102 SURFACE CON DENSERS. 
tubes equal to about one-half the diameter or depth of the condenser, for no other 
purpose than to reach the bottom of the condenser so it can be drained off by the 
pump. Draining the condensate off in stages after it has been condensed appeared 
to be more rational, and the test data confirm this belief. 
It is a matter of common occurrence with this condenser to have the tem- 
perature of the condensate and the temperature corresponding to the vacuum within 
a fraction of a degree, and where the temperature of the outboard delivery is kept 
within 5 to 8 degrees of the temperature corresponding to the vacuum, the con- 
densate continues to keep close to the temperature corresponding to the vacuum. 
If a wider range of temperatures be used, the results are very pronounced; that is, 
if the outboard delivery temperature is within 12 to 14 degrees of the temperature 
of the vacuum in the condenser, the condensate will invariably show 10 to 12 degrees 
higher than the temperature of the outboard delivery, thus showing that when 
pumping more circulating water through the condenser the temperature of the 
condensate still remains high. It has long been known that infiltration of air 
through cast-iron condenser shells is the source of great trouble and annoyance with 
high vacuums. In order to reduce this to a minimum we made tests without any 
coating on the shell, then with one coat of bitumastic enamel, and finally with two 
coats of enamel. ‘The effect of the final coat was an improvement of nearly 1 inch 
of vacuum under the same conditions. 
Particular attention is called to the air channel way in these condensers, it 
being formed of two plates running the entire length of the space between the tube 
sheets, these plates being spaced apart and having their sides perforated with small 
holes over the entire plate. This is done so as to prevent any air pockets being 
formed in any part of the condenser. Where the condensate pump and air ejectors 
are used, such as is shown on the proposed surface condenser for United States 
battle cruisers, it will be noted that the condenser head is cut away so as to clear 
the air-ejector connection. This makes a very simple connection and yet prevents 
the air pipe being flooded with water. In order to ascertain the height of the water 
in the condenser shell, a water gauge glass can be connected between the condensate 
suction pipe and the shell which will readily show the height of condensate, if any, 
within the condenser shell. 
It will be noticed in the plan referred to that small angle bars are placed in the 
lower edges of all the drain plates. These angle bars are not continuous but simply 
consist of short pieces with clear spaces of equal length between them so as to pre- 
vent the formation of a continuous sheet of water the entire length of the condenser 
running into the condensate space, which might interfere somewhat with the vapor 
and air entering the secondary zone. 
It is almost unnecessary to state that the design of a surface condenser for 
battleships and the like which are subjected to their maximum speeds for compara- 
tively short intervals of time is an entirely different proposition from that of the 
merchant ships, where the maximum speed is run the greater part of the time. In 
a case of the battleships we can afford to force a greater amount of water through 
the tubes than we can in the case of a merchant ship, for the reason that the cost 
