196

THE COLLIERY GUARDIAN.

July 24 1914

alternating wave crosses zero potential. The opening
usually allowed in an oil switch is an inch for every
thousand volts; thus 5,500 volts require 5Jin., a mini-
mum break of 3 in. being allowed for 2,000 volts. In
order to make an oil switch safe, it is essential to keep
a good head of oil above the contacts. Any bubble of
hot gas which may rise from the breaking contacts must
travel some distance before it reaches the surface. In
the Ellison switch (fig. 3) it has to travel along a roof
of coal material before reaching the upper chamber,
and the bubble is thus cooled below the ignition tem-
perature of any explosive mixture which may exist in
the upper chamber.

The Ellison switch is fitted with a conical bung, which
is easily withdrawn by the inspector when he wishes
to see whether the oil level is correct.

Pure mineral oil should be used, having no trace of
animal or fatty oils, resin or wax, or any other sub-
stance which will oxidise when heated up in contact
with air. It should be perfectly dry, and not liable
to absorb moisture. Its dielectric strength must be
high. Wellsaline oil has a flash point of about 360 degs.
Fahr., and a fire test about 420 degs. Fahr. The oil
should be perfectly neutral, and have no tendency to
alkalinity. It should not carbonise or coke, throw off
a film or sediment, and should not freeze. The oil
should be supplied in steel barrels, because when put
into wooden casks it is liable to absorb moisture.
Viscosity is important, because it has been found that
thick oils are best for low tension big currents, and
thin oil for high tension small currents.

There is considerable difference of opinion as to
whether it is advisable to use oil for direct
current controllers. Experience shows much greater
carbonisation of the oil with direct current switches.
The fingers and controller segments appear to come into
contact better with alternating current, and it has been
thought that alternating current has a repelling action
on the oil. Engineers are reluctant to use oil switches
for direct current, because they fear the puncturing
effect of a sharp potential rise or surge. That depends,
of course, on the amount of self induction in the circuit.
For example, if the coil of a brake magnet is connected
to the circuit, then there is danger in opening the switch
contacts under oil. Mr. Mickie, of the Thermal

s /*</?

Fig. 3.—Section through Tank of Ellison
Oil-break Switch, showing Head of Oil,
Position of Contact below Slate and the
Filling Hole.

Syndicate Limited, of Wallsend, has investigated the
action of oils when acted upon by electric arcs, and as
might have been expected, has found that there is
44 cracking ” of the oils and that a light petrol explosive
vapour is formed. Clearly the -kind of oil required for
switches is one that does not “ crack ” easily.

It is most important that all the connections which go
below the surface of the oil should be solid, so as to
prevent syphoning. If cables are carried into the oil,
the oil will syphon back along the strands, as the
standard test of the Bureau of Mines (Washington,
U.S.A.) for permissible switches for use in gaseous mines
is to cover the contacts with oil to 25 per cent, of the
depth designated by the manufacturer. Then to open
and close the switch fifty times, using full load direct
current, the switch being, of course, in an explosive
atmosphere.

Details of Oil Switches.

Oil switches are usually made with the moving con-
tact to fall, and the advantage of this is that it does
away with springs. On the other hand there is a slight
disadvantage in that more oil space is required. If the
switch has a rising contact, the fixed contacts can be
fastened in the bottom of the tank, and thus the cable
connections can be made underneath as shown in fig. 4.
The crank handle used by the Union Electric Company
is worthy of mention, because, at the end of the travel,
the movement is slow and thus there is no chance of
shock to the porcelain insulators, which is a considerable
advantage, as will be seen from fig. 5. In the action
of switching on, the movement of the switch contacts
is at first slow, and gradually increases to a maximum,
which is attained when the hand lever has completed
an arc of 90 degs. From this point, the movement
gradually becomes slower as the movement of the hand
lever is continued, until the full-on position is reached.
During the action of switching on, the maximum lever-
age is obtained when the greatest power is required,

i.a., when pressing the contacts together. On the other
hand, when switching off, the maximum leverage is
obtained at the commencement of the movement.

Usually the contacts of oil break switches are V-
shaped, as shown in fig. 3; but some makers employ
parallel surfaces for the contacts. Such a contact is
shown in fig. 6. The blade is of hard rolled copper of
rectangular section, and is pressed between contact
blocks of the same material. These blocks are carried
by strips of copper, and are pressed together by flat
springs. The advantages claimed for this method, are
that the contacts rub together, and the moving contact
has to move about 1 in. before the final break is made.
It has thus been enabled to get up a certain amount of
speed and can break a larger amount of current with
safety. The final break takes place on two fingers,
which project a little longer than the rest. They are
alongside the main contact fingers, and thus no lateral
adjustment of the switch blade is required.

The time that a set of contacts will last depends as
much upon the kind of circuit controlled by the switch
as upon the number of times the switch is operated.
For example, when switching a circuit in which there
is a static transformer, there may be a heavy current
rush, due to the switch being closed at the peak of the

Fig. 4.—Oil-break Switch with Connection
to Bottom of Tank.

Fig. 5.—Diagram of Movement Rate of Switch

wave. These current rushes are more frequent than is
generally supposed; even a voltmeter transformer has
been known to take as much as 80 times its normal
current. Messrs. Reyrolle and Company Limited have
met the difficulty by designing a switch so that a small
carborundum resistance is placed across the switch con-
tacts immediately prior to the switch blade being pushed
home. This allows a small magnetising current to flow
to the primary of the transformer, and so gives the
magnetising current a gentle start.

Switch Boxes.

Cast iron is much used for boxes, although there is
now a tendency to use sheet steel. Ordinary cast iron
is porous, and allows oil and gas to pass through, but
this defect can be minimised by galvanising. Even
steel boxes have to be galvanised, in order to close the
riveted joints. Pressure is sometimes considerable.

Three-phase switches are sometimes specified to have
one tank per phase; but for medium-sized switches it is
quite as good practice to have one tank for the three
phases, provided that insulating barriers are fitted
between the phases. A single tank gives larger clear-
ance to the earthed metal and a larger volume of oil per
switch, and it is easier to handle a single tank.

Three-ply wood has been found the most satisfactory
material for lining oil tanks and for barriers, because it
can be made very thin without danger of warping. It
is good practice to place the three-ply lining away from
the wall of the tank, so as to provide a layer of quiescent
clean oil between the wood and the tank. This is also
an additional protection against the arc spreading to the
metal case.

There are three methods in use for making switch
boxes flame-proof. The first is to use wide and rough
machined flange joints; the second to use windows fitted
with sheets of gauze; and the third is known as ‘ ‘ plate
protection.” The first method is the one most favoured
in this country, and it seems to be efficient. Messrs.
Reyrolle and Company Limited and the Phoenix
Dynamo Company have repeatedly tested such joints in
explosive gases, and they have been satisfactory. The
flanges are usually l|in. wide, and they are rough-
planed, so that the tool marks cross each other. The
Union Electric Company turn the flanges and cut
grooves in them about in. deep and j-in. pitch. By
setting the box and the lid in different positions on the
face-plate of the latter, these grooves cross each other
approximately at right-angles, and thus a number of
pin holes are provided for the exit of the gases.

With regard to the second method, it is of interest
to note that at the Bureau of Mines (Washington) and
Messrs. H. H. Clarke and R. W. Crocker have carried
out an elaborate series of tests on electric switches for
use in gaseous mines.* They state that explosion-proof
windows can be made with enough heat-absorbing
material to be quite safe. They commend a design of

Fig. 6.—Contact of Knife
and Fingers.

Fig. 7.—Ellison Oil-break Triple-pole
Switch, fitted with Connection-box,
Screw - lowering Gear for Tank,
Ammeter and Volt-meter.

switch in which the relief opening area was 12-64 sq. in.,
as against unoccupied space within the casing of
0*8 cu. ft. Fine gauze of 70 mesh, reinforced by per-
forated metal sheets, appears to be an efficient device
for absorbing heat.

The third method (“ plate protection ”) is used in
Germany, but it is mainly employed for motors. It is
stronger than gauze, and so far as one can ascertain,
is satisfactory. The AEG Electric Company use it.

Switch boxes for carrying alternating currents at low
frequencies should have the iron frame of the switch
so divided as to reduce eddy currents; and the division,
if possible, should be between the phases. The con-
nections of the coils of the three phases should be so
arranged that if any eddy currents are set up, they will
neutralise one another.

Wing nuts make the best fastenings for the lids of
switch boxes, because a steel bar can be placed between
the wings in order to tighten them, and so a spanner
is unnecessary. The bolt of a wing nut is hinged, and
the end of the thread should be burred over, so that
the nut cannot come off and be mislaid.

Armour glands are important for switch boxes, etc.,
for mines, because of the regulation regarding electrical
continuity of the earthed part of the system.

Headroom has to be specially considered for oil
switches, because the switch parts are uncovered by the
action of lowering the tank, or by withdrawing the parts

*1913, Bulletin No. 68 of the Bureau of Mines.
Washington, U.S.A.