Euler solver in BCs
 

Hi,
I'm trying to implement a boundary conditions that links pressure to outflow using a differential equation (Windkessel) and therefore need to solve it. A simple 1st order backward Euler solver should be enough and, fortunately, one is available in OF.

I've never added a new BC but I tried a take an existing one, use the outflow on the selected patch to calculate the pressure and export it.

I was wondering : what it the best existing BC to start from ? Has anyone an example a the use of a solver in a BC function ?


Best regards,

Florian

Dear Hiroshiman

The following pdf by Håkan Nilsson in Chalmers University is so helpful.

How to implement a new boundary conditions:
http://www.tfd.chalmers.se/~hani/kur...yCondition.pdf

Also, take a look at his homepage for more tutorials:

http://www.tfd.chalmers.se/~hani/index.html
http://www.tfd.chalmers.se/~hani/kurser/OS_CFD_2009/

Hi Kia,
thank you for the links. Actually I've already read the Chalmers guide for the programmation of BCs.
Currently I'm using the codedFixedValue BC with a code that calculates p from phi on the outlet.
I still have a few problems : how can I access the mean value of the pressure on the patch ? I'm able to do it for phi but not for p (surfaceField vs volume Field).

I must solve something like :
d(p-phi*R)/dt=f(p,phi) to impose p on the outlet, is it possible to use the ddt solver of openfoam ? I coded an Euler solver but I'd like to use a maximum of integrated functions...


Regards,

Florian

Hi,

Hiroshiman, did you manage to get it to work?
Here is how you can access pressure.
Did you have to solve the ode at every internal iteration in a timestep?

I am using resistance boundary conditions, where the pressure outlet is simply the product of a resistance value and the flowrate at the outlet.

I have coded it with codedFixedValue but am not getting the same results as those I find in Fluent.

Any pointers would be appreciated,

Kind regards,
Andreas

HI, It's been a long time, but has anyone solved this problem? I should be doing the exact same thing.

Thanks!