The Measurement and Reduction of Microphonic Noise 

 in Vacuum Tubes 



By D. B. PENICK 



The microphonic response of different types of vacuum tubes to the same 

 mechanical agitation covers a 70 db range of levels. Tubes of the same type, 

 on the average, cover a range of about 30 db. These response levels are 

 too sensitive to minute variations in testing conditions to be measurable 

 with any great precision, but values which are reproducible to within 5 db 

 are obtainable with a laboratory test set comprising a vibrating hammer 

 agitator, a calibrated amplifier, and a thermocouple galvanometer indicator. 

 Sputter noise is made measurable by frequency discrimination methods. 



Minimum microphonic disturbance under given service conditions is 

 attained by using the less microphonic types of tubes which are available, by 

 selecting the quieter tubes of a given type for use in positions sensitive to 

 mechanical disturbance, and by protecting the tubes from mechanical and 

 acoustic vibration. Examples of quiet triodes are the Western Electric 

 No. 264B (filament) and No. 262A (indirectly heated cathode). Indirectly 

 heated cathode type tubes are intrinsically less microphonic than filamentary 

 types. Further microphonic improvement in the tubes themselves is made 

 difficult by requirements for favorable electrical characteristics. Well 

 designed cushion sockets can reduce microphonic levels by as much as 30 db, 

 and other methods of cushioning, more expensive and less compact, can 

 extend the reduction even farther. Sputter noise can be eliminated almost 

 entirely in most types of tubes by commonly applied design features and 

 manufacturing methods. 



A MAJOR problem which has had to be met by every engineer who 

 has designed a high gain ampHfier is that of eHmination or reduc- 

 tion of noise. Noise of one kind or another, extraneous to the desired 

 signal, is always present in any amplifier, and sets a lower limit on the 

 smallness of the signal which can be amplified without intolerable 

 interference. In many experimental and commercial amplifiers, the 

 technical and economic obstacles to noise reduction necessitate a 

 compromise between inherent noise level and sufficient volume range 

 for ideal reproduction. The possible sources of this noise are numerous 

 and include power supply, faulty contacts, insulation leaks, pick-up 

 from stray fields, and many other disturbing elements. Among the 

 most persistent types of noise, however, requiring particularly careful 

 design for their elimination or satisfactory reduction, are three which 

 originate in the vacuum tubes themselves, namely, fluctuation noise, 

 microphonic noise, and sputter noise. 



Fluctuation noise has been treated at some length by Schottky,^ 



1 W. Schottky, Ann. der Phys., v. 57, p. 541, 1918; v. 68, p. 157, 1922. Phys. Rev., 

 V. 28, p. 74, 1926. 



614 



