[chengtun]

I am trying to modify the case of wallBoiling which is OpenFOAM-dev/tutorials/multiphase/reactingTwoPhaseEulerFoam/RAS/wallBoiling
I only changed the fluid properties and geometry parameters and unfortunately got the bug of Negative initial temperature T0 after some iterative steps.
The initial temperature is 303.15K for both gas and liquid. I tried heat flux q in T.liquid for both zero and several thousand. They all ended in bug of Negative initial temperature. More details are as follows.
--> FOAM FATAL ERROR:
Negative initial temperature T0: -35.6173826

From function Foam::scalar Foam::species::thermo<Thermo, Type>::T(Foam::scalar, Foam::scalar, Foam::scalar, Foam::scalar (Foam::species::thermo<Thermo, Type>::*)(Foam::scalar, Foam::scalar) const, Foam::scalar (Foam::species::thermo<Thermo, Type>::*)(Foam::scalar, Foam::scalar) const, Foam::scalar (Foam::species::thermo<Thermo, Type>::*)(Foam::scalar) const) const [with Thermo = Foam::hRefConstThermo<Foam::rhoConst<Foam::specie> >; Type = Foam::sensibleEnthalpy; Foam::scalar = double; Foam::species::thermo<Thermo, Type> = Foam::species::thermo<Foam::hRefConstThermo<Foam:: rhoConst<Foam::specie> >, Foam::sensibleEnthalpy>]
in file /home/cpx/OpenFOAM/cpx-5.0/OpenFOAM-dev/src/thermophysicalModels/specie/lnInclude/thermoI.H at line 54.

FOAM aborting

#0 Foam::error::printStack(Foam::Ostream&) at ??:?
#1 Foam::error::abort() at ??:?
#2 Foam::heRhoThermo<Foam::rhoThermo, Foam::pureMixture<Foam::constTransport<Foam::speci es::thermo<Foam::hRefConstThermo<Foam::rhoConst<Fo am::specie> >, Foam::sensibleEnthalpy> > > >::calculate() at ??:?
#3 Foam::heRhoThermo<Foam::rhoThermo, Foam::pureMixture<Foam::constTransport<Foam::speci es::thermo<Foam::hRefConstThermo<Foam::rhoConst<Fo am::specie> >, Foam::sensibleEnthalpy> > > >::correct() at ??:?
#4 Foam::phaseSystem::correctThermo() at ??:?
#5 Foam::ThermalPhaseChangePhaseSystem<Foam::HeatAndM assTransferPhaseSystem<Foam::MomentumTransferPhase System<Foam::twoPhaseSystem> > >::correctThermo() at ??:?
#6 ? at ??:?
#7 __libc_start_main in "/lib/x86_64-linux-gnu/libc.so.6"
#8 ? at ??:?
Aborted (core dumped)

I tried to set temperature constraints in fvOptions for gas and liquid. The following example is for liquid. But it doesn't work.
temperature_constraints
{
type limitTemperature;
selectionMode all;
active true;
min 273;
max 600;
phase liquid;
}
I also tried to delete the judgment for negative temperature in OpenFOAM-dev/src/thermophysicalModels/specie/thermo/thermo/thermoI.H line 52-57
if (T0 < 0)
{
FatalErrorInFunction
<< "Negative initial temperature T0: " << T0
<< abort(FatalError);
}

But it still shows the same error. I am really confused why this bug appear and why the deletion doesn't work. When I open the file /OpenFOAM-dev/src/thermophysicalModels/specie/lnInclude/thermoI.H at line 54, the judgment has already been deleted.
Could anyone give me a hint? Thanks!

[zhutaihang]

Hi,
I'm using the rhoPimpleFoam solver. I had a similar problem as you. I watched some threads and found my solution. It is working well so far.

My simulation failed at the second iteration due to negative initial temperature T0. I firstly reduced the tolerance for pressure in fvsolutions by three orders. It worked for about 15 iterations before it failed again. then, I doubled the time interval, which makes the maximum Courant about 1. The problem seems to be solved. my simulation runs for hundreds of iteration and it is still working.

I wish that it helps.

[chengtun]

Thank you so much for your help!

[calmpeace]

Sorry were you able to solve this? Thanks

[Robin.Kamenicky]

Hi Tun,

in general the reactingTwoPhaseEulerFoam can be quite unstable. The physics is complex and depends on lot of parameters. If you need further help I would ask you for more information about your computation.

Btw, was the fvOptions file read properly at the beginning of your simulation?

Kind regards,
Robin

[Scram_1]

I don't think increasing the time step is the right approach. Even though it makes the solver chug along and spit out results, some of the physics that occurs in the short time interval will not be captured by increasing the time step. Even I'm facing a similar issue (I'm using reactingParcelFoam btw) in that when I keep the time step at 1e-8, I can clearly see the droplets in my domain condense (which is what is expected) but the simulation crashes in 5 - 6 time steps. When I increase the time step to 1e-6 (which is small but not as small as 1e-8), the simulation runs without crashing but the mass transfer is such that the droplets initially evaporate before they can start condensing which is something that should not occur. So, probably tweaking the BC's (pressure, temperature) seems like a better bet to solve this issue.

[zhutaihang]

Originally Posted by Scram_1

I don't think increasing the time step is the right approach. Even though it makes the solver chug along and spit out results, some of the physics that occurs in the short time interval will not be captured by increasing the time step. Even I'm facing a similar issue (I'm using reactingParcelFoam btw) in that when I keep the time step at 1e-8, I can clearly see the droplets in my domain condense (which is what is expected) but the simulation crashes in 5 - 6 time steps. When I increase the time step to 1e-6 (which is small but not as small as 1e-8), the simulation runs without crashing but the mass transfer is such that the droplets initially evaporate before they can start condensing which is something that should not occur. So, probably tweaking the BC's (pressure, temperature) seems like a better bet to solve this issue.

Hi,

I agree with you. It somehow worked for a specific case of mine, but I don't think it is a good approach. Investigating the boundary condition is definitely a better method.