DPM- brownian diffusion

March 12, 2009, 16:40

fluent can not simulate brownian diffusion for particle tracking in DPM mode, what should I do if I want to fix it ?

March 13, 2009, 04:33

fluent can do this. But you must set the Energy mode on. That is "Define--> Energy". The Brownian needs the gas temperature.

March 13, 2009, 09:46

I did it, but the results are not good, it seems fluent can not consider it properly for some reasons!

HenrikS · March 16, 2009, 07:57

I think this might be a time step size problem..? There is no comment in the manual about the time step size to be used in simulations with Brownian motion w/ DPM, but if you check out some of the references for this approach it seems this has to be taken into account (as is in fact acknowledged in Fluent's online technical support solution case 978).

The time step size should be considerably larger than the time scale on which individual collisions occur (which is generally not a limitation), but also considerably smaller than the particle relaxation (response) time (which may very well be a limitation, as the particles for which Brownian motion is important are generally small and have low particle relaxation times).

I suggest you try adjusting the time step size (or the Step Length Factor, if you're doing steady simulations) accordingly and see if the results get better?

Hope this helps!

/Henrik

March 12, 2009, 16:40	DPM- brownian diffusion	#1
Sean Guest Posts: n/a	fluent can not simulate brownian diffusion for particle tracking in DPM mode, what should I do if I want to fix it ?

March 13, 2009, 04:33	Re: DPM- brownian diffusion	#2
long zhengwei Guest Posts: n/a	fluent can do this. But you must set the Energy mode on. That is "Define--> Energy". The Brownian needs the gas temperature.

March 13, 2009, 09:46	Re: DPM- brownian diffusion	#3
Sean Guest Posts: n/a	I did it, but the results are not good, it seems fluent can not consider it properly for some reasons!

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
Brownian Motion UDF Code	Confuser	FLUENT	3	August 13, 2021 01:14
Brownian force model in DPM	akr	FLUENT	1	January 2, 2010 09:16
particle's Brownian diffusion	Sean	FLUENT	1	July 30, 2009 23:52
Brownian Diffusion	HK	FLUENT	0	December 3, 2006 20:02
Brownian motion	Sank	Main CFD Forum	7	September 11, 2004 16:36

March 16, 2009, 07:57		#4
HenrikS Member Henrik Ström Join Date: Mar 2009 Posts: 33 Rep Power: 17	I think this might be a time step size problem..? There is no comment in the manual about the time step size to be used in simulations with Brownian motion w/ DPM, but if you check out some of the references for this approach it seems this has to be taken into account (as is in fact acknowledged in Fluent's online technical support solution case 978). The time step size should be considerably larger than the time scale on which individual collisions occur (which is generally not a limitation), but also considerably smaller than the particle relaxation (response) time (which may very well be a limitation, as the particles for which Brownian motion is important are generally small and have low particle relaxation times). I suggest you try adjusting the time step size (or the Step Length Factor, if you're doing steady simulations) accordingly and see if the results get better? Hope this helps! /Henrik