NEW DEVELOPMENTS IN HIGH VACUUM APPARATUS. 191 



With regard to the weights figuring upon the same quantity of air in both cases, the 

 twinplex runs from 2 to 2J^ times greater weight than the non-condensing ejectors with 

 turbine-driven condensate pumps, not approximately 7 to 1 as stated in the paper, which evi- 

 dently is comparing a very large capacity old-type twin with an ejector of limited capacity. 

 In the case of many cargo vessels, the items of space and weight are not of prime impor- 

 tance. 



It must be remembered also that the steam ejectors, on account of lack of elasticity, 

 must be installed in multiple units, and I would like to specially impress upon the members 

 the fact that the vacuum pump handles air and not steam, and the old basis of attempting to 

 rate pumps upon the quantity of steam handled must, before long, give way to an intelligent 

 selection of size based upon the quantity of air used. Referring again to Plate 68, assume 

 that the quantity of air is correctly given for 30,000 pounds steam as 15 pounds per hour 

 and for 90,000 pounds as 30 pounds per hour; the fact is that if the latter turbine, having a 

 normal steam constunption of 90,000 pounds, is operated at partial load, requiring only 30,- 

 000 pounds of steam, the air leakage will remain 30 pounds as given for the 90,000-pound 

 condition, provided it is not greatly increased owing to additional leakage at partial loads 

 as is the case in many designs of turbines. 



The President : — Is there any other gentleman who wishes to offer any comments on 

 this paper? 



Mr. William W. Smith, Member: — Mr. Kothny has contributed a very valuable pa- 

 per. It is original, covers the subject thoroughly, and contains much information which is 

 useful in condenser design. It is thought that more attention should be paid to' the scien- 

 tific design of condensing plants, especially so in the case of cargo vessels. There are sev- 

 eral elements that require to be considered in this connection, such as the velocity O'f the cir- 

 culating water in the tubes, the surface used, the quantity of cooling water, the temperatures 

 prevailing and the heat units to be transferred. 



In this connection, the table given by Mr. Kothny on page 175 is interesting, because he 

 states the different vacua which you might say are good practice, and which are recom- 

 mended for different temperatures of circulating water. It is quite important to note that 

 the vacuum is principally dependent on the temperature of the cooling water, and that ex- 

 tremely high vacua cannot be obtained with warm cooling water such as prevails under 

 tropical conditions. 



In this paper Mr. Ko'thny gives a vacuum of 2 inches, absolute for sea water of 70° F. 

 That appears to ]x the universal practice, and is considered a good rating for a marine 

 plant. If this design condition prevails — that is, 70° F. — then you will get different vacua 

 at the other temperatures of sea water. Very often the owners, in specifying the condensing 

 plant, will call for a vacuum of 38.5 inches at 85 degrees air temperature, which, of course, 

 is quite impracticable. 



Mr. Katzenstein in his discussion brought up the point of maintaining the vacuum at the 

 absolute pressure of a half inch of mercury for marine turbines. Turbines are not usually 

 designed to utilize as high degree of vacuum, and it would not be of much value if it were 

 produced. Most turbines will expand the steam to about 28.5 or 28.75 inches at full power. 



It should also be noted in this table that 60 degrees is the lowest temperature of sea 

 water given. Of course the temperature actually encountered will go much below this, and 



