by Anthony Neal Watkins, anthony.n.watkins@nasa.gov; William E. Lipford, william.e.lipford@nasa.gov and Kyle Z. Goodman, anthony.n.watkins@nasa.gov, all NASA

For engineers developing and designing new aerospace vehicle concepts, the determination of the transition from laminar flow to turbulent flow on an aerodynamic surface is of vital importance. It can be critical for accurate drag estimation, for example, and there are efforts underway to design wing shapes and vehicle concepts that can delay this transition for drag reduction leading to decreased fuel usage.

Current research at NASA Langley Research Center is focused on developing a more efficient method for making these measurements on larger scale models tested at the National Transonic Facility (NTF) using Temperature Sensitive Paint (TSP) combined with a Carbon Nanotube (CNT)-based heating layer. The NTF is capable of operating at cryogenic conditions (down to 115 K), providing higher Reynolds number conditions that most closely resemble those of flight. In 2008, Dr. Robert Kilgore, a researcher at NASA Langley and a former CSA president, wrote a short review of the history and purpose of the cryogenic wind tunnel (www.cryogenicsociety.org/resources/cryo_central/wind_tunnels).

The transition from laminar flow to turbulent flow is most often indicated by the temperature difference on the surface caused by variations in the heat transfer rates between these flows. And laminar flow will generally have a lower heat transfer rate to the surface than turbulent flow.

There are several methods for making these surface temperature measurements. Unfortunately, at cryogenic conditions these methods suffer from various disadvantages that render them of limited use. A viable alternative is TSP, a method that engineers have used in cryogenic wind tunnels for over 20 years. TSP consists of a luminescent dye that is dispersed within a binder. When the TSP coating is illuminated with light of an appropriate wavelength, the dye in the binder will luminesce. This luminescence varies with temperature, and this technology has been used in many aerospace and non-aerospace applications to provide global surface temperature measurements.

In traditional transition detection experiments at cryogenic facilities, a temperature step is introduced into the tunnel to enhance the natural temperature change due to transition. Depending on flow temperature and local Mach number, this can be on the order of 0.1 K or less. This step is usually accomplished by rapidly changing the liquid nitrogen injection rate into the tunnel in either a positive (less nitrogen flow, resulting in a temperature ramp-up) or a negative (more nitrogen flow, resulting in a temperature ramp-down) direction. While quite effective in increasing the temperature experienced on the model, this can add a significant cost in terms of data acquisition time and facility operation. There can also be a significant change in the local flow conditions during the step.

Researchers have presented recent work, however, combining TSP with a CNT-based heating layer that can locally increase the temperature of the model surface when current is flowed through it. This method was demonstrated at the German Aerospace Center, DLR, but the CNT layer suffered degradation and ceased operating at 130 K. Researchers at NASA Langley have extended this operating range with a CNT-based heater capable of operating down to at least 77 K.

Verification of the TSP/CNT system was performed in laboratory testing on several samples. The process consisted of developing application techniques to ensure even heating, as well as determining the CNT survivability at cryogenic temperatures. Figure 1 shows representative heating (imaged using an IR camera) on a 0.093 m2 plate after applying increasing power to the CNT layer. The heating is fairly consistent over the surface, though there is some variation due to slight differences in application. To test the survivability of the TSP/CNT system at cryogenic temperatures, the NASA team immersed samples in liquid nitrogen until the temperature equilibrated to 77 K. The researchers observed no physical damage to the samples, and the CNT layer was still functional with no loss of efficiency.

Once the application techniques were determined, a panel of approximately the same size was coated with the TSP/CNT system and cooled in a large cryogenic chamber equipped with a window for imaging. The engineers then cooled the chamber to 110 K (the approximate lower temperature limit for the NTF) and luminescence from the TSP was measured as power was applied to the CNT layer.

Figure 2 contains typical results, showing that the CNT layer is functional with relatively even heating. Researchers took these measurements at five distinct spots on the panel, showing minimal temperature variation at higher power levels (~200 W) and measurable changes in the luminescence with small amounts of power applied (22W).

Final verification of the TSP/CNT system was performed on a High Speed Natural Laminar Flow (HSNLF) airfoil in the NASA Langley 0.3 m Transonic Cryogenic Tunnel (0.3 m TCT). The 0.3 m TCT is generally considered a pilot tunnel for the NTF in which new measurement techniques and initial vehicle concepts can be tested before transitioning to the larger NTF tunnel.

The HSNLF airfoil was constructed from aluminum with a chord of 0.165 m and a span of 0.330 m. The upper surface of the airfoil was coated with the TSP/CNT system and electrical excitation of the CNT layer was provided by parallel conductors placed about 12 mm from the end plates. The inset of Figure 3 shows the painted model.

Determination of the transition location was accomplished by first establishing the flow inside the test section of the tunnel. Then a series of images were acquired to collect the TSP luminescence. After several images are collected, power is applied to the CNT layer and the TSP layer is imaged further.

Figure 3 shows results that researchers recorded using one of the images acquired with the CNT layer on and rationing it with one of the TSP images obtained without power applied to the CNT layer. The team obtained the images at a temperature of 200 K and a flow velocity of Mach 0.7. In these images, the tunnel flow is left to right. The lighter regions indicate laminar flow and the darker regions are indicative of turbulent flow. These images were acquired by applying different amounts of power to the CNT layer, and even the lowest power setting (83 W) still shows the transition location. The team acquired similar data at various Reynolds number conditions down to 110 K.

The researchers also compared the method with traditional temperature step methods of increasing the liquid nitrogen injection rate. Figure 4 shows a comparison of the methods that indicates that the CNT layer provides similar results to temperature steps produced using both a “fast” injection rate and a “slow” injection rate.

However, if the fast injection method is employed, both the Mach number and the Reynolds number change by more than 10 percent. While this change was not seen using the CNT heating layer or the slow injection method, this data could be acquired in about 10 s with the CNT layer as opposed to more than 30 s with the slow injection method.

This increase in operating efficiency could provide a significant savings in tunnel operation costs, especially when moving to the larger NTF facility. Due to the success of this testing, a series of tests at the NTF are planned on larger scale models to gauge the increase in efficiency and data quality that can be achieved using the TSP/CNT system. ■