STEADY STATE LOUD SPEAKER MEASUREMENTS 



151 



mouth. This curve differs considerably from the one obtained at a 

 distance of 12'. The marked depression in the curve at 750 cycles 

 checks very closely the first interference frequency as calculated for a 

 piston radiator approximately 30" in diameter and allowing for a 

 slight contraction of the radiating surface as the frequency is increased 

 (which assumption would be quite reasonable for the horn), the 

 irregularities at the higher frequencies are also explained in the same 

 manner. For the piston, however, the minimum pressure point would 

 be zero, which fact indicates that the wave-front at the horn mouth 

 either is not plane or is not of uniform intensity over the radiating 

 surface. Below 1,000 cycles the average trend of this curve is very 

 nearly parallel to the axis of abscissae while as noted for the curve 



Fig. 8 — Polar curves showing response (expressed relative to axis response) of 

 3J^" piston diaphragm loud speaker at various angles from perpendicular to center 

 of diaphragm and 12 feet away. 



obtained at a distance of 12', there is a very definite downward slope. 

 This is as would be expected if there were an increasing concentration of 

 the sound field about the axis as the frequency increased. The fact 

 that there is such a varying concentration is shown by the data on 

 Fig. 4. These curves were obtained with the condenser transmitter 

 at a distance of 12' in each case, but with a line from the center of the 

 horn mouth to the center of the transmitter making various angles 

 with the horn axis as specified. In making these measurements, the 

 transmitter remained fixed and the horn was rotated upward in a 

 vertical plane about the center of the mouth. It is apparent from 

 these curves that as the angle is increased the response at the higher 

 frequencies becomes lower, while at lower frequencies the change is 

 slight. The irregularities in the 45° curve are probably due to inter- 



