{Contributed by H. Luttrell Godden)

Amongst the many modern improvements, both in effect produced and in economy of production, the use of gas obtained from light oils or spirits, both for lighting and heating purposes, cannot be passed over.

This is especially the case in dealing with country districts, where are found many isolated mansions and other private residences, and where coal gas or current for electric lighting are unobtainable unless from special producer plants erected for each individual case.

The idea of using oil or spirit, both on account of its initial cheapness and facility of transport, is not by any means new, but it is only within the last few years that practical results have been realised.

The general principles involved are the same in all systems, and may be stated as follows: -

The inflammable oils and spirits consist of various hydrocarbons and their derivatives. These, if heated to a sufficient temperature, will vaporise into gases, which, in combination with a sufficient proportion of oxygen supplied from the air or some other source, will burn with a more or less luminous flame.

Thus the results mainly depend on (1) temperature, and (2) supply of oxygen.

The production of gas from the heavier oils may here be disregarded, this being somewhat analogous in process and plant required to coal gas, and attention confined to that obtained from the lighter oils and spirits which are readily vaporised, such as petrol, benzol, turpentine, etc.

Petrol is generally used for the purpose, both on account of its cheapness and the low temperature at which it can be vaporised.

The process may be briefly described as Carburetting Air, and consists of passing air through a carburettor, in which it takes up a proportion of the vapour of the oil or spirit with which it comes in contact, forming an inflammable mixture of air and vapour.

It is obvious that, if the temperature required to vaporise the oil or spirit were above that of the atmosphere, the oil or spirit would condense and be deposited when the produced cooled down in passing through the service pipes. This has been, and still is, a cause of serious trouble in many forms of apparatus; but in a good carburettor the petrol is vaporised at a very low temperature, far below that of the surrounding atmosphere, and consequently the mixture of air and gas is stable, provided the components do not separate by gravity, which might happen if kept stored for any length of time. As, however, the plant is usually only working while gas is being consumed, the gas is always in motion from producer to burner, and trouble from this source avoided.

One of the chief difficulties that has been encountered lies in the fact that petrol itself is a mixture of various hydrocarbons of a series, some of which vaporise more readily than others, so that if the air is simply passed over or through the liquid spirit the lighter hydrocarbons are all taken away first, and the gas produced is too rich; while later on, when only the heavier and less readily vaporised spirit is left, the air does not take up a sufficient proportion, and the gas is too poor perhaps to burn at all, while a residue of oil or spirit remains unfit for the process and must be disposed of otherwise.

This difficulty led to the adoption of a "metrical" carburettor, by means of which only the quantity of petrol required to carburet a given volume of air was measured into the carburettor as that volume of air was passed through, and the whole was vaporised.

Again, the temperature of the carburettor must not be allowed to fall too low, or the spirit will cease to vaporise at all. Consequently the thermal units taken up in vaporisation must be replaced, and this is done by radiation from without, such as by placing the carburettor in a jacket of water, glycerine, or oil, which is kept in circulation.

The amount of power required to drive a plant of from 25 to 100 burners is very small, as a pressure of only 3 or 4 inches is quite sufficient, and this power is usually obtained by a weight or by a small water or hot-air motor. .

The gas resulting is poorer in luminosity than coal gas, but gives a much higher flame temperature, and for lighting requires to be used with incandescent fittings, while for cooking or heating it gives an almost colourless and odourless clean flame.

The general faults to be found in most makes of petroleum gas plants are - (a) that the gas produced varies greatly in quality; or (b) that constant expert attention is required to obviate this.

The former (a) arises from the two chief difficulties previously mentioned not being satisfactorily overcome, and consequently the richness of the gas varies considerably according to the temperature of the air, and is very marked between summer and winter, while the variation of specific gravity in the liquid spirit in the carburettor may cause similar trouble at any time.

In most plants an endeavour has been made to overcome these troubles by a regulator which can be set to vary the length of time during which the air is in contact with the spirit, but this method brings in the objection (b) of requiring constant and expert attention to adjust the regulator.

A third drawback still is sometimes apparent through the spirit partly condensing in the service pipes, due to the carburetting process being carried on at too high a temperature.

The ideal plant should produce a gas of constant quality under all conditions, should be automatic in working, so as to require no skilled attention, and be of extreme simplicity in all its parts.

It has been claimed of various designs that they have overcome the above difficulties, but only one, and that quite recently invented, approaches the ideal, that of Mons. Benoit de Laitte, of which a short description and sketch (Fig. 187A) are appended.

This plant consists of the usual parts - (1) a small reservoir R, containing about one pint of petrol, which feeds the carburettor; (2) the carburettor C itself;