1878.] the Constants of the Cup Anemometer. 



287 



Y in terms of v and F. Conversely, if we nave a sufficient number of 

 values of V, v, and F, we can determine a, {3, and 7. 



Unfortunately, there is much difficulty in obtaining some of these 

 values. We cannot produce wind of known velocity, and must substi- 

 tute for it the translation of the anemometer through the air with a 

 known speed.* 



The most convenient mode of doing this which occurred to him was 

 the attaching the anemometer to a whirling machine. In this case, 

 however, the rotation of the apparatus causes an air- vortex, whose 

 motion must be subtracted from that of the arm to give what is 

 assumed as the effective Y. The means by which this vortex current 

 was measured are described; they show that it is exceedingly irre- 

 gular, and that it is accompanied by a radial current still more dis- 

 orderly. We get v with sufficient accuracy, bat it is otherwise with F. 

 Of this there are four kinds acting in these experiments. 1. That due 

 to the weight of the moving parts of the anemometer ; 2. That caused 

 by the action of a brake, intended to diminish v in respect of Y; 3. 

 That produced by the lateral pressure of its axle on its supports, pro- 

 duced by the wind, &c. ; and, 4. That due to the centrifugal force 

 arising from the circular track of the anemometer's centre, which in 

 the actual arrangement pressed the axle outwards. The modes of 

 measuring these are described ; but this part of the work proved far 

 from satisfactory. The ground where the apparatus was established 

 was affected by tremors from the action of machinery, which made 

 all these frictions variable; and as he had the use of it for a very 

 limited time, it was impossible to repeat the work with the precautions 

 taught by experience. 



The constant a was determined by connecting a strong clock- spring 

 with an arm of the anemometer and that of the whirling machine. 

 When the whirl was rotated the anemometer tended the spring till its 

 elastic force equalled the air's pressure on the cups. Then a brake- 

 friction was applied much beyond the power of the spring, which 



kept the anemometer in its place ; Y was determined, and the tension 



p 



given by a graduated circle fixed on the axle. Then a— . 



Y 2 



These measures showed that equation (I) contained no notable term 



* In 1845 it occurred to Dr. Eobinson to carry an anemometer on a railway, for 

 which full opportunity was offered by a valued friend, Mr. Bergin, then Secretary 

 of the Dublin and Kingstown Railway. He gave it up after carefully considering the 

 disturbing influences and the precautions that were required. The space to be 

 traversed should have no curves ; should be perfectly unenclosed ; should not be 

 very long ; and at each end of it an anemometer should be established to keep 

 record of any wind. The experimental instrument should be 20 or 30 feet before 

 the engine, so as to be clear of the air which it drives before it, and should be carried 

 by a platform so formed as to present little resistance, and 10 or 12 feet above it. 



