CFD Online Discussion Forums

CFD Online Discussion Forums (
-   OpenFOAM Running, Solving & CFD (
-   -   How to simulate particles in a gas flow (

sara December 1, 2008 03:17

Hi there Sorry to ask such a
Hi there
Sorry to ask such a general question but I'm really lost.
I want to simulate liquid particles in a gas flow. The particles are so small that they behave like solid particles. So I think they have to be lagrangian particles.
The gas flow carries the particles along. The geometry includes some obstacles. The gas flows obviously flows around the obstacles (fibres) while the particles are supposed to stuck to the fibres as soon as they touch them.
I started using interFoam and funkySetFields to insert the particles. However, even only one very small particle spreads out over a large area and doesn't behave at all like a droplet, much less like a solid particle.
Using icoLagrangianFoam doesn't seem possible either, since it doesn't work with OpenFoam1.5 which is the one I'm using.
Well, that's the problem.
Has anybody ever done something like this before or has any ideas how I could proceed?
Thanks, Sara

jeboone December 10, 2008 12:59

I am also looking at dispersio
I am also looking at dispersion of particles in a gas flow. It appears that the twoPhaseEulerfoam solver is the best fit for the problem. Is there a discription of the examples in the tutorial available?

sara December 10, 2008 22:00

Hi James I started with twoPh
Hi James
I started with twoPhaseEulerFoam, too. But this solver treats both phases continously. If this is what you want you'll be alright with it. But my particles should be treated as discrete particles, in a lagrangian way. I found the icoLagrangianFoam solver which is quite nice, but only as a demo since it doesn't really resemble the real word.
At the moment I'm going for interFoam and inject droplets using funkySetFields. Not sure though if this is the right way...
Cheers, Sara

sachin January 5, 2009 16:47

Hi all, any success in simula
Hi all,
any success in simulating particles in gas flows
In my case my flow is dilute and particle particle collision is neglected ...started with icoLagrangian Foam but failed
Any suggestions...

kcjarvis56 January 19, 2009 23:47

Hi all, You may be able to us
Hi all,
You may be able to use rhoTurboTwinParcelFoam which is a tutorial included with OF-1.5. I found this to be a good example for particles and a good starting point to make a solver for my application. Hope this helps.

luchen2408 May 8, 2014 18:09

hello,sara, Do you have some process? which solver you take at last. I am also puzzled about the solver. I maybe have the same question as yours, liquid particles flow along the gas and abosored by the wall.

LaSerpe May 9, 2014 10:40

I'm simulating a similar problem, ie water droplets impingement for icing simulation on airplanes.
The solver I'm using is uncoupledKinematicParcelFoam or icoUncoupledKinematicParcelsFoam as well.
though they're not so easy to set up at first, they seems to suit well for your problem


luchen2408 May 10, 2014 17:02

hello,LaSerpe, I need to define the droplet like solid particles and I also define the diameter of the droplet,and porous media will included in the solver. I also define the wall boundary just like rebound, absorb. exactly, I think I just perform the simulation with gas first, then that I can define a source to input particles. but It seems not to get it. besides, I think I can take sprayFoam in my simulation.Do you think so?

LaSerpe May 12, 2014 03:26

never used sprayfoam before, so I can't help on it.
I'm using icoUncoupledkinematicFoam, in this case you need first to solve the flow, (I use simplefoam for stationary flows) and then you can run the lagrangian solver.
with icouncoupledkinematicfoam you can set the cloud proprieties by editing the proper file in ../constant/
There you can specify parameters as particle diametre, density and so on, and there you can also define the type of wall interacton you want (you can set rebound, stick etc).


luchen2408 May 26, 2014 07:20

hello,LaSerpe, you mean that I have to run the case two times. first, I have to run the simulation with simpleFoam, then add the cloud properties and run thelagrangian solver (icoUncoupledkinematicFoam solver)? I can run one case with two solver?

LaSerpe May 26, 2014 08:25

yes, first you have to run simplefoam and then the lagrangian solver, be careful because before running the lagrangian solver you will have not only to include cloud proprieties but you will also have to set the proper control dictionaries.
Usually I set two different folders and when I get the aerodynamic solution I just copy the field to the lagrangian directory and run the second solver from there. This helps me to keep things clear ;)


luchen2408 May 27, 2014 09:52

Thanks,LaSerpe, but I am not sure if the particles will affected by the gas flow if I run the simulation separately? even I run the gas simulation first, then I run the lagrangian solver.

LaSerpe May 27, 2014 10:09

IcoUncoupledKinematicParcelFoam is a one way coupled solver, this means that the flow affects particles trajectories but particles do NOT affects the flow motion.
This allows you to run simplefoam and then the lagrangian solver.
Of course you will have to copy the resultant field (calculated with sompleF) to the directory where you will lunch icoUncoupled..Foam.
As I said you can run both the solver in the same directory if you like.

I've been using this procedure for the last year and I can assure you that, if every dictionaries is properly set, particles will be affected by the gas ;)


All times are GMT -4. The time now is 00:52.