Incompressible steady aerodynamics using a standard finite element code

Abstract

Data were also acquired at a - 4-deg angle of attack, which allowed more complete measurement of pressure side heat transfer values. Comparing the -4-deg (pressure side) data with the + 4-deg (suction side) data showed that for the roughened airfoil the pressure side experienced lower heat transfer than the suction side. In computer codes, heat transfer from airfoils is often estimated by using cylinder-in-crossflow heat transfer values in the leading edge region and flat plate heat transfer values farther aft. Figure 6 shows the IRT data for the dense 2, 0-deg, without spray condition compared with the cylinder and flat plate heat transfer values. The heat transfer in the stagnation region for the dense 2 roughened airfoil agrees fairly well with Frossling's5 smooth cylinder laminar flow solution. Moving downstream on the airfoil, the heat transfer drastically increases, reaching a maximum level near s/c of 0.035, and then decreases to a level fairly consistent with turbulent flow flat plate heat transfer values.6 The measured Frossling numbers at specific Reynolds numbers are somewhat higher than their respective flat pjate turbulent values. However, the higher measured heat transfer may be due to the increase in surface area caused by the roughness elements (3-7% increase on each gauge for the dense roughness patterns) that was not taken into account in the data analysis. It may be mentioned here that the maximum heat transfer is in the same general region, if slightly aft, of ice horn growth observed during glaze ice accretion.7 Conclusions Local heat transfer measurements from a roughened NACA 0012 airfoil were successfully obtained in flight and in the NASA Lewis icing research tunnel using the method and apparatus described in this work. Major conclusions resulting from this study are as follows. 1) The addition of roughness to the airfoil surface drastically increased the heat transfer downstream of stagnation. The roughness elements disturbed the laminar boundary-layer flow and in some cases caused a transition to turbulent flow. 2) Comparison of the flight and tunnel roughened surface data showed that the general effect of increased turbulence was a slight increase in heat transfer, especially at the higher Reynolds numbers. 3) Generally, the roughened surface airfoil cases showed the suction side heat transfer monotonically increasing with angle of attack.