Problem in fvschemes divSchemes cannot use Gauss linearUpwind

fippo_dk · November 16, 2011, 03:02

how do your results compare to measurements? are you under or over predicting?

In my case I over predict from 10 to 20 degrees

uli · August 2, 2012, 14:34

hi all

I noticed that both in both tutorial cases and posted fvschemes there is always

Code:

div(R) Gauss linear;

Is that because there is no flux (phi = rho U) in that term and should it be left like this?

I tried

Code:

div(R)          Gauss limitedLinear 1;
div(R)          Gauss QUICK;
div(R)          Gauss MUSCL;

and it gave me similar results without any problems/errors.

thanks
Uli

makaveli_lcf · August 2, 2012, 14:40

Hi! It is for other turbulence model where you use Reynolds stress

uli · August 2, 2012, 15:12

Thanks for your reply.

So using a k-Epsilon model does not involve R? But how is it possible that I get different results after switching the scheme for div(R)?

makaveli_lcf · August 2, 2012, 15:18

You wrote you get the SAME results. You can easily check which schemes you solver uses, just set "default none;", and you will get errors until you define all schemes.

uli · August 21, 2012, 18:34

hi, sorry for the late reply.

I wrote "similar" results since the plots are slightly different. I removed the schemes and it still worked, thanks for your help.

August 2, 2012, 14:34		#22
uli New Member Join Date: Jun 2012 Posts: 25 Rep Power: 2	hi all I noticed that both in both tutorial cases and posted fvschemes there is always Code: div(R) Gauss linear; Is that because there is no flux (phi = rho U) in that term and should it be left like this? I tried Code: div(R) Gauss limitedLinear 1; div(R) Gauss QUICK; div(R) Gauss MUSCL; and it gave me similar results without any problems/errors. thanks Uli

August 2, 2012, 14:40		#23
makaveli_lcf Senior Member Dr. Alexander Vakhrushev Join Date: Mar 2009 Posts: 189 Rep Power: 6	Hi! It is for other turbulence model where you use Reynolds stress __________________ Best regards, Dr. Alexander VAKHRUSHEV Lehrstuhl für Modellierung und Simulation metallurgischer Prozesse / Chair for Modelling and Simulation of Metallurgical Processes Department Metallurgie / Department of Metallurgy Montanuniversität Leoben / University of Leoben Franz-Josef-Str. 18 A - 8700 Leoben Österreich / Austria Tel.: +43 3842 - 402 - 3125 http://smmp.unileoben.ac.at

August 2, 2012, 15:18		#25
makaveli_lcf Senior Member Dr. Alexander Vakhrushev Join Date: Mar 2009 Posts: 189 Rep Power: 6	You wrote you get the SAME results. You can easily check which schemes you solver uses, just set "default none;", and you will get errors until you define all schemes. __________________ Best regards, Dr. Alexander VAKHRUSHEV Lehrstuhl für Modellierung und Simulation metallurgischer Prozesse / Chair for Modelling and Simulation of Metallurgical Processes Department Metallurgie / Department of Metallurgy Montanuniversität Leoben / University of Leoben Franz-Josef-Str. 18 A - 8700 Leoben Österreich / Austria Tel.: +43 3842 - 402 - 3125 http://smmp.unileoben.ac.at

Thread Tools
Show Printable Version Email this Page
Display Modes
Linear Mode Switch to Hybrid Mode Switch to Threaded Mode

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
DivSchemes Gauss limitedLinearV and interfaceCompression	nicasch	OpenFOAM Running, Solving & CFD	1	July 12, 2010 10:26
FvSchemes	sega	OpenFOAM Running, Solving & CFD	2	February 15, 2010 11:07
DivSchemes limitedLinearV and interfaceCompression	nicasch	OpenFOAM	0	February 28, 2008 10:33
Divschemes limitedLinearV 1 and interfaceCompression	nicasch	OpenFOAM	0	February 27, 2008 12:24
FvSchemes from 12 to 11	fedegavo	OpenFOAM Running, Solving & CFD	1	January 20, 2006 13:49

November 16, 2011, 03:02		#21
fippo_dk New Member Felipe Join Date: Apr 2010 Posts: 13 Rep Power: 5	how do your results compare to measurements? are you under or over predicting? In my case I over predict from 10 to 20 degrees

August 2, 2012, 15:12		#24
uli New Member Join Date: Jun 2012 Posts: 25 Rep Power: 2	Thanks for your reply. So using a k-Epsilon model does not involve R? But how is it possible that I get different results after switching the scheme for div(R)?

August 21, 2012, 18:34		#26
uli New Member Join Date: Jun 2012 Posts: 25 Rep Power: 2	hi, sorry for the late reply. I wrote "similar" results since the plots are slightly different. I removed the schemes and it still worked, thanks for your help.