CFD Online Logo CFD Online URL
www.cfd-online.com
[Sponsors]
Home > Forums > OpenFOAM Programming & Development

"phi" variations in OF

Register Blogs Members List Search Today's Posts Mark Forums Read

Reply
 
LinkBack Thread Tools Display Modes
Old   December 10, 2014, 01:05
Default "phi" variations in OF
  #1
Senior Member
 
Tushar Chourushi
Join Date: Jul 2009
Location: IIT-Indore, India
Posts: 319
Blog Entries: 1
Rep Power: 9
Tushar@cfd is on a distinguished road
Dear All,

I was wondering how to relate these two "phi" variations of OpenFOAM.

In simpleFOAM solver "phi" is represented as:

Code:
phi=linearInterpolate(U) & mesh.Sf();
And, in icoFoam solver "phi" is represented as:

Code:
phi = (fvc::interpolate(U) & mesh.Sf())
                + fvc::ddtPhiCorr(rUA, U, phi);
What is the significance of additional term (color)?


-
Best Regards!
Tushar@cfd is offline   Reply With Quote

Old   December 10, 2014, 08:30
Default
  #2
Member
 
Thomas Vossel
Join Date: Aug 2013
Location: Germany
Posts: 45
Rep Power: 4
ThomasV is on a distinguished road
Have a look at this: fvc::ddtPhiCorr(rUA, U, phi)
ThomasV is offline   Reply With Quote

Old   December 11, 2014, 00:12
Default
  #3
Senior Member
 
Tushar Chourushi
Join Date: Jul 2009
Location: IIT-Indore, India
Posts: 319
Blog Entries: 1
Rep Power: 9
Tushar@cfd is on a distinguished road
Dear Thomas,

I had a look to the post but still It has not clear my actual doubt on it's significance. I think you should look at the following post, similar discussion was carried out but not much information on its significance is revealed in the same.

DdtPhiCorr

-
Best Regards!
Tushar@cfd is offline   Reply With Quote

Old   December 11, 2014, 04:24
Default
  #4
Member
 
Thomas Vossel
Join Date: Aug 2013
Location: Germany
Posts: 45
Rep Power: 4
ThomasV is on a distinguished road
Ok - if that wasn't enough for you I can just give you some theory about the solver's background:

Hrvoje Jasak, PhD 1996, PDF of thesis posted at: Error analysis and estimation in the Finite Volume method with applications to fluid flows.
Henrik Rusche, PhD 2002: Computational fluid dynamics of dispersed two-phase flows at high phase fractions.

Maybe have a look at Rusche's PhD on page 110 and onwards - there he starts talking about how to calculate the fluxes but I can't check in detail right now if the given correction term is mentioned in this part of the text...

Oh and by the way - icoFoam recently was rewritten to the following form:

Code:
         // --- PISO loop    66 
    67         for (int corr=0; corr<nCorr; corr++)
    68         {
    69             volScalarField rAU(1.0/UEqn.A());
    70 
    71             volVectorField HbyA("HbyA", U);
    72             HbyA = rAU*UEqn.H();
    73             surfaceScalarField phiHbyA
    74             (
    75                 "phiHbyA",
    76                 (fvc::interpolate(HbyA) & mesh.Sf())
    77               + fvc::interpolate(rAU)*fvc::ddtCorr(U, phi)
    78             );
    79 
    80             adjustPhi(phiHbyA, U, p);
Still pretty much the same though...
ThomasV is offline   Reply With Quote

Old   December 11, 2014, 05:00
Default
  #5
Senior Member
 
Tushar Chourushi
Join Date: Jul 2009
Location: IIT-Indore, India
Posts: 319
Blog Entries: 1
Rep Power: 9
Tushar@cfd is on a distinguished road
Quote:
Originally Posted by ThomasV View Post
Ok - if that wasn't enough for you I can just give you some theory about the solver's background:

Hrvoje Jasak, PhD 1996, PDF of thesis posted at: Error analysis and estimation in the Finite Volume method with applications to fluid flows.
Henrik Rusche, PhD 2002: Computational fluid dynamics of dispersed two-phase flows at high phase fractions.

Maybe have a look at Rusche's PhD on page 110 and onwards - there he starts talking about how to calculate the fluxes but I can't check in detail right now if the given correction term is mentioned in this part of the text...

Oh and by the way - icoFoam recently was rewritten to the following form:

Code:
         // --- PISO loop    66 
    67         for (int corr=0; corr<nCorr; corr++)
    68         {
    69             volScalarField rAU(1.0/UEqn.A());
    70 
    71             volVectorField HbyA("HbyA", U);
    72             HbyA = rAU*UEqn.H();
    73             surfaceScalarField phiHbyA
    74             (
    75                 "phiHbyA",
    76                 (fvc::interpolate(HbyA) & mesh.Sf())
    77               + fvc::interpolate(rAU)*fvc::ddtCorr(U, phi)
    78             );
    79 
    80             adjustPhi(phiHbyA, U, p);
Still pretty much the same though...
Dear Thomas,

First of all, it is an open forum right? why being rude?

Secondly, You have not understood my question properly. If you re-read my post "the colored term" seems to be the problematic term. What is the actual significance of the term?
I would be very glad, if you could relate this colored term (fvc::ddtPhiCorr(rUA, U, phi); ) with the actual physics of the problem. Also by googling, the expression for this term can be formulated. But the actual problem is to relate this term with the physics, right? Now suppose if this term doesn't have any effect on the fluxes so, why it is mentioned in the "icoFOAM" solver? Right?

-
Best Regards!

Last edited by Tushar@cfd; December 11, 2014 at 06:13. Reason: Incorrect sentence
Tushar@cfd is offline   Reply With Quote

Old   December 11, 2014, 08:39
Default
  #6
Senior Member
 
Tushar Chourushi
Join Date: Jul 2009
Location: IIT-Indore, India
Posts: 319
Blog Entries: 1
Rep Power: 9
Tushar@cfd is on a distinguished road
Dear Thomas,

It's for your information please refer to the following post:

DdtPhiCorr

It is clear now that the colored term is "correction" which improves stability of "icoFOAM" solver for some cases.

-
Best Regards!
Tushar@cfd is offline   Reply With Quote

Old   December 11, 2014, 11:00
Default
  #7
Member
 
Thomas Vossel
Join Date: Aug 2013
Location: Germany
Posts: 45
Rep Power: 4
ThomasV is on a distinguished road
Ehm - I had no intention of being rude. Sorry if you got that impression...

I think I misunderstood your question or at least now I'm a little bit confused what extactly you did want to know. The link you just wrote down here was mentioned as first answer in the topic which I initially suggested you to read in my first posting here so I thought you had read that already and wanted to know something else. As you asked for the "significance" of the term I thought you wanted to know how much of an impact this has and not what its general meaning is. That is why I then gave the links to the theory behind the solver where I guess at some point this kind of correction and its reasons / impact are mentioned. I couldn't find the right spot by just skipping through it and looking this part up is nothing that is done quickly (it might even not be in there) so that's why I just was able to give you the background to have a look at for yourself...
ThomasV is offline   Reply With Quote

Old   December 12, 2014, 00:06
Default
  #8
Senior Member
 
Tushar Chourushi
Join Date: Jul 2009
Location: IIT-Indore, India
Posts: 319
Blog Entries: 1
Rep Power: 9
Tushar@cfd is on a distinguished road
Dear Thomas,

It's all Okay, No problem. I am also Sorry for not clearly briefing my question which led to misunderstanding. Anyways, we all are here to clarify our research doubt in a good manner and open forum like this is a good initiative.

-
Best Regards!
Tushar@cfd is offline   Reply With Quote

Reply

Thread Tools
Display Modes

Posting Rules
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is On
HTML code is Off
Trackbacks are On
Pingbacks are On
Refbacks are On


Similar Threads
Thread Thread Starter Forum Replies Last Post
FloEFD - Geometry Variations sys999 FloEFD, FloWorks & FloTHERM 3 March 27, 2014 21:27
How to integrate "phi" on a patch ubaid OpenFOAM 0 March 18, 2010 08:24
flow in ducts with streamwise variations of radius yonghong yan Main CFD Forum 1 July 31, 2002 00:14
variations on the equation of continuity 98.4F Main CFD Forum 0 August 10, 2001 07:22
Capturing Large Variations using FINITE DIFFERENCE Yogesh Talekar Main CFD Forum 1 August 20, 1999 12:21


All times are GMT -4. The time now is 19:01.