ARCHITECTURAL ACOUSTICS 



107 



Similar experiments in different rooms varying in volume 

 from 65 to 9,300 cubic metres gave a family of rectangular 

 hyperbolae as shown in Fig. 2, curves i to 7 being on one scale, 

 8 to 1 1 on a second, and 12 on a third. The parameters k are 

 found to be proportional to the volumes of the rooms, the 

 mean value of kjv being -171. 



The value of kjv will depend on the intensity of the initial 

 sound, and Sabine standardised his source as follows. Four 

 organ-pipes similar to the source were placed in different parts 

 of the room sufficiently far apart for it to be a justifiable assump- 

 tion that when all were sounding together the intensity was 

 four times that of the single pipe. Experiments were carried 



10 



6 



CO 



§6 



CO 



c 



2 

 Q3 



ID 20 30 40 50 60 70 80 QO 100 110 120 130 140 150 

 120 180 240 300 B60 420 



640 720 000 1080 1260 



Total absorbing mat^rlnj 

 Fig. 2. — Three scales are employed for the volumes by groups 1-7, S-ii, and 12. 



out to find the comparative duration of residual sound when 

 I, 2, 3 and 4 pipes were sounded. These yielded results which 

 showed that the sound decayed logarithmically and hence the 

 intensity I was found to be lo million times the minimum 

 intensity for audibility, Iq. 



It is to be noted that the initial intensity of the sound in 

 the room is slightly reduced by the introduction of the cushions, 

 and when allowance is made for this the value of kjv for I = io® Iq 

 is -164. 



Thus Sabine's experiments give for the time of reverberation 



■164 V 



/ 



a 



