Modelling DPM with periodic boundary conditions in a pipe

[scabo]

Hi

I want to model particle laden flows in a periodic pipe. I have set periodic boundary condition for the fluid as periodic. But i want to use DPM in a periodic fashion. How can i do it? Basically i want to model only a short section of pipe and give periodic boundary conditions to both the flow and the solid particles. How can I do it? Any help is greatly appreciated..thanks

[Rob]

That should work, https://ansyshelp.ansys.com/public/account/secured?returnurl=/Views/Secured/corp/v251/en/flu_ug/flu_ug_sec_periodic.html  but with steady particles I'm not sure how you'd introduce them. I assume they're neutrally buoyant?

[scabo]

Hi, thanks. I read the docu- what does this line mean: 

• Steady particle tracks can be modeled only if the particles have a possibility to leave the domain without generating incomplete trajectories. Can you explain this line? Particles will be changing positions with time rght? I think particles can be introduced throught DPM.

• Also, aftr modelling periodic lengths can we expand the periodic section into a long section in order to see the particle trajectory?

[Rob]

First point - the particles must all leave the domain through a boundary (escape/trap) and not time out somewhere in the middle of the section. 

No. You can plot repeated section but it's the same section repeated which should include the DPM trajectories. If you want to see the tracks down the long pipe you can try using injection files once the flow has been solved using the periodic flow. 

[scabo]

SO, if i want to model particle deposition on the pipe surface, that is not possible using this periodic BC? 

[scabo]

Also if i inject DPM in a streamwise periodic flow, how will the particles behave? Will the particles also have periodic BC attached to them? Or will the particles behave normally?

[Rob]

Particles will behave normally. Given the particles will settle over time, that's not periodic.  You're also not going to pick up any duning effects with DPM, you'll want Eulerian or Rocky for that: I'd favour the latter but neither will be periodic.