[torhans]

Hi

I have essentially a pipe with a controlled composition of 3 gases in one end and in the other end I have a semipermeable membrane, permeable to one of the gases (O2). Everything is laminar and there is no internal reactions. I want to control the O2 concentration or O2 flow in the semipermeable end.

I thought I could use reactionFoam with chemistry turned off and then using totalFlowRateAdvectiveDiffusive boundary. But it turns out that it can only be used with les problems.

What is my best approach.

Thanks
Torben

[Henning86]

Hi


I think you have to modify the solver and the solver needs to be compressible. set phi and mueff zero at the membrane in yeqn.h. This would result in an O2 concentration higher than 1. So you have to modify phi in pEqn.h and UEqn.h aswell. If the membrane is a boundary a new boundary condition should be enough.

MassfluxOverall = Sum (massfluxSpezies)

And the O2 massflux is zero.

I hope i understand the problem correctly could you provider a sketch?

Best regards,

Henning

Sry written with my phone

[torhans]

Thanks Henning

I expected that writing a new boundary condition was what I had to do. So I'm learning how to do that.

But why do you think that the solver must be compressible?

In the meantime I have attached a small (simplified) sketch of the problem. I guess massflux should be zero for all other species than O2.

Regards
Torben

[Henning86]

Hi,


i understood that you only have a channel with a predefined massflux. In the scenario with a gas reservoir a incompressible solver should be sufficient.

with the mass fluxes you are correct.

So you need a boundary condition for U.

O2 should be zeroGradient
The other species should be fixedValue 0

Best Regards

Henning