112 



BELL SYSTEM TECHNICAL JOURNAL 



TABLE I 



Average Overload Points of Different Kinds of Program Measured 

 AT THE Output of the 94B Amplifier 



* These tests antedated the new standards, and the values given are in db with 

 respect to a reference point based on a single frequency calibration of .006 watt 

 in 600 ohms. 



TABLE II 



Spread of Overload Points Whose Averages Are Given in Table I 



Character of Program R-M-S V.l. 



Male Speech 6.1 db 



Female Speech 4.6 



Piano 3.6 



Brass Band 4.0 



Dance Orchestra 3.7 



All Types 7.3 



Peak V.I. 

 3.7 db 

 2.5 

 4.9 

 3.9 

 2.4 

 5.9 



either volume indicator. However, it will be noted that with the 

 r-m-s instrument the average overload point for speech is about 2 db 

 lower than for music, while there is no significant difference with the 

 peak instrument. This undoubtedly is because speech waves have a 

 higher "peak-factor" (ratio of peak to r-m-s values) than music. 



Table II shows the spread of the overload points (difference between 

 highest and lowest values) for the various tests on each type of program 

 whose average is given in Table I. Most of the types of program show 

 a significantly narrower spread for the peak than the r-m-s instrument. 

 For comparison with values taken from Figs. 5 and 6, discussed above, 

 these spreads should be divided by 2 to show the difference between 

 the lowest and the mean values. 



It is concluded from the tests just described that the disadvantage 

 in using r-m-s instead of peak-reading volume indicators for controlling 

 volumes to avoid aural distortion due to overloading, is substantially 

 none when the overloading device does not have too sharp an over- 

 loading characteristic, and only slight when it does overload sharply. 



