Coupling to a material code

by **jano** on Tue Feb 10, 2009 7:40 pm

I would like to couple a CFD code to a material code. The coupling would be on the energy equation only.
To begin, I would like to implement a basic problem in an existing CFD code: the heat transfer equation inside a solid.
dT/dt = a d²T/dx²
Would it be feasible quicly in Free_CFD? If yes, could someone give me some guidelines on how to do this efficiently?

Thanks,
Jano

For the numerical method, I was thinking to an implicit formulation + Runge-Kutta for the time marching problem. But CFD is not my speciality, so any advice would be greatly welcome!

**Sponsor**

by **esozer** on Tue Feb 10, 2009 7:50 pm

Heat equation would be very easy to accommodate within Free CFD. Just as a first step (without modifying the code), molar mass of the fluid can be set to a really large value and viscosity can be set to zero. When right thermal conductivity is given, it would be OK.

But, a little work in the code to handle solid conduction wouldn't be hard and it would be much more efficient of course. After all, equation is already there being solved implicitly. It just has convection and other fluid related stuff that needs to be excluded.

I need to know more about the coupling but it sounds like and easy step too.

How does the heat conduction within a solid and the coupling fit in together? If you can let me know more specifics, I can offer help. This sounds like an interesting application.

by **jano** on Tue Feb 10, 2009 8:09 pm

Thank you for your quick answer!

Let me present myself, it will help. I am a postdoc in material modeling. I am modeling the response of thermal protection systems for re-entry capsules (like Apollo). Until now, I was using commercial tools, but I came up with an equation system that I cannot solve anymore with commercial tools.
So, my idea was to write a material code including heat transfer, some chemistry (the material is pyrolysing)... Of course, I would rather avoid starting from scratch. And in the future, I would like to couple the material code to a CFD code. Why this coupling? The material is heated by the surrounding flow (hypersonic entry). The temperature at the wall is given by a balance of the impinging aerothermal flux and the flux transfered inside the material. In other words, the material acts as a sink term at the wall.

To begin, developping only the material part would be great. I was planning to spend more or less 3 months on this first part...

by **esozer** on Tue Feb 10, 2009 8:30 pm

OK, this sounds familiar. I have a friend here in UofM who has done his PhD on exactly the same topic. He was also interested in utilizing Free CFD in the future. Three months for the first part sounds like enough, if not more than enough. I will think of a way to incorporate the solid part to the code. That should be useful in many cases. You can maybe get familiar with the code a bit in the mean time. I'll contact you when I come up with something.

by **jano** on Tue Feb 10, 2009 9:04 pm

This sounds good... indeed, I think that it would be very interesting to have such a coupling for many applications. I am ready to develop the material part with your guidelines. I have to do it anyway, either on the basis of your code or on the basis of another.
BTW, I think that I know your friend. Is he from Quebec?

by **esozer** on Tue Feb 10, 2009 9:19 pm

I think he is

by **phsieh2005** on Tue Oct 27, 2009 9:18 am

Hi,

Just wondering if there is any progress on this front. I am interested in conjugate heat transfer with both fluids and solids in the model.

phsieh

by **esozer** on Wed Oct 28, 2009 5:28 pm

We didn't get to work on it since I was finishing up my dissertation. But it is becoming clear that one of the most wanted features is conjugate simulations. So I will give it priority. Follow the feed on the main page and you will know as soon as that feature is added.

**Sponsor**