MEASUREMENT OF NOISE KAYE 175 



measurers or recorders, but for a number of reasons their use has 

 been ahnost entirely discontinued in favor of electrical microphones, 

 particularly those of the condenser or electrostatic type. 



The condenser microphone consists essentially of a metal dia- 

 phragm tightly stretched and mounted parallel and very close to a 

 metal plate, the gap being only about one one-thousandth inch. The 

 thin section of air inclosed materially adds to the stiffness of the 

 stretched diaphragm so that its resonant frequency is very high — 

 usually above the normal range of acoustic frequencies. To the con- 

 denser formed by the plate and diaphragm a potential difference of 

 about 200 volts is applied through a high resistance. Sound waves 

 incident on the diaphragm cause it to vibrate, resulting in variations 

 of the capacity of the condenser, which, in turn, produces across the 

 series resistance an alternating electromotive force which can be 

 readily valve-amplified and so measured, by means of a rectifier 

 (such as a thermo junction) and microammeter. Alternatively the 

 wave form may be examined by a cathode-ray oscillograph. The 

 condenser microphone, though somewhat insensitive, enjoys the ad- 

 vantage of a fairly uniform response over the acoustic range of 

 frequencies, and thus provides a useful standard instrument, which 

 can be calibrated in absolute units. 



At the National Physical Laboratory and elsewhere in Great Bri- 

 tain such calibrations have usually been carried out by direct sub- 

 stitution with the Rayleigh disk in a simple sound-field. In the 

 United States, until recently, particular attention has been paid to 

 calibration by means of the thermophone, an instrument in which 

 the alternate heating and cooling of a gold or platinum leaf by a 

 fluctuating electric current produces calculable pressure oscillations 

 within a small inclosure. It is now generally recognized that these 

 two types of calibration lead to different results, and that the Ray- 

 leigh disk is required for the normal use of a microphone in free 

 air, and the thermophone if measurements are to be made of the 

 oscillatory pressure in a small inclosure, such as an artificial ear 

 canal. 



THE MEASUREMENT OF NOISE 



It was, I think, Lord Kelvin who said that once we find out how 

 to measure a thing, we begin to learn something about it. As regards 

 noise, however, it is evident from the foregoing that the question of 

 its measurement is one of some complexity, involving not only physics 

 but physiology and psychology. Nevertheless, as far as the physical 

 aspect goes, it is clearly desirable that there should be a consensus of 

 opinion on the choice of a system of physical quantities. They 

 should be preferably of an absolute character, so as to assist inter 

 alia: 



