research: Transition Control At Mach 8 By Helium Gas Injection

Mike Etz

We have found that low levels of helium injection near the leading edge stabilizes the turbulent boundary layer, apparently causing a reverse transition to laminar flow to take place. These results have important implications for thermal insulation of hypersonic engine surfaces, controlling mixing at high Mach number and influencing shock structure in shock-wave boundary layer interactions.

Using our new hypersonic facility, we have made preliminary studies of the injection of helium into a flat plate boundary layer over a range of Reynolds numbers. Figure 1 illustrates very recent experiments at Princeton in which we have demonstrated that helium injection into a transitional boundary layer delays transition indefinitely, apparently by lowering the Reynolds number in the near-wall region and cutting off turbulent production.

Figure 2 shows a CO2 enhanced scattering image obtained 432mm from the leading edge of a flat plate model in a Mach 8 flow. The boundary layer is turbulent, with a Reynolds number based on momentum thickness Req = 3200. Figure 1b shows images obtained at the same location but with helium injected into the boundary layer at a location 87mm from the leading edge and at a temperature which matches the wall recovery temperature. The dramatic difference due to the helium injection is clearly evident. Interestingly, injection of air at the same location with the same momentum flow rate has no significant effects, and it produces boundary layer images similar to those shown in Figure 1a. While the phenomenon illustrated if Figure 1b is far from understood, it appears that the injection of helium effectively reduces the Reynolds number near the wall (the kinematic viscosity of helium is a factor of 8 higher than that of air at the same temperature). Transition is delayed, and no turbulent mixing takes place. Surveys show that the low density region remains trapped near the wall (see Figure 2), apparently because mixing has been inhibited.

A key question is how long this effect will persist downstream, and if a similar injection into a fully turbulent flow will cause reverse transition to a more laminar-like state. The phenomenon has important implications for heat transfer to the wall from the external flow: if the wall region is effectively laminar, the wall will be insulated to a much greater extent than if the flow is turbulent.