Effect of Reynolds Number

In the mid-1950s, the NACA encountered an unexpected aerodynamic scale effect related to the long fuselage forebodies being introduced at the time. This experience led to one of the more important and last­ing lessons learned in the use of free-spinning models for spin predic­tions. One particular project stands out as a key experience regarding this topic. As part of the ongoing military requests for NACA support of new aircraft development programs, the Navy requested Langley to conduct spin tunnel tests of a model of its new Chance Vought XF8U-1 Crusader fighter in 1955. The results of spin tunnel tests of a 1/25-scale model indicated that the airplane would exhibit two spin modes.[509] The first mode would be a potentially dangerous fast, flat spin at an angle of attack of approximately 87 degrees, from which recoveries were unsat­isfactory or unobtainable. The second spin was much steeper, with a lower rate of rotation, and recoveries would probably be satisfactory.

As the spin tunnel results were analyzed, Chance Vought engineers directed their focus to identifying factors that were responsible for the flat spin exhibited by the model. The scope of activities stimulated by the XF8U-1 spin tunnel results included, in addition to extended spin tunnel tests, one-degree-of-freedom autorotation tests of a model of the

XF8U-1 configuration in the Chance Vought Low Speed Tunnel and a NACA wind tunnel research program that measured the aerodynamic sensitivity of a wide range of two-dimensional, noncircular cylinders to Reynolds number.[510] The wind tunnel tests were designed and con­ducted to include variations in Reynolds number from the low values associated with spin tunnel testing to much higher values more repre­sentative of flight.

With results from the static and autorotation wind tunnel studies in hand, researchers were able to identify an adverse effect of Reynolds number on the forward fuselage shape of the XF8U-1 such that, at the relatively low values of Reynolds number of the spin tunnel tests (about

90,0 based on fuselage-forebody depth), the spin model exhibited a powerful pro-spin aerodynamic yawing moment dominated by forces produced on the forebody. The pro-spin moment caused an autorota­tive spinning tendency, resulting in the fast flat spin observed in the spin tunnel tests. As the Reynolds number in the tunnel tests was increased to values approaching 300,000, however, the moments produced by the forward fuselage reversed direction and became antispin, remaining so for higher values of Reynolds number. Fundamentally, the researchers had clearly identified the importance of cross-sectional shapes of mod­ern aircraft—particularly those with long forebodies—on spin charac­teristics and the possibility of erroneous spin tunnel predictions because of the low test Reynolds number. When the full-scale spin tests were conducted, the XF8U-1 airplane exhibited only the steeper spin mode and the fast, flat spin predicted by the spin model that had caused such concern was never encountered.

During and after the XF8U-1 project, Langley’s spin tunnel per­sonnel developed expertise in the anticipation of potential Reynolds number effects on the forebody, and in the art of developing methods to geometrically modify models to minimize unrealistic spin predic­tions, caused by the phenomenon. In this approach, cross-sectional shapes of aircraft are examined before models are constructed, and if the forebody cross section is similar to those known to exhibit scale effects at low Reynolds number, static tests at other wind tunnels are

conducted for a range of Reynolds number to determine if artificial devices, such as nose-mounted strakes at specific locations, can be used to artificially alter the flow separation on the nose at low Reynolds number and cause it to more accurately simulate full-scale conditions.[511]

In addition to the XF8U-1, it was necessary to apply scale-correction fuselage strakes to the spin tunnel models of the Northrop F-5A and F-5E fighters, the Northrop YF-17 lightweight fighter prototype, and the Fairchild A-10 attack aircraft to avoid erroneous predictions because of fuselage forebody effects. In the case of the X-29, a specific study of the effects of forebody devices for correcting low Reynolds number effects was conducted in detail.[512]