|
[Sponsors] |
March 10, 2017, 04:03 |
Imbalances in Output file Warning
|
#1 |
Senior Member
Join Date: Mar 2013
Location: Germany
Posts: 357
Rep Power: 14 |
Hi All
Have anyone faced this issue and what is it behind it ? I did a RANS computation for simple pipe flow with a diffuser at the end of the pipe. Inlet is given as Mass flow BC'n Outlet is given Avg Static Pressure =0 Reference Pressure =1 atm And the case was ran for 5000 iterations And I was running it on a cluster. So when I checked the .out file it showed that Domain Imbalances +--------------------------------------------------------------------+ | Normalised Imbalance Summary | +--------------------------------------------------------------------+ | Equation | Maximum Flow | Imbalance (%) | +--------------------------------------------------------------------+ | U-Mom | 1.2845E+01 | -0.0282 | | V-Mom | 1.2845E+01 | 0.0633 | | W-Mom | 1.2845E+01 | -0.0661 | | P-Mass | 5.8000E-02 | 0.0090 | +----------------------+-----------------------+---------------------+ Does it has anything to do with the simulation or should we consider this for some convergence issue or something related to it ! Kindly let me know. |
|
March 10, 2017, 06:00 |
|
#2 |
Super Moderator
Glenn Horrocks
Join Date: Mar 2009
Location: Sydney, Australia
Posts: 17,665
Rep Power: 143 |
Most simulations are converged by the time the imbalances reach that figure. But you should check in your case whether your simulation is converged by doing a sensitivity analysis on convergence criteria.
Also: CFX rarely requires 5000 iterations to obtain convergence. Something like 200 is more normal. You probably don't have the solver set up very well. |
|
March 10, 2017, 06:11 |
|
#3 |
Senior Member
Join Date: Mar 2013
Location: Germany
Posts: 357
Rep Power: 14 |
Thats true you are right ! By sensitivity analysis what do you mean ? my residual was given 1e-5 and the RMS values seems to be stable and within a 10 % deflection. Also Yes I ran 5000 just to see how the changes are from 500 !! And since its a small job and in cluster, it takes less than a couple of hours
But as per your suggestion this imbalances are fine ? |
|
March 10, 2017, 07:07 |
|
#4 |
Super Moderator
Glenn Horrocks
Join Date: Mar 2009
Location: Sydney, Australia
Posts: 17,665
Rep Power: 143 |
So a simulation at either 1e-4 or 1e-6 RMS and compare it to your 1e-5 result. If they are the same then you have sufficiently converged. If they are different you need to use a tighter convergence criteria.
I said for most simulations your convergence looks fine. But there are exceptions which require tighter or looser convergence. That is why you do the sensitivity check to determine what your case requires. |
|
March 10, 2017, 07:37 |
|
#5 |
Senior Member
Join Date: Mar 2013
Location: Germany
Posts: 357
Rep Power: 14 |
Perfect Solution !! Thanks a lot for your valuable comments !!
Respect !! |
|
Thread Tools | Search this Thread |
Display Modes | |
|
|
Similar Threads | ||||
Thread | Thread Starter | Forum | Replies | Last Post |
[swak4Foam] swak4foam building problem | GGerber | OpenFOAM Community Contributions | 54 | April 24, 2015 17:02 |
[swak4Foam] swak4Foam-groovyBC build problem | zxj160 | OpenFOAM Community Contributions | 18 | July 30, 2013 14:14 |
[swak4Foam] funkySetFields compilation error | tayo | OpenFOAM Community Contributions | 39 | December 3, 2012 06:18 |
[blockMesh] error message with modeling a cube with a hold at the center | hsingtzu | OpenFOAM Meshing & Mesh Conversion | 2 | March 14, 2012 10:56 |
"parabolicVelocity" in OpenFoam 2.1.0 ? | sawyer86 | OpenFOAM Running, Solving & CFD | 21 | February 7, 2012 12:44 |