Flow Control for Oblique Shock Wave Reflections

Background

Shock wave / boundary layer interactions can be encountered on transonic wings, on the turbine blades of jet engines, in the inlet of a supersonic jet and in many more situations. Because of their importance, they have been investigated for over decades now, but still the underlying processes are not fully understood and controlling these interactions remains difficult. However, new advances in experimental (Particle Image Velocimetry / Infrared Thermography) and numerical techniques (Large Eddy Simulations), make it possible to resolve more details than ever before and provide us with very detailed information of the interaction process.

My focus

My focus is on the oblique shock wave reflections occurring in supersonic jet intakes. The shock poses a strong adverse gradient and can cause the boundary layer to separate. A separated boundary layer not only increases the drag of the engine, but also reduces its performance and can in a worst case scenario cause the engine to unstart. Flow control strategies are therefore being developed to reduce the amount of separation by stabilizing the incoming boundary layer. Within my project two cases can be distinguished:

• Laminar / transitional boundary layer: To reduce the drag contribution of jet intakes it is beneficial to keep the boundary layer laminar for as long as possible. Laminar boundary layers are, however, also prone to separation and are not suitable for passing through shock waves. The question is therefore where and how to trip the laminar boundary layer into a turbulent state.

• Turbulent boundary layer: Turbulent boundary layers can also separate when the shock wave is strong enough or when the flow has to pass through a series of shocks. Micro-ramp vortex generators offer a promising flow control device within this context. They create two streamwise vortices that transport high-momentum fluid towards the wall, creating a fuller and more stable velocity profile.

Progress and Objectives

The experiments on the micro-ramp vortex generators are completed and the challenge is now to analyze all the data. The advantages of micro-ramps are clear in terms of separation probabilities and turbulence intensities. The exact working principles are, however, not yet fully understood. I am currently working on correlating flow features that enter the interaction with the instantaneous amount of separation. If the working mechanism of these ramps is better understood, then it should also be possible to design a more efficient ramp.

Regarding the effects of transition on the interaction, the basic elements are now in place. The models have been designed, were mounted in the tunnel and tested successfully. Methodologies were developed for processing the infrared thermography data and at the moment I am carrying out the first PIV experiments. Since the boundary layer is very thin, difficulties are expected with the seeding conditions, the time separation of the images and reflections coming from the model. These challenges will be dealt with in the coming 4 months.

Publications

- R.H.M. Giepman, G. de Giovanni, F.F.J. Schrijer and B.W. van Oudheusden, (2012) “Experimental Investigation into the Effects of Micro-Ramps on the Separation Bubble of Shock Wave / Boundary Layer Interactions”, 20th International Shock Interaction Symposium, Stockholm, August 2012

SR71 Blackbird jet intake

PhD Candidate: Rogier GiepmanDepartment: Section: Supervisors:

Promoter: Start date: Funding:

Aero

spac

e En

gine

erin

gAWEPAerodynamicsB. van Oudheusden & F. SchrijerF. Scarano01-12-2011FP7

Mounting a micro-ramp (MR) upstream of the interaction:

• The micro-ramp is most effective along its centre line, at 50% span the flow behaves virtually as if there is no ramp

• On average there is no separation taking place along the centre line. Instantaneously there still is, however a factor of 4 less than without a ramp

• Maximum turbulence intensity levels along the centre line are reduced by 10-15%

M>1 M<1

Supersonic mixed-compression intake

Interaction topology

x

zcentre

25%50%

Sonic lineSeparation bubble

Particle Image Velocimetry

U/U

∞U

/U∞

U’/U

∞

Along the centre line

Separation bubble

Turbulent interaction controlled by micro-ramp vortex generators

Effects of boundary layer transition on the oblique shock wave reflection

Micro-ramp vortex generator

To investigate the effects of transition on the interaction, one first has to determine the transition location itself, which is not trivial in supersonic flows with boundary layers less than 1 mm thick. Since, a turbulent boundary layer has a higher rate of heat transfer than a laminar boundary layer, this part of the plate will cool down faster in the tunnel. This can be visualized by means of infrared thermography and to increase the contrast, the plate is heated to about 65 degrees before the run.

The temperature distribution, however, provides a delayed representation of the transition location. Techniques were therefore developed to convert the temperature distributions into a heat flux distribution, which does show the instantaneous transition location. This delivers us the transition location xT as a function of the temperature at the transition location TT. A linear trend can be observed between the two variables. The colder the plate, the more downstream the transition location.

Before run t=17 s

Transition strips