[teymourj]

Dear all,

I am trying to use Discrete Phase Model (DPM) to look at particle behavior in the solution of my Single Rotating Reference Frame (SRF) which contains my countinues fluid.

As long as SRF has rotating frames, therefore a Centrifugal and Coriolis term will be added to the equation of force balance of the particles. What I need to do, is writing a UDF which calculates these two forces. Then compile it with my model, so that it cancels those two additional forces.

I wrote a UDF based on UDF manual section 2.5.2 file:///usr/ansys_inc/v120/fluent//f...udf/node67.htm as follow, but it did not work out and I get error when I want to compile it. I was wondering if anyone can help me with that.

/* UDF for computing the force due rotating refrence frame on a particle */

#include "udf.h"
#define ROHR 1.5

/* density ratio of particles (roh/roh_p) */
#define OMEGA 1.78

/* angular velocity [rad/sec] */
/* Calculate some of centrifugal & coriolis force on a particle */
/* axis of rotation is "Y" and there are two force components in "X" and "Z" */ /* Fx= (1-ROHR)*OMEGA^2*x + 2*OMEGA*(u_z,p - ROHR*u_z), Fz= (1-ROHR)*OMEGA^2*z + 2*OMEGA*(u_x,p - ROHR*u_x) */

DEFINE_DPM_BODY_FORCE(particle_body_force,p,i)

{
real bforce=0;
if(P_TIME(p)>=TSTART)
{ if(i==0)

bforce=(1-ROHR)*OMEGA^2*P_INIT_POS(p)[0] + 2*OMEGA*(P_VEL(p)[2] - ROHR*rel_vel[2]);

else if(i==2) bforce=(1-ROHR)*OMEGA^2*P_INIT_POS(p)[2] + 2*OMEGA*(P_VEL(p)[0] - ROHR*rel_vel[0]); }

else bforce=0.0;

/* an acceleration should be returned */
return (bforce/P_MASS(p));


Thank you for your help in advance.

-teymourj

[teymourj]

I found the solution to this problem. Read the MRF/SRF limitations for tracking particles in ANSYS FLUENT user's guide at:

// User's Guide :: 1 // 10. Modeling Flows with Moving Reference Frames // 10.3. Flow in Multiple Moving Reference Frames // 10.3.1. The Multiple Reference Frame Model

Note that the particle track in rotating reference frame for inert particles is MEANINGLESS (see the user's guide). You should use the TUI command to track particles and extract their data in absolute reference frame.

Best,

Teymour

[mingersai]

THANK YOU SO MUCH!!!!!!
That's why I can not get anyting. With rotating reference frame.
Ansys customer support seems to made a mistake, I had asked them whether I can inject particles without being affected by rotation reference frame and they said no......

I had been stucked here for a long time.

Thanks for your help!

Originally Posted by teymourj

I found the solution to this problem. Read the MRF/SRF limitations for tracking particles in ANSYS FLUENT user's guide at:

// User's Guide :: 1 // 10. Modeling Flows with Moving Reference Frames // 10.3. Flow in Multiple Moving Reference Frames // 10.3.1. The Multiple Reference Frame Model

Note that the particle track in rotating reference frame for inert particles is MEANINGLESS (see the user's guide). You should use the TUI command to track particles and extract their data in absolute reference frame.

Best,

Teymour

[mahayheidari]

can you explain it for me ?
i have problem in it yet

[teymourj]

@mahayheidari: Please be specific about your problem and what you are not understanding and I will provide you inputs.

This thread is very old and I am not sure what you need me to explain. Explain your problem from scratch.

[mahayheidari]

I have a prticle in a fluid
And the particle have transitional and rotational movement
I simulate it
But the particle just have transitional movement
And it cant rotate
I want to know what should i do
Thanks

[teymourj]

What are the models you are using to simulate the flow field and particle respectively in FUENT?

[mahayheidari]

There is a canal that a circular particle move in it
If you want i will sent the picture

[mahayheidari]

Dynamic mesh
Remeshing
Smoothing
6dof

[teymourj]

Well to be honest these informations are not helpful for me to help you.

This thread was about changing the Reference Frame for particle tracking of particles using DRW model coupled with the SRF or MRF model. In the SRF or MRF model RANS equations are solved in rotating reference frame and particles are tracked in that frame.

It seems that you are using Dynamic mesh model in which the movement of particles should be physical by default.

Hope this helps.