Help with multiphysics

[Anonymous]

Hi I am Asif

I need help simulating, fluid flow through a porous medium. I created a solid cylinder in inventor with a inlet and outlet. The fluid I want to use to perform the simulation is air. I want to find the pressure on the porous medium, with a velocity of 20cm/min coming out the outlet. I having problems trying to simulate this scenario. I look at the tutorials provided, but it does not help. Can you please assist me with this task.

[Joey.X]

I have uploaded react2.fem file which is the model only without result. 

And uploaded react2b.zip which inculdes everything, if  you are using  window-7, right click to "extract all", then open .fem file. 

[Anonymous]

Hi Joey

Thanks alot for all your help

I made a adjustment to the model. The model needed a new output. I made changes and I got the convergence error again. The only parameters i need is 20 in/s and 14.7 psi on the outlet. Can you please help me with this. I put the new model inside the dropbox as the name of reactor 3. The results I am looking for is the pressure at the inlet and the effect of the pressure on each of the different material.

Thank you

[Joey.X]

if your intention is put 14.7 psi at outlet(a single outlet only), it is suggested to set surface to zero psi pressure BC which is much easier  to converge in solver, theoretically the result pattern should has no difference except the pressure is offsetted with 14.7 psi. PS, you can interpret result  as gauge pressure as well.  Let me know if this trick works for you. 

[Anonymous]

I tried what you said, but I still have the converge problem. 

[Anonymous]

I uploaded a new model name reactor 5. I still have the convergence problem. I tried what you told me but I still get that error. Can you help me fix it 

[Joey.X]

In most recent uploaded mode React5.fem, the boundary conditions are specified redundantly, which causes problem over-constraint and solution divergence. A golden rule is, don't define multiple BCs on same surface, i.e, pBC, ioBC, vBC are exclusive in any single surface.

[Anonymous]

So should I name one as inlet. Than the other surface that I want to be a outlet, just put velocity bc? Will this help me find the pressure on the inside and pressure at the outlet?

[Anonymous]

I mean pressure at the inlet

[Joey.X]

For better numerical convergence, I would recommend setting one end as velocity BC and the other end as pBC, set to zero if the model has one inlet and one outlet, I stated the reason in previous posts from this thread. And the velocity BC set to fully constraint, i.e, make them zero in two tangential directions. 

PS, exclude boundary layer mesh at the inlet and outlet surfaces if you have applied boundary layer meshes. 

[Anonymous]

Okay but in the actually model, the only thing I know it the velocity and pressure at the outlet. So by doing it this way will it help me find the pressure inside and at the inlet?

[Joey.X]

This is a typical modeling issue, the physcial model has to be converted to reasonable numerical model in sumilation, and (probally) the simulation results may need some treatments such as simple offset. 

Physically you know the pressure and velocity at one outlet only, in simulation setup, to make solution stable numerical model, you can specify vBC at outlet as what you know, and zero pressure at inlet (this is a assumption). simulation will calculate compelete fluid domain including all nodal velocity and pressure, you may get the pressure value not consistent on outlet as you already know, then you can make an pressure offset in pressure result. Hope this helps.

[Anonymous]

I tried what you said, but I still got a convergence error. When I try to set the pressure zero at one surface, it says I can not do that. 

[Anonymous]

I do not know what to do about the divergence error.

I uploaded the new model, where I made the boundary conditions simple.