Great harm was done to the gas engine in its early days by the premature publication of the results of an experiment in using gas which had not been thoroughly cleaned. It was claimed that only primary cleaning was necessary, and that the fine dust left by this simply went into the cylinder and came out again without doing any harm. A few months' longer trial, however, soon showed that while most of the dust followed this course, enough of it adhered to the oily surfaces of the cylinder and piston rod to cause the rapid destruction of those parts, and it was then realized, once for all, that the gas must be cleaned, not only in primary washers, but must be scrubbed until the amount of dirt left in it was a negligible quantity. It is said that in present practice the gas passing into these cylinders is cleaner and more free of dust than the air we breathe.

This is not the place at which to deal with the question of gas cleaning, but it seems to be a fact that a positive and reliable solution of this problem, which could be guaranteed, was first obtained by the German engineer Theisen, and after the invention of his apparatus the development of gas-driven engines was much more rapid than before.

Adapting The Engine To Variable Loads

In Germany the method has largely been adopted of holding open the inlet valve during a greater or less portion of the compression stroke, and so, in fact, cutting down the capacity of the air cylinder when higher pressures were required than those for which the engine was designed. But in this country this plan seems not to have found favor, except with the Wm. Tod Co. of Youngstown, and in general the gas cylinders have been made large enough to blow the maximum pressure normally required, any change from this being met by adjustment of quantity or quality of gas mixture and enduring the consequent loss of efficiency.

This is for the reason that when a furnace gets "tight" and the pressure goes up, it is frequently, if not generally an indication that it needs more air and if this be not supplied a scaffold is liable to form causing weeks or months of trouble.

Therefore the plan of reducing the volume of the blast in order to attain the increased pressure needed does not commend itself to furnace men; it is good for the engine, but likely to be very bad for the furnace.

Adapting The Engines To Run Double Acting With Tandem Cylinders

The difficulties which prevented this development in the early days were those which came from the intense heat acquired by any surface of small curvature, such as a piston rod, unless it was water-cooled. The head and barrel of gas-engine cylinders have been water-cooled since the very beginning, but on account of the mechanical difficulties the early engines were made single acting and their pistons were not so cooled but made long and allowed to get their cooling effect from the barrel of the cylinder and from the air in the crank chamber. This, with the plain flat piston heads without projections, enabled them to run without getting so hot as to cause premature explosions.

When it became necessary to obtain more working strokes per cycle the trunk piston was discarded and the regular piston and piston rods of the standard engine substituted, except that these were both made hollow and water-cooled by flexible or trombone connections to the cross-head or to the rear end of the piston rod. This having been brought into successful practice, the next step was to add a tandem cylinder and run the piston rod through to its piston, and with the experience based on the previous work this was soon brought into successful operation also.

In the four-stroke cycle single-acting gas engine only one stroke in four is a working stroke instead of every one, as in the steam engine. Therefore by putting two double-acting cylinders tandem and having their working strokes alternate one with another two working strokes per revolution were obtained as in the steam engine. This resulted in an enormous reduction of the irregularities of power development and of turning moment, from which the early single-cylinder gas engine suffered.