620 



BELL SYSTEM TECHNICAL JOURNAL 



of these exists, each of which would agitate the tube in a different way. 

 However, from tests made with a variety of mounting arrangements 

 for the tube under test and a variety of degrees of intensity and points 

 of appHcation of forms of impact agitation, it may be concluded that in 

 practical set-ups these factors may be varied widely without changing 

 the general nature of the microphonic level measurements greatly. 

 That is, the form and breadth of the distribution curve and the scatter- 

 ing of the points on the reproducibility chart for any typical group of 

 tubes are likely to be quite similar to Figs. 1 and 2, respectively, for 

 almost any practical impact agitator. 



Although the general nature of the results obtained with various 

 combinations of these agitator and mounting arrangements is about the 

 same for all of them, there are certain particular dififerences, which 

 show up chiefly in two characteristics. One is that the mean noise 

 level of a group of tubes is in general not the same for different mount- 

 ings and methods of agitation. That this must be true is fairly ob- 

 vious and needs no comment. The other is illustrated in Fig. 4, 

 which is a correlation chart showing typical results of measurements 



5 



CO 

 UJ-I 



Q.11J 



Q. tD 



zQ 



15 



ujz 

 ^-o 



Fig. 



30 32 34 36 38 40 42 44 46 48 50 52 54 56 58 



TEST ON APPARATUS RACK 



MICROPHONIC LEVEL IN DECIBELS BELOW 1 VOLT 



4 — Comparison of two tube mountings with a constant, artificial, mechanical 

 stimulus (100 No. 102F Tubes). 



