Cavity Flow Model Convergence Problem

benderbending · February 18, 2017, 15:08

Hi all,

Im new and recently start learning cfd. Now im modelling a cavity flow in a rectangular duct for academic purpose. My problem is to get the convergence result, my RMS residual value is always above 1e^-4 and became oscillating after about 50 iteration around 2-3 e^-4. Im using steady state simulation. I have read some tips and FAQ about the convergence problem, but i have no idea what to do.
I have do some things like use local timestep factor and make finer mesh. This model is really simple, but i think i my mesh is the cause of this. This is my first cfd model, and I quite frustrated because i cant get the converged result.

Please give any suggestion to improve the convergence or fix on the mesh (pict attached).

Thanks for your consideration, sorry for my english.

ghorrocks · February 19, 2017, 18:47

Is this a single phase flow? What Reynolds number?

Add the residuals to your results file and rerun the simulation. Then look at the residuals in CFD-Post. This will show you where the high residuals are, and might give you some clues as to how to fix it.

But I know what the answer is already because this question comes up all the time

..... You have vortex shedding off one of the sharp edges on your geometry. Your options are:
1) Use a larger physical time step size, if that does not work then:
2) you are forced to use a transient simulation (because your simulation is not steady state anyway).

benderbending · February 20, 2017, 05:43

Quote:

Originally Posted by ghorrocks

Is this a single phase flow? What Reynolds number?

Add the residuals to your results file and rerun the simulation. Then look at the residuals in CFD-Post. This will show you where the high residuals are, and might give you some clues as to how to fix it.

But I know what the answer is already because this question comes up all the time

..... You have vortex shedding off one of the sharp edges on your geometry. Your options are:
1) Use a larger physical time step size, if that does not work then:
2) you are forced to use a transient simulation (because your simulation is not steady state anyway).

Hi ghorrocks,

Yes, i used ideal gas and the Re number is 10 million.
Did you mean "add the residual" is by rerun the solver again? So the residual graph continue from the former run?
Im looking the highest residual from the .out file and the max residual is on the corner and the edge of rectangular above the duct.
I already rerun the simulation with some physical timestep variation but its just same like before. Maybe ill try the transient simulation later.

Thx for the advice

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
Transition SST model - convergence problem with 2nd order discretization	xara	FLUENT	4	May 27, 2013 14:16
Heat Transfer convergence Problem mit a 1/6 compressor model	chili023	CFX	1	January 15, 2011 07:41
Problem in cavity flow problem	saad	Main CFD Forum	4	November 1, 2007 08:45
Problem with Convergence at high flow rates	Syed	Siemens	1	April 10, 2007 16:18
fluid flow fundas	ram	Main CFD Forum	5	June 17, 2000 22:31

February 19, 2017, 18:47		#2
ghorrocks Super Moderator Glenn Horrocks Join Date: Mar 2009 Location: Sydney, Australia Posts: 17,854 Rep Power: 144	Is this a single phase flow? What Reynolds number? Add the residuals to your results file and rerun the simulation. Then look at the residuals in CFD-Post. This will show you where the high residuals are, and might give you some clues as to how to fix it. But I know what the answer is already because this question comes up all the time ..... You have vortex shedding off one of the sharp edges on your geometry. Your options are: 1) Use a larger physical time step size, if that does not work then: 2) you are forced to use a transient simulation (because your simulation is not steady state anyway).