Bug caused by CrankNicholson scheme

cbeck · September 3, 2008, 18:56

Using the Crank-Nicholson Scheme in compressible transient solvers causes a unit problem in the pressure correction.

I can reproduce the problem with the coodles tutorial. Can anybody else confim this?

Version:
/openfoam-extend/branches/OpenCFD_Release/OpenFOAM-1.5_patch/ (a couple of weeks old)

kind regards,
Christian Beck

henry · September 4, 2008, 04:41

Thanks for the bug report. I have fixed the problem and running the coodles tutorial at the moment. If all is well I will upload the patch to our 1.5.x git repository (see http://www.opencfd.co.uk/openfoam/download.html for download details) later today.

H

carsten · September 12, 2008, 10:41

I observed the same problem with lesInterFoam: If I switch to Crank-Nicholson, a dimension error appears.

Bye,

Carsten

henry · September 12, 2008, 12:07

Even when using the version I fixed and uploaded to our git repository?

H

carsten · September 15, 2008, 06:32

Uhhhh. Difficult to say, sorry. I can't reconstruct exactly when I did check-out and I fear to check it out again, because I don't want to loose my settings/changes, which were necessary for compilation on our machine.

Anyhow, I will try to recompile in another directory and will report my findings.

Bye,

Carsten

carsten · September 15, 2008, 12:02

O.k., problem still exists. I checked out a new tree this morning and still have the problem.

You can recreate it with the lesInterFoam-tutorial, simply switch to CrankNicholson and it will die.

Bye,

Carsten

henry · September 15, 2008, 12:48

OK I will look into it...

H

virginie_e · March 27, 2009, 05:43

Hello,

I also have a bug when trying to use the crankNicholson scheme with interTrackFoam. However, it is not a dimension problem. The error message is rather:

attempt to read beyond EOF
file: /users/V1117324/OpenFOAM/v1117324-1.5-dev/run/tutorials/interTrackFoam/hydrofoil/system/fvSchemes::ddt(rho,U) at line 21.
From function ITstream::read(token& t)
in file db/IOstreams/Tstreams/ITread.C at line 64.
FOAM exiting

Do you know what this means? Thank you.

Virginie

alberto · March 28, 2009, 02:48

Hi Virginie_e,
that's not a bug but a typo or an error in the specification of the scheme. Did you specify a number between 0 (Euler) and 1 (pure Crank Nicholson) after the scheme name?
If this is not the case, please post your fvScheme file in the appropriate forum. This one is specific for bugs

Best,

September 3, 2008, 18:56	Using the Crank-Nicholson Sche	#1
cbeck New Member Chris Beck Join Date: Mar 2009 Posts: 4 Rep Power: 17	Using the Crank-Nicholson Scheme in compressible transient solvers causes a unit problem in the pressure correction. I can reproduce the problem with the coodles tutorial. Can anybody else confim this? Version: /openfoam-extend/branches/OpenCFD_Release/OpenFOAM-1.5_patch/ (a couple of weeks old) kind regards, Christian Beck

September 4, 2008, 04:41	Thanks for the bug report. I	#2
henry Senior Member Join Date: Mar 2009 Posts: 854 Rep Power: 22	Thanks for the bug report. I have fixed the problem and running the coodles tutorial at the moment. If all is well I will upload the patch to our 1.5.x git repository (see http://www.opencfd.co.uk/openfoam/download.html for download details) later today. H

September 12, 2008, 10:41	I observed the same problem wi	#3
carsten Member Carsten Thorenz Join Date: Mar 2009 Location: Germany Posts: 34 Rep Power: 17	I observed the same problem with lesInterFoam: If I switch to Crank-Nicholson, a dimension error appears. Bye, Carsten

September 12, 2008, 12:07	Even when using the version I	#4
henry Senior Member Join Date: Mar 2009 Posts: 854 Rep Power: 22	Even when using the version I fixed and uploaded to our git repository? H

September 15, 2008, 06:32	Uhhhh. Difficult to say, sorry	#5
carsten Member Carsten Thorenz Join Date: Mar 2009 Location: Germany Posts: 34 Rep Power: 17	Uhhhh. Difficult to say, sorry. I can't reconstruct exactly when I did check-out and I fear to check it out again, because I don't want to loose my settings/changes, which were necessary for compilation on our machine. Anyhow, I will try to recompile in another directory and will report my findings. Bye, Carsten

September 15, 2008, 12:02	O.k., problem still exists. I	#6
carsten Member Carsten Thorenz Join Date: Mar 2009 Location: Germany Posts: 34 Rep Power: 17	O.k., problem still exists. I checked out a new tree this morning and still have the problem. You can recreate it with the lesInterFoam-tutorial, simply switch to CrankNicholson and it will die. Bye, Carsten

September 15, 2008, 12:48	OK I will look into it... H	#7
henry Senior Member Join Date: Mar 2009 Posts: 854 Rep Power: 22	OK I will look into it... H

March 27, 2009, 05:43		#8
virginie_e Member Virginie Ehrlacher Join Date: Mar 2009 Posts: 52 Rep Power: 17	Hello, I also have a bug when trying to use the crankNicholson scheme with interTrackFoam. However, it is not a dimension problem. The error message is rather: attempt to read beyond EOF file: /users/V1117324/OpenFOAM/v1117324-1.5-dev/run/tutorials/interTrackFoam/hydrofoil/system/fvSchemes::ddt(rho,U) at line 21. From function ITstream::read(token& t) in file db/IOstreams/Tstreams/ITread.C at line 64. FOAM exiting Do you know what this means? Thank you. Virginie

March 28, 2009, 02:48		#9
alberto Senior Member Alberto Passalacqua Join Date: Mar 2009 Location: Ames, Iowa, United States Posts: 1,912 Rep Power: 36	Hi Virginie_e, that's not a bug but a typo or an error in the specification of the scheme. Did you specify a number between 0 (Euler) and 1 (pure Crank Nicholson) after the scheme name? If this is not the case, please post your fvScheme file in the appropriate forum. This one is specific for bugs Best, mm.abdollahzadeh and smog like this. __________________ Alberto Passalacqua GeekoCFD - A free distribution based on openSUSE 64 bit with CFD tools, including OpenFOAM. Available as in both physical and virtual formats (current status: http://albertopassalacqua.com/?p=1541) OpenQBMM - An open-source implementation of quadrature-based moment methods. To obtain more accurate answers, please specify the version of OpenFOAM you are using.

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
CrankNicholson errors	lr103476	OpenFOAM Running, Solving & CFD	13	February 2, 2017 11:45
CrankNicholson returns wrong dimensions and sound speed when mesh is moving	lillberg	OpenFOAM Bugs	6	May 4, 2009 18:53
Forces caused by fluids	anja	OpenFOAM Running, Solving & CFD	20	May 8, 2006 07:37
Adiabatic compression caused by gas ram	Chris	FLUENT	0	December 13, 2005 05:42
Probelm with CrankNicholson scheme in channelOodles	maka	OpenFOAM Pre-Processing	2	November 2, 2005 11:09