Presence of Water in the Cylinder.—It may sometimes happen that water may find its way into the cylinder with the gas by reason of the bad arrangement of the piping. It may also happen that water may enter the cylinder through the water-jacket joint. Again, the presence of water in the cylinder may be due to condensation of the steam formed by the chemical union of the hydrogen of the gas and the oxygen of the air, which condensation is caused by the cool walls of the cylinder. The water may sometimes accumulate in the exhaust pipe and box, when they have been improperly drained, and may thus return to the cylinder. Whatever may be its cause, however, the presence of water in the cylinder impedes the starting of the engine, because the gases resulting from the explosion are almost spontaneously chilled, thereby diminishing the working pressure.
If electric ignition be employed, drops of water may be deposited between the contacts, thereby causing short circuits which prevent the passing of the spark.
If there be no drain-cock on the cylinder, the difficulty
of starting the engine can be overcome only by ceaseless attempts to set it in motion. The leaky condition of a joint as well as the presence of a particle of gravel in the cylinder-casting, through which the water can pass from the jacket, is attested by the bubbling up of gas in the water-tank at the opening of the supply tube. These bubbles are caused by the passage of the gas through the jacket after the explosion. If such bubbles be detected, the cylinder should be renewed or the defect remedied. In order to obviate any danger, the stop-cocks of the water-jacket, which have already been described in a previous chapter, should be closed while the engine is idle.
Imperfect Ignition.—The difficulties encountered in starting an engine, and caused by imperfect ignition, vary in their nature with the character of the ignition system employed, whether that system, for example, be of the electric, or of the incandescent or hot tube type. Frequently it happens that in starting an engine a hot tube may break. If the tube be of porcelain the accident may usually be traced to improper fitting or to the presence of water in the cylinder. If the tube be of metal, its breaking is caused usually by a weakening of the metal through long use—an accident that occurs more often in starting the engine than in normal operation, because the explosions at starting are more violent, owing to the tendency of the supply-pipes to admit an excess of gas at the beginning.
A misfire arising from a faulty tube in starting may be caused by an obstruction or by leaks at the joints or
in the body of the tube itself, thereby allowing a certain quantity of the mixture to escape before ignition. This defect in the tube is usually disclosed by a characteristic whistling sound.
A tube may leak either at the bottom or at the top. In the first case, starting is very difficult, because the part of the mixture compressed toward the tube will escape through the opening before it reaches the incandescent zone. In the second case, ignition may be simply retarded to so marked an extent that a sufficient motive effect cannot be produced. An example of this retardation, artificially produced to facilitate the starting and to obviate premature explosions, is found in a system of ignition-tubes provided with a small cock or variable valve (Figs. 74 and 75).
Fig. 74.