but as this heated CO2 gas continues up through the deep fuel bed, its heat liberates more carbon and enables the latter to attach itself to the extra particle of oxygen of the CO2, which breaks off and joins with this free carbon, so as a result we now have two quantities of CO, or 2CO. (Figure 3.)

Chemically combining two elements produces heat; chemically disassociating two elements produces cold, and disassociating the CO, takes up latent heat, and the 2800 degrees resulting from the original combustion of the oxygen and the carbon is materially reduced. This latent heat is restored when the CO gas is finally burned to CO, either in a furnace or in the cylinder of an engine. I particularly refer to this cooling effect, which is technically known as an endothermic action, because it is made use of in the following manner to prevent the producer fire from reaching a temperature at which its ash will clinker. Either steam or carbonic-acid gas is admitted to the producer fire with the incoming air in quantities bearing a fixed proportion to the amount of gas being taken out of the producer. The rate at which the gas is made is a function of the heat of the producer; that is, the more gas made, the hotter the fire tends to be. As we all know, passing steam over an incandescent fuel bed decomposes it into hydrogen gas. This decomposition takes up latent heat and so has a cooling effect upon the fire and enables it to generate the required amount of gas without allowing its temperature to rise to a point where the silica, iron, sulphur and other components of the ash will fuse into a clinker. The same holds true of carbonic-acid gas which breaks up into CO and O. This CO2 gas is taken in small quantities from the engine exhaust. It is claimed for this last method that it performs its cooling effect without liberating hydrogen gas, which varies directly with the amount of steam used, and as it becomes a part of the outgoing gas may change the quality of such gas.

The firing of the producer is extremely simple. Those built on the up-draft type are loaded from a hopper at the top. The down-draft producers are open at the top and through these openings are fired directly. Now that the producer manufacturers have mastered the subject of controlling the fire and have determined the proper means for working out the ash, the labor required to run a producer is very slight. This is also

because the coal and ash handled per horse-power is about onefourth of that required by steam. In the smaller producers having a grate diameter of less than 6 feet, shaking or revolving grates are used; the latter seems to be the preferable form. In the larger diameter producers of the up-draft type, watersealed grates are decidedly preferable, because of the ease with which the ash can be taken out. Furthermore, these producers can be operated continuously, while the producers equipped with grates can run only about two weeks.

When it comes down to the question of anthracite and bituminous coal, a large field for discussion is opened up. In the past, two general types of producers have been developed. The up-draft producer can be used for handling anthracite coal, because the latter has no hydrocarbons which would afterward condense as tar in the outgoing gases that are taken off the top of the fire and over into the cooling and scrubbing apparatus and from there to the point where the gas is to be used. The well-known down-draft producers developed by Messrs. Loomis and Pettibone for gasifying bituminous coal, wood, and so forth, carry the hydrocarbons liberated from the freshly-applied fuel down through deep fire-beds, the heat of which is sufficiently high to crack these hydrocarbons into fixed gases.

Several installations of up-draft producers for gasifying bituminous coal have been made in the past, and are run at a high temperature to make as little tar as possible. The tar that is made is condensed and beaten out by mechanical tar extractors. A further development of this method is one in which an updraft producer is run at as low a temperature as possible, so that most of the hydrocarbons are preserved, enriching the gas by the resulting marsh and methane series. These series are largely burned to a soot in the high-temperature fires of the down-draft producers. This low-temperature up-draft method results in a very large amount of tar, and to handle the same an improved type of mechanical scrubber and tar extractor has been developed, having been fashioned after the best devices of this kind used abroad. It is proposed to atomize this tar in the fire-box of an auxiliary boiler, which furnishes the steam to run the auxiliaries of the producer and engine plant. It iɛ interesting to note that the calorific value of the tar in an

average bituminous coal represents about eight per cent of the total B. t. u. contained. This last type of producer has a water-scaled grate, and it is claimed that practically any grade of coal can be gasified in it successfully. The fact that gas producers throw out no active gas or smoke while in operation is an important consideration, too, in localities burning soft coal or in places where a smokestack is objectionable.

Pea coal is going out of this market. Standard producers can not handle smaller than No. 1 anthracite buckwheat, but with a properly-designed plant bituminous coal becomes available, and it is stated by some engineers well up in the art that slack bituminous coal and run of mine can be gasified, in which event a material saving is not only possible in bulk but also in the price of the fuel over that required by a corresponding steam plant.

Having now before us the nature of the apparatus, we are prepared to consider its application to central stations. In presenting this matter to you, I am going to limit these considerations to three conditions of service.

(a) A new moderate-sized central station about to be projected.

(b) The already existing small central station, whose steam layout is so uneconomical as to make it profitable to substitute a more economical plant.

(c) The already existing large steam central station, with its capacity, as well as that of its feeder system, overtaxed. Each of these problems has different factors and must be considered separately.

Case A. A medium-sized new station about to be projected. A comparison of the cost of operation under certain conditions is given in Figure 4. The fact that the station can be located near the centre of the town because of the absence of smoke is one advantage in its favor; on the other hand, the inability of the gas-engine plant to carry heavy overloads is an offset. To determine if such an installation is worth while, it is necessary to take into account all the conditions and include them in an equation the result of which should determine accurately just which type of plant can operate most profitably under the local conditions.

Case B. Now, as concerns the second class of stations (b) to which gas engines are applicable. We will consider the smail central station of, say, 500 kilowatts, running under uneconomical steam conditions and struggling to meet the interest on its bonded indebtedness; perhaps just squeezing through a small sinking fund or surplus account. Let us take a more encouraging view of the situation and assume that capitalists are sufficiently inter

ENGINE

ested in the situation to put up enough money to rehabilitate the plant, provided the current can be made cheaply enough to earn a good return. An engineering report is called for and the following facts are shown:

Assume that if we discard the old plant, a new 500-kw steam plant can be installed at $50 per kilowatt or $25,000, or a new gas plant installed for $75, or $37,500. For the sake of brevity I am going to assume that oil and waste, repairs, general expenses, labor, and so forth, will be the same in all cases and simply consider what will be saved on the coal-pile by installing more economical apparatus, charging the new plant with the interest, depreciation, and so forth, on its investment.

cent..

Coal.

Total.

Annual saving over present plant............. Reduction in cost per kw-hour (33% per cent loadfactor).

$1,250

1,250

1,875

(8 lbs. per kw-hr.) (4 lbs. per kw-hr.) (2 lbs. per kw-hr.)

It will be seen that the installation of a new steam plant will effect a saving of 16 per cent on the total cost of delivering the current to the meter, while the gas plant will save 23 per cent. The writer has in mind a small station that had defaulted in the interest on its bonds while running as a steam plant; a gas-producer plant was substituted, the defaulted bond interest was made up, a dividend earned on the common stock, and the plant sold to a large public utilities company at a good profit.

Case C. Let us assume that the peak load on the main station exceeds its normal capacity, that the feeder system to one of its largest centres of distribution located two miles away is short of copper and needs increasing. The additional investment chargeable for buildings, boilers, engines, generators, switchboard equipment, transformers and underground conductors and substation to supply 1000 kilowatts will equal the cost of a local substation, with producers, engines, electric generators and switchboard for a gas-engine plant. Let us say that in both cases the investment will be at the rate of $120 a kilowatt. We will assume that the steam plant will deliver its current to the