LES help

pksri · April 19, 2016, 08:05

Dear all,

Presently I am working on my master thesis to study about re-transition, At present I am trying to do a LE simulation for simple inclined flat plate in a rectangular channel considering base of the channel as a inclined plate(x = length, y = height, z = width). I am completely new to LES and finding very hard to workout and getting a required results. I am following the LES best practices guide from ANSYS. I have few questions and doubts. Let me first tell what I have did.

1. I did grid convergence study using SST transition model. Here I have problem that in one of your post you mentioned z+ should be less than 20. If I maintain this condition I am not achieving transition using the sáme SST model. In my default model I have z+ around 85 and I am achieving proper transition.

2. I used my default model for LES ( which has y+ = 0.15) and as per best practice suggestion I performed standard turbulence simulation. Using TUI i initialized the inlet condition "\solve\initialize\init-instantaneous-vel" command. changed my model to LES , Inlet BC I used vortex synthesizer with default value and my turbulence intensity and hydraulic diameter as 3 and 0.2 resp. I used WALE model, scheme PISO, gradient LEAST SQUARE , pressure SECOND ORDER, momentum CENTRAL DIFFERENCING, time step 1e-4, performed 7000 time steps. started averaging after L/U reach three repetitions (L= hyd.length, U = mean.Vel ). Is my LES setup is right?

3. If you could provide me some guidance in solution monitoring and post processing the results of LES.?

I am looking forward for your support and guidance

Sincerely,
Kumar

pksri · April 19, 2016, 19:23

Dear CFD Experience user,

It would be great help, if you could please guide me somehow.

This is my time averaged shear stress and I dont see any transition with my simulation or fluctuations. The shear stress is averaged along the flate plate from time step 0.25-0.65s.

Looking for your reply.

Kind Regards,
Kumar

davidwilcox · April 20, 2016, 01:55

For LES, read phillip schlatter's thesis. It covers a lot in detail and he even did something on transition. It is not exactly finite volume because of the gauss-lobatto collocation points (spectral element stuff) but it's a worth a shot. Regarding the z+ issue with the SST, is your aspect ratio of the element size near wall close to 1? you might have to adjust the spatial step size not just in the z direction but also in the streamwise direction because of the high velocity gradients near the wall.
In any case, if you have access to Piomelli's paper titled "Large eddy simulations in 2030 and beyond", you can find some interesting results with the SST model for a flow over a cylinder. The experimental drag coefficient,Cf, was measured to be 1.15. The steady-state SST gives Cf =0.78. with 3D URANS, the results got better. So have a look at it.

April 19, 2016, 08:05	LES help	#1
pksri New Member Kumar Join Date: Feb 2016 Posts: 23 Rep Power: 10	Dear all, Presently I am working on my master thesis to study about re-transition, At present I am trying to do a LE simulation for simple inclined flat plate in a rectangular channel considering base of the channel as a inclined plate(x = length, y = height, z = width). I am completely new to LES and finding very hard to workout and getting a required results. I am following the LES best practices guide from ANSYS. I have few questions and doubts. Let me first tell what I have did. 1. I did grid convergence study using SST transition model. Here I have problem that in one of your post you mentioned z+ should be less than 20. If I maintain this condition I am not achieving transition using the sáme SST model. In my default model I have z+ around 85 and I am achieving proper transition. 2. I used my default model for LES ( which has y+ = 0.15) and as per best practice suggestion I performed standard turbulence simulation. Using TUI i initialized the inlet condition "\solve\initialize\init-instantaneous-vel" command. changed my model to LES , Inlet BC I used vortex synthesizer with default value and my turbulence intensity and hydraulic diameter as 3 and 0.2 resp. I used WALE model, scheme PISO, gradient LEAST SQUARE , pressure SECOND ORDER, momentum CENTRAL DIFFERENCING, time step 1e-4, performed 7000 time steps. started averaging after L/U reach three repetitions (L= hyd.length, U = mean.Vel ). Is my LES setup is right? 3. If you could provide me some guidance in solution monitoring and post processing the results of LES.? I am looking forward for your support and guidance Sincerely, Kumar

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
LES Filtering: how are the small and large scales equations solved?	atmcfd	Main CFD Forum	38	March 14, 2016 15:52
Grid difference between wall modeled LES and wall resolved LES	hityangsir	Main CFD Forum	9	April 17, 2015 12:59
OpenFOAM LES capability questions	siw	OpenFOAM Running, Solving & CFD	3	January 6, 2015 09:50
Turbulence dampening due to magnetic field in LES and RAS	eelcovv	OpenFOAM	0	June 8, 2010 12:35
Some Questions about LES.	Bin Li	Main CFD Forum	2	February 20, 2004 10:58

April 20, 2016, 01:55		#3
davidwilcox Senior Member david Join Date: Oct 2012 Posts: 142 Rep Power: 14	For LES, read phillip schlatter's thesis. It covers a lot in detail and he even did something on transition. It is not exactly finite volume because of the gauss-lobatto collocation points (spectral element stuff) but it's a worth a shot. Regarding the z+ issue with the SST, is your aspect ratio of the element size near wall close to 1? you might have to adjust the spatial step size not just in the z direction but also in the streamwise direction because of the high velocity gradients near the wall. In any case, if you have access to Piomelli's paper titled "Large eddy simulations in 2030 and beyond", you can find some interesting results with the SST model for a flow over a cylinder. The experimental drag coefficient,Cf, was measured to be 1.15. The steady-state SST gives Cf =0.78. with 3D URANS, the results got better. So have a look at it.