SURFACE CONDENSERS. 107 
of tubes and the outside shell which, in our opinion, is an advantage as it allows more 
easy access of the steam to the lower banks of tubes. 
The high heat transfer obtained in the test quoted by the author is higher than any 
obtained on land installations with which the writers are familiar. We would, however, 
hesitate to credit the condenser design with this higher heat transfer until the relation 
between the heat transfer on 5é-inch and 1-inch tubes has been accurately determined 
under similar test conditions. 
Mr. W. W. Situ, Member:—Mr. Lovekin has presented a most interesting paper. 
He brought out the points that condensers should be designed more scientifically, in 
which I entirely agree with him. There are basic formulae available for designing con- 
densers, and I think there would be considerable advantage in using them. 
All of the condensers designed in the engineering department of the Federal Yard 
are designed from basic formula and not from the square feet or surface per shaft horse- 
power. I might also call Professor Bragg’s attention to this fact, and that we also use 
a standard form for making these computations, so that the computer or designer cannot 
go astray in making the design for his condenser. However, it is permissible to design 
condensers on the basis of square feet of surface per shaft horse-power where the con- 
ditions are standard. If your water rate is the same as previously assumed, and if your 
water temperatures and other variables are the same, then it is permissible to use the 
figure of square feet per shaft horse-power, and it is quite as accurate as the other method, 
since the constant was originally determined by the scientific method. 
Mr. Lovekin has called attention to the unlimited supply of circulating water. It 
is the general practice to supply about eighty times the condensate for turbine vessels, and 
about forty times for steam engine vessels. There is considerable difference in the 
practice between these two types of machinery, and it would seem that the quantity of 
water for steam engine practice might be increased to some extent. 
In connection with the tests presented by Mr. Lovekin, I understand that the shaft 
horse-power and the quantity of steam condensed were not actually measured, but were 
estimated from curves of previous tests that were made. In view of this I do not think 
we can consider these results as entirely conclusive, since his tests were not based al- 
together on accurate measurements. 
In the table given by Mr. Lovekin he gives the heat per pound of steam as 988 in 
one case, and 999 in the other case. It would appear that these results are rather high 
for reasonably good turbines, and there is some question as to whether the heat units 
in the condensate were deducted in getting these figures. 
I have made a number of computations for similar turbines and found that the 
British thermal units per pound of steam range from 940 to 950, and in this table by 
deducting the British thermal unit per pound of condensate—that is, 82 degrees minus 
the 30 degrees, giving 52 degrees—and deducting this gives us a value of 958 and 949 
which would appear to me to be more nearly correct. 
Mr. Lovekin gives the coefficient K as 645 and 759, which is sometimes used. On 
the other hand, I think it is more convenient to use a formula for the heat transfer, in 
which K is equal to a coefficient C times the square root of the velocity since the velocity 
has such a large effect on the transmission. Converting Mr. Lovekin’s figures into the 
coefficient C by using the square root of the velocity, we get 307 in one and 306 in the 
other case. Both of these values are very high, and it would be interesting in this con- 
