4 Ckamv, Measuteinefit of Air Velocities. 



not exceed about lo in. of water while velocities of 40 ft, to 

 70 ft. per second are common, a good deal of trouble may 

 be experienced, and the characteristics of ordinary centri- 

 fugal and propeller fans are often falsified by the- error 

 referred to. 



Now Heenan and Gilbert, in 1895,* published some 

 careful tests on various orifices or tips with the object of 

 discovering one that would reduce these disturbing effects 

 to a minimum. The tips which they used are illustrated 



Fig. I. 



in their paper, and the method of test consisted in attaching 

 the tip to a long arm, which was rotated in a circular tank 

 7 ft. in diameter, provided with baffles to reduce the move- 

 ment of air in the tank. Free communication was estab- 

 lished by means of a special tube between the tip under 

 test and a recording water gauge. In this way, if there 

 were no induction the reading of the gauge should cor- 

 respond with the pressure due to the centrifugal force ot 

 the air in the rotating tube. From the speed the value 

 of the latter could be exactly calculated, and thus the 

 inductive effect was deduced. The best results were 

 * PiOL.Inst.C.E.,\'o\. cxxiii., pt. i. 



