946

THE COLLIERY GUARDIAN.	May is, 1917.

CURRENT SCIENCE

Economy of Coke Gasification.

Mr. J. E. Enright (Gas Journal) reports that the
carburetted hydrogen plant installed at Tralee, for the
gasification of coke, has now been in operation for a
period of 10 months, with the following results:—Car-
buretted hydrogen gas made, 15,225,100 cu. ft.; coke
used, including boiler fuel and stand-by fuel, 363 tons
9 cwt.; tar used in plan^, 10,197 gals. ; wages,
£128 12s. 6d. ; working out at an average cost of 9d.
per 1,000 cu. ft. into the holder. During the same
period the quantity of coal gas made was 14,885,900
cu. ft., and the coal carbonised was 1,357 tons. The
working of the plant throughout has been most satis-
factory ; and, quite apart from the financial point of
view, the worry and anxiety that it has saved would
alone have been worth the capital outlay. There are
other important economies effected, such as in the
renewal, repair, and maintenance of retorts, the
abnormal wear and tear on the whole carbonising plant
in keeping up excessive heats, the number of stokers
required, and the many other expenses incidental to
the retort house. The operation of the plant is so
simple that one man can work it, and, when the
holders are full, he closes it down, and is then finished
until the next morning. Often he completes his work
as early as two o’clock in the afternoon.

With regard to the distribution and consumption of
the mixed gas, it has been found that the burner
adjustment for a 50 per cent, gas will suit for any
mixture up to all carburetted hydrogen gas. On two
different occasions last winter, carburetted hydrogen
gas was sent out exclusively, and the town was solely
depending on the carburetted hydrogen plant for its
supply of gas. During this period, an extra quantity
of tar was employed, together with some coal in the
generator. This was used with a view to increasing
the hydrogen and methane in the gas, and reducing
the carbonic oxide (this can be kept well below 18 per
cent.). By these means, they were able to make a gas
averaging upwards of 400 British thermal units. The
mixed gas varies between 475 and 500 British thermal
units; but a variation of anything between 400 and
500 British thermal units causes no trouble or com-
plaint whatever once the burners or stoves are adjusted
for the mixed gas; and Mr. Enright would have no
hesitation in sending out straight carburetted
hydrogen gas of 400 British thermal units, made from
coal and tar or a mixture of coal, coke, and tar, with
a carbonic oxide content of not more than 20 per cent.
With gas engine trials, it has been found that the
b.h.p. was not affected by the variation of the British
thermal units; but the consumption of gas worked out
practically in accordance with the British thermal
units—that is to say, the consumption of gas of 400
British thermal units would be 20 per cent, more than
with 500 British thermal units.

Coal Gas for Internal Combustion Engines.

The high price of petrol, and the increasing restric-
tions placed upon its use, have induced motor users
to turn again to the question of utilising ordinary coal
gas as a fuel for internal combustion engines. The
Commercial Motor, in a recent issue, describes and
illustrates a valve for the necessary reduction of pres-
sure when charging the engine. This valve is in use
in connection with the service cars on the Neath tram-
ways, where coal gas has been used on the system for
some years. The principle involved in the construc-
tion of the valve is that of a leather diaphragm, which
is inflated on pressure above the desired amount, and
deflated when that pressure is not reached; the action
of the diaphragm closes or opens respectively a valve
on the gas supply pipe. The gas first passes through
a filter on its way to a small conical valve, which opens
or closes the inlet to a chamber which, besides con-
taining the mechanism of the special device, serves as
a reservoir containing about J cu. ft. of gas. The exit
is by means of a branch pipe, coupled up to the engine
induction. The reservoir is covered by a leather
diaphragm stiffened in the centre by perforated brass
washers. The centre of the diaphragm is coupled by
means of links and a lever to the small valve already
mentioned. A light spring rests on the top of the
leather diaphragm, which is adjustable for strength
by a screw. The object of any regulation of the pres-
sure exerted by the spring is to make it possible to
maintain gas at varying pressures, and to obviate diffi-
culties which might arise owing to frictional resist-
ance of the mechanism. A flat spring held lightly
at one end by a bracket in the base of the chamber
tends always to close the valve. The excess of pres-
sure above that desired lifts the leather diaphragm,
closing the valve. So soon, however, as the pressure
in the chamber drops below that which it is desired
to maintain, the spring above the diaphragm will
depress the leather and open the valve.

Estimating the Heat Value of Similar Fuels.

In a paper recently delivered before the Society of
Chemical Analysts (Cheap Steam), Mr. Proctor Smith
lays great stress on the need for ascertaining the
calorific values of fuels so as to obtain figures which at
a glance enable the buyer or user to have a ready
means of classifying them, and to obtain a comparative
if not scientifically accurate knowledge of the amount
of heat which can be obtained from each fuel for a
fixed sum of money. He suggests as a basis for com-
parison the number of British thermal units which
can be obtained for Id. from each different class of
fuel.

Mr. Smith is careful to state that in this comparison
the different grades of fuel must be compared sepa-
rately, that is to say, a nut coal cannot be compared
with a slack, as the price is necessarily higher on the
former owing to the amount of work in screening, etc.
Taking, however, the coals in their different grades,
i.e., nuts, through-and-through, slack, smudge, etc.,
very valuable information may be obtained which will
constitute a trustworthy method of assembling the
fuels in the order of their relative values.

AND TECHNOLOGY.

Mr. Smith emphasises the usefulness of knowing the
mechanical condition of a fuel, and suggests that the
coal should be put through different sizes of sieves,
and by this means a faithful record of the condition
of the fuel can be kept. For instance, in an ordinary
steel works gas producer, the coal remaining on a
1| in. sieve would well serve the purpose, and that
which passes through a J in. sieve (fine slack) could be
weighed, and the percentage on and through the sieve
thus ascertained.

Mr. Smith includes in his paper an example on the
lines suggested by himself, of a report of a coal
analysis for the purpose of keeping a record : —Calcu-
lated calorific value, 12,380 British thermal units per
lb.; colour of ash, reddish-brown; coking test, good
strong coke; a good clean fuel, free from shale, yields
a long clear flame, particularly suitable for boiler use;
clinker well burnt, free from much unconsumed carbon
and showing very little signs of fusion; mechanical
condition very good.

Estimating Washable Products in Gaseous Mixtures.

Mr. N. P. Broadbridge (Gas World) proposes a sim-
plified method of determining the percentage of wash-
able products in gases, based on the principle that
when a hydroxylic solvent, such as ethyl-alcoho], con-
tains in solution hydrocarbons immiscible with water,
these latter are thrown out, when water is added, with
which the alcohol is miscible in all proportions. A test
constructed on this principle would be carried out in
the following manner : —Three absorption bottles are
arranged in series, the first two immersed in a freez-
ing mixture. The size of the bottles is such that
50 c.c. of alcohol will produce a seal of, approximately,
1 in. This amount is accordingly introduced into the
first tw’O, while a similar quantity of water is placed
in the last, to retain the small amount of solvent which
vaporises with the gas, the freezing mixture serving
to reduce this vaporisation to a minimum.

A convenient quantity of gas, determined by its con-
tent of washable products, is passed through the bottles
. at a rate of from 1 to 2 cu. ft. per hour. After this,
all three are drained out into a tall stoppered cylinder
of 400 c.c. capacity, and further rinsed with a few c.c.
of distilled water, which is added to the contents of
the cylinder. The volume is then made up to 400 c.c.
with water, and the contents of the cylinder allowed to
stand for half an hour or so, in order that the hydro-
carbons, insoluble in dilute alcohol, may rise to the
surface, where their volume is determined. Other
liquids may be used for the determination, providing
they embody the requisite requirements—viz., low
viscosity, moderate volatility, miscibility with water,
and solvent power for benzol. Rectified pyridine is
one such liquid, and the quality should be of the 90
per cent, at 140 degs. Cent, variety. It should be
previously tested by the usual method for oily matter
by dehydration, distillation, and the dilution of the
first few c.c. of distillate.

It should be noted that the extent of dilution of the
alcohol—viz., 1 to 4—should not be departed from, as
otherwise, if the mixture is too concentrated, benzene,
etc., would be held in solution ; or, if water is added in
excess of this ratio, the total mixture becomes incon-
veniently large, and presents occasion for minute
globules of hydrocarbons to become entangled, and
therefore remain in suspension. These precautions
apply more especially to the estimation of very small
quantities, as, for instance, in a washed or partly-
washed gas.

OiGlectrics io Electrostatic Fields.

The problem whether dielectrics will change their
dimensions when exposed to strong electrostatic
fields has received little experimental investigation.
One would expect some change, although not more
than a small change probably. Righi and Quincke
made some experiments 30 years ago, and Cantone and
Pozzani followed the matter up in 1900; but the results
were not concordant. Resuming these investigations
recently, L. Bouchet (Comptes Itendus de V Academic
des Sciences—abstracted by Engineering) first studied
the expansion in a direction normal to the electric
field. He placed cylinders of glass, paraffin, or
ebonite between two concentric cylinders of brass,
filling the annular space left with water; the system
formed a tubular condenser. On the top of the
cylinder of the dielectric material he placed a glass
plate, and on this a leiis; when the electric field was
excited by an influence machine (Voss) the cylinder
expanded in the direction normal to the electric field,
and pressed the glass plate against the lens, causing
displacement of the interference rings. The length
of the cylinders was about 40 cm., the diameter 1-5 cm.,
the thickness up to 2 mm. The expected dilatations
were observed, but they amounted to 1 part in 1010 or
less. The expansion was proportional to 1/E, how-
ever, E being the Young modulus of elasticity, an
interesting relation. On the other hand, the numerical
results were not what they should have been for perfect
dielectrics, and glass and ebonite (half-hard), which
disagreed most, are, indeed, far from being considered
perfect dielectrics. There seemed to be no direct-
external electrostatic pressure effect on the length of
the insulator. In the second series of experiments,
Bouchet studied the effect of the electric field in the
direction of the electric lines of force. The electro-
static pressure effect would probably be more marked
in this case, and there might be superposed either a
contraction or a dilatation of the dielectric owing to
the internal stress. The materials chosen for these
experiments were pure Para rubber, vulcanised or
unvulcanised, and, further, glass. Discs were formed,
15 cm. in diameter, up to 0-65 cm. in thickness, and
.soldered (with paraffin) to discs of brass; on the top
of each disc was placed a piece of aluminium or tin-
foil, the brass and foil thus forming the condenser elec-
trodes. Changes in the thickness of the disc were again
measured by the method of interference. In the case
of the vulcanised rubber, the observed contraction

(amounting to 22 x 10-6 cm. maximum) was somewhat
larger (although less than 100 per cent, larger,
certainly) than the calculated contraction, and
increased with the square of the field intensity; the
unvulcanised rubber seemed to conduct the current,
and behaved less satisfactorily. In order to ascertain
whether the direct electrostatic pressure (attraction
between the top and bottom of the condenser) was an
important factor, Bouchet piled five of the vulcanised
rubber condensers upon the top of one another, to
form a multiple condenser; in that case the direct
attraction effect should be the same as in a simple con-
denser, because the attractions upwards and down-
ward would balance one another in the intermediate
condensers (of the multiple arrangement). The
observed contractions were double those of the experi-
ments with the simple condenser. Thus the contraction
due to the internal electric stress would appear to be
much stronger than the direct pressure effect.

SUBSIDENCE RESULTING FROM
MINING*

By L. A. Young and H. H. Stoek.
(Continued from page 910.)

Underlying Rocks.

The physical character of the rocks immediately
underlying the mineral deposit is of great importance.
Frequently coal beds are underlaid with beds of clay
of such consistency that it will not support the pillars
when the weight upon them is increased by the open-
ing of rooms. The pillars are slowly pushed into the
clay, while the • clay is forced into the rooms which
have been mined. Similarly, when water reaches clay
beds underlying the coal, the clay may be softened and
forced into the rooms by the weight of the pillars, and
a subsidence results. The term “ creep ” is very com-
monly applied to such a movement.

Very few tests have been made upon the bearing
power of the clays occurring in mines, but numerous
tests have been made upon clays and soils upon the
surface. Owing to the importance of not placing upon
the clay floor of a mine a burden which shall exceed
the bearing power of clay, which is usually much less
than the compressive strength of coal, the following
values (Baker) are of interest: —

Safe bearing power in
tons per sq. ft.

Kind of material.

Minimum. Maximum.

Rock—the hardest—in thick layers

in native bed ............  200-0

Rock equal to best ashlar masonry... 25-0
Rock equal to best brick masonry... 15-0
Rock equal to poor brick masonry... 5-0
Clay in thick beds, always dry .... 6-0

Clay in thick beds, moderately dry... 4-0
Clay in soft beds ................. 1-0

Gravel	and	coarse sand,	well

cemented .................... 8-0

Sand,	dry,	compact, and well

cemented .................... 4-0

Sand, clean, dry .................. 2-0

Quicksand, alluvial soils, etc....	0-5

30-0

20-0

10-0

8-0

6-0

2-0

10-0

6-0

4-0

1-0

The data on clay given in the table are not for fire-
clay, and no data have been obtainable which are the
results of observations upon the supporting power of

such; clay of the character and occurring under con-
ditions similar to those found in coal mines.

Overlying Rocks.

The study of subsidence due to mining operations
involves particularly a consideration of the rocks over-
lying the mineral deposit. Lack of uniformity in the
overlying measures is the rule, not the exception, and
this fact must be recognised in all attempts to formu-
late theories and rules. The effect of different condi-
tions of the overlying beds is well illustrated by two
examples in England. At Sunderland, where the
measures contain 50 per cent, of hard rock beds, seams
at a depth of from 1,400 to 1,800 ft. have been worked
for 70 years without reference to the surface. On the
other hand, in the Midland and South Yorkshire coal
fields, where the cover is composed largely of soft
shales, the effect of workings at as much as 2,000 ft.
is appreciable on the surface.

Investigations of the thickness and physical char-
acter of each overlying bed are fundamentally neces-
sary to the accurate study of subsidence in any dis-
trict. Much of the data as to the behaviour of various
strata that can be secured will be at best only rela-
tive. However, the more data that can be secured the
fewer will be the variables with which the investigator
must deal.

Practically every theory of subsidence which has
been advanced, when analysed, involves some funda-
mental principle of mechanics. The beds may be sub-
ject to tension, compression, bending, or shear.
Samples of the various rocks may be tested in the
laboratory in order to secure data to be used in the
study of each problem. The great difficulty of obtain-
ing specimens which will be representative of the
section undet investigation is largely responsible for
the scarcity of data along certain lines, n'otably those
concerning the strength of rock in tension and in
bending.

Data on the strength of the rocks that are of import-
ance in the study of subsidence have been collected and
published by Bunting, who stated that numerous tests
of various stones have proved that sandstones take
permanent sets for the smallest loads, whereas granite
and limestones are nearly perfectly elastic. It has
also been proved by tests on various stones that the
modulus of elasticity in compression is practically the
same as in cross bending, but no fixed relation has
been determined of the compressive, tensile, or shear-
ing strength of the various kinds of stone. The shear-
ing strength of sandstones and slates per sq. in. is

* From University of Illinois Engineering Experiment
Station Bulletin No. 91.