[vinz]

Dear foamers,

I have some domain filled with air and I would like to inject another gas into the domain at an inlet patch.

This gas has got different thermoPhysicalProperties but I don't want to take into account any reaction.

So I just want to be able to inject a different gas at the inlet from what I have in my domain.

Is there a way to do this in OpenFOAM? does groovyBC would be a mean for instance?

[nimasam]

some suggestion
1- two phase flow solver , for example interFoam
2-reactingFoam, with out combustion

[vinz]

This afternoon I tried compressibleMultiphaseInterFoam which looks interesting.

However since it is based on interFoam I am not sure that it takes into account gas mixing like it is the case into twoLiquidMixingFoam for instance?

[tomf]

reactingFoam without would work. It does however assume a Schmidt number of 1 if I am not mistaken. This may be inaccurate. interFoam or similar will not have mixing.

Regards,
Tom

[vinz]

Ok, thanks i will give it a try.

[vinz]

Ok, So I tried the whole day to get something working within reactingfoam without much success unfortunately.

In fact, the simulation is running for some thousands iterations but it stops at some point without any apparent reason. The timePrecision is increased while the deltaT is decreased radically and the code crashes:
"Increased the timePrecision from 637 to 638 to distinguish between timeNames at time 0.00154733"

I tried both laminar and turbulent with the same result, although turbulent is running a bit longer.

I tried to activate/deactivate combustion in combustionProperties file but the result is the same.
I also tried to activate/deactivate chemistry in chemistryProperties file but then again the result is the same.

My mesh is good according to checkMesh.
I don't see any specific problem on my boundary conditions, especially since the simulation is running for some time.

The only problem I could think of is that the simulation seems to crash when speed is increasing and flow becomes supersonic. Could that be the reason?

Any help would be highly appreciated. It seems that reactingFoam is a good candidate so I would be really happy to make it work.

[tomf]

For supersonic flow I would suggest rhoReactingFoam. This uses density based thermodynamics which seems to cope with supersonic/transonic flow better.

Regards,
Tom

[vinz]

I will try it as well.

This week-end I figure out the problem was coming from the fvSchemes and the fvSolution files.
Simulation is working now with the upwind scheme and GAMG solver instead of PCG for pressure.
I have to see if I can still run with another scheme than upwind which is not the most accurate. Maybe using rhoReactingFoam will help.

[rohitpurdue]

Hi Vinz,

I am also trying to simulate mixing problem using reactingFoam an my simulation too crashes after running for around 0.1 seconds that is its time precision starts increasing and it goes upto 630 and then blows up. It would be very helpful if you can help me figure out this problem.


Thanks,

Rohit

[vinz]

Did you try the schemes and solvers I was indicating in my previous message?

[rohitpurdue]

Hi Vinz,

No I have not tried those but I want to try them, as I am new to CFD can you please tell me where did you use those upwind schemes? If possible can you share your fvscheme script. I have been trying to run this case for last 2 weeks continuously without any success. I really appreciate your help and guidance.

Thanks,

Rohit

[jherb]

Have you checked if your concentrations become unbounded? Do you see warnings in the log output? Have you chosen bounded schemes for the bounded variables?