534

THE COLLIERY GUARDIAN.

March 15, 1918.

revolutions could be obtained over a period of from
one to two minutes. Figs. 4 and 5 show the arrange-
ment fitted to a Simplex indicator.

The apparatus consists of a steel tube that can be
. easily.fixed to the indicator by a clamp ; it carries at
its extreme end a pulley running on ball-bearings
driven at a constant speed by gearing and a governed
clockwork through a flexible shaft. The ordinary
reciprocating drum on the indicator is replaced by a
pulley running on ball bearings, and a belt of paper
6 ft. long runs on the two pulleys. Variable gearing is
provided, and there is a considerable range of speed in
the clockwork, the proper speed being adjustable
between 2 ft. and 10 ft. per minute.

A governor is provided, so that any predetermined
speed is constant. The pencil of the indicator marks
the paper as it passes over the drum, and an auxiliary or
signalling pencil also marks the paper by means of a
small magnet controlled by a push-button in the hand
of the operator, so that the instant of starting, stopping,
end of stroke of the piston, etc., can be given relatively
to the indicator record. Signals can also be given
autom itically from a contact-maker on the shaft of the
engine.

It will be seen that the paper moves under the pencil
continuously, and does not reciprocate as in the ordinary
indicator, and that the movement of the paper is pro-
portional to time and not to the volume swept by the
piston. The pencil line, therefore, does not return
on itself, but gives a series of waves or cycles, the
lengths of which, are inversely proportional to the speed
at which the crankshaft is revolving.

The mean pressure can be readily calculated, but the
card has to be specially divided so as to give readings
representing constant differences of volumes swept by
the cylinder. A very large number of diagrams have
been taken and compared with those taken by an
ordinary indicator, and a'though the indicator used is
a small one, very consistent lesults have been obtained.

The signalling pencil gives a continuous line along the
top of the record, which is useful for obtaining the
atmospheric line.

In indicating a winding engine, the paper is set to
travel its whole length in a few seconds longer than the
winding cycle, and the indicator and signal pencils are
set against the paper.

The operator watches the piston rod and presses the
button twice on the engine starting, at the same time
starting a stop watch. He. also presses the button at
the end of every stroke, and this is recorded on the
paper. This is important, as although the earlier strokes
can be easily deciphered while steam is on, the period
of running free and braking with the steam does not
give a definite indication on the diagram of the end of
the stroke. The signal marks the end of each stroke,
and shows where steam is against the engine. At the
end of the run, a signal is marked on the paper, the
stop watch hand is stopped, and the time taken is
recorded.

In working out the diagram, it is cut in two, giving a
paper 6 ft. long, and pinned to a wooden straight-edge.
The length of the paper between the signals “start”
and “stop” is measured, and when divided by the
number of seconds of the run this gives the length of
a division of paper repi esenting one second. The length
is stepped out on the paper, and the time taken by each
stroke can be obtained, as also the space of time from
the start to the beginning of each stroke.

After the mean pressures are calculated, they are
plotted out, and compared with the relative times,
as shown in fig. 6. The revolutions per minute are also
plotted in the same way.

From these curves the indicated horse-power at any
moment, the velocity of the cage and accelerations, both
positive and negative, can be calculated (fig. 7).

The weight of steam used can also be calculated in
the usual way from the diagrams, but this is, of course,
only an approximation, as the “ missing quantity,” or
the difference between the weight of steam shown on
the diagram and that actually used, has to be esti-
mated. By taking this “ missing quantity ” to be a
fourth of the total amount, or a third of that shown on
the diagram, a conservative estimate of the steam used
can be obtained, the maximum rate of flow being par-
ticularly interesting.

It does not take a draughtsman very long to obtain
the curves shown on figs. 6 and 7, but in order to get
the steam consumption, diagrams have to be taken from
the four cylinder ends and about sixty cards worked
out, entailing at least two days’ work.

This gear can a’so be used for finding the maximum
load and variations of speed of continuously-running
engines (fig. 8). It is readily digged up, and can be
attached to a winding-engine and a complete card taken
in a. quarter of an hour.

This apparatus is not claimed to be a novelty, as the
Mathot in-lira* or gives a somewhat similar diagram.
Similar results are given of some German tests in Messrs.
McCulloch and Filters’ book on Winding Engines and
Winding Appliances* but the apparatus is not described.
The apparatus described in this paper has been in
regular use since-1903.

Inking* Chronograph.

This instrument records short intervals of time, and
is very useful for taking the speed cycles of winding
engines, rolling mills, and lifts, and other apparatus
having variable speeds. It consists of a good centre-
seconds watch, having at the end of the seconds hand a
little cup full of ink. On pressing a knob, a spot of ink
is left on the dial, and a large number of consecutive
readings can be recorded at short intervals. It is par-
ticularly useful for plotting the acceleration curves of
electrical winding engines. It is a rare instrument, and
is used for astronomical work. That used by the writer
was made in 1877.

The Measurement of Compressed Air.

Although the electrical plant at a colliery is strictly
controlled by Home Office Regulations, and it is a
general practice to have ammeters, voltmeters, leakage
indicators and recorders, and circuit breakers to give

* 1912, page 245.

power measurements and to control the circuits, very
few compressed-air plants are fitted with anything more
than a pressure gauge, which is frequently of a type
that is liable to give inaccurate readings. This is all
the more remarkable as compressed air is probably of
greater importance to most collieries than electricity.
This statement holds good, particularly in South York-
shire, as regards inbye haulage, there being a wide-
spread feeling among mining engineers that electricity
should only be used underground under the shaft pillar
and in main intake airways.

It is probable that the amount of steam used for
compressing air largely exceeds that used for generating
electricity at the average colliery, even when the surface
plant and main haulages are electrically driven.

Great strides have been made in the last few years in
the simplification of means of measuring and recording
the flow of compressed air, and remarkable improve-
ments have been made in the accuracy of these
instruments. In fact, air-flow indicators and recorders
that will give accurate readings at full load within
1 per cent, of the mean can be obtained at a reasonable
cost.

Venturi and Orifice Measurements.

The Venturi meter is, perhaps,the most refined method
of making air and water measurements, the readings
being extremely accurate and the fall of pressure over
the instrument very low. The thin-edged orifice is also
a s’mple and accurate means of measurement, but has
the disadvantage that it requires a larger drop of
pressure than the Venturi tube, so that the loss of
pressure may be as great as 1 per cent, at full load. It
is not, therefore, so suitable for leaving in the pipe-
circuit for taking daily readings. It is, however, very
easily made, and the calculations for its construction

Fig. 8.—Mean Pressure Curve of Ordinary
Steam Engine.

An —									
	M E AN	PRESSURE					
40							
30							
2.0							
to							
							

□	to SCCOXOt ZO	JQ





4oa<

Rate of- Flow per 8-imch Meter-valve 0 +.

f.TeO

ft *

<0

£ m

Too

Cu8‘

/yffiT ' l/x? 7

2ooa

Fig. 9.—Results of Calibration Tests on
Sentinel Meter Valves.





and use are quite within the mathematical attainments
of the mining engineer. The flow of air is proportional
to the square roots of the fall of pressure across the
orifice, the absolute pressure of the air measured, and
the reciprocal of the absolute temperature.

As to take all these data into consideration would
involve a complication, it is quite possible for practical
purposes to read only the fall of pressure over the
orifice, a constant being used to include a standard
pressure and temperature. The scale of the instrument
can then be calibrated to give the weight of air or the
volume (either at working pressure or atmospheric
pressure) passing in unit time, the usual unit being the
number of cubic feet of free air compressed per minute.

Meter Valve.

The writer has fortunately been able to arrange with
Messrs. Beiliss and Morcom Limited, for the testing of
two meter valves, the readings being taken by com-
parison with standard instruments on their test plate.
These valves were made by Messrs. Alley, MacLellan
and Company Limited, and consist of well-made full-
bore valves, fitted with an accurate scale to show the
travel of the valve disc. A manometer is fitted, which
indicates against a fixed mark when there is a pre-
arranged fall of pressure across the valve. The valve
is closed until this float shows that this fall of pressure
is obtained, and the reading on the scale when compared
with a* chart gives the flow of air, corrected for both
pressure and temperature. The test readings of the
two 8-in. valves are shown in fig. 9, the result showing
a quite remarkable accuracy.

It is certain that the general use of some form of
air meter is of the greatest importance, both for proving
the condition and efficiency of the air compressors and
the leakage in the mains.

All these instruments are affected by pulsations in
the air supply, and should be placed in the outlet side
of ample receivers.

The intention of this paper is to call attention to
various means of maintaining the efficiency and safety
of colliery power plants, particularly with a view to the
reduction of coal and power consumption.

MAGNESITE AND MAGNESITE BRICKS.

A paper on this subject was read by Mr. W. Donald
(Glasgow) at the ordinary monthly meeting of the
Ceramic Society, on March 9, at Stoke-on-Trent.

The supplies of magnesite are at present obtained
from Greece and India. Other available supplies are to
be found in South Africa, Australia, Canada and
California, while in Europe deposits occur in Russia,
Norway, Sweden and Lapland, apart from the well-
known deposits in Austria and Hungary.

The refractoriness of magnesite bricks has relations
with the density, porosity, tensile strength, resistance
under load conditions at high temperatures, and
chemical composition. For the control of these factors
it is necessary to acquire full knowledge of the raw
magnesite both as regards mineral impurities and the
varied effects of different degrees of heat acting for
different periods and in oxidising or reducing atmo-
spheres, followed by slow or rapid coolb g. The effects
of eliminating certain impurities, and adding certain
others should also be known, and the effects of furnace
heats on the products. Of great importance, too, is
the detailed microscopic examination, which furnishes
information not otherwise obtainable, particularly as
to the course taken by alterations, the causes of which
can often be traced in this way.

After describing at some length the mode of occurrence
of the Austrian magnesite deposits, and the chief
characteristics of the materials and the bricks made
from them, the author proceeded to deal similarly with
Greek magnesite, which (like the magnesites from
India, Canada, etc.), has a finely crystalline structure
as contrasted with the massive crystalline structure of
the Austrian magnesite. The average analysis of
calcined Greek magnesite shows 90*67 per cent, mag-
nesia, 3*02 lime, 0*71 ferric oxide, 0*69 alumina, and 4*20
silica, but much of the rock is pure enough to give up to
95 per cent, magnesia on calcination. Chemically the
Greek magnesite is purer than the Austrian material,
but the iron compound which constitutes the bulk of
the impurity in the latter is very evenly distributed
throughout the mass, whereas the corresponding con-
stituent in Greek magnesite is very irregularly dis-
tributed, as are also the alumina and silica, though lime
runs uniformly through most of the raw magnesite.
Irregularity of distribution of impurities is even greater
in Canadian magnesite.

In sections of Greek magnesite bricks fired at 1,750
degs. Cent, and 1,850 degs. Cent, it was evident that the
face next the heat was quite fused, and half an inch from
that face the brick showed alterations of crystals from
the small to the large crystal, but no bad effect of the
heat was visible where the large crystals had been
formed. Sections of Austrian magnesite bricks fired at
the same temperatures showed no sign of being fired
at a high temperature.

In Austrian magnesite bricks the magma has inter-
penetrated each particle, and is distributed through the
whole mass of the brick. In Greek magnesite bricks no
such interpenetration of the particles by the magma has
occurred, but as the brick with each presentation of a
new face after spalling is converted bit by bit into the
larger crystal formation, the impurities present in the
magma often cause a frothing with great porosity. The
silicates of magnesia are bulky and highly viscous, and
having a melting point at 1,550 degs. Cent, they must be
still almost immovable as a molten liquid at 1,650 degs.
Cent. Hence it seems necessary to introduce ferrous
iron.

Dr. A. Scott has shown that whilst magnesia com-
pounds and lime compounds have melting points ranging
between 1,500 degs. Cent, and 1,550 degs. Cent., com-
pounds, including both the magnesia and the lime, melt
at at least 150 degs. Cent, lower. Hence lime should be
eliminated in every way possible, or its equal distri-
bution should be facilitated by giving the necessary
consistency to the magma.

On the whole, the result of using Greek magnesite
bricks of British make in basic open-hearth furnaces,
pig iron mixers, Talbot furnaces, and electric furnaces,
is satisfactory from the steelmaker’s point of view.
But difficulties are still experienced with regard to
spalling, and inability to withstand the corrosive action
of the basic slag, and the great strain that arises in the
roofs of electric furnaces.

Generally speaking, two types of magnesite brick are
made in Great Britain. One is made from finer material,
and the bricks weigh about 72-76 cwt. (sometimes to
80 cwt.) per thousand, the other being made from less
finely ground material, and the bricks weighing
82-86 cwt. (sometimes to 88 cwt.) per thousand. In
sections, the lighter bricks show finer particles and on
the whole a more complex colour scheme, whilst in the
heavier bricks there is greater light and shade between
particle and magma and between some particles and
others. Comparison with sections of calcined.magnesite
shows that a considerable percentage of the particles in
British-made bricks remains in the state of calcined
magnesite, and that the impurities surround rather than
interpenetrate them.

Indian and Greek magnesite calcined at 1,250 degs.
Cent., showed minute granular forms; Indian magnesite,
calcined at 1,750 degs. Cent, in an electric furnace,
showed alteration, which seemed greatest where impurity
was present. In Greek magnesite, fired to about
1,900 degs. Cent., the particles around the impurity
were completely altered, and the alteration spread in
all directions from such centres. Greek magnesite,
after prolonged calcination at a lower temperature in a
modern shaft kiln, showed the change in the mineral
particles practically throughout the mass. The mineral
crystals in these magnesites resembled those seen in
microscopic sections of Austrian magnesite bricks, and