
[Sponsors] 
Energy equation in rhoCentralFoam (revisited) 

LinkBack  Thread Tools  Search this Thread  Display Modes 
May 31, 2016, 08:54 
Energy equation in rhoCentralFoam (revisited)

#1 
Senior Member
Join Date: Sep 2015
Location: Singapore
Posts: 102
Rep Power: 10 
Hello everyone,
I am a bit confused about the energy equation implementation in rhoCentralFoam. Particularly, why does the viscous term 'sigmaDotU' appear in the (supposedly) Euler equation? Code:
surfaceScalarField sigmaDotU ( "sigmaDotU", ( fvc::interpolate(muEff)*mesh.magSf()*fvc::snGrad(U) + (mesh.Sf() & fvc::interpolate(tauMC)) ) & (a_pos*U_pos + a_neg*U_neg) ); solve ( fvm::ddt(rhoE) + fvc::div(phiEp)  fvc::div(sigmaDotU) );  Kuragnov flux scheme is derived for strictly hyperbolic equations which are the Euler equations in this case  'a_pos' and 'a_neg' are calculated using the eigenvalues of Euler equations and hence, the flux scheme should not be applied to the viscous term 'sigmaDotU' However, when I modified the solver by removing the viscous term from this equation and introducing it in the internal energy/enthalpy equation later, the results were not as good as the current solver (more diffusion, shocks were not sharp, unexpected flow separation). So, I am guessing that whatever the current solver is doing is correct but I can't seem to understand why. There are two other threads that I managed to find asking the exact question but both remain open: 1. Viscous terms in rhoCentralFoam 2. energy equation in rhoCentralFoam I am hoping that someone would point in the right direction. Thanks to all! 

April 30, 2018, 11:46 

#2 
Member
K
Join Date: Jul 2017
Posts: 97
Rep Power: 8 
Did you find any answer for your questions ? I have the same exact questions.


May 2, 2018, 01:10 

#3 
Senior Member
Join Date: Sep 2015
Location: Singapore
Posts: 102
Rep Power: 10 
Hi Mary,
I am afraid that I do not have a complete answer but let me give my opinion and, hopefully, someone will confirm/correct it. The first thing to note is that if the case is inviscid, then muEff=0. So, the term 'sigmaDotU' does not have any effect on the results for inviscid flows. The second thing to note is that the energy equation is solved using a predictor step for rhoE: Code:
fvm::ddt(rhoE) + fvc::div(phiEp)  fvc::div(sigmaDotU) Code:
fvm::ddt(rho, e)  fvc::ddt(rho, e)  fvm::laplacian(turbulence>alphaEff(), e) a_pos and a_neg are essentially interpolation factors to compute an intermediate state at each face based on the pos and neg values obtained by TVD reconstruction. So, (a_pos*U_pos + a_neg*U_neg) can be seen as some intermediate velocity Ustar. This intermediate velocity is used in the calculation of 'sigmaDotU'. Code:
surfaceScalarField sigmaDotU ( "sigmaDotU", ( fvc::interpolate(muEff)*mesh.magSf()*fvc::snGrad(U) + (mesh.Sf() & fvc::interpolate(tauMC)) ) & (a_pos*U_pos + a_neg*U_neg) ); Cheers, USV 

July 5, 2018, 06:52 

#4 
Member
K
Join Date: Jul 2017
Posts: 97
Rep Power: 8 
Thanks a lot USV. Your explanation clarifies my question very well.


Tags 
energy equation, openfoam, rhocentrafoam 
Thread Tools  Search this Thread 
Display Modes  


Similar Threads  
Thread  Thread Starter  Forum  Replies  Last Post 
Energy equation issue  zytra  FLUENT  0  October 20, 2014 22:34 
Need experience comment on cavitation combine energy equation!!!  lehoanganh07  FLUENT  0  September 24, 2014 11:28 
Energy equation in OpenFoam  nwpukaka  OpenFOAM Programming & Development  2  June 24, 2014 03:40 
energy equation in rhoCentralFoam  nakul  OpenFOAM  0  October 10, 2010 16:07 
SIMPLE and energy equation convergence  Fabio  Main CFD Forum  0  June 1, 2007 07:06 