while the best results were always obtained both with plain coal
gas and with mixtures of water gas with coal gas when the rate of
flow was adjusted until a 3-inch flame in the chimney was obtained.
Mr, >houbridg'e properly objected that when the gas was sent out
into the district the gas manager would not test the gas under
this favourable condition, but would use the Referees' Table
Photometer, and adjust the rate of flow to give a i6-candle flame,
and therefore that this latter method of testing should be adopted in
these experiments. All the results quoted in the present paper
were therefore obtained by adjusting the rate of flow to give a
i'6-candle flame and then calculating the results to a 5 cubic feet
.When using the water gas it was soon found that the addition
of small proportions resulted in but little gain, but that, as the pro-
portion of water gas was increased, the gain in candle-feet per ton
(volume of gas per ton x illuminating power -=- 5) became more and
The conclusion deduced from the experiments with horizontal
retorts was that an addition of about 40 per cent, of water gas
during the first three hours of carbonisation of each charge is the
most suitable proportion of water gas to employ, and this was
confirmed in the following year by experiments conducted under
more satisfactory conditions with inclined retorts.
Mr. Shoubridge having completed the erection of a new bench
of 70 inclined retorts in the early part of the year 1901, the upper
rqouthpieces of these were provided with pipes for the intro-
duction of water gas, and a long series of experiments were then
carried out with extremely satisfactory results. Although little or
rki' Change was detected in the composition of the tar, a notable
enrichment of the water gas was effected, and the results make it
perfectly clear that a gas manager who has been supplying a
i6-candle gas can, by simply putting in a blue water gas plant and
utilising 40 per cent, of this gas in the retorting, turn out between
14,000 and 15,000 cubic feet of 14.5 candle gas per ton, without
any alteration in his heats or general procedure. Even with coke
ai a high price, the cost of water gas made by the Dellwik process
should, riot. exceed 3d. or 3jd. per thousand cubic feet, and with
water gas at the higher figure an economy of 2$d. per ton of coal
carbonised can be effected.
UTILISATION OF WATER GAS.
O M OO O W>
oo i^ vb '* -
b V 1 ^
W -^-oo r^.. M
Tj- CO <*
00 t^ 1 ^- OC'!
k W-* VOX
N 1-1 i- N Tf
ON oo o
iO O O CO
: vb ^ : ^ 00
288 UTILISATION OF WATER GAS.
The principal results obtained are shown in the following table,
but the author is of opinion that by introducing hot instead of
cold water gas, and by a more careful proportioning of the rate of
flow of water gas to the rate of evolution of gas from the coal in
the retort, results yet more favourable can be obtained.
The Discussion was taken part in by the following members:
Mr. G. R. Love, Mr. E. H. Millard, the Chairman, Mr. W. Graf ton,
Mr. T. Glover, Mr. W. R. Herring, Mr. Charles Hunt, Mr. S. Y.
Shoubridge, and Mr. J. W. Helps.
The author replied, and a vote of thanks was accorded to him.
Communications from Mr. Thomas Holgate and Mr. D. H. Helps
have appeared in the technical press since the Congress, and are
incorporated in the proceedings.
"THE AUTOMATIC LIGHTING AND EXTINGUISHING
OF STREET LANTERNS."
Paper by A. ROTHENBACH, Jun.
FOR many years past, attempts have been made in the direction of
lighting and extinguishing street lamps by some automatic or
mechanical means, which would be more reliable and less expensive
than the present system.
With the introduction of the Welsbach burners for the public
lamps, arose the desire to light them in a way similar to that usual
with electric, incandescent, and arc lights ; and from that time date
most of the trials made in this direction, especially those in which
electricity is used.
Separate wires were drawn between the lamps, and small devices
fixed, by means of which the valves could be opened and closed
by the electric current, and at the same time the gas ignited,
either by electric sparks, or platinum-black, or by a wire brought
to red heat.
In course of time, however, many disadvantages showed them-
selves, such as : (i) The breaking of the wires by the weight of
snow or some other cause; (2) entanglement with other wires, es-
pecially those of trolley lines ; and (3) changes of temperature, etc.,
causing the oxidation of the contact-buttons, and otherwise in-
fluencing the small electric devices. In many of these instances, it
is difficult to locate the defect which causes the interruption of the
current. Putting the wires underground only increased the cost
of installation, without improving the situation.
A second method was to light and extinguish by means of com-
pressed air. This system has the disadvantage that leaks or
obstructions in the small pipes to be used, and which have to be
laid underground, would, until found, cause a great expense, and
until repaired, would put quite a number of lamps out of use.
A third way to accomplish the desired end was tried with an
apparatus influenced by the difference in the pressure of the gas;
but it was found difficult to procure the required difference in a
distance of many miles, especially where the pipes leading into the
houses are connected with the same main and in towns where the
streets are hilly. This system has failed altogether, as proved by
an attempt made in Brussels to measure the gas consumed during
the hours of day and night with one and the same meter.
290 AUTOMATIC LIGHTING AND EXTINGUISHING
A fourth method, using an hydraulic apparatus based upon the
difference in the gas pressure or compressed air, proved a failure
also on account of the evaporation of the fluid and the influence
Attempts have also been made in other directions.
There is, in the author's opinion, but one way to solve this
problem, and that is by the use of some device whereby each lamp
can be lighted and extinguished independently of the others, so that,
in the event of a failure through any cause, one lamp only, and
not a whole section, will be affected.
The Gas Engineers' Association of Switzerland had requested
the author to show and explain such an apparatus made by the
Actien-Gesellschaft fiir autom. Ziind und Loschapparate in Zurich,
which Company has now, after experimenting for three years,
brought it to such a perfection that it seems to answer all demands.
This apparatus consists of a clockwork, of the very best quality,
which cannot be influenced by changes of temperature. It
is hermetically enclosed in a brass box, containing the valves
(separated airtight from the movement to prevent gas escape and
explosion), which are set in motion by the spring of the clock.
The apparatus is placed in the centre of a wrought-iron support
specially constructed, which can be fitted to every description of
lantern. The whole has a neat appearance, and throws no shadow.
The movement itself runs twenty days, but should be wound up
every fortnight. Once within that time, at least, lamps have to
be cleaned and the lighting-hours changed ; so, without extra ex-
pense, the winding and time changing can be attended to by the
The advantages of this system are the following :
(1) Each lamp can be lighted and extinguished separately
and at any designated time.
(2) Any number can be lighted and extinguished together
within a few minutes' time.
(3) The mantles of the Welsbach lights are better preserved,
because the apparatus opens the valve gradually to let
the air escape, thus preventing an explosion, and be-
cause the jar caused by the knocking of torches against
the lamps is done away with.
(4) A great many lights can be extinguished about midnight.
Some cities allow these to burn all night, because a
third round on the part of the men would add more
to their wages than the amount saved in the gas con-
The apparatus can be furnished in five different forms :
(i) One with a simple stopcock.
OF STREET LANTERNS. 291
(2) One with a regulating-cock, through which the lantern can
also be lighted and extinguished at any time by turning
the lever. This has also an arrangement with which
the action of the movement can be detached, and is
suitable for such towns as during moonlight or summer
months suspend lighting altogether.
(3) One for lighting and extinguishing two to three flames
(4) One which will light two flames, and extinguish them'
singly, at different times.
(5) One which can light and extinguish twice within twenty-
By means of these different arrangements, it is not necessary to
do away with any lanterns now in use, or employ help for such
The apparatus is used in prominent cities like Zurich, Geneva,
Lucerne, and Winterthur, over a thousand being in use or in course
The paper was accompanied by illustrations.
The Discussion was postponed until the following day (p. 292).
The meeting was then adjourned.
WEDNESDAY, 4th SEPTEMBER, 19O1.
Mr. WILLIAM FOULIS, Vice-Chairman, in the Chair.
DISCUSSION ON MR. ROTHENBACH'S PAPER.
The Discussion was taken part in by Mr. J. L. Chapman, Mr.
Charles Carpenter, Mr. T. Holgate, Mr. G. R. Love, and the
Mr. A. Kilchmann replied on behalf of the author, to whom a
vote of thanks was accorded.
"THE PRINCIPLES OF CONSTRUCTION OF A PROPOSED
MODERN GASHOLDER FOR AMSTERDAM."
Paper by J. VAN ROSSUM DU CHATTEL.
THE conditions for the construction were the following : The
capacity of the holder (to be erected on a very poor subsoil) should
be 100,00 cubic metres, or about 3^ million cubic feet. Piles
must be used to give the necessary stability. The diameter
must be about 60 metres, or 200 feet. The piles must bear a
maximum weight of 10 tons each, their normal length being 14
metres. The water level is ij metres below the level of the ground,
where the holder is to be built. The indifferent nature of the
ground, and the high cost of a good foundation, make it necessary
to reduce as much as possible the total weight, and to exclude a
tank made of brickwork or concrete.
Only a wrought iron or steel tank containing a minimum weight
of water could, therefore, solve the problem. An ordinary tank,
with a flat bottom for a four-lift holder, with a diameter of about
60 metres, would contain 29,000 tons of water; and with the holder
weighing about 2000 tons, the total weight would be 31,000 tons.
This shows that, with regard to economy in weight, it was necessary
PRINCIPLES OF A PROPOSED GASHOLDER FOR AMSTERDAM. 293
to take into consideration the weight of the water. These diffi-
culties, it was thought, could be met by the construction of a
tank after the patent of Professor Intze, or by an annular tank
above the ground. For the latter construction less material is re-
quired, and consequently it is cheaper and more desirable also from
a general point of view. Care, of course, must be taken that no
gas can enter into the interior space of the tank, which is intended
to be used as a store room. It is therefore necessary that the
inner roof of the tank or intervening central space should be
covered with water.
In order to make this large .store room suitable for heavy
materials, it was so arranged that a locomotive and train could pass
under the tank and the walls, the door openings being made suffi-
This is, so far as the author knows, the largest annular tank ever
constructed under these conditions, and the character of the soil
and the vibrations induced by trains passing under the tank must
be taken into consideration. The supposition that, due to the nature
of the subsoil, the tank will sink at one side 20 centimetres, or about
8 inches, made the calculations very complicated.
If the tank sinks on one side more than 8 inches, means are
provided to put it straight again with wedges; and these, it is
clear, may also be used for a sinking of less than 8 inches. To
prevent this sinking becoming more than 8 inches, 80 lifting
apparatuses, put together under the stays, will come into action,
after the tank has been emptied. Great care was taken with the
foundation to avoid, so far as possible, the use of these contrivances.
Testing piles clearly proved that the subsoil was of a varying
character, and that the length of the piles at different places must
vaiy between 40 and 60 feet. On the piles which are below the
lowest water level a bed of concrete with old iron rails is built.
The wall on which the tank is to be erected will be made so
that there are 40 door openings through which the train may pass.
Moreover, these walls, instead of being made as thick as if a low
stress was allowed, will be made of superior brickwork, with the
best bricks set in Portland cement, so that a higher stress may be
allowed. The calculations are for a pressure of about 250 kilo-
grammes per square metre; and for a maximum lateral strain on
the bricks of 15 kilos, per square centimetre. The piles will bear
a maximum weight of 7000 kilos, each. In order to render a turn
table with radial rails possible in the centre of the store room under
the tank, so that the waggons may be discharged in all directions,
it was not desirable to have any support under the tank, or any
centre pier. This problem was rather difficult to solve.
Instead of covering the intervening space or roof of the tank with
a few inches of water, a larger quantity was used, 40 centimetres
294 PRINCIPLES OF A PROPOSED GASHOLDER FOR AMSTERDAM.
or about 16 inches depth of water resting on the cover. The
weight of this water gives compound stresses on the 40 oblique
struts, and, from these, on the 40 vertical stays of the inner mantle
of the tank. The vertical stays give additional strength to meet
the pressure of the water in the annular tank, so that a much
lighter construction sufficed. The roof of the central space is made
of sheet iron, resting on 40 radial horizontal beams, supported by
the 40 vertical stays, these last serving also to support the curved
plates and transmitting the water pressure from the roof to the an-
nular space. To support the vertical stays struts run from the
horizontal beams, and give, by their shearing force, an exterior
bending stress. The 40 horizontal beams meet in a central ring
or star, and the 40 rays of this star are joined by hinges to the
beams. This construction simplifies the calculations; and this is
desirable, as lateral compressions are to be expected on account
of the nature of the subsoil. The principal dimensions, given in
Exterior diameter of tank ... ,,.''" ... 6130
Height ... ,... .... ' 989
Width of the ring, as far as the front of the stay 215
Diameter as far as exterior of stays ... ... 57
Height to the roof sheets ... ... ... 918
Total length of roof beams ..." ... ... 55
Height of stays to the beams ... 863
Distance from the face of the curved sheets to
the front of stays . . . ... ... ... 12
(The paper gives further an introduction to the investigations
and calculations regarding the forces acting on each part of the
The calculations for the tank must take account of the follow-
1. In addition to the weight of material from which the holder
and tank are constructed there is a uniform load of 60
centimetres of water.
2. The tank sinks 20 centimetres over one side, the whole
gas-pressure working. In this case a uniform load of 50
centimetres of water, and a wedgewise load at one side
zero, at the opposite of 20 centimetres of water, have
to be considered.
3. The tank is filled, but without gas in the holder; and there
is a uniform load of 40 centimetres of water on the
cover or roof of the tank. This case has to be calcu-
lated separately, as some struts undergo the full pressure
from the outside, but a much weaker bending and shear-
ing pressure from the inside.
PRINCIPLES OF A PROPOSED GASHOLDER FOR AMSTERDAM. 295
4. During the simultaneous filling of the roof and the annular
space, the degree of filling must be found at which there
is the maximum outward bending of the struts.
Of very great interest is the simultaneous filling of the cover
of the central space and the annular tank. Both must be filled at
the same time. The tank having a capacity of 4545 cubic metres,
and the cover of 974 cubic metres, the rate of filling must be in
the ratio of 4.67 to i. The cover having two concentric spaces,
the filling of both parts has to take place in proportion to the
surfaces; that is as i : 8.4.
Special boxes with overflows, dividing the water in the required
quantities are therefore made for the purpose. The same care has
to be taken in emptying the tank.
Very important, also, is the construction of the hinge joints in the
horizontal beams. The calculations show that they have to bear
a vertical load of 6640 kilos. ; and there is also a maximum hori-
zontal force of 136,281 kilos.
The following members took part in the Discussion : the Chair-
man, Mr. Charles Hunt, Mr. Charles Carpenter, and Mr. W. Wood.
The author replied, and a vote of thanks was accorded to him.
A correspondence has appeared in the technical press between
Mr. F. S. Cripps and the author, and is reported in the proceedings.
'THE PRODUCTION OF ILLUMINATING GAS FROM
Paper by F. SCHNIEWIND.
THIS paper describes the progress made in the United States and
Canada in recovering illuminating gas from by-product coke ovens.
It discusses its bearing upon the smoke problem of large cities,
and gives particulars of various allusions to the subject in past
literature. It deals with the fuel supply of large cities, and gives
figures showing the comparative amounts of bituminous coal and
anthracite coal used in some American cities for the year 1900.
Ifc then gives a general description of the combined coke oven
and gas process, compares it with ordinary gas retort practice, and
gives a description of a plant of 100 coke ovens of the latest type
of the United Coke and Gas Co., including the system of coal and
coke handling, the arrangement of gas mains, the condensing plant,
the treatment of the tar produced, and the methods adopted for
the further enrichment of the rich gas by the benzole extracted
from the poor gas.
It then proceeds to discuss the principles of the dry distillation
of coal in coke ovens, and gives figures as to the yields of gas, tar
and ammonia, etc., of various American coals in use. It details
the quality of the gas made during the various periods of the
coking process, and gives figures showing that the operating results
approximate very closely to those obtained in the various tests
made. The question of heat balance is then carefully discussed,
and comparisons made of the heat distribution in products of
distillation from Otto Hoffman Ovens, and ordinary gas retorts.
The subject of the enriching of coke oven gas is then carefully
discussed, and tables given showing the distribution of illuminants
in international coal gas. The author then deals with the applica-
tion of coke plants to the gas supply of large cities, and gives figures
showing the approximate gas consumption of a city of 400,000
inhabitants supplied by a coke plant. The fluctuation in gas
consumption is again introduced, and the methods of meeting it
by means of auxiliary producer plants, auxiliary water gas plants,
and combined blue water gas and producer plants are discussed.
The author concludes by claiming for the system serious con-
PRODUCTION OF ILLUMINATING GAS FROM COKE-OVENS. 297
sideration in the solution of the smoke problem, and argues that
it is capable of forming a central station for the supply of light,
heat, and power.
The following members took part in the Discussion: Mr.
Livesey, the Chairman (Mr. Foulis), Mr. S. O. Stephenson, Dr.
Revay, Mr. Charles Hunt, Mr. James Barrow, Mr. W. R. Herring,
and Mr. W. W. Hutchinson.
Dr. Revay replied on behalf of the author, to whom a vote of
thanks was accorded.
Dr. Schniewind has also replied by letter to the remarks on his
paper, and Mr. Charles Hunt has sent a communication.
THE DESTRUCTION OF GAS PIPES BY MEANS OF
Paper by W. LEYBOLD.
THE author gives the durability of the pipes used for the distribu-
tion of gas in towns as from 25 to 50 years. He states that there
are certain influences sometimes at work which may considerably
shorten their lifetime. A new danger has, however, been intro-
duced through the construction of electric tram lines viz., electro-
lysis. In Germany the electric current passes into the wires from
the generating stations at a pressure of about 500 volts, and returns
to the station by means of the rails. As the rails give a certain
resistance, part of the current will pass through the earth into the
gas and water pipes. The author took steps to discover whether
in any pipes, near which any electric tram lines ran, the electric
current was in existence, and he found that in water pipes laid at
a distance of 6 kilometres from the nearest electric station consider-
able tensions were found; and this was also the case in gas pipes
in the town at night when no electric trams were running. This
is accounted for by the theory that the cast iron pipes, with lead
as the jointing metal, lying in the damp ground, produce a galvanic
action. When the tram lines were working the tension in the
pipes varied from 0.2 to i volt to as much as 4.65 volts near the
generating station. All the electricity was supplied from one
works, the working line being 100 kilometres long, and the annual
consumption of current 13 million kilowatt hours.
It is known that, by the electric current in the presence of saline
solutions, metals can easily be dissolved. The ground in Hamburg
contains small quantities of chloride of sodium, to the extent of
0.006 to 0.04 per cent. ; the electric tramway authorities also use
salt, etc., for melting the snow in winter time ; this affords, therefore,
an opportunity for the eating up of iron in the earth by the electric
current in the presence of the solution of salt,
In April, 1899, an escape of gas was found in a street near the
electricity works, at a spot where the cars pass at intervals of three
minutes, there being two lines of tramway rails. On investigation,
it was found that the service pipes passing at right angles beneath
the rails were corroded immediately underneath them, penetration
being discovered in nearly every case. The pipes were covered with
canvas soaked in boiled tar. In many cases blisters were found
between the iron and the tar, which were filled up with a green<
DESTRUCTION OF GAS PIPES. 299
solution, protochloride of iron, and it was inferred that the wrapping
of boiled tar and canvas favours the destruction. The pipes were
taken up and replaced with others, but after the expiration of
seven to eight months the destruction again showed itself as before.
The importance of great care being taken to reduce the currents
passing into the pipes is emphasised, and various methods are
mentioned for securing this result. The rails should be of high
conductivity, with sufficient transverse section, and with the points
of contact well joined together by soldered copper wire. The use of
thermite is also recommended, as is also the fixing of insulated
return transmission cables in many places for the carrying of the
current back to the works. This has been done in Hamburg, with
the result that the tension existing in the gas pipes has been reduced
to 0.45 volts. The cast-iron pipes in the town have not been
perceptibly affected. Allusion is made to electrolytic damage done
to pipes at Erfurt; and this, it is stated, was principally due to
the rails used for the tramway being too light for the purpose.
Mention is made of the rules for the protection of gas and
water pipes drawn up by the German Electrical Technical Associa-
tion, particulars of which are to be published shortly. The author
gives it as his opinion that gas and water works have a right to
demand that the Electric Authorities should do everything in their
power to protect the pipes.
Mr. Livesey at this point again took the Chair and opened the