LOUDNESS 



421 



The circuit shown in Fig. 15 was employed to generate and control 

 the reference tone and the sounds to be measured. Vacuum tube 

 oscillators were used to generate pure tones, and for complex tones and 

 other sounds, suitable sources were substituted. By means of the 

 voltage measuring circuit and the attenuator, the voltage level 

 (voltage level = 20 log V) impressed upon the terminals of the re- 

 ceivers, could be determined. For example, the attenuator, which 

 was calibrated in decibels, was set so that the voltage measuring set 

 indicated 1 volt was being impressed upon the receiver. Then the 

 difference between this setting and any other setting is the voltage 

 level. To obtain the intensity level of the sound we must know the 

 calibration of the receivers. 



FEEDBACK 



CONTROL 



I AWWVW— 



r 



OSCILLATOR 

 1000 

 CRa 



ATTENUATOR 



— AMPLIFIER 



Z^ 



7^ 



THERMO- 

 COUPLE 



CALIBRATED VOLTAGE 

 MEASURING CIRCUIT 



o HEADPHONES 



I ;v\^/vvvA/^ 



OSCILLATOR 

 FREQUENCY 



FEEDBACK 

 CONTROL 



ATTENUATOR 



AMPLIFIER 



Fig. 15 — Circuit for loudness balances. 



The observers were seated in a sound-proof booth and were required 

 only to listen and then operate a simple switch. These switches were 

 provided at each position and were arranged so that the operations of 

 one observer could not be seen by another. This was necessary to 

 prevent the judgments of one observer from influencing those of 

 another observer. First they heard the sound being tested, and im- 

 mediately afterwards the reference tone, each for a period of one 

 second. After a pause of one second this sequence was repeated, and 

 then they were required to estimate whether the reference tone w^as 

 louder or softer than the other sound and indicate their opinions by 

 operating the switches. The levels were then changed and the pro- 

 cedure repeated. The results of the tests were recorded outside the 

 booth. 



The typical recording chart shown in Fig. 16 contains the results of 

 three observers testing a 125-cycle tone at three different levels. Two 



