CFD Online Discussion Forums

CFD Online Discussion Forums (
-   FLUENT (
-   -   Problem with DPM model (

cwflying January 29, 2003 20:54

Problem with DPM model
Hi everyone,I am now using DPM modele to simulate particle movement in a hydrocyclone. In terms of hydrocyclone, it has one velocity inlet and two outflow outlets( I have set the flow weight already). First I simulate fluid flow in hydrocyclone using RSM modele( because it is strong swirling flow) then add particle injection at inlet,assuming particles having the same velocity with fluid flow at inlet, using DPM modele. The problem is: when I track particles, I find the particles only escape at inlet, not two outlets!

(two axisymetrical dimension,inlet velocity for example u=0,v=-0.2m/s, w=2m/s)

Jurgen January 30, 2003 03:56

Re: Problem with DPM model
Some ideas: are the outlets set to escape in Boundary Conditions? You can also specify the boundary at which the particles escape. I think that is done under Report-> Discrete Phase Model. Hope this helps, Jürgen

cwflying January 30, 2003 05:43

Re: Problem with DPM model
I have already set the escape boundary conditions at two outlets and inlet, the wall B.C is set to reflect.

dag January 30, 2003 05:50

Re: Problem with DPM model
Try setting your inlet face as reflect instead of escape. I would initially guess the centrifugal field is large enough to instantly move your particles out through the inlet before they enters the domain.

In addition Fluent has a bug when it comes 2-D acisymmetric particle tracking in cyclones (and probably other type of geometries), where particles lighter than the fluid move towards the wall. You can easily check if this is part of your problem with a 3-D calc of the same geometry (take a 5 deg rotational symmetric 3 D example of the same geometry).

cwflying January 30, 2003 06:01

Re: Problem with DPM model
Thank you, I have also tried setting inlet B.C to be reflect, but it still does not work. And in my case particle is about 2000kg/m~3, fuild flow is water.

bala January 31, 2003 12:25

Re: Problem with DPM model
Hi, First you solve for the continuous phase. After some iterations, just enable the DPM model and try.

Regards, bala

All times are GMT -4. The time now is 10:10.