How does openFoam calculate residuals?

sahm · March 1, 2010, 14:10

I have solved a case with the density based solver rhoPisoFoam. When I checked the residuals for convergence, I noticed that the residuals for Rho were zero in every time step. Does anyone know how it can be? Does any body know how does OpenFoam calculate the residuals in every timestep?
Since the Piso loop in solver solves the rho, I thought this can be the result. But rho is being solved several times in piso loop, so it is changing, and this can be the method for calculation of residuals?

Houthuys · October 18, 2017, 17:27

Quote:

Originally Posted by sahm

Does any body know how does OpenFoam calculate the residuals in every timestep?

Did you find answer to your question?

I know how the linear solver residual is calculated. But how the SIMPLE residual is calculated?

krikre · October 20, 2017, 06:41

I'm interested in this topic and will try to contribue what I can.

The correction of rho is outside of the PISO loop.

https://github.com/OpenCFD/OpenFOAM-.../rhoPisoFoam.C

See the:

Code:

#include "rhoEqn.H"

I'm guessing this is the rhoEqn.H. The fvm term is implicit and the fvc is explicit. I assumed that this simple equation

Code:

solve(fvm::ddt(rho) + fvc::div(phi));

reduces to an assignment operation. Correct me if I'm wrong.

Houthuys, if you could exlpain the linear solver residual, I'd be grateful. I found this on the OF+ site. Do you understand this?

I guess $\bar{ \textbf{x}} = \frac{1}{N}\sum_i^N x_i$ and $\vert\textbf{x}\vert = \sqrt{\sum_i^N x_i^2}$
but I still don't understand the logic behind the calculation of

.

krikre · October 26, 2017, 17:24

Anyone looking at this thread should also see
Residuals for convergence of segregated solvers

March 1, 2010, 14:10	How does openFoam calculate residuals?	#1
sahm Senior Member Seyyed Ali H.M. Join Date: Nov 2009 Location: Utah Posts: 107 Rep Power: 15	I have solved a case with the density based solver rhoPisoFoam. When I checked the residuals for convergence, I noticed that the residuals for Rho were zero in every time step. Does anyone know how it can be? Does any body know how does OpenFoam calculate the residuals in every timestep? Since the Piso loop in solver solves the rho, I thought this can be the result. But rho is being solved several times in piso loop, so it is changing, and this can be the method for calculation of residuals?

October 20, 2017, 06:41		#3
krikre Member Kristjan Join Date: Apr 2017 Location: Slovenia Posts: 36 Rep Power: 8	I'm interested in this topic and will try to contribue what I can. The correction of rho is outside of the PISO loop. https://github.com/OpenCFD/OpenFOAM-.../rhoPisoFoam.C See the: Code: #include "rhoEqn.H" I'm guessing this is the rhoEqn.H. The fvm term is implicit and the fvc is explicit. I assumed that this simple equation Code: solve(fvm::ddt(rho) + fvc::div(phi)); reduces to an assignment operation. Correct me if I'm wrong. Houthuys, if you could exlpain the linear solver residual, I'd be grateful. I found this on the OF+ site. Do you understand this? I guess $\bar{ \textbf{x}} = \frac{1}{N}\sum_i^N x_i$ and $\vert\textbf{x}\vert = \sqrt{\sum_i^N x_i^2}$ but I still don't understand the logic behind the calculation of .

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

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
Calculate aerodynamic coefficients with openfoam using only opensource programs	Xwang	OpenFOAM	20	May 20, 2016 11:26
How to view residuals during OpenFOAM iteration?	ericnutsch	OpenFOAM	2	October 5, 2009 15:12
[Commercial meshers] How To Use OpenFOAM Software To Calculate Pump?	renyun0511	OpenFOAM Meshing & Mesh Conversion	2	August 30, 2009 21:49
Modified OpenFOAM Forum Structure and New Mailing-List	pete	Site News & Announcements	0	June 29, 2009 05:56
Best way to calculate residuals	Morteza	Main CFD Forum	1	August 19, 2006 23:44

October 26, 2017, 17:24		#4
krikre Member Kristjan Join Date: Apr 2017 Location: Slovenia Posts: 36 Rep Power: 8	Anyone looking at this thread should also see Residuals for convergence of segregated solvers