldom increases the total length of time 
to grind when there are several duplicate pieces in a lot, as two 
arbors may be employed and the operator can insert one arbor in 
the work while the machine is grinding the work mounted on the 
other arbor. 

Number of Operators 

The economical operation of a grinding machine presents very 
varied problems. It is sometimes an advantage for one man to run 
two machines; this is generally on long pieces where the time of 
actual grinding far exceeds the time of placing the work in the 
machines. There are, however, some jobs where two men can work 
very successfully on one machine. This is the case with short, large 
bushings that are driven on an arbor, in whicn case the time of 
changing the arbor from a finished piece to an unground piece equals 
or exceeds the actual time of grinding. When these conditions exist, 
the operator and the machine are non-productive at least half of the 
time, and this non-productive period can be reduced to a minimum by 
a helper to assist the operator. 



In any machine shop or department of a manufacturing plant where 
tools for manufacturing operations are made, a properly designed and 
equipped disk grinder should be considered almost indispensable; for 
a large portion of the operations most commonly done with a file, 
and many that are considered surface grinder, milling machine or 
shaper jobs, can be done better and quicker, and at less cost for files, 
cutters, etc., with a disk grinder. 

As a simple example we will take the case of a piece of tool steel 
needed, say, for a box tool, a back rest, a cutter, or a forming tool, to 
be, say, % inch thick, 1 inch wide, 2 inches long, ends and sides straight 
and square all around. Probably the bar steel *4 inch by 1 inch will 
be enough oversize to grind on a disk to exact size, but not enough 
oversize to work with a milling machine, shaper or surface grinder. 
Even if larger stock, say 5/16 inch by 1% inch, or a forging, is used, 

Induttrial Prttt, A'. F. 

Fig-. 15. Snap Gage Finished on Disk Grinder 

it is only necessary to rough one flat side and one edge down fairly 
close to size and finish all over on a disk grinder. For squaring the 
ends of one piece like this, and bringing it to exact length, the saving 
in time over the common way is considerable. Suppose this piece 
has to be hardened, and after hardening must fit a certain space. 
It will need truing up after hardening, and here again the disk 
grinder proves its adaptability. 

Regarding the degree of accuracy obtainable with a disk grinder, 
an example may be of interest. An experienced toolmaker was with 
an exhibit of disk grinders at a fair. Having p ] enty of time on his 
hands, he employed a part of it in grinding up six steel pieces, each 
a one-inch cube. He got the pieces planed roughly in the bar, a little 
oversize, and sawed off a little long. In his spare time he ground 
them to 1-inch cubes, measuring them with a 1-inch micrometer 
caliper. When he had finished with them, there was no point on any 
of the cubes that varied more than 0.00025 inch from 1 inch. Packing 
them together with any combination of sides, the greatest variation 

MACHINERY, June, 1904. 



from 6 inches as measured with a 6-inch micrometer was 0.0005 inch. 
All the sides of all the cubes were so nearly square with each other 
that no error could be detected with a hardened steel square. In 
grinding these cubes no fixture or clamp of any kind was used. They 
were laid on the swinging table, against the rib, and pressed against 
the wheel with the fingers. 

A few examples of the application of the disk grinder to tool-room 
work will give a general idea of its application. Suppose a snap gage 

Fig. 16. Work to be Gaged 

such as shown in Fig. 15 is to be made. With the disk grinder the 
gage can be finished all over, sides, edges and ends, and corners beveled 
or rounded. In hardening, the gage springs somewhat, but can easily 
be squared again on the disk grinder. We are now ready to grind 
the notch to size. Lay the piece on the swinging table, with the back 
edge against the rib, the wheel being in the notch. The piece is now 
ground on both sides without turning it over. This will make the 
faces of the notch parallel with each other, which they might not be 
if the piece were turned over. By the use of an end measure gage 
the snap gage in Fig. 15 is now easily completed. 


A" s 



lndutnal Prtst. K.Y 

Fig. 17. Snap Gage Ground to Size on Disk Grinder 

In a certain shop a JOD came up to be done in the turret machine. 
A number of cast-iron pieces, of the shape shown in Fig. 16, were to 
be machined. There were eight different sizes of pieces and three 
dimensions made to gage on each piece, making 24 dimensions in all. 
The largest dimension on the largest piece was about four inches. 
The smallest dimension on the smallest piece was about % inch. A 
few thousand pieces of each size were to be made. Extreme accuracy 
was not required; a variation of 0.001 inch was allowable. A tool- 
maker was given the job of making a set of snap gages. 

Taking the figures, he made 24 end measure pieces from 5/16-inch 
round drill rod, hardened them, and marked the size. He then cut 



the gages from %-inch thick sheet steel, as shown in Fig. 17. The 
working faces were hardened and ground to the end measure pieces 
on the disK grinder, and the edges squared and the corners rounded 
in the same machine. The gages were not touched with a file except 
to smooth off the edge in the bottom of the notch. 

Pig. 18. Form of Hollow Cast-iron Block used for Test 

The examples given indicate the use of the disk grinder as a tool- 
room machine. This machine, however, is also efficient for removing 
large amounts of metal in a short time. The efficiency of the machine 
for this purpose depends largely upon the kinds of disks used. Tests 
were made at the shops of the Gardner Machine Co., Beloit, Wis., to 
determine the comparative efficiency for grinding cast iron by differ- 

U*. 3 

Machinery, X.Y. 

Fig. 19 Hollow Cast-iron Block used for Test 

ent kinds and makes of disks, such as are commonly used in connec- 
tion with disk grinders. In the following table the different kinds of 
disks are indicated by figures: 

Xo. 1 indicates the Gardner improved abrasive disk No. 126. No. 5 
is the regular No. 24 commercial emery cloth. No. 6 is the same in 
emery paper. Xos. 2, 3, and 4 are disks of excellent quality as com- 
pared to commercial emery cloth. 



The disks tested were all 20 inches in diameter and all excepting 
Nos. 5 and 6 were No. 16 grain. The grinding was done on the ends 
of hollow blocks of cast iron, as shown in Figs. 18 and 19. The area 
ground at the end of blocks was 5 square inches. Reducing the blocks 
one inch in length indicated the removal of 5 cubic inches of metal. 
The grinding was all done on the same machine by the same operator. 

The micrometer stop at the back of the table was set to grind- off 
a fixed amount, usually 0.050 inch, and the twelve blocks ground to the 
stop. The stop was then moved back 0.050 inch and the operation 
repeated until the blocks became too warm for efficient grinding, when 











Wn-i o 



l| 1 



o c 


Time Us 





<u a < 


cubic in 
per m 

1 5 s 

c Q o nj 

"58 s 

P 4) r" 

y</2 e 

Life of 
Based on 










100. % 



































9.4#i 0.7# 










they were cooled, and the time of grinding and the amount of metal 
removed, was noted. This was repeated until the disk was worn out or 
the blocks all ground up. In the latter case, new blocks were substi- 
tuted and the operation continued until the disk was worn out. By 
reversing the blocks they were ground down until the wheel touched 
the handles on both sides. During this test several hundred pounds 
of these blocks were converted into cast-iron chips. 

It will be noted in Table III that it was necessary to use a Hunt- 
ington emery wheel dresser on all disks tested except Nos. 1 and 3. 
The dresser was used whenever the surface of the disk became dull 
and glazed so that it would not cut cast iron readily. The use of a 
dresser shortens the life of the disk, but it is absolutely necessary. 



Grinding a Large Crankshaft* 

A leading English chainmaker some time ago sent to the Norton 
Grinding Co., Worcester, Mass., a rough-turned crankshaft to be ground 
to the dimensions given in Fig. 20. The conditions given were that the 
throw must be % inch plus or minus 0.001 inch and that the keyway 
shown in Fig. 20 should line up exactly with the highest point of the 
eccentric. The keyway was already in the shaft when received. The 
following method was pursued in preparing the crankshaft for the 

Two cast-iron blocks, Fig. 22, were planed to the dimensions given, 
and one side, E in Fig. 23, was scraped to a surface-plate. A squaring 
chip was then taken across a lathe face-plate and the plate was rigged 
with blocks and parallels as in Fig. 23. The surface E of the parallel 
B was also scraped to a surface-plate. When the large hole was bored, 
the block A. Fig. 23, was against parallel C, and when the small hole, 
or eccentric hole, was bored, A was moved along parallel B and block 
D was inserted. Tissue paper was used in both settings to insure 
actual contact. The large holes were bored 0.015 inch larger than the 
finished diameter of the crankshaft ends. After boring the small holes, 
a 1-inch arbor was forced into the small holes and the 60-degree center 
holes were turned with a lathe tool. The truth of these 60-degree holes 
was tested by means of a ground cone point and red lead. A tapped 
hole and setscrew completed each block. 

The shaft was now prepared for the blocks by grinding each end 
a wringing fit for its block. Before doing this, the center holes in the 
shaft were tested and scraped to a 60-degree cone point, to insure a 
perfectly round shaft when ground. 

The next operation was to correctly locate the keyway. For this, 
two blocks, A and B, Fig. 21, were made. A is a 1-inch block that 
tapped lightly into the keyway and projected a short distance, as shown. 
B is a block planed to micrometer gage, and of such a height as to 
bring the center line of the keyway and the center line of the crank- 
shaft into a plane parallel to the planer surface C, Fig. 21. The proper 
height of B was easily found by means of micrometer measurements 
and deductions. Having made A and B, Fig. 21, the whole job was 
taken to a newly-planed planer table and the end blocks were placed 
on the crankshaft. A was then placed in the keyway and the crank- 
shaft turned until A rested on B. With tissue paper under the end 
blocks D, Fig. 21, and between A and B, adjustments were made until 
all the papers held fast. The blocks D were then made secure by means 

* MACHINERY, March. 1907. 



o o 



00 00 










of the setscrews E. After a final test with the tissue papers, the crank- 
shaft was ready to have the eccentric ground. This was done on an 
18-inch by 96-inch Norton plain grinder. The fillets on the eccentric 
were also ground at the same time. 

Fig. 22. Fixture for Grinding Ore 

The length of throw was tested in the grinder by means of a Bath 
indicator and a 1-inch B. & S. disk, and found to be within the required 
limits. When the eccentric was completed, the end blocks were re- 

Fig. 23. Method of Boring the Fixture used for Grinding the Crankshaft 

moved and the remainder of the crankshaft was ground on its own 

Grinding Kinks* 

In the following are described some of the kinks used by toolmakers 
in grinding; these kinks were contributed by Paul W. Abbott. 

* MACHINERY, December, 1908. 




Fig. 24 is a hand grinding rest which is very handy for use on the 
universal grinder. It is adjustable up and down for height, and is 
used for hand grinding circular and straight form tools, sharpening 
metal slotting saws, formed cutters, etc. Fig. 26 shows the application 
of the hand rest to the grinding of saw teeth in a blank. The tooth 
rest used in connection with this operation is shown in Fig. 25. These 
saws are first ground on an arbor, the old teeth being ground off, 
leaving a perfect circle. The operator then puts on this device, set- 
ting the tooth rest so that the teeth will be about *4 inch apart, and 
grinds around by hand, not quite bringing each tooth to a sharp point. 
On the last nine or ten teeth he evens up any inaccuracy in the 
spacing, the wheel being trued off to the exact shape of tooth space 

Fig. 27 shows a device for accurately sharpening formed cutters up 
to 3 inches diameter, which is used when the cutter grinder has 
another job in it, or could be used to advantage where there was no 
surface or cutter grinder. The device consists of the cast-iron slide 
B, at the end of which is a tapped hole C, with a small fillister head 
screw which holds the various sizes of bushings which fit the holes 
in the cutters. On the same end is the index pin D, which is adjust- 
able back and forth. In operation, the hand rest shown in Fig. 24 is 
also used, and the pins A are lined up parallel with the forward travel 
of the wheel, and so that the cutting face of the wheel is on a line 
with the center of the bushing. The cutter is then slipped on over 
the bushing and the index pin is set so that the required amount will 
be ground from the face of the tooth. The operator brings the wheel 
up to the proper position and then pushes the slide forward until the 
wheel has reached the bottom of the tooth space; he then withdraws 
the slide and indexes to the next tooth, and so on, tooth after tooth. 
It will be noticed that the index pin rests against the back of the tooth, 
which means that upon the previous milling of the teeth depends the 
accuracy of the grinding; but on the standard cutters furnished by 
numerous concerns this spacing will be found accurate enough. 

Fig. 28 is a center for the head-stock for holding small forming 
tools of odd size, or threaded pieces which are to be ground on the 
periphery. The tools are simply clamped to the face of the center, 
and trued up by an indicator. Fig. 29 is a device for the tool grinder 
for grinding snap gages, where there is no surface grinder for this 
class of work. The shank of this device is made to fit the head-stock, 
and the gages are clamped to it by a small strap and two screws. 
This fixture revolves while in use, and the jaws of the gage are ground 
by feeding a thin wheel in and out by hand. Revolving the device 
insures perfectly straight gage faces. Fig. 30 shows a center for 
the universal grinder for holding a standard line of large end milling 
cutters with threaded holes, while sharpening. The head-stock is 
swung around at right angles to the ways, and with a long support 
for the tooth rest (Fig. 31), which is bolted to the platen, the cutters 
are ground very handily by throwing in the feed and grinding one 
tooth, and then, before the wheel comes back, indexing to the next 
tooth, and so on. 




Fig. 32 shows a hardened roller which is ground all over, and Fig. 33 
the fixture for the universal grinder for grinding the sides of this 
roll. This plate was made of cast iron, with both sides ground and 
with each hole ground to 0.0005 inch over standard size. Each hole 
has a *4-mch set-screw, as shown at A. In operation, the plate is 
fastened to the face-plate by a draw-back rod, and the head-stock is 
swung around at right angles. As the plate revolves, 16 rolls are 
ground at once, first on one side, and then the plate is turned and 
the other side ground, the rolls being made to standard length by 
using a depth gage. The hardened roll shown in Fig. 34, which is 
used on swaging machines, is held by the centers shown in Fig. 35 
antf 36, when being ground. Fig. 35 is the head-stock center cupped 
out on the end to fit the beveled end of the roll. This center drives 
the roll by friction, the pressure being obtained by the spring tail- 
stock. Fig. 36 is the tail-center, which is in two parts, the inner 
spindle running with the roll and being adjusted by the screw in the 
end so that the thrust is taken by the ball B, the tapered portions 
C just clearing each other. Other methods of grinding rolls are shown 
in Figs. 37 to 41. One example of grinding is shown in Fig. 37, and 
its center in Fig. 39. The roll is driven by a pin on the center, which 
engages with a corresponding hole in the work. A better method is 
to center me roll and then in one end drive a square 60-degree punch, 
using the square center shown in Fig. 41 for driving the work while 
grinding. Another good method for hollow rolls, such as shown in 
Fig. 38, is to use a lo-degree square center, such as shown in Fig. 40, 
the end of which just enters the hole. 

Figs. 42 and 43 show two end mills. The smaller one is fastened 
inside of the larger when in use, and when in position rests against 
the bottom of the hole and projects outside a definite distance. The 
length D is standard in all these mills. Fig. 44 shows the fixture for 
grinding two pairs of these mills at a time, so that the same amount 
will be taken off of both the short and long ones. Threaded bushings 
E fit the larger size mills, and F, the smaller. The collars O are of 
such thickness that the cutting face of the smaller mill is brought 
into the same plane as the larger, and so when grinding an equal 
amount is removed from the face of each mill. The plate is held to 
the face-plate by a draw-back rod. The head-stock is swung at right 
angles, and with the fixture revolving, the wheel traverses back and 
forth across the faces of the mills. The mills are then taken to a 
cutter grinder and backed off. 

Fig. 45 shows a small crankshaft, and Fig. 46 the fixture for grind- 
ing the pin. The bearings are first ground on centers in the usual 
way. The fixture is of cast iron and is held to the face-plate by screws 
and dowel pins. In the making of this fixture the hole H was ground 
out to the size of the bearing, and then the fixture was correctly located 
and doweled to the regular face-plate. The crank,, while being ground, 
is held by the set-screws J and the screws K f which are set against 
the crank on either side. 

The grinding of formed cutters, similar to the one shown in Fig. 47, 

40 No. 38 GRINDING 

so that they will be interchangeable, is very interesting. The error 
limit is 0.00025 inch. The grinder used is a Norton universal tool 
and cutter grinder. After hardening, the cutters are first ground to a 
definite thickness. For this operation they are held against the face- 
plate by a draw-back chuck. TTie next operation is grinding the bev- 
eled sides, which is accomplished by holding the cutters against a 
small face-plate by a draw-back chuck. The correct angle of bevel is 
obtained with the protractor, and to get the correct diameter of the 
bevel sides, and to insure that the bevel sides stand exactly in the 
same relation to each other, the gage shown in Fig. 48 is used. This 
gage is hardened and ground all over, and the two gaging points L are 
set a predetermined distance apart and as near the same height from 
the platen as mechanical means can make them. It is obvious that 
cutters which are all ground the same thickness, and which will pass 
through this gage with the beveled sides both touching the gage points 
with equal pressure, will interchange within pretty close limits. The 
operator grinds one bevel side at a time, trying the work every little 
while in this gage; when one side passes through the gage the cutter 
is turned around and the other bevel ground. For grinding the radius 
on the periphery and bringing the cutter to the correct diameter, the 
radius grinding fixture shown in Fig. 49 is used. The dovetailed base 
M is fitted to the platen of the grinder, and upon this base is a sliding 
base N which is pivoted to M by a bolt 0. Upon the base N there is 
an auxiliary platen P which can be adjusted back and forth by the 
screw Q for getting the proper radius. This auxiliary platen is made 
the same as the machine platen so that the regular head- and tail- 
stocks will go on it. A cutter is placed on a special arbor and the 
platen P adjusted to give the correct radius. The wheel is then brought 
up and the cutter is ground to the correct diameter, the curved face 
being obtained by swinging the base N back and forth by hand in an 
arc of a circle, with bolt as a center. 

Another ingenious scheme is shown in Fig. 51. Three or four pieces 
similar to the one shown in Fig. 50 were to have the holes ground out. 
"With an independent 4-jawed chuck this would have been easy, but 
there was no such chuck; and as there would never be any more of 
these pieces to be ground the fixture for doing the work had to be 
inexpensive. The face-plate could not be used, as the pieces were 
smaller than the hole in the face-plate. The operator thought awhile, 
and then hunted around a few minutes and found a large washer 7?, 
tapped two holes in it, filed up the sheet steel strap 8, and with a 
couple of machine screws was ready to begin. The washer was first 
put in the universal chuck and the outer side ground. One of the 
pieces was then clamped in place, and after putting on the internal 
grinding attachment it was ready to be ground. 

Selection of Wheel* 

The following little hints regarding grinding, taken from a booklet^ 
issued by the Norton Co., Worcester, Mass., will prove of value to 
all who have to do with grinding machines and grinding. 

* MACHINERY, August, 1908. 


Don't believe that all materials can be ground equally well with one 
and the same wheel. 

Get the proper wheel for the work. 

You would not expect to turn all kinds of lathe work with one tool 
having only one form of cutting edge. The grinding wheel is a tool 
for cutting. 

Different shapes of work, different kinds of metal, require different 
cutting edges when grinding as well as when turning. Different grades 
and grains of wheels are required for different kinds of work. 

Grinding wheels are numbered from coarse to fine, and graded from 
soft to hard. The grade is denoted by the letters of the alphabet from 
E to Z. 

Don't decide on the wheel without knowing the work. 

Spindle speed and character of the material, shape of work to be 
ground, and surface of wheel in contact are prime factors. 

In cylindrical grinding, speed of work, diameter of work and depth 
of cut must all be reckoned with in the selection of the right com- 
bination of grain and grade. 

The condition of the machine affects the efficiency of the wheel. 
Heavy machines with large wheel spindles and massive wheel support 
call for a wheel different from those for lighter machines with smaller 

Don't order a certain grade of wheel merely because that grade is 
used on similar work in another plant. 

Don't use a hard wheel to economize it is production you are after. 

A hard wheel is more likely to change the temperature of the work 
or to become glazed than a soft one; furthermore, it requires more 
power to do the same amount of worK, 

It is a common error to assume that a wheel for grinding steel and 
cast iron, chilled iron and hardened steel must be as fine as the 
surface desired. A coarse wheel will produce a fine finish if the proper 
relations between grade, depth of cut, speed of work, speed of wheel, 
etc., are observed. 

When grinding brass and the softer bronzes, the wheel must be as 
fine as the finish required. Bronzes with "manganese" or "phosphor" 
permit the use of coarser wheels. 

Don't get a wheel made for soft steel for use on hard steel. 

For a fine finish on hard stock, a coarse wheel may be necessary, 
and the harder the stock, the coarser the wheel. 

When ordering wheels, don't forget the diameter, width, style of 
face, arbor holes, description of work, speed of spindle, and the number 
and letter denoting the combination of grain and grade, if known. 

The width of the wheel should be in proportion to the amount of 
the material to be removed with each revolution of the work. 

If you reduce the width of the wheel you must use a finer feed, and 
consequently do less work. 


Never mount wheels without flanges. 

Flanges should be at least one-third the diameter of the wheel: one- 

42 No. 38 GRINDING 

half is recommended. Flanges should be concave never straight or 

Use fiber or rubber wasliers a trifle larger than the diameter of 
the flanges, or flanges with soft metal facings. 

Hooded machines are desirable when practicable. 


Don't start work on a new wheel until you are sure it runs true. 

Always have a wheel dresser handy for truing wheels for off-hand 

Never use a dresser on wheels that grind circular work on centers. 

For truing wheels used on plain cylindrical and universal grinding 
machines, cutter and reamer grinders, etc., the diamond is recommended. 
To obtain the best results it is absolutely necessary. 

Never attempt to true a wheel for circular grinding unless the dia- 
mond is held in a rigid toolpost on the table of the machine. You 
cannot do good work with such a wheel when it is trued "by hand." 

To get a truly ground surface you must keep the face of the wheel 


Don't start grinding until you know the speed is right not "near 
enough," but right. 

Even a slight variation in speed may be the cause of success or 
failure of any wheel. 

Failure is sometimes turned into success by merely changing the 
speed of either the wheel or work. 

Speed up the spindle as the diameter of the wheel is decreased. 
Approximately the same peripheral rate should be maintained as the 
wheel wears down. 

Complaint is sometimes made that wheels appear to be softer toward 
the center. Usually this is because the same surface rate of .speed is 
not maintained as the wheel is reduced in diameter. This causes the 
wheel to wear away faster and appear softer. It is also true that while 
the grade of the wheel may be uniform throughout, yet the smaller 
line of contact due to the smaller diameter will cause the wheel to 
appear softer. 

Increasing the speed of a grinding wheel gives the effect of a harder 
wheel; decreasing the speed gives the effect of a softer wheel. 

For surface grinding it is customary to run wheels at a somewhat 
slower rate of speed than for general grinding. A speed, of 4,000 to 
5,000 surface feet is usually employed. 

Wheels are run in actual practice from 4,000 to 6,000 feet per minute. 

General Suggestions 

Transferring a wheel worn down to a small diameter from a large 
machine to a small one is good practice. 

Keep the tickets or tags which are sent on the wheels in a record 
book, so that if a wheel is not satisfactory, reference can be made 


to order number when making complaint. It is equally valuable as 
a reference when ordering duplicate wheels. 

Don't use the wrong wheel on a job because it will require a few 
minutes' time to change wheels. A stop-watch will prove to you that 
changing wheels is cheaper. 

There is seldom a case where one and the same wheel can be used 
on all work without a greater loss of time than the change of wheel 
would involve. Many times the time saved in grinding a single piece 
more than pays for changing the wheel. 

Considerable difference in diameters of work will affect the cutting 
quality of a wheel on any given material. 

A successful wheel on the small diameters may work much slower 
on the larger diameters. 

The wheel most suitable for work of very large diameter may wear 
away too fast on work of smaller diameter. 

A suitable wheel for small diameters may cause chatter on pieces 
of large diameters. 

Don't grind circular work dry. 

A good wheel will grind in water, soda water or oil. 

Water keeps the wheel working cool and increases grinding produc- 

Soda water keeps the work and the machine from rusting. 

Oil in soda water increases the wheel's effectiveness. 

The particles from a grinding wheel do not adhere to steel. Don't 
let any one convince you to the contrary. 

Grinding is profitable for removing stock as well as for finishing. 

Keep the face of the wheel true and parallel with axis of spindle. 

Vibration makes grinding wheels wear. 

Keep all rests adjusted close to the wheel, otherwise work is liable 
to be caught and injury result. 

Keep boxes well oiled and adjusted. 

When practicable, indicate on each machine the revolutions of spindle 
and size of wheel to be run upon it. 

Don't disregard the setting up instructions that go with the grinding 



To figure, with any degree of accuracy, the cost of commercial wet 
grinding, requires considerable experience in the use and management 
of the machine, in order to be as closely approximated as lathe work. 
There seems to be a great difference in operators, due partly, no doubt, 
to the fact that the grinder has not yet become as generally used as 
other standard machine tools. A great many operators seem to be 
afraid to push their machines, and spend a good deal of time in useless 
calipering. They seem to forget that if they have several thousandths 
of an inch to take off a piece and are feeding in one or two thousandths 
of an inch at each reversal of the machine, they need not caliper until 
within one or two thousandths of an inch of size. Another class seems 
to think that because grinding is a finishing job, it must be nursed. 

As a matter of fact, there is no machine which so rapidly and accu- 
rately responds to the touches of an operator as the wet grinding 
machine. Of course, there are delicate pieces and certain shapes which 

MicHinery, K. Y. 
Fig. 52. Plain Cylindrical Piece to be Ground 

have to be carefully handled, but the usual run of work is so simple 
that any good apprentice can be put on it and taught in a short time. 

The work usually comes from the lathe with approximately 1/64 
to 1/32 inch stock to be removed. The work is then completed by a 
few reversals of the grinding machine with a feed nearly the full 
width of the wheel, and a cut of two to four thousandths of an inch 
until nearly up to size, and then a much slower traverse per revolution 
for finishing, according to the kind of finish desired. To obtain the 
best speed, the limits required on the lathe must not be made too 
narrow, from 1/64 to 1/32 inch being admissible for ordinary work, and 
more on large work; for the facility of the grinder in finishing work 
is far in excess of the lathe, and the latter must be relieved of all the 
finishing possible. 

To figure the actual time for removing stock on the grinder, we 
must take into account the longitudinal traverse of the wheel for 
each revolution of the work, the surface speed of the work and the 
depth of the cut. The latter must be varied according to the nature of 
the material, greater or less according to whether it is hard or soft; 
and the traverse per revolution of work is lessened if a fine finish is 
desired. The shape of the piece also somewhat affects both of these 

* MACHINERY, October, 1906. 



-r 1 *;* 

points, as long, thin pieces require a slower tra- 
verse and lighter cuts. 

Take, for instance, the plain piece, Fig. 52; 
material, hardened steel. For this a work sur- 
face speed of 15 feet, or about 37 revolutions per 
minute would be suitable. Assuming we have 
a wheel 18 inches in diameter and iy 2 in ch face, 
a traverse of two-thirds the face of the wheel or 
one inch per revolution of work is usual. This 
would require 12 revolutions to pass the length of 
the piece, plus 1 revolution for clearance, or for 
dwell if there happens to be a shoulder. This 
would make, roughly, three reversals a minute. 

On a medium-sized machine, an automatic feed 
equivalent to a work reduction of about 0.002 
inch would be suitable, or a reduction of about 
0.006 inch per minute. If the work came with 
an average allowance of 0.030 inch for grinding, 
it would require theoretically 5 minutes' actual 
grinding time to rough this piece down. To this 
must be added the time for handling the work, 
adjusting the machine and back-rests (in this 
case only one rest would be used), calipering the 
work and finishing. This time will amount to as 
much as the grinding time with most operators 
(most of it being taken up in finishing), which 
would make the actual time about ten minutes 
apiece. As a matter of fact, work of this size is 
actually being ground at the rate of seven or eight 
pieces per hour. 

If a fine finish were desired, a higher work 
speed and slow r er traverse would be required. For 
a very fine finish a work speed of 45 feet surface 
speed and traverse of 1/6 inch per revolution would 
be suitable for finishing, with, of course, a very 
much smaller feed. This change in the work and 
traverse speed could be made when the work is 
nearly up to size, and would probably require 
about three minutes. If the piece were of soft 
steel, a deeper cut could be taken and a wider 
traverse, a cut of 0.003 inch and a traverse nearly 
up to the width of the wheel being admissible. In 
grinding long shafts it is necessary to allow pro- 
portionately more time for adjusting back rests and 
for calipering, to insure that the piece be straight. 
This often takes twice the actual grinding time. 

Now let us look at the more complicated piece, 
Fig. o^. This will have to be done on a larger machine, and 
machines are slower to handle. This piece is a piston rod of 

) t 


: 1 












46 No. 38 GRINDING 

steel. We will use for this a 20-inch wheel of 2% inches face. A suit- 
able traverse for this wouM be 2 inches per revolution and a surface 
speed of 15 feet would make about 19 revolutions for the part 3 inches 
in diameter, and about 15 for the part 3% inches. The figures would 
be about as follows: 

Total amount to be removed, 0.060 inch; amount per reversal, 0.004 
inch; number of reversals required, 15. 
3 inches diameter, to cross once, 1 1/5 minute; total for 

15 reversals 18 minutes 

3% inches diameter, to cross once, 1% minute; total for 

15 reversals 22 minutes 

Tapers, both, to cross once, 2/5 minute; total for 15 re- 
versals 6 minutes 

Setting up and adjusting 10 minutes 

Total 56 minutes 

If it be desired to put a radius on the wheel and grind the fillets 
at shoulder A, about 10 minutes more shouM be allowed; and if there 
were more than one piece to be done, considerable time could be saved 
in setting for the tapers. 



In 1902 some important tests of the strength of emery wheels were 
undertaken at the Case School of Applied Science, Cleveland, Ohio, 
under the direction of Prof. H. Benjamin. Fifteen wheels of various 
makes were tested to destruction. The results of these tests are 
given in the following. 

Most manufacturers of this class of wheels test them for their own 
information, but the results are not generally given to the public. 
At the Norton Emery Wheel Works all wheels are tested before 
leaving the shop at a speed double that allowed in regular service, 
and occasionally wheels are burst to determine the actual factor of 

Emery-wheel accidents are not uncommon, but can usually be traced 
to the carelessness of the operator. One common cause of failure is 
allowing a small piece of work to slip or roll between the wheel and 
the rest. 

The wheels selected for the experiments were all of the same size, 
being sixteen inches in diameter by one inch thick, and having a 
hole one and one-half inch in diameter. The object of the experiment 
being to determine the bursting speed of such wheels as are actually 
on the market, emery wheels were obtained through various outside 

* MACHINERY, July, 1903. 


parties without indicating to the agents or manufacturers the use to 
be made of them. In this way wheels of six different makes were 
obtained, the label on each wheel showing usually the maker's name, 
the grade number or letter, the quality of emery, and the speed 
recommended for use. As shown in Table IV, giving the results, the 
working speed varied in the different wheels from 1,150 to 1,400 revo- 
lutions per minute, the average being about 1,200 revolutions per min- 
ute. For a diameter of sixteen inches this corresponds to a peripheral 
velocity of about 5,000 feet per minute. The table also shows that 
the fineness of the emery varied from ten to sixty, the average being 
about thirty. 

The wheels were held between two collars, each six and one-eighth 
inches in diameter and concaved, so as to bear only on a ring three- 
fourths of an inch wide at the outer circumference. 

Fig. 54. Various Ways in which Emery WTieels Burst 

Table IV shows the results of the experiments in detail, and needs 
but little explanation. The illustrations in Fig. 54 show character- 
istic fractures, and the appearance of various wheels after bursting. 
Wheels numbered 1, 2, and 3 in the table were of one make, and showed 
a remarkable uniformity in strength. Nos. 4, 5, 8, and 9 were all 
made by one firm; the two latter wheels were of finer grain than the 
others, and showed a correspondingly greater strength. Nos. 6 and 7 
contained a layer of brass wire netting imbedded in the emery, and 
were about one-third stronger than the average of the ordinary wheels. 
The wheels numbered 10 and 11 were the weakest among those tested, 
but had an apparent factor of safety of between five and six. Nos. 12 
and 13, of still another make, burst at about the average speed. Wheels 
Nos. 14 and 15 were so-called vulcanized wheels, containing rubber in 
the bond, and intended for particularly severe service. These showed, 
as was expected, rather more than the average strength. 


An examination of the last two columns in the table shows that the 
wheels burst at speeds varying from two and one-quarter to three 
and three-quarters of the working speed, and accordingly had factors 
of safety varying from five to thirteen. 

It is, then, apparent that any of these wheels were safe at the speed 
recommended, and would not have burst under ordinary conditions. 
At the same time, considering the violent nature of the service and 
the shocks to which they are exposed, it would seem that the factor 


>Jo of 


No. of 





Revs, per 

Feet per 

Revs per 

Feet per 


4 5 


























19i 200 








































Tests 6 and 7; wheels made with wire netting; tests 14 and 15, with 
vulcanized rubber. 

of safety for emery wheels should be large. In comparison with those 
generally used in machines, a factor of eight or ten would seem small 
enough. It may also be said that such a variation in strength between 
wheels of the same make and grade, as for instance, that between 
Nos. 4 and 5, indicates a lack of uniformity which causes distrust. 
The fractures were in the main radial, as may be seen from Fig. 54, 
the wheel splitting in three, four or five sectors as might chance. It 
may be assumed that these radial cracks started from the rim, where 
the velocity and stress were greatest, but it is a fact worthy of notice 
that in nearly every instance the cracks radiated from points where 
the lead bushing projected into the body of the wheel. 




JA* 31 T346 

!|Jnu& 4 */ 1 

D * 


7 .\\ AM 7 9 

|\!Ar U lAJ 


LD 21-100m-7,'40 (6936s) 

YC 53944 



No. 13. Boilers and Chimneys. 


nl' B 

H < 'him 

No. 14. Locomotive and Railway Data. 

- 1 , 


No. 15. Steam and Gas Engines. Sat- 
urated Steam; H ; . Steam 


. i ; H o r s e p 
No. 16. Mathematical Tables. Squares 

t ioi 

Soli for Solv- 


No. 17. Mechanics and Strength of Ma- 
terials.- -Work; < Vntri 


Mat P] 

No. 18. Beam Formulas and Structural 
Design. ormulas; 

uli of Si ruct urn 1 Shapes . 


.Sparing; S|. 

No. 19. Belt, Rope and Chain Drives. 

I Mi: 

Tahles i Mrivin.u 

No. 20. Wiring- Diagrams, Keating' and 
Ventilation, and Miscellaneous Tables. 

Typical Motor Wiring- : 

i'peY Wir. 

nts, Metrie < 'onve 

Metals, Draft in.u 


journal, originator of the Reference and 

editions the Shop Edition, $1.00 a year; 

ih<> Ifnifiri-iif Edition, $2.00 a year, and the 

No. 4. Reamers, Sockets, Drills and 
Milling- Cutters. Hand K-um-rs: Sh.-ll 


Mills and Angular 

No. 5. Spur aearing 1 .- ! and 

Circi 'ns of Spui' Ge 

Tiii.i. mtograph 

Tables; Rolling Mill Gearing.; Strengtl 
Spur Gears; Horsepower Transmitted 
Cast-iron and Rawhide Pinions; Design of 
Spur Gears: Weight of Cast-iron, 
lie Gearing. 

No. 6. Bevel, Spiral and Worm Gear- 
ing 1 . Rules and >vel 
Gears; Strength of Jievel G 
of L: -s; Kules and Formulas 
Spiral Gearing; Tal 

for Cutters for Spiral 
! Worm 

No. 7. Shafting 1 , Keys and Keyways. 

Hoi- ms and 

: WoodruJ 

l>Upil'X .1 

No. 8. Bearing's, Coupling's, Clutches, 
Crane Chain and Hooks. Pillow 


No. 9. Spring's, Slides and Machine 

Details. Formula: 



No! 10. Motor Drive, Speeds and Feeds, 
Taper Turning, and Boring Bars. !' 

Tools: Cutting 

per., | 

! Tools; ' ; 

No. 11. Milling 1 Machine Indexing-, 
Clamping 1 Devices and Planer Jacks. 

Tables for Millii ine Index 

Spirals; An 


No. 12. Pipe and Pipe Fitting's. Pipe 
Thr iron Fittings;. 

MACIIIXEUY, the monthly mechanical 
Data Sheet Series, is published in four 
the Engineering Edition, $2.00 
Forrifin Edit a year. 

The Industrial Press, Publishers of MACHINERY,' 
49-55 Lafayette Street, New York City, U. S