EDEM-Acusolve Coupling Simulation

Danish Imran

Hello. I have been trying to simulate an inline separator separating particles from water using EDEM and Acusolve. Boundary condition set on the inlet of the pipe is 2 m/s using the constant function. I followed the flow setup from here 'https://2021.help.altair.com/2021.1/hwsolvers/acusolve/topics/tutorials/acu/acu_2000_intro_cfd_c.htm#acu_2000_intro_cfd_c' with only changing some of the boundary conditions and physics model to EDEM Bidirectional material. The particle radius size in EDEM is 1000mm and 0.001 in EDEM bidirectional material. What should I change to make the particles flow forward? The failed simulation can be seen at this link 'https://youtu.be/FYDCxh0tQvM'. Any help would be much appreciated. Thank you.

Geo_21939

Hello Imran,

Adding to Ravi's comment. Besides the dimensions and particle radius that seem to be rather large, I have some further comments on your setup:

1) I do not see a reason of having all the solid parts in the Acusolve setup:

Only the (Boolean-subtracted) fluid domain is needed to compute the flow field (i.e. the last solid part in the part tree).

 

2) Why is one of the solid parts assigned with water-particle material?

3) Transition effects modeling is probably not needed here:

4) In the Physics models of your EDEM setup, you have two Lift models active in the "Particle Body Force" tab:

  a) There is no need to have any models here, since all the forces are computed based on the Acusolve/EDEM coupling. We would like to put models for drag and/or lift in this tab if we did not run a coupled Bidirectional simulation (like in this example tutorial: https://2021.help.altair.com/2021.1/hwsolvers/acusolve/topics/tutorials/acu/acu_6100_intro_cfd_r.htm#acu_6100_intro_cfd_r )

  b) All the drag and lift models that you desire to have in your simulation have to be inserted during the material definition in HWCFD, as shown below:

Since you have spherical particles, we do not expect to have any significant lift forces forming on the particles, I would expect the default drag model (Schiller-Nauman) to be enough here.

 

5) I think these tutorials will be of great help to setup your case:

a) If you want to have a Bidirectional coupling: https://2021.help.altair.com/2021.1/hwsolvers/acusolve/topics/tutorials/acu/acu_6102_intro_cfd_r.htm#reference_usb_3yl_tnb

b) If you would like to do a Unidirectional coupling, i.e. to simplify things and not consider the effect of the particles to the flow: https://2021.help.altair.com/2021.1/hwsolvers/acusolve/topics/tutorials/acu/acu_6101_intro_cfd_r.htm#acu_6101_intro_cfd_r

 

Regards,

Geo

 

 

 

 

 

Ravi Teja Chaganti

Hi Imran,

Can you share more details of the problem that you're trying to solve such as actual physical conditions and the boundary conditions that you've used in your model. If possible, please share the model files, so that we can have a look at it. Also, we would appreciate if you don't post videos with misleading titles on a public platform such as YouTube before making sure that the actual failure of simulation is not because of some user error.

Thanks,
Ravi

Danish Imran

Ravi Teja Chaganti said:

Hi Imran,

Can you share more details of the problem that you're trying to solve such as actual physical conditions and the boundary conditions that you've used in your model. If possible, please share the model files, so that we can have a look at it. Also, we would appreciate if you don't post videos with misleading titles on a public platform such as YouTube before making sure that the actual failure of simulation is not because of some user error.

Thanks,
Ravi

Hello Ravi,

Im sorry for the inconvenience caused on my part. I have already changed the title of the video to avoid spreading misleading information. The hyperwork file can be found in this link: Inline Separator Test. Im not really sure if EDEM files can be opened on another PC.

Thank you,
Imran

Ravi Teja Chaganti

Danish Imran said:

Hello Ravi,

Im sorry for the inconvenience caused on my part. I have already changed the title of the video to avoid spreading misleading information. The hyperwork file can be found in this link: Inline Separator Test. Im not really sure if EDEM files can be opened on another PC.

Thank you,
Imran

After reviewing the model, it seems like this is a modeling error and the scale of the models doesn't seem right. The pipe diameter is 100m and the inlet velocity is 2m/s and the length of the pipe is about 2 km. Please make sure that the scale of the model is right. While importing the geometry into HyperWorks CFD, you have the option to set the target units. Please set it to MKS and also set the EDEM model in MKS with the right scale. The particle diameter that you have is 2 m and CFD-DEM coupling is not the right choice for such large bodies. Please update these and let us know about the status.

 

Thanks,
Ravi

Geo_21939

Hello Imran,

Adding to Ravi's comment. Besides the dimensions and particle radius that seem to be rather large, I have some further comments on your setup:

1) I do not see a reason of having all the solid parts in the Acusolve setup:

Only the (Boolean-subtracted) fluid domain is needed to compute the flow field (i.e. the last solid part in the part tree).

 

2) Why is one of the solid parts assigned with water-particle material?

3) Transition effects modeling is probably not needed here:

4) In the Physics models of your EDEM setup, you have two Lift models active in the "Particle Body Force" tab:

  a) There is no need to have any models here, since all the forces are computed based on the Acusolve/EDEM coupling. We would like to put models for drag and/or lift in this tab if we did not run a coupled Bidirectional simulation (like in this example tutorial: https://2021.help.altair.com/2021.1/hwsolvers/acusolve/topics/tutorials/acu/acu_6100_intro_cfd_r.htm#acu_6100_intro_cfd_r )

  b) All the drag and lift models that you desire to have in your simulation have to be inserted during the material definition in HWCFD, as shown below:

Since you have spherical particles, we do not expect to have any significant lift forces forming on the particles, I would expect the default drag model (Schiller-Nauman) to be enough here.

 

5) I think these tutorials will be of great help to setup your case:

a) If you want to have a Bidirectional coupling: https://2021.help.altair.com/2021.1/hwsolvers/acusolve/topics/tutorials/acu/acu_6102_intro_cfd_r.htm#reference_usb_3yl_tnb

b) If you would like to do a Unidirectional coupling, i.e. to simplify things and not consider the effect of the particles to the flow: https://2021.help.altair.com/2021.1/hwsolvers/acusolve/topics/tutorials/acu/acu_6101_intro_cfd_r.htm#acu_6101_intro_cfd_r

 

Regards,

Geo