[Ric_]

Good morning,

I am simulating the fluid dynamics of a tube that contains a porous medium. A gaseous mixture (mainly CO, H2, CO2 and small amounts of inert gases) enters the tube at 473K and is heated up by the presence of an external heating medium (that surrounds the tube and is assumed to be at a constant temperature of 240°C).

I would like the porous medium to have the same temperature as the fluid along the entire tube. In other words, I would like to impose "Isothermal" in the "Heat transfer" option of the "Solid Models" tab, and then insert the variable temperature ("T") instead of a constant value (the temperature of my solid changes and it is equal to the one of the fluid). When I try to do it, I get the following message:

Parameter 'Solid Temperature' in object '/FLOW:Flow Analysis 1/DOMAINefault Domain/SOLID MODELS/HEAT TRANSFER MODEL' is not allowed to be assigned an expression value that depends on variables. It must be assigned a numeric value, or an expression that resolves to a constant value.

I have read that it is it possible to change the rules of a simulation, but I was unable to do it. Could anyone explain in detail how I should put the dependency of the "Solid Temperature" to the option "ANY" and then make my file read my new RULES?

I have attached my ccl file in the zipped folder.

Thanks a lot,

Riccardo

[Opaque]

Do you want to have the whole solid fraction of the porous domain at a single uniform temperature, or locally at the temperature of the fluid?

For the former, add a source to the solid side with the proper coefficients and you can force it to be any temperature you want,

For the latter, just use a very large interface fluid-solide heat transfer coefficient in the porous settings.

Either model should be trivial.

[ghorrocks]

And, adding to Opaque's comments:

You cannot just edit the RULES file to add new functionality. Editing the RULES file is almost always a really bad idea. Leave the RULES file as it is an do any modelling you want in the supported interfaces - either CFX-Pre, CCL or user fortran.

[Ric_]

Good morning,

Thanks a lot for the reply.
The solution proposed by Opaque is very simple, but effective!

Thanks,
Riccardo