ON METHODS OF DETERMINING THE DRYNESS OF STEAM. 409 



the general result obtained when the salt test has been used is that in 

 ordinary cases there is very little moisture in the steam. Some trials 

 made lately by Professors Kennedy and TJnwin gave the same result. 

 On the other hand, some trials simultaneously with the salt method and 

 one of the other methods, such as the wire-drawing or separating method, 

 have given this result : that the moisture in the steam indicated by the 

 salt method was only about one- tenth as great as that shown by the 

 other methods, and that not in steam taken from a steam pipe at a 

 distance from the boiler, but in steam taken directly out of the boiler 

 itself. 



It is clear that the salt test can only indicate boiler water mechanically 

 mixed with the steam by ebullition, not moisture produced in other ways. 

 It might be expected, therefore, that the salt method would give a larger 

 dryness fraction than other methods. Whether the great difference is 

 entirely attributable to this cause, or in part to inherent difficulties in the 

 salt method, cannot at present be ascertained. It would rather seem that 

 the Escher method, which apparently has never been tried, would give 

 the best results, if used in a case where the trial extended over a week or 

 more. 



A subordinate question is this. Some engineers have thought that 

 the salt method was specially suitable for boiler trials because it gave 

 directly what may be called the mechanical priming, although other 

 methods were more suitable for engine trials where what is wanted is a 

 knowledge of the absolute quality of the steam. This view seems to be 

 founded on a misapprehension. In determining the efficiency of a boiler 

 it is necessai-y to know how much of the feed leaves the boiler as steam 

 and how much as water. If, in steam taken directly from the boiler, there 

 is more moisture than corresponds to mechanical priming, that is as much 

 a deduction from the evaporation of the boiler as if it were mechanical 

 priming. The product of the boiler is so much wet steam, and the value 

 of this steam depends on its dryness. The total heat utilised depends on 

 the dryness. It is not at all important how the moisture came into the 

 steam provided it is there. The only point to attend to in a boiler trial 

 is that the steam tested should be taken directly from the boiler, and not 

 from a steam pipe in which condensation due to radiation may have been 

 going on. It is equally desirable that in an engine test the steam should 

 be taken from the steam pipe very near to the engine. 



Tests to Determine the Relative Trustworthiness of Different 



Methods. 



Test of the Wire-drawing Calorimeter. — The instrument tried was one 

 received direct from Mr. Barrus. It seemed extremely desirable to get 

 some kind of absolute test of the trustworthiness of this instrument, as it 

 is much the simplest to use for nearly dry steam. Especially it seemed 

 desirable to ascertain whether any correction for radiation or other loss of 

 heat would be necessary. This can only be tested by using steam of 

 known condition. The direction given in the instructions accompanying 

 the instrument is to test it with the steam from a boiler not doing much 

 work, the steam being then assumed to be dry saturated steam. But this 

 is inconclusive. 



To test this instrument, a small superheater was constructed, heated 

 by a row of gas jets. Then there is definitely superheated steam in both 



