THE COLLIERY GUARDIAN

AND

JOURNAL OF THE COAL AND IRON TRADES

Vol. CXVI.

FRIDAY, DECEMBER 13, 1918.

No. 3024.

COOLING ELECTRIC MOTORS.’

By P. A.

The progress realised in dynamo design during the
last few years is gradually reducing the problem of a
further advance to the solution of a more and more-
complicated thermal problem. In the early days th?
size of dynamos and motors was limited by considera-
tion of sparking or regulation; in comparison with the
present machines, these were much larger, and it
seldom mattered whether the cooling-surface was well
arranged or not. The introduction of carbon brushes,
commutating poles, low-hysteresis iron, a better supply
of high grade insulating materials, high speed prime-
movers and machinery, and a better knowledge of
electro technique, has enabled designers to reduce the
size of electric machines to such an extent that the
natural cooling surface has now so shrunk that it can-
not radiate the heat generated in motors and dynamos
without the he-lp of artificial ventilation.

The electric machine is necessarily built of materials
which vary greatly in their heat conductive and resist-
ing properties. The usual and now available insulat-
ing materials, which form the vital parts of the
machine and are the essence of its reliability, do not
withstand a high temperature. Such substances as
micanite and oilcloth soften at about 200 degs. Fahr.
It follows, therefore, that the current carrying parts
should not exteed this ultimate temperature. On the
other hand, the temperature of the surrounding air
may be as high as 100 degs. Fahr. It thus means
that the safe permissible temperature rise of the insu-
lation should not exceed 90 degs. Fahr. This is
indeed a very small difference of temperature, and con-
tributes to the difficulty of the problem. The insulat-
ing materials used at present are bad conductors of
heat, and the windings from which the heat should
be dissipated as quickly as possible are lagged with the
very materials which render 'this process more difficult.
This is almost equivalent to the conditions that would
be prevalent if the steam engineer had nothing but
metals with which to lag the live-steam carrying parts.
The skill of the physicist and inventor should b?
applied to produce insulating materials capable of
withstanding a higher temperature rise. Pending
advance in that direction, the designer has to fall back
on three methods to improve the cooling and conse-
quently reduce the size of electric machines. He has
either (1) to drop artificially the temperature of th?
medium in which the machine has to run; (2) improve
the orthodox way of air ventilation; (3) develop new
methods and use cooling fluids having a higher specific
heat coefficient. Steps are being taken in this direc-
tion, especially in the case of large generators, where
the problem is one of dissipating hundreds of kilowatts
in one single unit, by introducing water or oil cooling.

For smaller machines, a closer study of air ventila-
tion should be undertaken. It is, however, important
to break down prejudice which exists in many users,
and even in designers’ minds concerning forced ven-
tilation. Such prejudice arises from an entire miscon-
ception regarding the reason for its use, and false
apprehension as to the proportion of power necessary
to drive air-cooling devices.

In the process of ventilation of an open machine
one may assume that the heat generated is carried off
by the natural diffusion of the molecules in contact
with the heated or heat-conducting parts of the
machine, the eddying motion due to the rotating move-
ment of the armature or parts, and, lastly, by metallic
conduction and radiation. These two last-named
causes account for very little. The effect of the arma-
ture rotation is to induce a constant cooling effect so
long as the peripheral velocity and temperature diffe-
rence between the air and the heated part is constant.
It is obvious that the problem is to allow a constant
supply of fresh air to come into contact with the rotat-
ing and other parts of the machine, so as to renew the
particles of air which have been heated and have been
removed from contact with these parts by the com-
bined forces of gravity and rotation.

In modern machines, in which the windings are con-
gested and complicated, vortex movements take place,
and some of the heated molecules of air are drawn in
again into contact with the armature; this tends to
diminish the cooling effect, and causes a higher rise
of temperature. In order to prevent this taking place,
many modern machines are provided with an indepen-
dent fan mounted on the armature shaft, by which
means a positive draught is created through the
machine. In passing through the machine, however,
the cooling air is heated and the difference of heat
potential between the air and the machine decreases,
so that one part of the machine has a tendency to be
warmer than the other. If the cooling effected by the
fan is not sufficient, the practice has been to put in
a bigger fan and so increase the air velocity. Other
designers have improved matters by retaining the same
size of fan, increasing the surface of contact by mak-
ing longitudinal holes in the armature, and providing
passages between the field coils, etc.

* From paper read before the Mining Institute of
Scotland.

MOSSAY.

Mr. B. G. Lamme, chief engineer of the American
Westinghouse Company, has given some very interest-
ing figures respecting the velocity of the air in large
machines. He mentions 5,000 to 6,000 ft. per minute
as being fairly common, and in some cases speeds of
10,000 ft. per minute, which corresponds to 125 miles
per hour. Such figures are abnormally high. In the
author’s opinion, these high air current velocities in
many cases betray a weakness in the design of the
machine. Air propelled at that speed has a tendency
to move as a solid mass. The same layers are brought
into contact with the surfaces to cool; the bulk of the
volume of air which passes through the machine is
ineffective by reason of the inherent bad heat conduc-
tivity of the air; and the power absorbed by the fan
is out of proportion with the ventilation obtained.

The author has found cases where the fan employed
to cool the machine was so injudicially constructed as
to create a vortex that kept the air contained in th?
machine practically in a closed circuit, with the result
that the machine would have probably run cooler with-
out a fan. How much can be sinned in that direction
can be shown by some experiments which the author
carried out that tend to show that the emissivity of
a heated surface rapidly reaches a maximum with an
air draught of 3,000 ft. per minute.

There is very little data available at present regard-
ing the power which a fan should absorb per unit of
heat dissipated under the conditions prevailing in

Fig. 1.—Section through Schroder Totally-enclosed Motor.

hlllHUIHHIIIII

EllHHWliiHttWWmiUHII'.ilfltWIUilffl?

12

12

,6

V8
13\

5

electrical machines. Some authors mention 1 kw. for
every 5 kw., whilst others mention 1 for every 8 kw.
In some enclosed machines the author has obtained
as much as 12 kw. per kw. power of the fan. These
figures, however, are misleading, because it is difficult
to judge under what conditions they were obtained.
The pressure necessary to send a sufficient quantity
of air through the windings varies considerably accord-
ing to the design of the machine. In some cases a
| in. head of water is sufficient to drive this air
through; in others it will require 3 in. of water.

With increasing losses in the machine, the power
necessary to carry off the heat increases enormously
if the apertures remain the same. In order to double
the quantity of air, it would require about eight times
the power previously used. It is in this direction that
great improvement can be expected. New shapes of
windings can be devised, and new ways of building the
essential mechanical parts, which would greatly en-
hance the efficiency of the cooling.

Unfortunately, the efficient fanning of electric
motors brings with it considerable drawbacks. In
most cases the fan ventilation is somewhat of an after-
thought, and the machine is transformed thereby into
a form of powerful vacuum cleaner. The air which is
forced into the machine, unless it be filtered before-
hand, carries particles of dust, which, by reason of
the tortuous passages—and in many cases unavoidable
change of velocity of the air currents—will be deposited
in the machine and gradually choke it, thereby reduc-
ing the cooling efficiency of the fan. In addition to
this defect, oil from the bearings, or oil carelessly
spilt by overfilling, will creep or be drawn in and
thrown by centrifugal force on to the windings, where
it saturates and perishes the insulation. Ninety per
cent, of the breakdowns in electric machines can be
ascribed to this cause, and this has in a certain
measure justified the prejudice which exists against
forced ventilation.

These are the reasons why the author became con-
vinced long ago that some day the problem of building
totally enclosed motors that would compete in price
with open machines would have to be faced. The
advent of the war has, if anything, precipitated this
conclusion. Totally enclosed motors have been
required in large quantities for munition factories,
chemical works, etc., and numerous inventions have
been made that tend to show that the future will see
great developments in this kind of motor.

If the problem of cooling open machines is in itself
complicated, that of cooling a totally enclosed machine
is even more difficult. In the open type machine most
of the devices deal with a particular way of forcing
cool air through the machine. In the case of totally
enclosed machines of the usual type, practically the
whole of the heat generated inside must pass from the
heated part to the contained air, from the air to the
inside of the case, from the inside of the case to the
outside, and thence to the outside air. In other words,
the heat has to pass through a layer of bad heat-con-
ducting material. By inserting a fan inside the
machine, however, a circulation of the internal air is
created, and this air will carry heat out, by convection,
to the outside casing.

It appears from experiments that the air inside
reaches a uniform temperature, due to the churning
effect of the fan. The total heat which can be radiated
from such a motor is proportional to the surface of
the motor, and it is obvious that this surface does
not increase in the same ratio as the cubic contents of
the motor, whereas the permissible output of an open •
type machine is nearly proportional to the total
volume.

From this it will be seen that the ratio of the price
of totally enclosed motors to the price of open type
motors of the same power and speed is a quantity
which grows rapidly with the size of the machine. In
fact, it may be said that the restriction of output by
totally enclosing a large machine in the orthodox way
is sufficiently severe to make this type of motor for
large power impractical, on account of unwieldy size,
weight, and initial cost. In addition, the efficiency
of a large totally enclosed motor of the orthodox type
will be much lower than the efficiency of an open type

machine of the same horse-power, and the difference
is specially marked at light loads.

Attempts have been made to overcome these objec-
tions, and a half-way solution has met with a con-
siderable amount of success. The pipe-ventilated motor
to which reference will now be made is not, however,
in reality an enclosed motor. The necessary trunking
is sufficient in a great many cases to prevent the use
of this type. The fresh air has to be conveyed to the
motor by means of pipes, and unless this air is filtered
beforehand, the motor may be affected adversely by
moisture or dirt being drawn into it. This method
would be prohibitive in coal mines and chemical fac-
tories, where there is a deleterious atmosphere, since,
when the motor is shut down a diffusion of this atmo-
sphere with the air in the motor will take place.

A good example of pipe ventilation' is that of the
British Thomson-Houston Company Limited.

Several methods of cooling totally enclosed motors
have been proposed, and might be classified under two
headings, namely : (a) totally enclosed motors having
a cooling device external to the motor; and (b) self-
contained cooling devices. In the first type the same
volume of air is passed repeatedly through the
machine, drawn in at one end and expelled at the
other, the circulation being completed between stag-
gered ribs formed in the box-shaped foundation plate.
In this manner the heat is transmitted to the surround-
ing foundations. The ribs may be hollow, and air or
water may flow through them. This Type of construc-
tion would appear to be efficient, but it involves con-
siderable extra cost and bulk, since the complicated
foundation plate becomes a necessary adjunct to the
motor.

A more successful means of carrying out the idea of
an external radiator has been lately put on the market
by Electrometers Limited, of Openshaw, Manchester.
It is in various forms. In the first form the radiator
consists of a box with end pieces designed to fit on
vertical trunk openings in the motor end covers, and
through this box a number of thin metal tubes pass in
a vertical direction. Inside the back end cover of
the motor a fan is fitted on the armature shaft, and
this causes a circulation of air through the machine
and radiator box. The heated air communicates the
heat from the machine to the walls of the tubes, which
are cooled by natural radiation.

Another form of this invention consists in construct-
ing the radiator box sides of thin sinuous metal, with