Supersonic dispersion in gases II. Air containing water vapour

Abstract

It has long been known from incidental observation that humidity affects the propagation of sound in air. Not only is there an increase in the velocity of sound waves, but, refraction due to irregularities apart, there is a change in the attenuation, so that distant sounds may often be heard more clearly under conditions of high humidity than when the air is dry. Similar effects of humidity on the absorption of sound in reverberation chambers were reported by Sabine (1929), but it was left to Knudsen (1931, 1933) to show that the absorption of sound in air as derived from reverberation experiments at audible frequencies reaches a maximum (as the humidity increases) at first and then falls, and that when the frequency is changed the position of the maximum also moves. This anomalous behaviour of sound in moist air was pursued into the supersonic range of frequencies by Pielemeier. In 1937 (after the present work was commenced) he reported two humidity regions exhibiting abnormal velocity, with sources of sound covering the range 500-100 kc./sec. This year (1939) he has compared his measured velocities with those given by a formula of Miller, based on outdoor experiments, which indicates a steady rise of velocity with humidity. Whereas the earlier experiments of Reid (1930) and Ishii (1935) had shown no measurable departure from the Miller formula, Pielemeier attained values definitely higher at certain values of humidity and temperature.