June 23, 1916.

THE COLLIERY GUARDIAN

1185

M. Gobert proposed to employ such a process in tubes
similar to those used by Poetsch, but of undulated or
helicoid form as regards the inner tubes; the liquid finds
escape through small holes into the outer freezing tubes.
For various reasons, however, this arrangement is
unsatisfactory.

Mr. Chas. Schmidt has suggested increasing the speed
of the gas after vaporisation, and thus forming a wet
vapour which will continue to expand while flowing. In
order to do this, he couples together several freezing
tubes, which are successively traversed by the same
current of wet gas. This method would also have the
advantage of making the distribution easier by the reduc-
tion in the number of valves required. However,
expansion with cooling still remains problematical, and
this method has not found any more practical applications
than that of M. Gobert.

Liquid air has also been suggested, but as the evapora-
tion is still slower than that of ammonia it would ha-ve
little success. Other’ inventors suggest using the direct-
circulation of the cold gases, but if small diameter tubes
are used the speed of the gas would have to be so high
that the friction would be excessive with the resultant
heat, and with ordinary speed such wide freezing tubes
would be necessary that the method would become
impractical.

The same applies to the Koch process, in which method
the previously expanded gas is driven through the
freezing tubes, but as the expansion takes place in t<he
region of the upper strata, only a small portion of the
useful effect of the freezing plant is transmitted to the
ice wall. The writer understands that this system has
been tried in. Holland, but without success.

Resistance of Frozen Ground.

In the early stages of the application of the Poetsch
process, it was considered that frozen wet sand would
have the same resistance as that of rock salt, sandstone
or limestone, and Weitz gave 2,8501b. per sq. in. as the
theoretical resistance for frozen sand. Several tests were
made to determine the exact resistance of frozen
materials. The first were probably those undertaken in

rapidly as the temperature decreases. If the propor-
tion of water diminishes to two-thirds of that necessary
for saturation, the resistance is also reduced to approxi-
mately two-thirds that of saturated sand. Similarly, if
the proportion of water is only one-third, the resistance
is reduced to one-third.

In the tensile tests, the proportion of water plays a
far more important part. Saturated sand at about
-12 degs. Cent. (+10-40 degs. Fahr.) is able to resist
40 kilogs. per sq.cm. (5701b. per sq. in.). With only
two-thirds of the water required for saturation, the resis-
tance is about 25 kilogs. (3551b.). But if the propor-
tion is only about one-third, the resistance is almost nil.

Tests with Mixtures of Sand, Clay, and Water.

For mixtures, the quantities of which in weight of
water and clay exceed that of sand, the results are
almost the same; the sand grains are scattered in the
muddy mass, and do not modify its resistance. This
resistance is about two-thirds of the resistance of frozen
sand to compression. Beyond that the resistance
increases with the proportion of sand.

Under tensile bests, a mixture of half water and half
elav has almost the same resistance as pure water or
ice (about 12 kilogs. instead of 10 kilogs. per sq.cm.,
or 172 1b. to 1431b. per sq. in.). With sand, this
resistance increases much quicker than with compres-
sion. Pure loam gives in compression, as well as in
traction, the resistance equal to that for saturated sand.

Tests with Pure Ice.

All pure ice blocks were crushed under a pressure of
20 kilogs. per sq.cm. (2851b. per sq. in.). The resist-
ance in the tensile teste was found to be 10 kilogs. per
sq.cm. (1431b. per sq. in.). It may be remarked that
sometimes there exist very singular anomalies between
tests of the same series.

Applying the Freezing Process to Different Strata.

From the values obtained in such tests, it is possible
to conclude that the freezing process is suitable for all
kinds of water-bearing strata, but especially for those
which are not coherent, such as alluvial, tertiary, and

increased quickly at 50 kilogs. (7101b.), and more at
60 kilogs. (850 lb.). It may be possible that the plasticity
will diminish with the temperature, and tests will be
necessary at 20degs. Cent. (— 40degs. Fahr.).

Interesting verifications on the temperature of rocks
resulting in differences of conductivity have been made at
the Beeringen Collieries at about 275 m. (900ft.) depth,
namely

In frozen sand —19'5° to—21’4° C. (—3'1° to—6’5° F.)
Infrozenclay -16'8° to -18° C. ( + 1’8° to-0’4° F.)
In frozen lignite -12° to -13'8°C. ( 10’4 to + 7’2°F.)

Early Applications of the Poetsch Method.

In sinking the Archibald shaft, near Sthneidlingen
(Germany) to a seam of lignite, through 18 ft. of moving
water-bearing sand, horizontal boreholes driven about
20 in. above the lignite proved that round each tube
there was an ice circle of about 4 ft. 9 in. diameter, and
much thicker nearer the bottom of the tubes than at the
top. This was due to the fact that the most intense
cold is produced at the bottom- of the tube, near the
point where the brine leaves the descending tube and
flows into the ascending tube, and forms small conies
of frozen ground, which coalesce, making finally a large
block, the size of which increases with the depth. Water
was encountered in the shaft when near completion. A
similar phenomenon occurred at the Shaft Emilie, No. 1,
Finsterwalde, and this circumstance caused great
uneasiness as to the exactitude of the opinions of
Poetsch and Weitz; and tests were made by the French
Government, who, at that time, intended to apply the
Poetsch method in driving a railway tunnel. Although
these tests were made in 1885, they have not lost their
importance, as the old theory of the frozen mass having
the form of a cone with a larger base at the bottom has
still been adopted in a pamphlet issued in 1912 in
Germany. These tests afforded the following informa-
tion with regard to ground which has to be frozen for
shaft sinking :—

Firstly, the lower end of the mass of ice resembles in
shape the base of a soda water bottle, the centre being
unfrozen, whilst the ice ring extends downwards. The

Fig. 3.—Evaporators and Brink Pump.

,W'W

- i

- Wi



Fig. 1.—Stream Condenser.

1885 by the French Government, under the supervision of
M. Alby,* and tests consisted in crushing in an
hydraulic press and also by submitting samples to a
tensile test. As the tests were of an industrial character
they lacked the precision of laboratory experiments. It
was, for instance, admitted that cubes from an enclosure
at n degrees and put immediately under the press
remained at n degrees. As the resistance increases with
the diminution of temperature the number so found is
certainly lower than the actual number. It has also been
admitted that cubes which had remained for several houis
in dishes under water at n degrees wore themselves also
at n degrees. Finally, it has been ascertained that frozen
cubes exposed to the air do not return to freezing point-
(32 degs. Fahr.) under one hour. Further tests were made
in order to give greater accuracy. A series of very exact
tests have been made in Germany for the application cf
the freezing process to- marine foundations. The values
obtained are noted in Table III.

Tests with a Mixture of Sand and Water.

The Alby tests gave convincing results at about
0, -10, -12, -14 and 25 degs. Cent. (32, +140, +10-40,
+ 6-8 and -13 degs. Fahr.). The saturation of the
sand was a mixture of about 165 grammes of water for
one kilog. of sand (1,115 grains per lb.). With 200
grammes (1,400 grains per lb.) of water, there was no
more saturation from the sand. The maximum resist-
ance was reached with saturated sand. If the propor-
tion of water diminishes in the sand, its resistance also
diminishes.

The tensile tests were made only at a temperature of
about -12 degs. Cent. ( + 10-4 degs. Fahr.). However,
these tests gave some difficulty as the gauges became
damaged under the pressure of the levers. With the
indications given for the compression, it is possible to
trace a curve giving, with an error of less than 5 kilogs.
per sq. cm. (721b. per sq. in.), the resistance to crush-
ing of frozen sand. The equation obtained from the
parabolic curve is y =' —0-153 x2 + 11 x + 20. It may
be noted that the resistance of the ground increases

* M. Alby. “ Note sur des experiences de congelation de
Terrain.” Annales des Ponts et Chaussies. Paris, 1887.

creteceous deposits which abound in sand, gravel, clay,
chalk, marl, etc. It gives the best results, however,
for sand, which is changed into a genuine hard sand-
stone. Clay does not become so hard under the effect
of freezing, so that a good temporary lining is required,
and only short lengths can be driven through without
tubbing, and in certain cases underhanging tubbing
must be adopted. It is not advisable to be too reliant
on the efficacy of very low temperatures, because in this
case the texture becomes schistous, accompanied by
slips.

The best results, however, have been obtained.in hard
red triassie sandstone, such as exists in Yorkshire. In
such ground freezing prevents water from leaking into
the shafts. The rocks themselves are frequently suffi-
ciently consistent to remain rigid during sinking, but the
quantities of water are -sometimes so enormous that
efforts to keep them down with pumps have been in vain.
These rocks, as a rule, are not suitable for cementation,
and many sections are not more consistent than quick-
sand. Frozen lignites have only a small resistance;
moreover, they are bad conductors of heat and cold,
and sometimes become heated by self-fermentation.
These circumstances have not always been recognised,
and it is quite possible that several accidents which have
been attributed to adverse circumstances were as a
matter of fact due to the presence of lignite. Rocks
impregnated with salt water need special attention, as
will be described in another paragraph. For very deep
shafts it is necessary to work with much cooler brines
than for smaller shafts. The ice wall near the bottom has
to resist a very much greater pressure, and, with the
lowering of temperature, its thickness, as well as its
resistance, increases rapidly. It was previously imagined
that ice would become plastic under the corresponding
pressure of great depths, but proof was obtained by the
results shown in shafts of 400 m. (1,310 ft.) in depth,
where it was -absolutely compact and free from fissures.
Tests recently made at the Beeringen Collieries, Belgium,
proved, however, absolutely-the contrary. A hollow sand
cylinder frozen to 10-6 degs. Cent. ( +12'9 degs. Fahr.)
showed only a slight deformation at a pressure of
40 kilogs. per sq.cm. (5701b. per sq. inch), but this

outer diameter of the ring is, however, smaller than
that in the centre of the pit. Therefore, the ice wall is
always weaker at the base than at any other part, and
it is absolutely necessary to keep the freezing tubes in
until solid, and, if possible, dry strata are reached. This
is most important, as the greatest pressure is at the base
of the ice wall, and the deviations of the boreholes may
easily produce weak points.

Secondly, at the top, the centre of the shaft remains
unfrozen until a certain depth is reached, as a good deal
of the cold is lost near the surface of the earth. In
large shafts the soft sand may extend to a depth of
260 to 365 ft., as was found in several of the collieries
in the Belgian Campine. This is favourable for sink-
ing, which, at the outset, advances quickly, and it is
advisable to commence the excavating as soon as the
ice wall is closed.

Thirdly, a protracted stoppage of the freezing plant
may become' very dangerous when the sinking has
almost reached the bottom of the shaft, since a stoppage
for even four days has been found sufficient to thaw the
ice wall and" allow irruption of water. In some cases
it is, therefore, more prudent to use two smaller ice
machines rather than one of larger size, one acting as a
stand-by in case of breakdown.

Formulae of the Ice Wall.

In order to determine the power of the freezing plant,
it is necessary to ascertain the thickness of the ice wall
required. For round shafts, M. Lebreton* gives the
following formula:—E = R	0 ’ w’iere'n

E = thickness of the ice wall; R = the diameter of the
unlined shaft; N = the resistance of the ground to
crushing, in kilogs. per sq. cm. (about 15 lb. per sq. in.),
and P = the external pressure.

For N — 2 P, the thickness would have to be infinite,
i.e., only shafts of relatively small depth could be sunk
by the freezing process.

* F. Lebreton. “ Memoire sur le methode de congelation de
M. Poetsch.” Annales des Mines. Paris, 1885.