[Rachel]

Dear All,

I would like to know if this solver is available for testing/preview/further development?

I am interested in a similar problem involving phase change and chemical reactions.

[sandy]

Dear all,

I have three questions about interPhaseChangeFoam solver:

1) In OF, it seems that the model fomulation in the file Kunz.C is different from the literature, for example,

in OF, it is:
Kunz::mDotAlphal():
( mcCoeff_*sqr(limitedAlpha1)
*max(p - pSat(), p0_)/max(p - pSat(), 0.01*pSat()),
mvCoeff_*min(p - pSat(), p0_) )

Kunz::mDotP():
( mcCoeff_*sqr(limitedAlpha1)*(1.0 - limitedAlpha1)
*pos(p - pSat())/max(p - pSat(), 0.01*pSat()),
(-mvCoeff_)*limitedAlpha1*neg(p - pSat()) )

where,
limitedAlpha1 = min(max(alpha1_,scalar(0)),scalar(1));
mcCoeff_ = Cc_*rho2()/tInf_;
mvCoeff_ = Cv_*rho2()/(0.5*rho1()*sqr(UInf_)*tInf_);


However, in literature, it is:

Mprod=(Cv/0.5*UInf2*tInf2)rho2/rho1*gamma*min[0,p-pv]
Mdest=(Cc/tInf)rho2*gamma2[1-gamma]

Are they same? What is the differences?


2) In the UEqn.H of interPhaseChangeFoam,
----------- UEqn.H ----------------
surfaceScalarField muf =
twoPhaseProperties->muf()
+ fvc::interpolate(rho*turbulence->nuSgs());
fvVectorMatrix UEqn
(
fvm::ddt(rho, U)
+ fvm::div(rhoPhi, U)
- fvm::Sp(fvc::ddt(rho) + fvc::div(rhoPhi), U)
- fvm::laplacian(muf, U)
- (fvc::grad(U) & fvc::grad(muf))
//- fvc::div(muf*(fvc::interpolate(dev2(fvc::grad(U))) & mesh.Sf()))
);

UEqn.relax();

if (momentumPredictor)
{
solve
(
UEqn
==
fvc::reconstruct
(
(
fvc::interpolate(interface.sigmaK())*fvc::snGrad(g amma)
- ghf*fvc::snGrad(rho)
- fvc::snGrad(pd)
) * mesh.magSf()
)
);
}
--------------------------------------------
What is the meanings ' UEqn.relax() ' ? In PISO, it still need some relaxation factors? wrong ...

3) If I want to simulate the full wet flows of a hydrofoil firstly, how should I specify the parameters? I just need to set Cc_ and Cv_ 0, right? But I still could not get the convergent results, why?

Could you help me to explain them? Thank you very much.

Regards,
Sandy
sandy.lee37@gmail.com

[sandy]

Hi man,

Now, I don't want to fight with the Liquid Hydrogen or flash evaporation again. I just want to make the foil caviting, but why my gamma equation can not be solved by interPhaseChangeFoam. I choose the oneEqEddy in the Les turbulent model and Kunz's interPhaseChange model during my simulating. I could not find what is the matter with them.

Please give a finger.

Thank you very much.
Sandy

[isabel]

In the interPhaseChange solver. In gammaEqnSubCycle.H, we have this line:

MULES::implicitSolve(oneField(),gamma,phi,phiGamma ,Sp,Su,1,0)

Does anybody know what means " oneField() " ?
And Sp, Su ?

[gschaider]

Originally Posted by isabel

Does anybody know what means " oneField() " ?

The name means "Questions about this class should only be asked once (and not in multiple unrelated threads) otherwise people will stop answering"

Another explanation can be found here: http://foam.sourceforge.net/doc/Doxy...1oneField.html

The second explanation is the right one, but I like the first one better

[sandy]

You are great! Gschaider. Sure, we should study from busy ants from onefield to another and others in everyday, right? And should have multihead but one .

In fact, I do feel confused to use this solver in these days .

In Kunz.C and phaseChangeTwoPhaseMixture.C of this solver, the variables mDotAlphal() and vDotAlphal() are const, so if I want to change the values Cc and Cv of the file transportProperties from 0 (namely, the full-wet flows) to 30000 and 900000 (namely, Kunz's model), the code will report error infomations. I would like to know how to avoid this problems?


Thanks.

[sandy]

Hi isabel, Sp and Su are the direct variables of vDotAlphal(), however, they are just indirectly connect with mDotAlphal().I think it is maybe a key to this solver.

[sandy]

Sp = (vDotvAlphal - vDotcAlphal) * gamma , and it will be solved implicitly.
Su = vDotcAlphal, and it is an explicit term in this equation.

But, why Su also includes the term divU*gamma in gammaEqu.H ? If the MULES::implicitSolver was chose, this term should be deleted, right?

Who knew it? Please help me out. Thanks.

[isabel]

I have studied the tutorial "Solve cavitating flow arroun a 2d hydrofoil using a user modified version of interPhaseChangeFoam", cut I don't understand what these mean:

vDotvAlphal
vDotcalphal
vDotvP
vDotcP

Does anybody know what these mean?

[sandy]

Originally Posted by isabel

I have studied the tutorial "Solve cavitating flow arroun a 2d hydrofoil using a user modified version of interPhaseChangeFoam", cut I don't understand what these mean:

vDotvAlphal
vDotcalphal
vDotvP
vDotcP

Does anybody know what these mean?

vDotvAlphal = (Cv*rho2/(0.5*rho1*sqr(UInf)*tInf))*min(p - pSat, 0)*{1.0/rho1 - gamma*(1.0/rho1 - 1.0/rho2)}

vDotcalphal = =(Cc*rho2/tInf)*gamma^2*{1.0/rho1 - gamma*(1.0/rho1 - 1.0/rho2)}

vDotvP = -{Cv*rho2/(0.5*rho1*sqr(UInf)*tInf)}*gamma*(1.0/rho1 - 1.0/rho2)

vDotcP = ={(Cc*rho2/tInf)*gamma^2*(1.0 - gamma)/(p-pSat)} *(1.0/rho1 - 1.0/rho2)

Am I right?

[isabel]

Thank you very much, sandy.
By the way, what you asked before about what is the meaning ' UEqn.relax() ' In PISO. This line is to adjust the coefficients so as to incorporate the underrelaxation coefficient, such that a solution to UEqn == source will produce a partly-relaxed version of U.
It is explained in this link:

http://www.cfd-online.com/Forums/ope...implefoam.html

[sandy]

I think, in UEqn.H, the red and blod terms which is

fvVectorMatrix UEqn
(
fvm::ddt(rho, U)
+ fvm::div(rhoPhi, U)
- fvm::Sp(fvc::ddt(rho) + fvc::div(rhoPhi), U)
- fvm::laplacian(muf, U)
- (fvc::grad(U) & fvc::grad(muf))
//- fvc::div(muf*(fvc::interpolate(dev2(fvc::grad(U))) & mesh.Sf()))
);

should be deleted, because they destroyed the original PDE. Am I right?

[eric_zh]

Dear Hamed,

have you got any progress in your simulation? I'm simulating something very similar, liquid propane release into atmospheric area. I started with dieselFoam with evaporation model on with no success, then I found your thread. If there is any experience you can share with me, it will be highly appreciated.

Regards,
Eric

Originally Posted by haghajani

Dear All,

How can I trace the vaporized liquid (changed phase from 1->0) in to air.
I am trying to study the effect of flash evaporation in high pressure liquid Hydrogen release. I used interPhaseChangeFoam for a simple case and the results seems reasonable. Now I want also see the vapor of hydrogen inside surrounding air.

Best regards,
Hamed Aghajani
hamed.aghajani@gmail.com
h.aghajani@kingston.ac.uk

[haghajani]

Dear Eric,
I skipped tracing vaporized liquid at that time, and haven't returned back yet.
To do that, I think, there should be an extra equation for vaporized liquid. Be cause of miscible behavior of vaporized liquid and surrounding gas, say, Air, the new solver should have capability to act as "reactingFoam" do and calculate the mixture thermoPhysical property in that way. Beside this it is better to add compressibilty and turbulence modelling as well.

I have no specific Idea how to do All this.

Seeking for every comment and help

Hamed

[Rachel]

Hello,

Have you looked at rhoReactingFoam ?
It has compressibility effect with multiphase reacting flows.

I am trying to use, however yet to figure out the details. The absence of tutorial is making it tough to create a case to test this solver.

[Rachel]

Forgot to mention it is a new solver in OF-1.6

[eric_zh]

Dear Hamed,

thanks for your reply and suggestion. I'll continue my simulation and see how it goes for the time being. I'll update you guys later if I made any progress.

have a nice day!
Eric

[eric_zh]

Hi Rachel,

Haven't update to 1.6 yet, I'm using 1.5.x. So have no idea how the new solver works! Probablly will take a look at it tomorrow.

Cheers,
Eric

[haghajani]

Rachel,

Thank you very much for your clue,
I try to shift to 1.6 and before that, OpenSUSE 11.1 .
Please Keep us update of any further progress,

B4N
Hamed

[sandy]

Hi Rachel, are there some cases or tutorials about interPhaseChangeFoam solver in OF_1.6.