gravity

Submitted by stefanb on Thu, 05/05/2011 - 15:45

Hi everybody,

I've got a question about gravity. I am trying to model a dense system with a density ratio of 6 by particles' to fluid's density. If I apply gravity, are the buoyancy effects taken into account? Well, I tried to look at the source of your coupling but I haven't found anything. If I was blind - then I'm sorry.

If it's not taken into account I would reduce the gravity taken the density ratio into account (in my example reduce to 8.2). Do you think that would be the best way?

So, looking forward to hear anything.

Kind regards,
Stefan

Forums:

ckloss's picture

ckloss | Thu, 05/05/2011 - 16:57

I think there may be a buoyancy model, but I am not sure. It's best to let Christoph Goniva answer this question. He's currently out of office but I guess he will answer to that next week

Christoph

tshan | Fri, 05/06/2011 - 14:41

Hi Stefan,

The buoyancy is not included in the particle-fluid interaction force model of CFDEM. To do this is simple, and you can modify the code and add the buoyancy into the force model.

Best,
Tong

cgoniva's picture

cgoniva | Mon, 05/09/2011 - 17:06

Hi Stefan,

As tong allready wrote, you can easily add the influence of gravity to the force model.

I would recommend to write a new "force model" which only accounts for the effect of gravity.

(then you do not have to touch the "orignal" force models and profit from bug fixes)

As the code is written in a modular way you can then superpose drag and gravitational forces by simply adding the models' "kewords" in the couplingProperties.

Cheers,
Chris

stefanb | Tue, 05/10/2011 - 09:47

Hi Chris,

I thought of modifying the gravity fix. Just using the fix and reducing the gravity by the density ratio of fluid and particle. Including this into one fix sounds to me the best way keeping computational cost in mind. So, if there aren't any reasons which will change my mind I will try to code it that way and if you are interested in let me know it.

Regards,
Stefan

ckloss's picture

ckloss | Tue, 05/10/2011 - 10:28

yes, that would be the easiest way yet not the most convenient one
If it's implemented as a coupling submodel, you can read in fluid density from the dictionary
If you change the fix parameter, you will have to change the gravity manually according to the fluid density

Christoph

cgoniva's picture

cgoniva | Tue, 05/10/2011 - 13:51

thus, for the general case of variable density (stemming from compressible fluid or multiphase fluid)
you will need to use a force model on CFD side.

Cheers,
Chris

rqwang | Thu, 05/19/2011 - 16:26

Hi,

So far, I can modify DiFeliceDrag.C to add or change the force models.

However, I would like to know the modular way to write it.

Should I change "forceModel.C", or reproduce DiFeliceDrag directory, change name to "BuoyancyForce", and do the corresponding change and then add the key word "BuoyancyForce" to the couplingProperties?

Thanks.

rq

cgoniva's picture

cgoniva | Thu, 05/19/2011 - 18:18

Hi rq!

you are right:

reproduce DiFeliceDrag directory
change name accordingly,
and do the corresponding changes in the *.C and *.H files
change the "keyword" in the *.H
use the new key word in the couplingProperties?

Cheers
Chris

ainntt | Wed, 12/05/2018 - 15:46

hi rq,

is the new CFDEM version buoyancy force inclouded or not? If so far not inclouded, can you please the preliminary code for the buoyancy force Upload again? because of invalid now.

thx
ainntt

ainntt | Wed, 12/05/2018 - 15:40

hi rq,

is the new CFDEM version buoyancy force inclouded or not? If so far not inclouded, can you please the preliminary code for the buoyancy force Upload again? because of invalid now.

thx
ainntt