ACOUSTICAL INSTRUMENTS 401 



Of the more common piezoelectric crystals tourmaline possesses a 

 characteristic which renders it peculiarly suitable for the absolute 

 measurement of sound intensities at all audio frequencies, in that it 

 may be so cut into slabs that a potential difference will be developed 

 between its lateral surfaces when it is subjected to a purely hydro- 

 static pressure. Because of this characteristic Sir J. J. Thomson ^^ 

 suggested its use for measuring pressures in gun barrels. Such a slab 

 of tourmaline, having dimensions small compared with the wave- 

 length, except for its low sensitivity, is the ideal microphone. Tour- 

 maline is mechanically strong and its activity is practically constant 

 under all atmospheric conditions. Resonant frequencies in the slab 

 lie far out of the range of audio frequencies so that the response at all 

 frequencies is the same and is easily determined from static or low- 

 frequency measurements. Unfortunately the sensitivity of such a 

 device is low, some 70 db below that of a moving coil microphone. 

 In spite of this low sensitivity it can be used for calibrating other 

 microphones if sound waves of rather high intensities are used and if 

 the electrical circuit is provided with a band-pass element transmitting 

 only frequencies in the immediate neighborhood of the measuring 

 frequency. A measuring system of this character, unlike the Rayleigh 

 disc, is not subject to disturbances from circulating air currents. 



Thermometric Microphones 



As the pressure variations in a sound wave are accompanied by 

 corresponding variations in temperature, corresponding electrical 

 currents will be generated by a resistance thermometer or a thermo- 

 couple when placed in the sound field. The temperature variations 

 are of the order of 0.0001° C. per bar acoustic pressure. The use of a 

 resistance thermometer (for measuring these periodic temperature 

 variations) was first investigated by Heindlhofer,^^ and more recently 

 by Friese and Waetzmann,^* who found that at a frequency of 1000 

 c.p.s. a wire 0.0004 cm. in diameter will undergo temperature varia- 

 tions equal to about 0.15 of the variations in the surrounding medium. 

 To derive an alternating electric current from the periodic resistance 

 variations that follow the temperature variations, a direct current 

 must be passed through the wire. The heat generated by this current, 

 unless it is kept down to an extremely small value, will set up convec- 

 tion currents around the wire and so greatly complicate the operation. 



The thermocouple is entirely free from this objection, but is not 



readily constructed so as to have a heat capacity as small as the Wollas- 



^^ Engineering 107, 543 (1919). 



"^w?z.. d. Phvsik 37, 247 (1912): 45, 259 (1914). 



^^Zs.f. Physik 29, 110 (1925); 31. 50 (1925); 34, 131 (1925). 



