Doubt in GAMMA scheme

Tushar@cfd · May 22, 2014, 07:46

Dear ALL,

In gamma scheme, the gamma coefficient is multiplied with "2" where gamma coefficient varies from 0.1 to 0.5.

Whereas in OpenFOAM, I am finding it difficult to understand why the gamma coefficient is divided by "2". Refer code below for OpenFOAM:

Code:

k_ = max (k_/2.0, SMALL)

Correct me if I am wrong.

dkxls · January 16, 2015, 12:27

Not the newest thread, I know. But, an OpenFOAM developer (i.e. Henry Weller) gave some clarifications on this issue:
http://www.openfoam.org/mantisbt/view.php?id=1491#c3545

Hope this is of help!

dkxls · January 16, 2015, 12:32

And I found another reference, where Hrvoje Jasak states basically the same:
http://www.cfd-online.com/Forums/ope...tml#post189665

Well, this inconsistency between the implementation and the original paper apparently leads every now and then to some confusion.

hjasak · January 17, 2015, 03:12

Actually, I changed it on purpose, after publishing the paper.

The natural range for k (see paper) is between 0 and 0.5. People were using the scheme with any number. The clever ones used the 0-1 range andwe ven got attempts to use 10!

Therefore, I have re-scaled the factor and bound it to 0-1.

In practical simulations, you will not be able to tell the difference.

Hrv

May 22, 2014, 07:46	Doubt in GAMMA scheme	#1
Tushar@cfd Senior Member T. Chourushi Join Date: Jul 2009 Posts: 321 Blog Entries: 1 Rep Power: 17	Dear ALL, In gamma scheme, the gamma coefficient is multiplied with "2" where gamma coefficient varies from 0.1 to 0.5. Whereas in OpenFOAM, I am finding it difficult to understand why the gamma coefficient is divided by "2". Refer code below for OpenFOAM: Code: k_ = max (k_/2.0, SMALL) Correct me if I am wrong.

January 17, 2015, 03:12		#4
hjasak Senior Member Hrvoje Jasak Join Date: Mar 2009 Location: London, England Posts: 1,905 Rep Power: 33	Actually, I changed it on purpose, after publishing the paper. The natural range for k (see paper) is between 0 and 0.5. People were using the scheme with any number. The clever ones used the 0-1 range andwe ven got attempts to use 10! Therefore, I have re-scaled the factor and bound it to 0-1. In practical simulations, you will not be able to tell the difference. Hrv Tushar@cfd, J-P, dkxls and 3 others like this. __________________ Hrvoje Jasak Providing commercial FOAM/OpenFOAM and CFD Consulting: http://wikki.co.uk

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
Why RNGkepsilon model gives floating error	shipman	OpenFOAM Running, Solving & CFD	3	September 7, 2013 08:00
Question about gamma intgerpolation scheme	titio	OpenFOAM Running, Solving & CFD	0	July 26, 2010 10:22
Differences between serial and parallel runs	carsten	OpenFOAM Bugs	11	September 12, 2008 11:16
Fundamental problem with Gamma scheme	dominik_christ	OpenFOAM Running, Solving & CFD	3	August 29, 2008 08:24
?to implement gamma differencing scheme	Gonski	Main CFD Forum	15	January 28, 2007 17:00

January 16, 2015, 12:27		#2
dkxls Senior Member Armin Join Date: Feb 2011 Location: Helsinki, Finland Posts: 156 Rep Power: 19	Not the newest thread, I know. But, an OpenFOAM developer (i.e. Henry Weller) gave some clarifications on this issue: http://www.openfoam.org/mantisbt/view.php?id=1491#c3545 Hope this is of help!

January 16, 2015, 12:32		#3
dkxls Senior Member Armin Join Date: Feb 2011 Location: Helsinki, Finland Posts: 156 Rep Power: 19	And I found another reference, where Hrvoje Jasak states basically the same: http://www.cfd-online.com/Forums/ope...tml#post189665 Well, this inconsistency between the implementation and the original paper apparently leads every now and then to some confusion.