36 THE DESIGN OF AN OIL ENGINE. 
was willing to stake his reputation. In order to start this engine steam would have 
to be supplied to the water jackets. This is undesirable for such a small engine, but 
the submarine of today is not the submarine of tomorrow. When the offer was 
made to design a submarine engine reference was made to such an engine as would 
be needed in the large ocean-going submarines in which a speed of 25 knots would 
be obtained. These submarines have been designed and the speed reduced to 19 
knots because of the inability to obtain the above speed. To produce steam needed 
to warm this engine will not present any great difficulty, and the engine will be 
greatly improved by an increase in power and an increase in mechnical efficiency. 
As soon as the engine is firing, water will be needed in the jackets and the engine 
will then follow the conventional Diesel in its cycle. 
Extreme emphasis has been laid on the fact that this design is not claimed to 
be mechanically perfect. If put into the hands of a first-class draughting force and 
the several details gone over with great care, the result would be a great improve- 
ment. But the contention is made that the designer should keep his hands off the 
general arrangement. 
The general arrangement will be a good place from which to begin a considera- 
tion of the engine as a whole. No air compressor is shown because it is felt to be 
very bad practice to couple the air compressor to the main shaft as is too often 
done. 
Let us assume that the air compressor be coupled to the main shaft of the en- 
gine. In such an arrangement a fixed amount of air is handled by the compressor 
for each revolution. This arrangement is used when air is used for starting the 
engine. The result is that the amount of air which should be compressed is a varying 
quantity. After the starting cycle is completed the air consumed for starting must 
be replaced. To meet this condition the capacity of the compressor must be in excess 
of that needed for normal operation. If the engine is run a long time without a 
starting cycle, then obviously an undue amount of work will be done in the com- 
pressor. If the air compressor be a separate auxilary plant, then this variation in 
air supply can be handled without any excess of high-pressure air being accumu- 
lated. Again, let us assume that the valves of the compressor are not working 
properly and that the vessel is on arun. If the compressor is direct coupled, all that 
can be done is to stop the engine and correcct the trouble. It may be very essential 
that this be not done. If the air compressor be a separate plant, then it will be pos- 
sible to speed up the compressor to make up for the leakage, and the vessel can be 
kept under way. The loss in the air compressor will be appreciable, but this loss will 
be far less than that caused by stopping the vessel. This explains the reason for not 
having the air compressor direct coupled to the main shaft. 
The engine is divided into two sections of three cylinders each with the coup- 
lings, reverse gear and fuel pump placed between the two sets of cylinders. It is 
felt that by this arrangement the stresses in the cam shafts will be reduced below 
that of the arrangement where the cam shafts are driven from one end. There is 
space available due to this arrangement for placing the fuel pump between the two 
