CFD Online Discussion Forums

CFD Online Discussion Forums (http://www.cfd-online.com/Forums/)
-   Main CFD Forum (http://www.cfd-online.com/Forums/main/)
-   -   solving moving airfoil flows at Re=5X10^4 (http://www.cfd-online.com/Forums/main/9727-solving-moving-airfoil-flows-re-5x10-4-a.html)

quarkz August 18, 2005 21:52

solving moving airfoil flows at Re=5X10^4
 
Hi,

I need to solve a moving (flapping) airfoil in 2d at Re between 2X10^4 - 1X10^5. mainly to get lift/drag. how should i approach it?

should i be using dns? Will it take months?

or should i be using RANS/LES with some turbulence models? in that case, what would be a suitable one?

Can some experts here give their view?

Thank you!

Adrin Gharakhani August 19, 2005 15:08

Re: solving moving airfoil flows at Re=5X10^4
 
If it is incompressible flow vortex methods would be ideal for this problem (you can do low resolution DNS at this Reynolds number range with no problem).

Adrin Gharakhani

Suriyanarayanan August 22, 2005 07:08

Re: solving moving airfoil flows at Re=5X10^4
 
hi,

Flapping aerofoil problems at reynold numbers of 10^6 have been solved using RANS, even with 0- equation model to reasonable accuracy, so yours should not be a problem. I suggest the spallart and allmaras or MENTER - SST as a good RANS model for your problem. However the most important thing to be noted is the amplitude of oscillation. While RANS models work reasonbly well for attached and light stall cases they fail miserably for deep stall cases. For deep stall perform LES or DNS.

You can get solutions in a matter of hrs depending on the type of method that you would be employing. Infact, using dual time stepping with convergence acceleration devices reduces the time to nearly half a day.It is a luxury for unsteady computations.

Suriya



All times are GMT -4. The time now is 15:52.