Author(s): M. Moeini; B. Khorsandi; L. Mydlarski

Linked Author(s):

Keywords: Acoustic Doppler velocimeter; Sampling frequency; Turbulence statistics; Jet flow

Abstract: The effect of the (user-adjustable) sampling frequency of acoustic Doppler velocimeters (ADVs) on the measurement of turbulence statistics in a turbulent jet has been examined. The results demonstrate a remarkable sensitivity to the sampling frequency of the instrument. Acoustic Doppler velocimetry is a popular method for the measurement of velocity in hydraulic engineering applications. It has been found that ADVs accurately predict the mean velocity; however their precision in making turbulence measurements has been questioned. The effect of noise in ADV turbulence measurements has been investigated at the default frequency of 25 Hz by Khorsandi et al. (2012). It is expected that the noise in turbulence measurements is furthermore influenced by user adjustable device parameters, such as the sampling frequency of the ADV. These parameters affect how often the ADV outputs velocities (by averaging pings) and as a result they will have an influence on the Doppler noise level. To this end, an experimental work focusing on the performance of ADV operating at different sampling rates has been conducted to evaluate how the turbulence statistics may be affected. The velocity field of a turbulent axisymmetric jet with a Reynolds number of 10,000 issued into a background of quiescent water was measured via a Nortek Vectrino ADV with user-adjustable sampling rates ranging from 10 Hz to 200 Hz. Results showed a noteworthy dependence of the second-order statistics on the sampling rate. The axial and lateral RMS velocities, as well as the Reynolds shear stresses, corresponding to higher sampling frequencies exhibited overestimated values compared with those measured at lower sampling rates, and also with those obtained using other measurement techniques, such as laser Doppler anemometry and hot-wire anemometry. Moreover, the effect of the sampling frequency is more significant on the turbulence statistics in regions where RMS velocities are larger. The turbulence statistics are less affected by the sampling frequency close to the edges of the jet, where the RMS velocities decrease. On the other hand, the mean velocity remained quite independent of the change in the sampling rate. The role of dissimilar shot-noise levels and number of pings averaged, which prevails in differing choices of sampling frequency, in the measurement of turbulent statistics will be discussed in the full paper. The effect of the ADV’s sampling volume size on the mean velocity and turbulence statistics will also be studied. The outcomes of present study will provide recommendations for the use of ADVs, and in the analysis of their outputs in various field and laboratory applications.

DOI: https://doi.org/10.3850/978-981-11-2731-1_125-cd

Year: 2018