CFD Online Discussion Forums

CFD Online Discussion Forums (https://www.cfd-online.com/Forums/)
-   NUMECA (https://www.cfd-online.com/Forums/numeca/)
-   -   Problems starting a fluid-particle interaction simulation (https://www.cfd-online.com/Forums/numeca/145068-problems-starting-fluid-particle-interaction-simulation.html)

Richard Beauchamp November 27, 2014 09:51

Problems starting a fluid-particle interaction simulation
 
Hi,

My name is Richard Beauchamp and I am a graduate student at Ecole
Polytechnique de Montreal. I am working on a project using Fine/Open v 3.1
from Numeca and I am having a weird problem: I cannot start a
Fluid-Particle interaction simulation.

The Std file says that the simulation crashes when it is time to do the first
Lagrangian iteration, which evaluates the forces applied on the particles. I get the following error: No such file or directory.


The solver runs a complete unsteady fluid only simulation without any problem. When performing a Fluid-Particle Interaction simulation, the steady and unsteady state flow only iterations run fine also.

I tried to first run a fluid only simulation, let it complete, and then use the results of this run as initial condition for my fluid-particle interaction simulation, but I always get the same error message.


Can anybody please help me?

.


Thank you,

colinda1 December 2, 2014 13:56

Dear Richard,

Are you well running the simulation in steady conditions first? The Lagrangian model in FINE/Open with OpenLabs has been extensively validated for steady runs.

In addition you may want to check your boundary conditions in the Fluid-Particle interaction Parameters page. If the reference frame for the velocity is set to Absolute and the velocity of the injected particles is set to 0 in all directions, there are no particles entering the domain.

Best regards,
Colinda.

colinda1 December 5, 2014 10:58

Hi Richard,
Did you manage to get your case running?

I realized when re-reading my post that possibly I could give some further clarifications to my previous remark: I may incorrectly have given the impression that setting the inlet boundary condition in relative frame would help by emphasizing the reference frame. This is not the case. The relative frame would apply to rotating machinery so does not apply in your case. In your case you could well use the absolute reference frame but it would be important to inject the particles with a certain non-zero velocity at the inlet.

I hope you can advance in your modeling. Please do not hesitate to post further questions if any remaining.

Richard Beauchamp January 22, 2015 11:08

Hi Colinda,

I am happy to tell you that my simulation works fine in steady state. The only thing that bothers me a little bit is the lack of ability to get results in the viscous sublayer and the buffer layer of the flow, as the speed of the particles calculated with fine is zero.

What I would mostly like to know is if there is a way to run an unsteady with the particles?

The last time I tried, which was the reason for my question back in December, the solver would just crash as it tried to do one lagrangian iteration.

Do I need to run a steady state fluid particle simulation first, and then run the unsteady simulation?

Thank you for your collaboration,

colinda1 January 23, 2015 14:30

Unsteady fluid-particle interaction
 
Dear Richard,

The Lagrangian module has not been tested in unsteady mode and its compatibility can not be guaranteed as indicated in the compatibility matrix in the documentation package.

Seen the error message you encountered I would tend to say this is indeed due to the current limitations in the implementation. When you first told us about your wish to make unsteady simulations, I have submitted the request to support Lagrangian module in unsteady mode. I will cross-check the status of this request and see whether it could be included in the planning for development and validation. I will let you know.

Best regards,
Colinda

Richard Beauchamp January 26, 2015 10:44

Fluid CO simulation
 
Well ok then. I guess that I will not be able to make unsteady fluid-particle interaction simulations.

However, I would like to know if it is possible to make a two-gas unsteady simulation, say with air and carbon dioxide?

Thank you for your collaboration,


All times are GMT -4. The time now is 14:36.