[OpenFOAM] Can't change timestep on a decomposed case with python scripts

Gambenius · May 6, 2020, 09:11

Hi everyone, need to postprocess a 140GB transient LES OpenFOAM case.
Basically what i'm doing is a bunch of 'for' cycles in python that pick a timestep, pick a slice plane, pick a parameter between U/p/vorticity, print the screenshot, then move the plane and so on, what i get is a lot of images of U/p/vorticity sliced in different positions and trough time.
I've tried using the script on a decomposed case (i have included casefoam.CaseType = 'Decomposed Case' in the .py file, paraview opens it as a decomposed case) but it won't cycle timesteps.
On a reconstructed case everything works fine except that it runs on a single core (I have a Xeon E5 2620 v3 CPU)
I am on windows 10 with paraview 5.7.0

TL;DR I need to cycle timesteps on a decomposed transient case with a python script, the one i made only works with reconstructed and not decomposed cases.

Thanks in advance

mrishi · May 21, 2020, 09:16

I am facing the exact problem as yours. It seems that for a decomposed case, the variable casefoam.TimestepValues only lists [0.0]. It outputs the correct list for a reconstructed case allowing for selection.

Insterestingly, even with TimestepValues information missing in our reader, if I use the TemporalStatistics filter, it successfully traverses all timesteps (evident from some warnings that pop up in the process).

May 6, 2020, 09:11	Can't change timestep on a decomposed case with python scripts	#1
Gambenius New Member Edoardo Gamba Join Date: Dec 2019 Posts: 6 Rep Power: 6	Hi everyone, need to postprocess a 140GB transient LES OpenFOAM case. Basically what i'm doing is a bunch of 'for' cycles in python that pick a timestep, pick a slice plane, pick a parameter between U/p/vorticity, print the screenshot, then move the plane and so on, what i get is a lot of images of U/p/vorticity sliced in different positions and trough time. I've tried using the script on a decomposed case (i have included casefoam.CaseType = 'Decomposed Case' in the .py file, paraview opens it as a decomposed case) but it won't cycle timesteps. On a reconstructed case everything works fine except that it runs on a single core (I have a Xeon E5 2620 v3 CPU) I am on windows 10 with paraview 5.7.0 TL;DR I need to cycle timesteps on a decomposed transient case with a python script, the one i made only works with reconstructed and not decomposed cases. Thanks in advance

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
Is Playstation 3 cluster suitable for CFD work	hsieh	OpenFOAM	9	August 16, 2015 14:53
Running Parallel on Windows using Python Scripts	amarkkassery	SU2 Installation	6	April 4, 2013 12:37
Free surface boudary conditions with SOLA-VOF	Fan	Main CFD Forum	10	September 9, 2006 12:24
Changing gambit file without change of case file??	Asghari	FLUENT	2	August 28, 2006 13:48
The energy conservation in a case of phase change	robin	Phoenics	2	April 6, 2002 00:09

May 21, 2020, 09:16		#2
mrishi Member Rishikesh Join Date: Apr 2016 Posts: 63 Rep Power: 9	I am facing the exact problem as yours. It seems that for a decomposed case, the variable casefoam.TimestepValues only lists [0.0]. It outputs the correct list for a reconstructed case allowing for selection. Insterestingly, even with TimestepValues information missing in our reader, if I use the TemporalStatistics filter, it successfully traverses all timesteps (evident from some warnings that pop up in the process).