A new counter signal processor for true time-averaged fluid velocity measurements

This paper evaluates the averaging procedures for 'burst-mode' laser-Doppler measurements and the corresponding bias corrections commonly used to obtain the mean and turbulent velocity information. It is explained and experimentally shown that most bias corrections transfer the usually measured particle-averaged information into the desired time-averaged information. Results obtained with different velocity bias correction methods, when applied to an idealized flow with well defined time scales, are presented. Comparisons are also made with an alternative averaging method which yields time-averaged information from the individual realizations. It is shown that this latter method brings the averaged velocity properties closer to the theoretically expected values than other 'bias-correction' methods.