THE MEASUREMENT OF ACOUSTIC IMPEDANCE 7 



and the interior of the tube was closed by a flexible piece of leather. 

 To prevent vibrations of the magnet from getting to the tube, the 

 magnet was held in position by flexible supports. The exploring tube 

 / was about 5" long with a 1/16" bore which led to the transmitter, T. 

 The voltages generated by the transmitter were measured with an 

 amplifier and an a.c. potentiometer. The potentiometer was used 

 because with it small voltages can be measured and errors due to har- 

 monics are avoided. The proper functioning of the apparatus was 

 determined by measuring the coefficient of reflection with no absorbing 

 material in the piston. Theoretically the reflection should then be 

 practically 100 per cent. The pressure ratios that were actually ob- 

 served were of the order of 12,000 which corresponds to a reflection 

 coefficient of 98 per cent. Evidently some extraneous pressures or 

 voltages were still present. However, no attempt was made to reduce 

 these further as the materials tested had a reflection coefficient con- 

 siderably less than this value. 



Experimental Results 



A brief study was made of the absorption of hair felt, as there is an 

 appreciable variation in the data given by various investigators on the 

 absorption frequency characteristic of felts of presumably the same 



1.00 

 .90 



.80 

 .70 

 .60 

 .50 

 .40 

 .30 

 .20 

 .10 



30 60 125 250 500 1000 2000 



FREQUENCY -CYCLES PER SECOND 



Fig. 3 — Power absorbed by hair felt 



4000 



type. After measurements on several samples it was evident that 

 concordant results could not be expected as the absorption varied con- 

 siderably with the packing of the felt. This point is illustrated by the 

 curves shown in Fig. 3. These curves were all obtained on the same 



