COMPUTATION OF THE COMPOSITE NOISE 



613 



the laboratory how the customers would handle the trays. The points 

 on the curve indicate the peaks that could be measured in the cafeteria 

 which had an average composite noise of 66.5 db. It will be seen that 

 the lowest peaks that could be measured satisfactorily were at 74 db 

 sound level. The rest of the curve was estimated using the statistical 

 methods referred to above. 



The curves in Figs. 1 and 2 are plotted on "arithmetic probability 

 paper." On this paper, cumulative normal distributions appear as 

 straight lines. 



"^ so 



72 



O 



i^.. 



12 5 10 20 30 40 50 60 70 80 90 95 98 99 



PER CENT OF PEAKS ABOVE LEVEL SHOWN BY ORDINATE 



Fig. 2 — Distribution of noise peaks from metal trays in a cafeteria. 



Effect of Room Characteristics 

 Generally the noise sources are at various locations so that it is 

 necessary to determine how much the noise from each is reduced 

 by its distance from the observing point assumed for the computation. 

 Since it is not the primary concern of this paper to discuss the dis- 

 tribution and decay of sounds in rooms, only a very simple approxi- 

 mate method of computing distance losses, based on the classical 

 theory of the steady-state distribution of sound in a room, is given 

 here. This method has been found adequate for practical purposes 

 in rooms having relatively simple geometric shape and large enough 

 dimensions so that the sound is diffused. For a more complete treat- 

 ment of room acoustics the reader should refer to the literature on 

 this subject.* 



The total steady-state intensity, It, at an assumed observing posi- 

 tion in a room consists of two parts: Ir, the reflected sound intensity 

 and /d, the direct sound intensity, so that: 



It = Ir -\- Id- 



