Particle size requirement for DPM and case of violation

trbprnz · October 3, 2017, 07:34

Hi, I have a Eulerian-Lagrangian DPM case where the particle size is larger than the grid size. As mentioned in various threads on cfd-online, this is not suitable for DPM as it does not fulfill the assumption of point masses and proper computation of any particle surface integral based forces such as drag is not straightforwardly possible.

However, Fluent still solves something. Does anyone know how exactly in such a case e.g. drag is being computed? Here there is no unique Re number as the fluid velocity may be different from one cell to another? Is this simply interpolated according to e.g. volume fractions?

I am aware of the macroscopic Particle Model; however, this is not an option for the time being.

In addition, is this size requirement stated explicitly somewhere in the documentation? It is not mentioned in the DPM limitations.

Many thanks, BR, Alex

mrishi · November 16, 2018, 13:12

Did you find it out? I have been asking this too.

What is being calculated on the finer grid?

trbprnz · November 21, 2018, 04:37

No, I just commented on it in my recent paper 10.1016/j.jnnfm.2018.11.005.
I think you basically introduce some additional error, which may be large depending on the forces considered and their numerical values. To evaluate that you probably would have to do a study comparing the DPM results with results obtained with e.g. the MPM or a DNS.

October 3, 2017, 07:34	Particle size requirement for DPM and case of violation	#1
trbprnz New Member Join Date: May 2015 Posts: 9 Rep Power: 11	Hi, I have a Eulerian-Lagrangian DPM case where the particle size is larger than the grid size. As mentioned in various threads on cfd-online, this is not suitable for DPM as it does not fulfill the assumption of point masses and proper computation of any particle surface integral based forces such as drag is not straightforwardly possible. However, Fluent still solves something. Does anyone know how exactly in such a case e.g. drag is being computed? Here there is no unique Re number as the fluid velocity may be different from one cell to another? Is this simply interpolated according to e.g. volume fractions? I am aware of the macroscopic Particle Model; however, this is not an option for the time being. In addition, is this size requirement stated explicitly somewhere in the documentation? It is not mentioned in the DPM limitations. Many thanks, BR, Alex

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
Particle tracking error	alchem	OpenFOAM Bugs	5	May 6, 2017 17:30
Mass conservation law violation in stationary case	vitmalin	CFX	3	September 3, 2007 14:32
Free surface boudary conditions with SOLA-VOF	Fan	Main CFD Forum	10	September 9, 2006 13:24
time step size requirement	turb	Main CFD Forum	0	February 21, 2005 22:57
ACCESS VIOLATION during autosaving case file	Ryan	FLUENT	1	April 20, 2004 11:32

November 16, 2018, 13:12		#2
mrishi Member Rishikesh Join Date: Apr 2016 Posts: 63 Rep Power: 10	Did you find it out? I have been asking this too. What is being calculated on the finer grid?

November 21, 2018, 04:37		#3
trbprnz New Member Join Date: May 2015 Posts: 9 Rep Power: 11	No, I just commented on it in my recent paper 10.1016/j.jnnfm.2018.11.005. I think you basically introduce some additional error, which may be large depending on the forces considered and their numerical values. To evaluate that you probably would have to do a study comparing the DPM results with results obtained with e.g. the MPM or a DNS.