XXXI.The air-bubble viscometer

Summary 1. (a) It was desired to find whether, and under what circumstances, the rise of a large bubble in a vertical tube containing liquid could be used as an indication of the viscosity of the liquid.2. (b) The effects of length of bubble and diameter of tube have been examined and the rate of rise determined for several oils of known viscosity and surface tension, for water and for glycerin in different tubes. A dimensional analysis of the problem has been made involving certain assumptions which seem to be capable of verification, in any particular case, from the appearance of the bubble.3. (c) Above a certain limit the length of the bubble is without appreciable effect on the rate of rise. For bubbles rising “slowly” in a given tube, i.e. at such a rate that not more than one thickened annulus of liquid appears between the bubble and the walls towards the rear end of the bubble, it has been found that the rate of rise is inversely proportional to the kinematic viscosity, provided that the surface tension is constant. An approximate determination has been made of the form of the function of r/a upon which, in addition to the viscosity, the rate of rise depends. In certain conditions, e.g. when light oils are examined in a tube of such a diameter as to give a convenient rate of rise, the rate may vary about fourteen times as fast, as the diameter and six times as fast as the surface tension, but these effects become very much less pronounced when viscous liquids are used in wide tubes.4. (d) The air-bubble viscometer may be used with confidence for the approximate comparison of viscosities of materials of the same class. A simple modification is suggested which enables a check to be made on the assumed constancy of surface tension.