Improvements to the Dynamic Range of Velocity Measurements Using DPIV

The accurate measurement of velocity fields containing large dynamic ranges is important because many problems of interest feature regions of both very high and low velocity. Examples include mixing tanks, jets injected into quiescent chambers, and stagnation regions behind bodies in high speed flow. However, accurate calculation of the velocity is currently limited to displacements greater than the total error of the scheme used (typically in the range of 0.01 to 0.05 pixels) and less than one-quarter of the window size. To counteract this difficulty, two new methods for improving the dynamic range of DPIV calculations have been developed. The first can be used with any double-pulsed time-resolved DPIV system where closely spaced frame pairs are captured at kilohertz rates. Displacement measurements can be made both within each frame pair (resolving high velocity regions), and between successive pairs (resolving low velocity regions). The two displacement fields are then reconciled, resulting in a single flow field measurement. The second method uses multiple laser pulses per camera exposure. Four laser pulses are required per measurement (two per camera frame), however, unlike the first method, kilohertz repetitions rates are not required. By carefully selecting the intervals between pulses, it is possible to associate the each cross-correlation peak with the correct delay time, and thereby simultaneously obtain velocity measurements over a much wider dynamic range. These two methods have been applied to synthetic and experimental data and their performance has been characterized through error analysis. Results indicate that both methods can increase the dynamic range by one to two orders of magnitude as compared to traditional techniques, while retaining similar total error and spatial resolution characteristics.