CFD Online Discussion Forums

CFD Online Discussion Forums (https://www.cfd-online.com/Forums/)
-   FLUENT (https://www.cfd-online.com/Forums/fluent/)
-   -   DPM stochastic collision (https://www.cfd-online.com/Forums/fluent/150056-dpm-stochastic-collision.html)

prasadsai31 March 14, 2015 06:53

DPM stochastic collision
 
Hello,
I'm simulating spray dryer. Using Stochastic collision model for collision of droplets. I want track the details ( like mass, diameter, velocity) of the collision partners. Is it possible using UDF or is there any other way. I'm Using Fluent 14.5.

Thanks.

`e` March 14, 2015 15:17

You could use the DEFINE_DPM_SPRAY_COLLIDE macro in a UDF to customise your collision of particles and output variables (such as mass, diameter and velocity) to a data file.

prasadsai31 March 16, 2015 02:25

Thanks e for your quick response.
What I'm looking is different thing, I explained below.
In spray drying particles will collide and some collisions results in agglomeration.
I want to use default collision model and a developed agglomeration model (not coalescence default model) to that. If I use DEFINE_DPM_SPRAY_COLLIDE macro this will not allow using default collision model. I have to write Whole program i.e. UDF for collision model.
I will use Collision model for particle collision and if i can identify collision partners and details of the partners that will be input to the agglomeration model.
In short I use inbuilt collision model and need data of collision partners than that data will be used to see particle will agglomerate or not.

`e` March 16, 2015 03:32

I don't believe you can call the default collision model when you use the DEFINE_DPM_SPRAY_COLLIDE macro, but let us know if you find a way (there might be a way if you find the function in the source code).

It sounds like you'll need to use the DEFINE_DPM_SPRAY_COLLIDE macro for your collision mechanics and have a couple of ways forward for coding the UDF:
1) find the source code for Fluent's collision model to copy & paste to start with (not sure if that's possible)
2) use the collision models listed in the theory guide (or in literature) and code them yourself

prasadsai31 March 16, 2015 04:09

Thanks e.
I have to try any of the 2 options. But writing the whole collision model as a UDF will be
pain taking.

`e` March 16, 2015 04:26

It'd be a little tedious to write the code for the collision model; although there's only a few equations required by the looks in the theory guide. But at least you can compare your code with Fluent's model (then proceed with developing the agglomeration model).

zizi63 August 17, 2016 18:33

Zizi
 
Dear all,
I have just started to work with openfoam and I want to simulate the motion of two kinds of particles in a fluidized bed. I am using DPMFoam when I define the properties of the second type particle in the kinematicCloudProperties and run the program the software gives me an error indicating that the StochasticCollisionModel is not defined in the dictionary!!!:confused:
this is while in the kinematicCloudProperties the StochasticCollisionModel none; code is used.
could anybody help me?:o

hjubaer March 19, 2017 23:01

Quote:

Originally Posted by prasadsai31 (Post 536413)
Thanks e for your quick response.
What I'm looking is different thing, I explained below.
In spray drying particles will collide and some collisions results in agglomeration.
I want to use default collision model and a developed agglomeration model (not coalescence default model) to that. If I use DEFINE_DPM_SPRAY_COLLIDE macro this will not allow using default collision model. I have to write Whole program i.e. UDF for collision model.
I will use Collision model for particle collision and if i can identify collision partners and details of the partners that will be input to the agglomeration model.
In short I use inbuilt collision model and need data of collision partners than that data will be used to see particle will agglomerate or not.

Hey Prasad,
have you been able to find a feasible solution to the issue you faced? I would really appreciate your feedback since I intend to do the same, if possible; use the built-in collision model and then extend on the agglomeration.
thanks

hjubaer March 26, 2017 19:39

Dear Prasad,
I am currently standing exactly there, where you also were back in March 2015. Have you found a solution around the tedious and "paint-taking" way of coding the stochastic model all by yourself?

To all who might be reading,
Is there any way to use the default collision model and then make FLUENT predict additional or different collision outcomes?
Your help would be greatly appreciated.
Cheers!


All times are GMT -4. The time now is 10:58.