Aero-optical measurements in a subsonic, turbulent boundary layer with non-adiabatic walls

Abstract

This paper presents experimental studies of aero-optical distortions due to a turbulent boundary layer over a range of subsonic speeds with the underlying wall both heated above and cooled below the adiabatic wall temperature. A statistical scaling model, based on extended strong Reynolds analogy is derived and shown to correctly predict experimentally observed results. The temperature mismatch between the flow and the wall was shown to have a profound effect on the level aero-optical aberrations, as the heated wall amplifies them and the cooled wall significantly reduces distortions. The importance of an inclusion of the pressure term in explaining the boundary-layer density fluctuations with the cooled wall is also discussed.

Studies of Density Fields in Non-Adiabatic Boundary Layers Using Wavefront Sensors

“…2 American Institute of Aeronautics and Astronautics Sontag and Gordeyev AIAA-2017-3835 Aero-optical effects caused by the boundary layers are increased by heating the wall and are decreased by cooling the wall. This was experimentally observed in [8], and the ESRA was implemented in an attempt to explain the experimentally observed results. It was shown that while the ESRA qualitatively explained the functional trends of OPDrms(ΔT), it incorrectly predicted them quantitatively.…”

“…When the same approach was applied to explain the aero-optical distortion for non-adiabatic boundary layers, using the Extended Strong Reynold's Analogy, the resulting model [8] was still able to qualitatively predict the experimentally-observed results, but started missing quantitatively. In [8] this issue was discussed and several mechanisms were identified to potentially affect the model predictions. One of the main mechanism is the pressure fluctuations inside the boundary layer, neglected by the Extended Strong Reynold's Analogy.…”

“…One of the main mechanism is the pressure fluctuations inside the boundary layer, neglected by the Extended Strong Reynold's Analogy. While on average they are several times smaller than the velocity-related effects for adiabatic boundary layers, it was speculated in [8] that for moderately-cooled walls it might become one of the main factors.…”

“…As a consequence, the resulting density fluctuation profile is only a function of the velocity profile and ΔT. For subsonic speeds and replacing fluctuating terms with rms terms and Favre-averaged with Reynold's-averaged [8], (4) where ΔT = Tw -Tr is the temperature difference between the wall, Tw, and the recovery temperature, Tr, U(y) is the time-averaged profile, urms(y) is the fluctuation velocity profile, and A(y) was introduced in [10] to take into account the stress integral distribution in the boundary layer. As A(y) ≈1, it will be neglected in the following analysis.…”

“…For moderately cooled and heated boundary layers the buoyancy effects can be neglected, so the velocity field, and the correlation length, Λρ(y), will be largely unchanged. The density field, however, will be affected by the temperature difference, ΔT, between the wall temperature and the flow recovery temperature [8]. Therefore, the resulting wall-normal variation of OPDrms for nonadiabatic boundary layers will be 2 ( , ) = 2 2 2 ( , ) ( ) .…”