simulating solid granules in CFX

sakalido · January 19, 2015, 18:11

Hi all,

I am trying to simulate the following problem in CFX 15.0.

500 granules (PE is the material) of 6mm diameter and overall weight of 500gr are sitting (floating) in water. At some point water starts to flow and with it these granules also flow. I will have turbulent flow.
I know that in FLUENT this can be done with DPM. and I am trying to use particle transport solids in CFX (to simulate Lagrangian particle tracking) but I don't know how I can start with granules sitting (floating in water) not injecting granules in the domain.

Can you give me some hints?

Thank you,
Sara

ocea · January 26, 2015, 09:29

Hi Sara,

I am also facing the same problem. In my case I want to distribute some particles on a solid surface and investigate the effect of jet impingement. Did you make any progress so far?

Ocea

ghorrocks · January 26, 2015, 18:21

Sara's question sounds possible in CFX with little difficulty. Have you done the tutorials? Or what specific thing is not working.

Ocea's question is a little more difficult - what is holding the particles there before the jet hits? Are they just resting there under gravity.

sakalido · January 29, 2015, 12:00

Glen,

Imagine I have a tank full of water (open to atmosphere) and the PE granules are sitting in this tank. We have a reservoir of water (again open to atmosphere) which is a little higher than the tank. The simulation starts when the water reservoir's opening to tank opens and allows water flow into the tank and then moves the granules.
So the start of the simulation is the point that these granules are already in the water and the opening form reservoir to the tank opens.

My problem is I don't know how I can simulate these granules without defining an injection point. In my simulation so far, the granules start to be injected in the water from a point with a certain velocity and mass flow.
I don't know how I can start from the point that some certain number of granules are already sitting in the water.

I did not find any tutorial related to this problem. Could you suggest any?

thank you,
Sara

ghorrocks · January 29, 2015, 18:06

Just looked in the documentation - it looks like you cannot specify initial locations for particles.

So I can think of a few work-arounds:
* You could do an initial simulation where you inject the particles and allow them to settle. Then you use this as initial condition for the actual run.
* You use user fortran to specify a user injection location at time step 0 with zero velocity and place them in the region you want them to start in.

And another thought: This model may be more suited to a eularian particle model simulation. Then the initial condition is easy to specify, and you can also set maximum packing fractions and many other parameters which will probably be useful which are not available in the lagrangian model.

basivi · March 9, 2020, 23:03

I have a similar problem, any progress or any suggestions about defining solid particles resting in a cylinder?

ghorrocks · March 10, 2020, 07:09

I think my post #5 explains it, don't you?

Note you cannot model particles resting on each other in CFX. CFX only supports very limited particle to particle interaction. You will need a Eularian model or a DEM model to model that.

basivi · March 11, 2020, 21:00

Thanks for the reply, I can able to do it in Fluent by using Eulerian model.

Reference: https://www.youtube.com/watch?v=W2y9rKXapds

January 19, 2015, 18:11	simulating solid granules in CFX	#1
sakalido Member Sara S. Join Date: May 2010 Location: NJ, USA Posts: 41 Rep Power: 16	Hi all, I am trying to simulate the following problem in CFX 15.0. 500 granules (PE is the material) of 6mm diameter and overall weight of 500gr are sitting (floating) in water. At some point water starts to flow and with it these granules also flow. I will have turbulent flow. I know that in FLUENT this can be done with DPM. and I am trying to use particle transport solids in CFX (to simulate Lagrangian particle tracking) but I don't know how I can start with granules sitting (floating in water) not injecting granules in the domain. Can you give me some hints? Thank you, Sara

January 29, 2015, 12:00	solid granules in fluid, no injection	#4
sakalido Member Sara S. Join Date: May 2010 Location: NJ, USA Posts: 41 Rep Power: 16	Glen, Imagine I have a tank full of water (open to atmosphere) and the PE granules are sitting in this tank. We have a reservoir of water (again open to atmosphere) which is a little higher than the tank. The simulation starts when the water reservoir's opening to tank opens and allows water flow into the tank and then moves the granules. So the start of the simulation is the point that these granules are already in the water and the opening form reservoir to the tank opens. My problem is I don't know how I can simulate these granules without defining an injection point. In my simulation so far, the granules start to be injected in the water from a point with a certain velocity and mass flow. I don't know how I can start from the point that some certain number of granules are already sitting in the water. I did not find any tutorial related to this problem. Could you suggest any? thank you, Sara

March 10, 2020, 07:09		#7
ghorrocks Super Moderator Glenn Horrocks Join Date: Mar 2009 Location: Sydney, Australia Posts: 17,870 Rep Power: 144	I think my post #5 explains it, don't you? Note you cannot model particles resting on each other in CFX. CFX only supports very limited particle to particle interaction. You will need a Eularian model or a DEM model to model that. basivi likes this. __________________ Note: I do not answer CFD questions by PM. CFD questions should be posted on the forum.

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
gas solid fluidizd bed with cfx	mohamad	CFX	5	July 28, 2010 16:15
Multiple Solid Domains - Interfaces	Scott	CFX	8	July 31, 2008 16:20
how to simulate 2d solid heat conduction using CFX	richard	CFX	3	March 24, 2008 08:27
gas solid fluidized bed with ansys cfx	mohamad	CFX	0	November 21, 2007 18:55
CFX 4.4 installation problem	Pandu Sattvika	CFX	1	December 1, 2001 05:07

January 26, 2015, 09:29		#2
ocea New Member Join Date: Jan 2015 Posts: 3 Rep Power: 11	Hi Sara, I am also facing the same problem. In my case I want to distribute some particles on a solid surface and investigate the effect of jet impingement. Did you make any progress so far? Ocea

January 26, 2015, 18:21		#3
ghorrocks Super Moderator Glenn Horrocks Join Date: Mar 2009 Location: Sydney, Australia Posts: 17,870 Rep Power: 144	Sara's question sounds possible in CFX with little difficulty. Have you done the tutorials? Or what specific thing is not working. Ocea's question is a little more difficult - what is holding the particles there before the jet hits? Are they just resting there under gravity.

January 29, 2015, 18:06		#5
ghorrocks Super Moderator Glenn Horrocks Join Date: Mar 2009 Location: Sydney, Australia Posts: 17,870 Rep Power: 144	Just looked in the documentation - it looks like you cannot specify initial locations for particles. So I can think of a few work-arounds: * You could do an initial simulation where you inject the particles and allow them to settle. Then you use this as initial condition for the actual run. * You use user fortran to specify a user injection location at time step 0 with zero velocity and place them in the region you want them to start in. And another thought: This model may be more suited to a eularian particle model simulation. Then the initial condition is easy to specify, and you can also set maximum packing fractions and many other parameters which will probably be useful which are not available in the lagrangian model.

March 11, 2020, 21:00		#8
basivi New Member Basivi Reddy Join Date: Mar 2017 Location: Tainan, Taiwan Posts: 10 Rep Power: 9	Thanks for the reply, I can able to do it in Fluent by using Eulerian model. Reference: https://www.youtube.com/watch?v=W2y9rKXapds