New biasing elimination method for laser-Doppler velocimeter counter processing

Abstract

A new method is introduced for correcting the biased statistics of measurements with laser-Doppler velocimeters operating in the individual realization mode and using counters for signal processing. Two variations of the new correction method are proposed. One is suitable for all uniformly seeded flows; prior to averaging, the individual velocity measurements are weighted with the corresponding residence time of the seed particles in the probe volume. The other variation is suitable for flows with nonuniform seeding in which the average particle separation time is small compared with the time scale of the turbulent fluctuations, and in this method the particle separation time is used to weight the individual measurements. In both cases, a large number of samples is required. An experimental setup for realizing the new biasing correction method is described for each variation. The residence-time weighting method is applied to mean-velocity measurements in a uniformly seeded round jet. The results are compared with uncorrected measurements and with results using the McLaughlin–Tiederman correction. The uncorrected results show biasing errors of up to 25%, and the McLaughlin–Tiederman correction works only in the center portion of the jet where the instantaneous flow direction is nearly constant. The McLaughlin–Tiederman correction is restricted conceptually to this flow situation and to uniformly seeded flows. In contrast, the new correction method is not restricted to any particular flow situation. It supplements the new burst-gating counter system described in a parallel paper to form an especially accurate and versatile LDV-signal processor.