Execution problems in dsmcFOAM

turanc · September 23, 2010, 14:24

I don't think many other people are using or have used the relatively new dsmcFOAM solver but for any that have, I'd like to ask if you've had problems getting the solver itself to actually complete. It seems like the tutorials run fine, and sometimes my own cases complete but often enough the execution just seems to halt. Debugging it by putting output statements in the code and recompiling shows that it often stops just after the output (to the log) of the number of particles inserted in a given time step. This line is in FreeStream.C, one of the inflow boundary models that comes with the dsmc library. This is the very last line of the inflow() function, with the next line being the movement of the particles as:

Cloud<ParcelType>::move(td);

which is in the evolve() function of DsmcCloud.C, and to which it never gets. The execution never seems to exit the inflow() function though sometimes and since there is literally no code (that I can see) between the end of inflow() and the beginning of move() , this makes zero sense.

If anyone else is using dsmcFOAM, I'd like to hear if you've seen inconsistencies like this. Thanks.

hafez · March 5, 2011, 23:49

I'm a master student and i'm a beginner in using this program. can you help me yo known

the basics of DSMCFOAM?

because i will use it in simulation of gas flow through Nano tube.

if we can talk on my email , i'll be grateful.

my email

rof_eng_abdelslaam@yahoo.com

thanks

ehsan · March 10, 2011, 11:04

Dear All,

You could get more information about dsmcFoam reading our paper in CAF journal:

linkinghub.elsevier.com/retrieve/pii/S0045793010001891

Regards

eemdee · April 10, 2017, 13:58

I'm having the same problem with my dsmc solver. It crashes right after the line that turanc points out. Has anyone come up with a solution?

edit: Sorry, should've read the paper before posting. The issue for me at least was that my deltat was too large. The paper notes that

"Good DSMC practice [1] suggests that the time-step should be a fraction of the mean free time (the mean time between collisions)at the free stream conditions."

decreasing my delta t by several orders of magnitude made dsmc.evolve() not get stuck.

September 23, 2010, 14:24	Execution problems in dsmcFOAM	#1
turanc New Member turanc Join Date: Sep 2010 Location: Colorado, US Posts: 1 Rep Power: 0	I don't think many other people are using or have used the relatively new dsmcFOAM solver but for any that have, I'd like to ask if you've had problems getting the solver itself to actually complete. It seems like the tutorials run fine, and sometimes my own cases complete but often enough the execution just seems to halt. Debugging it by putting output statements in the code and recompiling shows that it often stops just after the output (to the log) of the number of particles inserted in a given time step. This line is in FreeStream.C, one of the inflow boundary models that comes with the dsmc library. This is the very last line of the inflow() function, with the next line being the movement of the particles as: Cloud<ParcelType>::move(td); which is in the evolve() function of DsmcCloud.C, and to which it never gets. The execution never seems to exit the inflow() function though sometimes and since there is literally no code (that I can see) between the end of inflow() and the beginning of move() , this makes zero sense. If anyone else is using dsmcFOAM, I'd like to hear if you've seen inconsistencies like this. Thanks.

March 5, 2011, 23:49	i need your help	#2
hafez New Member abdelsalam Join Date: Mar 2011 Posts: 1 Rep Power: 0	I'm a master student and i'm a beginner in using this program. can you help me yo known the basics of DSMCFOAM? because i will use it in simulation of gas flow through Nano tube. if we can talk on my email , i'll be grateful. my email rof_eng_abdelslaam@yahoo.com thanks

March 10, 2011, 11:04	dsmcFoam paper	#3
ehsan Senior Member Ehsan Join Date: Mar 2009 Posts: 112 Rep Power: 17	Dear All, You could get more information about dsmcFoam reading our paper in CAF journal: linkinghub.elsevier.com/retrieve/pii/S0045793010001891 Regards

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
Needed Benchmark Problems for FSI	Mechstud	Main CFD Forum	4	July 26, 2011 13:13
Help required to solve Hydraulic related problems	aero	CFX	0	October 30, 2006 12:00
Some problems with Star CD	Micha	Siemens	0	August 6, 2003 14:55
Airpak: Problems with heat flow contour conditions	Ivan de Miguel Urain	FLUENT	1	November 16, 2001 10:36
Using Fluent for Geophysical Problems	Hassid Samuel	FLUENT	0	February 23, 2001 07:04

April 10, 2017, 13:58		#4
eemdee New Member Eric Emdee Join Date: Mar 2017 Posts: 6 Rep Power: 9	I'm having the same problem with my dsmc solver. It crashes right after the line that turanc points out. Has anyone come up with a solution? edit: Sorry, should've read the paper before posting. The issue for me at least was that my deltat was too large. The paper notes that "Good DSMC practice [1] suggests that the time-step should be a fraction of the mean free time (the mean time between collisions)at the free stream conditions." decreasing my delta t by several orders of magnitude made dsmc.evolve() not get stuck.