The local aerodynamic and heat transfer performance were measured in a rib-roughened square duct as a function of the rib pitch to height ratio. The blockage ratio of these square obstacles was 10 or 20 percent depending on whether they were placed on one single (1s) or on two opposite walls (2s). The Reynolds number, based on the channel mean velocity and hydraulic diameter, was fixed at 30,000. The aerodynamic description of the flow field was based on local pressure distributions along the ribbed and adjacent smooth walls as well as on two-dimensional LDV explorations in the channel symmetry plane and in two planes parallel to the ribbed wall(s). Local heat transfer distributions were obtained on the floor, between the ribs, and on the adjacent smooth side wall. Averaged parameters, such as friction factor and averaged heat transfer enhancement factor, were calculated from the local results and compared to correlations given in literature. This contribution showed that simple correlations derived from the law of the wall similarity and from the Reynolds analogy could not be applied for the present rib height-to-channel hydraulic diameter ratio (e/Dh = 0.1). The strong secondary flows resulted in a three-dimensional flow field with high gradients in the local heat transfer distributions on the smooth side walls.

The present experimental study is dealing with a detailed aero/thermal investigation of the turbulent flow inside a rib-roughened turbine blade cooling channel by means of Particle Image Velocimetry (PIV) and Liquid Crystal Thermometry (LCT). The main objectives of the paper are to provide detailed information about the behaviour of such a complicated flow, useful for its understanding, and to create a wide and reliable data base for numerical code validation. The measurements are carried out at engine representative Reynolds number within a scaled up model of a stationary straight cooling channel, with turbulent promoters (or ribs) installed on one wall. The ribs have an angle of attack of 90 deg with respect to the “mean” flow direction; their blockage ratio is equal to 30%. Detailed wall heat transfer distributions are presented. The main time-averaged flow features are identified and quantified; a number of rms characteristics are put in evidence and compared to the heat transfer distributions.