The effects of sidewall boundary layers in two-dimensional transonic airfoil testing were investigated using oil-flow or liquid crystal visualization techniques. Three different chord models were tested in order to clarify the sidewall effects and to seek a suitable aspect ratio of the airfoil. The oil-flow visualization data systematically reveal the surface flow patterns affected by the sidewalls and suggest a minimum aspect ratio for conducting reliable two-dimensional tests. The results of the liquid crystal visualization also show the three dimensionality of the transition behavior and the necessity of the high aspect ratio. In addition, investigations on effects of the sidewall boundary-layer suction and application of a sidewall interference correction produce significant results for improvement of airfoil testing by removal of the sidewall effects.

Flow visualization studies on sidewall effects in two-dimensional transonic airfoil testing

A correction to account for the sidewall boundary-layer effects in two-dimensional airfoil testing is presented by taking into consideration the nonlinear variation of the crossflow velocity across the width of the tunnel. The crossflow effects in the wind tunnel are represented in an empirical manner by considering the inviscid, compressible flow between a straight and a wavy wall. The analysis shows significant reduction in sidewall boundary-layer effects on airfoil midspan measurements with increasing aspect ratio of the model. Application of the correction to wind-tunnel data on airfoils demonstrated the method to be satisfactory in correlating shock location and also in giving good agreement between the measured pressure distribution and computed free air predictions. shed due to loss of lift in the boundary layer and a consequent change in the effective angle of incidence. The second approach, proposed independently by Barn well2 and Winter and Smith,3 assumes the changes in the boundary-layer thickness due to model-induced pressure field to have a significant effect on the flow over the airfoil. Using the small disturbance equation and accounting for the changes in the width of the flow passage, Barnwell presented a simplified correction for the measured forces in the form of a modified Prandtl-Glauret rule to account for the attached sidewall boundary-layer effects. Barn well's correction gave good agreement with the experiments of Bernard-Guell e4 at low Mach numbers. Sewall5 extended Barn well's approach to transonic speeds using the von Karman similarity parameter. Recently, an alternative simpler form of the correction encompassing both the Barnwell and Sewall methods has been proposed by Murthy.6 The Barnwell-Sewall correction appears to be effective in giving a satisfactory comparison between the measured and calculated pressure distributions on several airfoils tested in the NASA Langley 0.3-m Transonic Cryo- genic Tunnel.7"9 These studies suggest that the change in the sidewall boundary-layer thickness due to the airfoil pressure field can be a significant source of blockage correction, particularly at transonic speeds. The Barnwell-Sewall correction has been derived under certain assumptions of simplified boundary-layer treatment and linear variation of the crossflow velocity across the width of the tunnel. These assumptions imply that the airfoil chord is sufficiently large so that the effect of the sidewall boundary layers can be considered to be quasi-one-dimensional. Barn- well10 has shown recently that the linear crossflow assumption is justified provided (4d*/b)(b/c)2 is small. Hence, this assumption is likely to become less accurate when the width of the tunnel is much larger than the airfoil chord (i.e., for high- aspect-ratio models).

Effects of aspect ratio and sidewall boundary-layer in airfoil testing

Reynolds number effects noted from selected test programs conducted in the Langley 03-Meter Transonic Cryogenic Tunnel (03-m TCT) are discussed The tests, which cover a unit Reynolds number range from about 20 to 800 million per foot, summarize effects of Reynolds number on: (1) aerodynamic data from a supercritical airfoil, (2) results from several wall interference correction techniques, and (3) results obtained from advanced, cryogenic tests techniques The test techniques include: (1) use of a cryogenic sidewall boundary layer removal system, (2) detailed pressure and hot wire measurements to determine test section flow quality, and (3) use of a new hot film system suitable for transition detection in a cryogenic wind tunnel The results indicate that Reynolds number effects appear most significant when boundary layer transition effects are present and at high lift conditions when boundary layer separation exists on both the model and the tunnel sidewall

A summary of Reynolds number effects on some recent tests in the Langley 0.3-meter transonic cryogenic tunnel

The NDA's (National Defense Academy of Japan) 0.06-m*0.3-m cryogenic wind tunnel was constructed in 1985 for testing transonic airfoils and for other basic research of fluid mechanics. Stainless steel SUS 304 was chosen as the material of the pressure shell, and a centrifugal compressor was chosen as the compressor. External insulation was adopted for the tunnel. Although no information was available on problems of using SUS 304 as the externally insulated tunnel pressure shell at cryogenic temperature and the thermal conductivity of SUS 304 is worse than that of aluminium alloys, only eight thermocouples were installed to monitor the thermal condition of the shell. The original temperature control was achieved by manually controlling the mass flow of liquid nitrogen injected into the tunnel circuit, but that system was found to be inadequate because the settling time of the total temperature took about 15 min in the change of rotational speed of the compressor. The total pressure control systems were modified to simple automatic PID (proportional, integral, derivative) controls. As a result, the pressure control can be achieved almost perfectly, and the temperature control is also greatly improved as the settling time of temperature is greatly reduced. >

Instrumentation and operation of NDA cryogenic wind tunnel

Preliminary Test Results of NDA Cryogenic Wind Tunnel and Its System

An investigation was carried out on two CAST 10-2 airfoil models with chords of 3 in. and 6 in. To evaluate the extent of sidewall influence on airfoil tests at transonic Mach numbers. The tests were conducted in the Langley 0.3-m Transonic Cryogenic Tunnel two-dimensional test section equipped with an upstream sidewall boundary layer removal system which reduces the boundary layer displacement thickness to about 1 percent of model halfspan from an initial 2 percent without boundary layer removal. Test results have shown the changes in the location of the shock on the upper surface of the airfoil to be about the same for both models with and without sidewall boundary layer removal. Even though large differences were noted in the high lift characteristics of the two models, the sidewall boundary layer removal had little effect on the differences. These tests also served to validate the boundary layer removal technique and the associated Mach number correction required with upstream boundary layer removal.

C. B. Johnson

Papers

Investigation of sidewall boundary layer removal effects on two different chord airfoil models in the Langley 0.3-meter Transonic Cryogenic Tunnel