I am currently working on the simulation of the turbulent flow around a helicopter fuselage. I still have to weigh different options wrt. the turbulence models. Since the flow is separated at many locations, I don't think conventional LES (either with Smagorinsky or dynamic SGS) will work properly. Is there anybody who knows some good references about DES? It seems to be the way to go for separated flows. Respectfully.

Hi there,

Check the following paper: "Approach to wall modeling in large-eddy simulations." Nikitin, N.V.; Nicoud, F.; Wasistho, B.; Squires,K.D.; Spalart, P.R. Physics of Fluids, v12, n7, 2000, pp 1629-1632.

Abstract: In this study, the detached-eddy simulation (DES) model was applied without adjustment as a subgrid-scale (SGS) model in the large-eddy simulation (LES) of channel flow. Following this, simulations were conducted on personal computers and work stations. Overall, the results are stable and fairly accurate.

I hope this helps. Sincerely,

Frederic Felten CFD Laboratory, UT Arlington. http://utacfdb.uta.edu/~felten

I'd love to see you try DES. I'm currently working on spinning an aircraft using DES along with Kyle Squires (http://www.eas.asu.edu/~squires/) and Phillipe Spalart using Cobalt60 (http://www.va.afrl.af.mil/vaa/vaac/cobalt/Co_main.html). You can check out some of our work (a very shoddy web page) at: http://www.usafa.af.mil/dfan/des.htm. My dissertaion was putting DES into Cobalt60 (an AF developed unstructured solver) and running it on a supersonic base flow. You can get a pdf of it, and the AIAA paper at: http://www.usafa.af.mil/dfan/forsythe.htm. It will provide the original DES references. Kyle's web page has some stuff on flow over a sphere as well.

If you are a US citizen you may want to consider running Cobalt - it has been run succesfully on a few helicopters (not using DES though). If you are interested, I can get you started with it. It is unstructured, so grid generation is relatively easy.

Please feel free to drop me an e-mail. I think DES will do no worse than RANS, and perhaps quite a bit better.

Jim Forsythe USAF Academy

As means to promote discussion, can someone tell me how the DES approach differs from where LES started? - when approximate wall treatments were used because one could not resolve the boundary layer? Excluding obvious academic marketing potential does it justify yet another label?

Hi,

I'm looking to gather all relevant articles relating to hybrid RANS/LES (DES) methods whic have been applied to unsteady flows.

Thanks