position by bolts outside the main casting, instead of through the centre, the end covers pressing hard on the plates.

In every arrangement of exhaust silencer provision should be made for draining off the water which may collect through condensed gases or rain getting into the escape pipe. A convenient method is to arrange a moderate sized pocket in the form of a standpipe, arranged in the lowest position and fitted with a drain cock.

CHAPTER XXIII

GAS METERS

THE capacity of a meter is calculated by the number of lights, consuming 6 cubic feet per hour, it is designed to supply. Thus, a 5-light meter is capable of passing 30 cubic feet per hour, though meters are very often worked to a considerably higher capacity than that for which they are designed, and the dry form can be safely expected to give 11⁄2 times the duty stamped.

The size of service pipes depends upon the distance and size of the main gas supply. If, however, the pressure of the gas is above, or not less than, 1 inch water pressure, and the distance from the main not more than 30 yards, the above meter sizes will be found suitable. It is advisable rather to increase the size of pipes to the meter than decrease them, and gas pipes of less than inch bore should never be placed in the ground.

M

CHAPTER XXIV

GENERAL ARRANGEMENT OF ENGINE

FIG. 170 is an elevation and fig. 171 a plan of a 9 B.H.P. gas engine, giving a convenient arrangement of engine, tanks, &c.,

but the varying conditions of space and machinery to be driven make it impossible to fix any definite arrangement to suit all

cases.

For the foundation, brick, stone, concrete, or wood is used; where convenient, concrete makes a good foundation, and admits of holding-down bolts having a firm hold.

It will be seen that two cooling tanks are used; the inlet and outlet of the water to the cylinder and combustion chamber are shown by the direction of the arrows. A number of vessels arranged so that the hot water from the cylinder and combustion chamber returns to the top of the first water vessel, falling to the bottom and rising up the partition and into the top of the second vessel, and falling again to the bottom and to the engine, makes a most efficient method of circulating. The capacity of these vessels equals 56 cubic feet. The way in which a gas engine is fixed has frequently a great influence upon its future running. If the air necessary for combustion flows into the cylinder through a long, continuous pipe, indifferent firing of the charges will probably follow, causing great loss in power and troublesome explosions. Large gas engines should be provided with an air regulator, which in most cases is fixed near the engine, although much better results are obtained by arranging regulator on the end of air pipe. In some cases much longer air pipes are used than when testing engine at the works. For this reason air pipes should always be larger beyond the air box.

Gas standing in ordinary gas pipes for any length of time becomes mixed with air by diffusion. If the mixture of gas and air is not allowed to flow out before attempting to start an engine, a difficulty will be experienced; and it will be found the best plan to always test the gas before attempting to start for the first time, or after it has been standing for a few days, and a cock should be fixed in the gas pipe near the gas cock, with a pipe leading to the under-side of the water inlet pipe, with a small cock attached, which will not only serve the purpose of testing the quality of the gas, but may be left burning in frosty weather to prevent the water in the circulating pipe from freezing.

If, on lighting the gas at this cock, it burns with a blue

or Bunsen flame, it is diluted with air, and should be allowed to burn until the flame becomes the colour of an ordinary gas jet. It may so happen that, although the engine may run fairly well with a light load, yet with a full load the gas supply may be found insufficient; so that, the gas bag being emptied, the supply is drawn direct from the main, with explosions in the air and exhaust pipe following as a consequence because the gas and air are not mixed in the proper proportions in cylinder. In case the supply of gas is throttled-i.e. the main is not large enough to suit the pressure, an accumulation of water, or obstruction in the pipe; sometimes by negligence in making joints. On the other hand, it may be all right at night, when the pressure is higher, the explosions taking place only during the day. All lights connected to the same main will fluctuate considerably, so that the Bunsen flame for heating the tube may persist in lighting at the air holes. If the gas is insufficient, the simplest remedy is to increase the size of the main. The increase in size will, of course, depend on the size of the main gas supply, and the distance the engine is fixed from this supply. However, the lowest pressure should always be known before the engine is fixed. Engines placed at the top of a building are better off in this respect.

CHAPTER XXV

SCAVENGING METHOD

THE first successful engine using a positive scavenger was the six-stroke engine invented by Linford in 1881; but the first four-stroke Otto cycle engine with positive scavenger was that invented by Mr. J. H. Hamilton, B.Sc., and exhibited by Messrs. Wells Brothers at the Crystal Palace Electrical Exhibition in 1891 and 1892. This engine was a 14 H.P. NOM.,' and used 16.5 cubic feet of gas per 1.H.P. per hour when developing 27 B.H.P. The compression of the charge before ignition was 65 lbs. per square inch, and although using a differential piston the work absorbed in forcing the air through is, owing

to free passages, very slight, a reduction of lb. to 1 lb. per square inch from the indicator card will well cover it.

Although the advantages gained by exhausting the products of combustion have been disputed and re-disputed over and over again, there is no doubt that to get rid of the products as quickly as possible after the working stroke is completed is the right thing to do. Not only is the average pressure of the indicator diagram maintained more uniformly, but a greater power is developed for the same size of cylinder, with a marked decrease in the temperature of the cooling jacket water, and thus making the engine more suitable for hard continuous work. Under certain conditions, such as using a long exhaust pipe without a silencing chamber, when the pipe has become heated a partial vacuum is created (even with the ordinary valve settings this is very noticeable when the air and exhaust valves have been arranged in close proximity to each other). Several methods of eliminating the exhaust gases without the use of separate pumps have been used, the most noticeable being the Crossley-Atkinson Scavenging Method,' patented in 1893, and described in the Engineer,' December 1894.

Fig. 172 shows various positions of the crank pin with the valve settings, and fig. 173 is a weak spring diagram.

Assuming the crank pin to be in the position A at the time the exhaust valve opens, towards the termination of a working stroke, the exhaust valve is kept open until position B is reached. The exhaust valve in an ordinary 'Otto' engine would be closed about the point C when the piston is at the end of its stroke.

Fig. 172 shows approximately the crank pin in its various positions. The relative periods are also shown developed.

In the ordinary 'Otto' engine valve settings the air and gas enter during a half-revolution, the exhaust valve opens before the end of the outward stroke, and closes at the end of the instroke; but referring to fig. 172 it will be seen that in the scavenging engine the air valve opens before the end of the instroke and closes at the end of the out-stroke, and the exhaust valve opens before the end of the stroke but does not close until the piston has passed the dead centre, the air and exhaust valves being thus open together for about a quarter of a