36 SECTIONAL ADDRESSES 



the average of a trained group of observers) in an experimental chamber 

 with highly absorbent walls. Furthermore, while it may be possible in 

 the fundamental evaluation of the equivalent loudnesses of sounds or 

 noises in phons, for certain sources to be accommodated in the lagged 

 test chamber, it may not be practicable in other cases, whether on the 

 score of size or the impossibility of quickly starting and stopping the 

 source, so as to permit alternate listening with the reference tone. In 

 such cases, the sources have to remain outside the chamber and the 

 sounds are transmitted at will to the observer inside by means of a micro- 

 phone and loudspeaker system. A closer specification of the measuring 

 technique is at present being developed by the British Standards 

 Institution. 



' Phon {B.S.). A unit of equivalent loudness, defined as follows : 

 The standard tone shall be a plane sinusoidal sound wave train 

 coming from a position directly in front of the observer and having 

 a frequency of looo cycles per second. The listening shall be done 

 with both ears, the standard tone and the sound under measurement 

 being heard alternately and the standard tone being adjusted until it 

 is judged by a normal observer to be as loud as the sound under 

 measurement. The intensity level of the standard tone shall be 

 measured in the free progressive wave. The reference level shall be 

 taken to be that corresponding to an R.M.S. sound pressure of o- 0002 

 dyne per sq. cm. When, under the above conditions, the intensity 

 level of the standard tone is n decibels above the stated reference 

 intensity, the sound under measurement is said to have an equivalent 

 loudness of n phons (B.S.).' 



Reference may here be made to the criticism to which the phon scale 

 has been subjected in some quarters, in that it does not interpret 

 numerically one's sensations of loudness, a phon near the top of the 

 scale admittedly signifying a much bigger step than it does near the 

 bottom. Some workers in this country and the States have accordingly 

 attempted to set up a subjective scale of loudness units in which the 

 assumption is made that fractional estimation by an observer is possible, 

 so that, for example, it is claimed he can say when one sound is twice as 

 loud as another. A comparison of the mean scale so derived with the 

 phon scale suggests that such an estimated doubling of loudness is 

 equivalent to an increase of about 10 phons for moderately loud sounds. 

 While, however, the scale of equivalent loudness in phons may not be 

 subjectively ideal, it does rest on sound physical foundations and accepted 

 subjective principles. Furthermore the connection of the phon scale with 

 the decibel scale is of great practical value to the acoustical engineer, for 

 example in dealing with the transmission of air-borne sounds by walls. 



Noise Meters. 



It will be appreciated that the experimental conditions laid down for 

 the fundamental evaluation of sounds or noises in phons, are rarely likely 

 to be available for noise measurements in everyday circumstances, and 



