Dynamic stall on an airfoil

antm324 · February 27, 2015, 03:53

Hi,

hopefully a simple question for somebody to answer.

I am trying to use fluent to predict stall angle and post stall cl/cd.

I have been trying to run a 2d transient problem, both with DDES transition SST and the pressure and density models.

i am not yielding the results i would like to achieve. I am wondering if this is because i need to simulate this as a 3D problem?

I have very limited computational power and am running the student version of fluent at university.

Please if you can help or need any more information, please get in touch even if it just ideas i have been stuck for quite a while and do not have many resources at my university for CFD and no specialists.

Thanks
Anthony

antm324 · March 4, 2015, 10:22

has anybody got any information to input?

erlofg · March 18, 2015, 05:59

I don't know anything about the DDES model, but it should be possible with SST k-omega for instance. However, in the post-stall regime, the problem is highly unsteady. There is likely to be periodic vortex shedding requiring a fine mesh and small time step. Also, since cl and cd are surface forces it is important to properly resolve the boundary layer around the airfoil to get good accuracy. If possible, make sure that the y+ is around 1 in your first cells at the airfoil. And then you also need to make sure that the boundaries are far enough away from the airfoil, so that the flow is fully developed before it reaches the outlet.

Technically, there is to my knowledge no reason why this would be more difficult to simulate in 2D than in 3D so I think you should do fine. But note that you will probably not find steady solutions in the post-stall regime, so in order to get good statistics (mean drag etc.) you will probably have to do quite long simulations.

antm324 · March 18, 2015, 07:59

Hi Erlofg,

Thank you for your response. I have quite a few questions that i believe you may be able to answer if you are willing to help? Do you mind if i pose these questions to you? Ideally it would be best to have a live chat if possible and then i can post the discussion on the forum to also help benefit other users

Anthony

erlofg · March 18, 2015, 10:04

I'm not really an expert per se in these matters, but ask away and we'll see what I can answer.

It's better for me to take it here in the forums so that I may take some time to consider the answers...

er_trebbia · August 11, 2015, 12:27

Basically, you can't perform a DES in a 2d problem, because the LES part of your DES won't make any sense, since turbulence is 3d. Then, of course you have to use the 3d solver.

LuckyTran · August 11, 2015, 12:38

Quote:

Originally Posted by er_trebbia

Basically, you can't perform a DES in a 2d problem, because the LES part of your DES won't make any sense, since turbulence is 3d. Then, of course you have to use the 3d solver.

That's not exactly correct. There is such a thing as 2D turbulence (a degenerate state of turbulence), it just doesn't behave the same as 3D turbulence or ordinary turbulence.

2D DES/LES/DNS can be computed but the results are limited to simulating this degenerate 2D turbulence state and should not be used to even try to describe what happens in 3D turbulence.

er_trebbia · August 11, 2015, 12:42

Yes, but the problem here is an airfoil in post stall, if you hope to catch something like that with a 2D DDES, be my guest.

February 27, 2015, 03:53	Dynamic stall on an airfoil	#1
antm324 New Member Anthony McCarthy Join Date: Nov 2014 Location: United Kingdom Posts: 21 Rep Power: 11	Hi, hopefully a simple question for somebody to answer. I am trying to use fluent to predict stall angle and post stall cl/cd. I have been trying to run a 2d transient problem, both with DDES transition SST and the pressure and density models. i am not yielding the results i would like to achieve. I am wondering if this is because i need to simulate this as a 3D problem? I have very limited computational power and am running the student version of fluent at university. Please if you can help or need any more information, please get in touch even if it just ideas i have been stuck for quite a while and do not have many resources at my university for CFD and no specialists. Thanks Anthony

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
SU2 AOA optimization	454514566@qq.com	SU2	9	March 7, 2022 16:17
2D FFD Optimization	RLangtry	SU2	2	August 5, 2014 09:48
sliding mesh ? airfoil dynamic stall?	xiaomi	FLUENT	2	March 18, 2013 05:42
How to set the dynamic stall convergence criterion of airfoil?	shlgzz	FLUENT	0	January 10, 2013 21:45
dynamic stall simulation over an airfoil	mleahy	FLUENT	0	October 9, 2009 12:15

March 4, 2015, 10:22		#2
antm324 New Member Anthony McCarthy Join Date: Nov 2014 Location: United Kingdom Posts: 21 Rep Power: 11	has anybody got any information to input?

March 18, 2015, 05:59		#3
erlofg New Member Erik Löfgren Join Date: Feb 2015 Location: Umeå, Sweden Posts: 9 Rep Power: 11	I don't know anything about the DDES model, but it should be possible with SST k-omega for instance. However, in the post-stall regime, the problem is highly unsteady. There is likely to be periodic vortex shedding requiring a fine mesh and small time step. Also, since cl and cd are surface forces it is important to properly resolve the boundary layer around the airfoil to get good accuracy. If possible, make sure that the y+ is around 1 in your first cells at the airfoil. And then you also need to make sure that the boundaries are far enough away from the airfoil, so that the flow is fully developed before it reaches the outlet. Technically, there is to my knowledge no reason why this would be more difficult to simulate in 2D than in 3D so I think you should do fine. But note that you will probably not find steady solutions in the post-stall regime, so in order to get good statistics (mean drag etc.) you will probably have to do quite long simulations.

March 18, 2015, 07:59		#4
antm324 New Member Anthony McCarthy Join Date: Nov 2014 Location: United Kingdom Posts: 21 Rep Power: 11	Hi Erlofg, Thank you for your response. I have quite a few questions that i believe you may be able to answer if you are willing to help? Do you mind if i pose these questions to you? Ideally it would be best to have a live chat if possible and then i can post the discussion on the forum to also help benefit other users Anthony

March 18, 2015, 10:04		#5
erlofg New Member Erik Löfgren Join Date: Feb 2015 Location: Umeå, Sweden Posts: 9 Rep Power: 11	I'm not really an expert per se in these matters, but ask away and we'll see what I can answer. It's better for me to take it here in the forums so that I may take some time to consider the answers...

August 11, 2015, 12:27		#6
er_trebbia New Member Join Date: May 2013 Posts: 9 Rep Power: 12	Basically, you can't perform a DES in a 2d problem, because the LES part of your DES won't make any sense, since turbulence is 3d. Then, of course you have to use the 3d solver.

August 11, 2015, 12:42		#8
er_trebbia New Member Join Date: May 2013 Posts: 9 Rep Power: 12	Yes, but the problem here is an airfoil in post stall, if you hope to catch something like that with a 2D DDES, be my guest.