Hi! Dear all,

I have learnt CFD for two years, but I still have a limited knowledge of CFD. Because it is my first time to study CFD here.

I am trying to make a code to simulate the shock tube problem by using C language so far.

Now the problem that I face now is that how to set up the second diaphragm on the code.

So could you please help me and tell me the way to solve the problem???

Thank you very much!

Po.

I understood your question in such a way: to perform a 1-dimensional simulation of the shock-tube problem with a second diaphragm which is a local restriction, a model is needed to consider a wave interaction with this element.

So, you may try to treat this element as a special case of the border between 1D computational cells - a restriction of the "diaphragm" type, and solve the generalized Riemann problem to obtain computational fluxes (mass, momentum and energy) on the both sides of diaphragm on each time step.

This approach is rather straightforward to those familiar with Godunov-like methods. Generalized flux evaluation procedure must take into account characteristic properties of the governig equations, and certain hydraulic properties of the diaphragm as a local restriction. In other words, it is necessary to consider information brought by characteristics, and to fulfill the simple conditions of local quasi-steady flow through diaphragm itself.

The following conditions are commonly used: mass and enrergy conservation in steady flow, and a relation expressing hydraulic losses on diaphragm, for example a ratio of total pressures on each side (as a function of reference Mach number).

Goog luck!

Hi! Dr./Mr. Chernousov,

Thank you very much for your reply. Your suggestion is very valuable for me! I will try this way which you suggest me. One more thing, may I ask you when I have any further question about this subject?

Anyway, thank you very much!

Po.

Mr. Po!

You may ask me any questions via this site or (carefully) by E-mail.

Andrew.