108 SURFACE CONDENSERS. 
nection to know the condition of the condenser tubes at the time these tests were made, 
since the condition of the tubes, with respect to cleanliness, has a tremendous effect 
on the heat transmission. 
The water velocity given in the table is 414 feet per second. In this connection 
I think it is generally better in marine condensers with 34-inch tubes to use a water 
velocity between 5 and 7 feet per second. The higher water velocity in the tubes creates 
a turbulence in the water which materially increases the heat transfer. In a great many 
marine condensers, as has been stated, the water velocity is very frequently around 
3 feet per second. I do not know of any case where it is as low as 1 foot per seocnd, 
but even 3 feet a second is lower than can be used with advantage, in my opinion. 
Another point which has been brought out is the steam velocity in the condenser— 
that is, the velocity of steam, especially in the top rows of tubes of the condenser. The 
reason for reducing this velocity is so as not to have a large pressure drop from the top 
to the bottom of the condenser, and I think it is good practice to limit this velocity in 
the upper tubes of the condenser to about 100 feet per second. That can be done in 
a number of ways. It can be done by providing a large space for the steam to enter 
tubes which are equally spaced, or it can be accomplished by larger spacing of the tubes, 
or you can arrange large V-shaped openings into the tube tanks, permitting the steam 
to go down that way. The principal thing is to limit the velocity to about 100 feet per 
second. 
The velocity of water in the tubes necessarily determines the length of the tube, 
and we have found that about 12 feet length of tube for two pass condensers, giving a 
total length of 24 feet, generally gives very good results. When the tube length is 
reduced the velocity will also be reduced, unless an additional pass is added to the con- 
denser. You can get the same results with an 8-foot tube with three passes as you do 
with a 12-foot tube with two passes. 
The efficiency of a condenser depends a great deal on the air-pump efficiency. If 
the air pump extracts the air efficiently, so that there is no air blanket around the tubes, 
the heat transmission will be much higher than if there is an air blanket which prevents 
the proper transmission through the tube surface. 
In all condenser tests—that is, tests where you are not testing the condenser plant 
as a whole, but testing the condenser alone—the air-pump capacity should be greatly 
in excess of what would be otherwise in order to leave no question as to the air blanketing 
of the tubes. In a condenser test, this would materially change the results. In this 
connection, attention might also be called to the fact that in a great many modern con- 
denser plants in ships, not enough attention is paid to proper air tightness, and I do 
not believe the present practice in installing condensers in most shipyards is as good as 
it ought to be. I do not think that you can say that a condenser and a vacuum system 
is tight unless you fill up the system with water and put it under pressure. I do not 
believe it is possible to get a condenser air-tight unless you do that. The fact that small 
leakages have such a considerable effect in high vacuum condensers makes this feature 
very important. 
With reference to the question of coefficient of heat transfer, this depends on the 
cleanliness of the tubes, both inside and outside, and on the air contained in the condenser. 
Oil especially is one of the things which reduces the heat transmission efficiency in marine 
condensers. Theoretically, turbine condensers should be perfectly clean, but actually 
they are not always so, since most of these condensers are used for condensing the steam 
