low reynolds number, laminar model, not converging

c_web11 · July 21, 2016, 09:07

Hi

I'm running a transient simulation of an unsteady airflow over a flat plate using a laminar model, where the angle of attack varies between -15 and +15 degrees at a frequency of 40 hz. My airflow is based on a reynolds number of 1000.

My problem is that the solution will not converge, however it converged at a reynolds number of 10,000, so I'm fairly sure I have everything set up correctly.

Would a finer mesh help with this?

My time step is approximately 4*10^(-4), would having a smaller time step help the solution converge?

I'm stuck on how to proceed, so any help would be appreciated.

Thanks

ghorrocks · July 21, 2016, 19:42

Where did you get your time step size from? The most common answer to this question is "I guessed". And that is common to get the time step totally wrong.

Use adaptive time steps, homing in on 3-5 coeff loops per iteration and make sure you set the minimum time step size small enough that it never hits it. Then the solver will find the time step size it requires by itself. I bet that will help things along for you.

c_web11 · July 25, 2016, 08:08

Thanks for the reply.

I chose this time step size, so that angle of attack at the inlet changed by 1 degree for every time step, for this chosen frequency of oscillation.

I tried as you suggested, and yes there was definitely and improvement in the solution converging, however my simulation has been running over the weekend and still hasn't completed

July 21, 2016, 09:07	low reynolds number, laminar model, not converging	#1
c_web11 New Member Ciaran Join Date: Jul 2016 Posts: 2 Rep Power: 0	Hi I'm running a transient simulation of an unsteady airflow over a flat plate using a laminar model, where the angle of attack varies between -15 and +15 degrees at a frequency of 40 hz. My airflow is based on a reynolds number of 1000. My problem is that the solution will not converge, however it converged at a reynolds number of 10,000, so I'm fairly sure I have everything set up correctly. Would a finer mesh help with this? My time step is approximately 4*10^(-4), would having a smaller time step help the solution converge? I'm stuck on how to proceed, so any help would be appreciated. Thanks

Thread Tools	Search this Thread
Show Printable Version Email this Page	Search this Thread: Advanced Search
Display Modes
Linear Mode Switch to Hybrid Mode Switch to Threaded Mode

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
decomposePar problem: Cell 0contains face labels out of range	vaina74	OpenFOAM Pre-Processing	37	July 20, 2020 06:38
[OpenFOAM.org] OF2.3.1 + OS13.2 - Trying to use the dummy Pstream library	aylalisa	OpenFOAM Installation	23	June 15, 2015 15:49
Myong Kasagi Low Reynolds k-epsilon model coding	Ruonin	Main CFD Forum	2	December 5, 2012 18:39
k-epsilon model, low Reynolds number formulation	Märta-Karin	CFX	1	September 15, 2003 22:37
How low for low Reynolds model?	Jean-Luc	Phoenics	3	March 10, 2001 02:34

July 21, 2016, 19:42		#2
ghorrocks Super Moderator Glenn Horrocks Join Date: Mar 2009 Location: Sydney, Australia Posts: 17,893 Rep Power: 144	Where did you get your time step size from? The most common answer to this question is "I guessed". And that is common to get the time step totally wrong. Use adaptive time steps, homing in on 3-5 coeff loops per iteration and make sure you set the minimum time step size small enough that it never hits it. Then the solver will find the time step size it requires by itself. I bet that will help things along for you.

July 25, 2016, 08:08		#3
c_web11 New Member Ciaran Join Date: Jul 2016 Posts: 2 Rep Power: 0	Thanks for the reply. I chose this time step size, so that angle of attack at the inlet changed by 1 degree for every time step, for this chosen frequency of oscillation. I tried as you suggested, and yes there was definitely and improvement in the solution converging, however my simulation has been running over the weekend and still hasn't completed