|
[Sponsors] |
July 3, 2009, 12:35 |
Little question
|
#1 |
New Member
Franz Roman
Join Date: Jun 2009
Posts: 28
Rep Power: 17 |
Hi,
I wanted to know if there is a problem with my simulation if the residuals drop and the go up again, and down and up and down again, with an overall tendency to go down, until it finally reaches convergence to 1e-6. I mean, should the residuals drop constantly? If the oscillate, does that mean the problem is not well set up? thanks franz |
|
July 6, 2009, 00:56 |
|
#2 |
Super Moderator
Maxime Perelli
Join Date: Mar 2009
Location: Switzerland
Posts: 3,297
Rep Power: 41 |
If the residuals are decreasing, I assume you are gonna catch a convergence despite the oscillating.
But I would monitor a special variable such as force, massflow etc... and check if the monitor is also oscillating. If yes, then maybe you are handling an unsteady problem
__________________
In memory of my friend Hervé: CFD engineer & freerider |
|
July 8, 2009, 12:30 |
|
#3 |
New Member
Franz Roman
Join Date: Jun 2009
Posts: 28
Rep Power: 17 |
hi mAx,
thanks for your answer. When I posted this thread I had a case where the residuals oscillated about 4 times and finally converged (1e-6, bot not yet horizontal residuals) after around 5000 iterations. Now I am running another simulation in which I changed something in the system, nothing very special or tricky at all. the thing is now I also have oscillations with the tendency to drop, but I am already in 11000 iterations and 16 oscillations. the oscillations of all residuals are in phase and the all look exactly the same, only that slowly going down from one iteration to the next. My guess is, it will reach the 1e-6 criteria but maybe in 10000 more iterations or so. Is that also normal or ok? I am monitoring the massflow andthe graph seems also to be oscillating but in the order of 5e-8 to -2e-7 kg/s, very small I think. My system is quite simple: a velocity inlet in the horizontal direction, air flows then into a plenum chamber and then through a porous media in the vertical direction out through a pressure outlet. Can you explain me that about maybe my problem being unsteady? The airflow coming in is constant, the system and the grid (Hex map) is absolutely simmetric. Thanks again. Franz |
|
July 8, 2009, 13:13 |
|
#4 |
Super Moderator
Maxime Perelli
Join Date: Mar 2009
Location: Switzerland
Posts: 3,297
Rep Power: 41 |
*you can speedup the convergence with under-relaxation's factors
*regarding the unsteady case, check the von karman vortex... straight flowfield around a cylinder (could be computed with symetries) but at specified Re, the wake is oscillating
__________________
In memory of my friend Hervé: CFD engineer & freerider |
|
|
|
Similar Threads | ||||
Thread | Thread Starter | Forum | Replies | Last Post |
Unanswered question | niklas | OpenFOAM | 2 | July 31, 2013 16:03 |
Static pressure at outlet - Stupid question??? | Miry | CFX | 0 | September 16, 2007 18:14 |
Poisson Solver question | Suresh | Main CFD Forum | 3 | August 12, 2005 04:37 |
CHANNEL FLOW: a question and a request | Carlos | Main CFD Forum | 4 | August 23, 2002 05:55 |
question | K.L.Huang | Siemens | 1 | March 29, 2000 04:57 |