Oscillating of residual for SDR

Jihun, Lee · April 3, 2017, 23:51

3D
Steady State
Coupled Flow implicit 2nd order
Ideal gas
RANS
AUSM+ FVS
K-W SST

Using Polyhedral mesh & prism layer

Stagnation inlet 10MPa 1500K
Pressure Outlet 0MPa

Courant Number : 10 No ramp
AMG Linear Solver : Low On, Maximum Number of AMG Cycle Monitor 2
Expert Initialization : On
Solution Driver : Expert Driver On
CCA : On

TKE residual is Oscillating
Continuity residual : 6.866532e-04
X momentum residual : 6.292211e-04
Y momentum residual : 3.581249e-04
Z momentum residual : 7.172282e-04
Energy residual : 1.076242e-03
TKE residual : 1.066230e-06
SDR residual : 9.982527e-06

All Y+ treatment : Y+ Range 0.05~24

1. If Maximum Number of AMG cycles is 2, can I increase Courant number?
2. Doesn't matter Oscillating of SDR residual for convergence? Is it OK just decrease whether oscillating or not?
3. Can I judge that my simulation reach convergence(steady state), with above residual condition?

Thanks in advanced,

fluid23 · April 5, 2017, 16:02

1) I think that more than 5 cycles and you need to reduce your courant, so if you are only seeing 2 I would bump it up a bit. Maybe use a ramp so that it doesn't introduce new issues.

2) You cannot judge convergence solely based on residuals. You need to look at figures of merit as well, which depend on what you are actually modeling. In general, oscillating residuals are not a good thing. However, they are not a huge problem to solve. I would ditch the solution driver, that is probably what is causing it.

3) No, this plot is not enough to gauge convergence.

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
simpleFoam error - "Floating point exception"	mbcx4jc2	OpenFOAM Running, Solving & CFD	12	August 4, 2015 03:20
Unstabil Simulation with chtMultiRegionFoam	mbay101	OpenFOAM Running, Solving & CFD	13	December 28, 2013 14:12
pimpleFoam: turbulence->correct(); is not executed when using residualControl	hfs	OpenFOAM Running, Solving & CFD	3	October 29, 2013 09:35
calculation stops after few time steps	sivakumar	OpenFOAM Running, Solving & CFD	7	March 17, 2013 07:37
Orifice Plate with a fully developed flow - Problems with convergence	jonmec	OpenFOAM Running, Solving & CFD	3	July 28, 2011 06:24

April 5, 2017, 16:02		#2
fluid23 Senior Member Matt Join Date: Aug 2014 Posts: 947 Rep Power: 18	1) I think that more than 5 cycles and you need to reduce your courant, so if you are only seeing 2 I would bump it up a bit. Maybe use a ramp so that it doesn't introduce new issues. 2) You cannot judge convergence solely based on residuals. You need to look at figures of merit as well, which depend on what you are actually modeling. In general, oscillating residuals are not a good thing. However, they are not a huge problem to solve. I would ditch the solution driver, that is probably what is causing it. 3) No, this plot is not enough to gauge convergence.