MICROPHONIC NOISE IN VACUUM TUBES 



629 



ference as indicated in the equivalent input noise column of the table, 

 compares quite favorably with that of the triodes. 



From the point of view of the user of vacuum tubes, constrained to 

 work with available types, the most effective means of microphonic 

 noise reduction is the use of one of the quieter types of tubes which 

 have been described. In cases where noise difficulties are experienced 

 in existing apparatus not readily convertible to the use of a quieter 

 type of tube, however, some relief may be gained by selecting the 

 quieter tubes from a number of the type to be used. To be fully effec- 

 tive, the selection should be based on measurements made while the 

 tube is in the socket in which it actually works. Under such circum- 

 stances, the measurements are reliable to within about 5 db. 



Where selection in the field is not feasible, a smaller degree of relief 

 may still be gained by selection at the factory. The degree of effective- 

 ness of this method can be deduced from Fig. 4. Suppose, for example, 

 that quiet tubes for service on the apparatus rack are to be selected by 

 a test made on the continuous tapper. Choosing the best 25 of the 

 group as tested by the continuous tapper (those plotted above the 

 horizontal line in the figure), it is immediately obvious that when these 

 selected tubes are tested on the apparatus rack (compare abscissae in 

 Fig. 4) the worst tubes are somewhat quieter than some of those in the 

 remaining portion of the group. This is more clearly shown in Fig. 8 



J 90 



$ 80 



z 

 < 



£ 70 



< 60 



O 30 



20 



34 36 38 40 42 44 46 48 50 52 



MICROPHONIC LEVEL MEASURED ON APPARATUS RACK 

 (PENDULUM AGITATOR) IN DECIBELS BELOW I VOLT 



Fig. 8 — Effectiveness of selection of quiet tubes. 



