The flow field in a high aspect ratio cooling duct with and without one heated wall

Abstract

The flow in a high aspect ratio generic cooling duct is described for different Reynolds numbers and for adiabatic as well as non-adiabatic conditions. The Reynolds number is varied in a range from 39,000 to 111,000. The generic cooling duct facility allows for applying a constant temperature on the duct’s lower wall, and it ensures having well-defined boundary conditions. The high-quality, optical noninvasive measurement methods, namely Particle Image Velocimetry (2C2D-PIV, i.e., two velocity components in a plane), Stereo Particle Image Velocimetry (3C2D-PIV, i.e., three velocity components in a plane) and Volumetric Particle Tracking Velocimetry (3C3D-PTV, i.e., three velocity components in a volume), are used to characterize the flow in detail. Pressure transducers are installed for measuring the pressure losses. The repeatability and the validity of the data are discussed in detail. For that purpose, modifications in the test facility and in the experimental setup as well as comparisons between the different measurement methods are given. A focus lies on the average velocity distribution and on the turbulent statistics. The longitudinal velocity profile is analyzed in detail for Reynolds number variations. Secondary flows are identified with velocities of two orders of magnitude smaller than the longitudinal velocity. Reynolds stress distributions are given for several different cases. The Reynolds number dependency of $\overline{u'^2}$ and $\overline{v'^2}$ is shown, and a comparison between the adiabatic and the heated case is given. $\overline{u'^2}$ changes significantly when the lower wall heat flux is applied, whereas $\overline{v'^2}$ and $\overline{u'v'}$ almost stay constant.