Unsteady Boundary Layer Transition Measurements with Infrared Thermography

Abstract

The boundary layer transition location is a crucial design parameter in aerodynamics. The computational prediction of boundary layer transition in engineering applications is typically based on empirical models. These models require experimental measurements for calibration and validation purposes. For unsteady aerodynamic processes, the range of suitable boundary layer transition measurement techniques is traditionally limited to fast-response discrete sensor techniques, e.g. hot-film anemometry. Introduced in 2014, differential infrared thermography is an alternative approach to measuring unsteady boundary layer transition using thermal images acquired with an infrared camera. The application of this optical measurement technique reduces the experimental effort, but problems emerge when the temperature response of the surface under investigation is slower than the aerodynamic unsteadiness. The idea of differential infrared thermography is to evade this problem by subtracting subsequent thermographs and accrediting the largest difference to the moving boundary layer transition location. The working principle of the technique has been demonstrated in various experimental setups. In a heat transfer simulation, it has been shown that the image separation time between the subtracted thermographs should be as small as possible to avoid systematic measurement errors. An experimental validation of these results is performed in the present study, conducted at the DLR in Göttingen. In this study, the infrared radiation from a pitching airfoil model suction surface in the wind tunnel at Ma = 0.15 and Re = 1 ⋅ 10^6 is measured with an infrared camera. The pitching frequencies f and amplitudes a_1 are varied in a range from f = 0.25 Hz to f = 8 Hz and a_1 = 1° to a_1 = 8° to study their effects on the boundary layer transition measurements. The differential infrared thermography technique is applied with several different image separation time steps to confirm the findings of the heat transfer simulation. The time step size is optimized as a compromise between the measurement error and the signal strength. In the next step, two alternative boundary layer transition measurement schemes using infrared thermography are developed, based on a quasi-steady model and a heat transfer simulation result. The first newly developed approach involves the automated selection of the appropriate image separation time step for each data point over the motion period when applying differential infrared thermography. This adaptive approach outperforms the conventional fixed time step approach when applied to experimental data from the wind tunnel test. The second newly developed approach is termed local infrared thermography. When analyzing the thermographic measurement results at fixed model locations, the extraction of the extrema of the measured radiation signal yields instants of the motion period that correlate with the occurrence of boundary layer transition. It is demonstrated that the local infrared thermography approach can be extended to measuring two-dimensional boundary layer transition fronts. The analysis of these results provides insight into the origins of the differences between the results of differential infrared thermography and the reference measurements of the unsteady boundary layer transition location with a fast-response technique.