Calibration of the Preston tube and limitations on its use in pressure gradients

Preston's method of measuring skin friction in the turbulent boundary layer makes use of a circular Pitot tube resting on the wall. On the assumption of a velocity distribution in the wall region common to boundary layer and pipe flows the calibration curve for the Pitot tube can be obtained in fully developed pipe flow. Earlier experiments suggested that Preston's original calibration was in error, and a revised calibration curve has been obtained and is presented here. From experiments in strong favourable and adverse pressure gradients, limits are assigned to the pressure-gradient conditions within which the calibration can be used with prescribed accuracy. It is shown that in sufficiently strong favourable gradients the ‘inner-law’ velocity distribution breaks down completely, and it is suggested that this breakdown is associated with reversion to laminar flow. As an incidental result, values have been obtained for the constants occurring in the logarithmic expression for the inner-law velocity distribution.