NASA’s F-15B #836 is a two-seat version of the F-15, which is a high-performance, supersonic, all-weather fighter airplane. The F-15B is used as a test-bed aircraft for a wide variety of flight experiments. In support of this use, a flight-test fixture (FTF) was developed to provide a space for flight experiments in a region with known aerodynamic conditions. The FTF is a fully instrumented test article mounted on the center line of the bottom of the fuselage of an F-15B airplane. The FTF includes an interchangeable experiment panel and is 107 in. (2.72 m) long, 32 in. (0.81 m) high, and 8 in.(20.3 cm) wide, with a 12-in. (30.5-cm) elliptical nose section. The FTF has been used in many flight experiments during the past several years and can be modified to satisfy a variety of research requirements.

Figure 1. The inside of the Experiment Panel and the Hot-Wire Assembly are depicted in this photograph.

One method of measuring turbulent fluctuations of density and velocity across the compressible boundary layer of an aircraft surface in flight (which fluctuations give rise to Reynolds stresses) involves the use of a recently developed automated hot-wire anemometry system. Prior to the development of the automated hot-wire anemometry system, a method of measuring turbulent velocity fluctuations in flight had not been perfected and routinely used in NASA’s flight experiments, primarily because of the limitations of conventional anemometry systems. Conventional anemometry systems are characterized by difficulties in tuning, poor signal-to-noise ratios and low bandwidths at low overheat ratios, sensitivity to electromagnetic interference, and vulnerability to effects of cable capacitance. The automated hot-wire anemometry system is, more specifically, a constant-voltage anemometry (CVA) system that has been shown not to be subject to the aforementioned deficiencies of conventional anemometry systems. The CVA system was selected for flight testing on the FTF on the F-15B airplane.

Figure 2. These Power Spectral Densities were computed from hot-wire-anemometer measurements taken at mach 0.9, an altitude of 15,000 ft (≈4.6 km), and a local Reynolds number of 2.8 × 107.

It is essential to characterize the turbulent boundary layer in flight experiments because the length scales characteristic of turbulence in wind tunnel experiments are significantly different from those of turbulence in flight. Thus, flight measurements in turbulent boundary layers are necessary for validation of computational fluid dynamics (CFD) computer codes and for predicting transitions from laminar to turbulent flow. The specific objective of the flight tests was to validate the concept of CVA for measuring velocity fluctuations in turbulent boundary layers in flight.

The flight-tested CVA system included four hot wires (two wires of 5-µm and two of 10-µm diameter) mounted on probes on the aft panel of the F-15B FTF (see Figure 1). The measurements taken with the wires of each diameter were used to verify the sensitivity and repeatability of measurements. The hot wire probes were mounted under the airfoil skin and were extended just prior to collection of data. The probes were extended at an angle of 45° into the boundary-layer flow to distances of 1, 2, 3, and 4 cm from the skin. Data were taken simultaneously for each wire during a time interval of 32 seconds at each flight condition. After collection of data, the probes were retracted. The associated electronic circuits, (including the power supply, CVA subsystem, and data-acquisition subsystem) were installed aboard the FTF.

In the tests, data were taken under various flight conditions ranging between mach 0.6 and mach 1.3 at an altitude of between 15,000 feet (≈4.6 km) and 30,000 feet (≈9.1 km). At the time of reporting the information for this article, analysis of the data was underway. Figure 2 presents a sample of processed data obtained from the flight tests. The raw data were sampled at a rate of 50 kHz and low-pass filtered at frequency of 20 kHz. The plots in Figure 2 show that there was significant turbulence at distances of 1, 2, and 3 cm from the skin. The turbulence was considerably lower at 4 cm from the skin, indicating the edge of the boundary layer. The peaks of the power spectral densities occurred at frequencies between 1,000 and 1,200 Hz. The turbulent energy cascade is clearly indicated, at frequencies above 12 kHz, by approximately constant negative slope at all four boundary layer locations.

The turbulent data signals were analyzed to determine the degree to which they approximated Gaussian functions. Those obtained at the distances of 1 and 2 cm from the skin were found to be Gaussian, while changes from Gaussian to non-Gaussian were found to occur at distances of 3 and 4 cm.

This work was done by Angela K. Beaver and Donald S. Greer of Dryden Flight Research Center. For further information, access the Technical Support Package (TSP) free on-line at www.nasatech.com/tsp  under the Mechanics category. DRC-01-15