50 years of gas-phase velocity measurements in fire science: Revisiting the bidirectional probe

The bidirectional velocity probe, developed by McCaffrey and Heskestad in 1976, remains one of the most widely used instruments for gas-phase velocity measurements in fire research. Despite its extensive use, the original calibration methodology lacks a rigorous uncertainty quantification, and subsequent studies have not always produced consistent results. Therefore, a comprehensive experimental and analytical investigation of the probe’s uncertainty, angular response, and Reynolds number dependence was undertaken using the “Guide to the expression of Uncertainty in Measurement” (GUM) framework. Four probe diameters were calibrated in an open-jet wind tunnel across a Reynolds number range of 100 ≤ Re ≤ 8400 and incident flow angles between −60 ° and +60 ° relative to the stream direction. The results show deviations from the traditionally accepted asymptotic flow coefficient of 1.08. A phenomenological model was proposed to describe the probe’s angular sensitivity, enabling the integration of the uncertainty from the angular sensitivity into the overall uncertainty budget of the flow measurement. The combined expanded uncertainty in the flow coefficient was found to be approximately 2% with a coverage factor of k = 2. Furthermore, the uncertainty is expected to increase substantially when being used for dynamic fire tests. The study highlights the need for individual probe calibration and provides a detailed methodology for uncertainty propagation in experimental fire applications.