hello, Can shock wave propagation in 2D using an inviscid, segrageted solver simulate successfully in fluent 5/6? when the time going, the interface between hign pressure and low pressure have been becoming fuzzy gradully.why? Thanks a lot for helping me.

Coupled solver should be used for flows containing shock waves.

What do you mean by "fuzzy?". If the moving shock wave is produced in a shock tube, you can see transient wave structures, as unsteady shock wave propagating into the low pressure zone, unsteady expansion waves travelling into the high pressure zone, a contact discontinuity, etc. Please refer to typical Riemann shock tube problem for more details

regards,

Rajesh

Thank you for replying. 'fuzzy' means the distinctness of shock waves' interface become worse. You are right. I am trying to simulate shock tube. But I still don't know why coupled solver can ,but segareted solver cann't. Thanks again.

Please go through the Fluent manual (section: Choosing the Solver Formulation). You can find out the various onditions under which both solvers are used.

Regarding the shock tube problem, it is not strange that the distinction between the high pressure zone and the low pressure zone becomes worse as the time progresses. Within a short time after the diaphragm rupture, a typical wave system is formed inside the shock tube, which contains mainly four sections (please see a text book on compressible flows for a typical wave diagragm in a shock tube). Section 1 (low pressure zone) is the stagnant flow upstream of the moving shock, section 2 and section 3 are the mass motion (in the direction of shock wave motion) down stream the shock wave and section 4 (high pressure zone) is the stagnant flow upstream of the expansion wave. For an inviscid, adiabatic one-dimensional flow, the flow velocities and pressures in sections 2 and 3 are the same, but there will be an entropy change across a section, which is called a contact discontinuity. If you vizualise the flow properties (except veocity and pressure) in sections 2 and 3 you can see zones with different temperatures and densities. However, in real flows, this contact discontinuity will gradually thicken and transform to a contact region with continuous changes in temperature and density through out the zone, due to diffusion. You might've seen this in you simulation if you are after the real flow.

In inviscid, adiabatic one-d flow also, gradual changes in properties may be seen between sections 3 and 4 where the expansion waves exist. The expansion waves are spread over a length where you can see all the properties changing between the head and tail of the wave. You might've seen this if your simulation is for inviscid flows. In short, in a shock tube problem, the clear distiction in properties can only be seen across the moving shock wave and hence the shock wave is seen as a thick line in the low pressure zone.

Rajesh

Thanks for your regards.

I'm trying to simulate a two-phase shock tube by Eluer-Eluer Model. So, the first step , I have to get an only ideal gas shock tube's solution by segregated solver. Because the segregated solver provides Multiphase mixture models that are not available with the coupled solvers. On the condition that, I will be able to simulate a two-phase shock tube. But, now , at the first step ,I can't succeed . Can You tell me whether this way is feasible ,and, whether I have to add some UDF? What should I do?Can I achieve? Hoping for you help.

Thanks a lot.

jiangxuxu

Dear Jiangxuxu,

From fluent manual,

"The segregated solver traditionally has been used for incompressible and mildly compressible flows. The coupled approach, on the other hand, was originally designed for high-speed compressible flows. Both approaches are now applicable to a broad range of flows (from incompressible to highly compressible), but the origins of the coupled formulation may give it a performance advantage over the segregated solver for high-speed compressible flows." So you can very well go ahead with the segregated solver as that is the only option left for you. But the main concern is the convergence rate and accuracy of the results.

Anyway "something is better than nothing" to start with.

All the best

Rajesh