CFD Online Logo CFD Online URL
www.cfd-online.com
[Sponsors]
Home > Forums > Software User Forums > ANSYS > FLUENT > Fluent Multiphase

Solution diverges in packed bed granular simulation.

Register Blogs Members List Search Today's Posts Mark Forums Read

Like Tree1Likes
  • 1 Post By vinerm

Reply
 
LinkBack Thread Tools Search this Thread Display Modes
Old   May 9, 2020, 23:51
Default Solution diverges in packed bed granular simulation.
  #1
Member
 
Anshuman Sinha
Join Date: Oct 2018
Posts: 70
Rep Power: 8
Anshs is on a distinguished road
I'm trying to model a packed bed gas absorption model. I'm using the Eulerian multiphase model with the species transport equation.

I have defined the materials as mixture materials, with material 1 = Air ( Oxygen + Nitrogen ) and material 2 = Carbon ( carbon_solid + product_solid ). Material 2 is the granular material and secondary phase.

With the heterogeneous reaction being { Nitrogen + Carbon_solid = product_solid. }

Problem: The solution runs well for normal fluidization bed, but as soon as I turn on the 'packed bed' option the solution diverges, especially the continuity equation diverges first.

I'm using: Drag model = syamlal-obrien-para , void fraction = 0.4 and min fluidization velocity = 0.06 m/s

Screen Shot 2020-05-10 at 8.17.00 AM.jpg


Model: Bed zone = { volume fraction 0.6 Material 2 } and { 0.7 mass fraction species Carbon_solid. }

Screen Shot 2020-05-10 at 8.17.10 AM.jpg
Anshs is offline   Reply With Quote

Old   May 10, 2020, 13:23
Default Packed Bed Model
  #2
Senior Member
 
vinerm's Avatar
 
Vinerm
Join Date: Jun 2009
Location: Nederland
Posts: 2,946
Blog Entries: 1
Rep Power: 36
vinerm will become famous soon enough
A requirement of using packed-bed model in Fluent is to fix the velocities of the cell zone representing packed-bed to 0. So, you need to go to Cell Zone Conditions, enable Fixed Values for the cell zone representing the packed-bed, and set all velocities to 0. This is in addition to selecting packed-bed under secondary phase properties.
__________________
Regards,
Vinerm

PM to be used if and only if you do not want something to be shared publicly. PM is considered to be of the least priority.
vinerm is offline   Reply With Quote

Old   May 10, 2020, 21:09
Default
  #3
Member
 
Anshuman Sinha
Join Date: Oct 2018
Posts: 70
Rep Power: 8
Anshs is on a distinguished road
Quote:
Originally Posted by vinerm View Post
A requirement of using packed-bed model in Fluent is to fix the velocities of the cell zone representing packed-bed to 0. So, you need to go to Cell Zone Conditions, enable Fixed Values for the cell zone representing the packed-bed, and set all velocities to 0. This is in addition to selecting packed-bed under secondary phase properties.
I went to the cell zone condition and set the packed bed cell zone condition to zero velocity for both 'primary phase + secondary phase' as well as tried it by only setting zero velocity for the 'secondary phase', but the solution diverges for both. This time it writes 'Divergence detected in AMG solver: vof-1'


Also, I have tried modeling this particular system using the porous media approach. Where I defined 'wall surface reaction' in the porous zone.

I have a question in that,

1. Whether the 'site species' of the wall surface reaction get consumed or not?

If yes! Then the product of this surface reaction must diminish with time since the reactant is getting consumed up! But no such development is seen in my results!

I'll post the model in detail if required.

Thanks,
Ansh
Anshs is offline   Reply With Quote

Old   May 11, 2020, 04:45
Default Packed Bed
  #4
Senior Member
 
vinerm's Avatar
 
Vinerm
Join Date: Jun 2009
Location: Nederland
Posts: 2,946
Blog Entries: 1
Rep Power: 36
vinerm will become famous soon enough
You only need to fix secondary phase velocity field to 0, not the primary phase velocity field. If the case is diverging, it could be because of many other reasons.

You need to specify the surface concentration of site species under Mechanisms.
__________________
Regards,
Vinerm

PM to be used if and only if you do not want something to be shared publicly. PM is considered to be of the least priority.
vinerm is offline   Reply With Quote

Old   May 11, 2020, 05:48
Default
  #5
Member
 
Anshuman Sinha
Join Date: Oct 2018
Posts: 70
Rep Power: 8
Anshs is on a distinguished road
Quote:
Originally Posted by Anshs View Post

Also, I have tried modeling this particular system using the porous media approach. Where I defined 'wall surface reaction' in the porous zone.

I have a question in that,

1. Whether the 'site species' of the wall surface reaction get consumed or not?

If yes! Then the product of this surface reaction must diminish with time since the reactant is getting consumed up! But no such development is seen in my results!
I did set the site density to a value to 0.0005 kg mol/m2 in the reaction mechanism tab. But the concentration at the site seems to remain to fix since the adsorption reaction looks like coming to a steady-state, where the reaction product is getting fixed to a particular value.

{ **The geometry of this problem remains the same as the original problem in the post, just instead of O2+N2 I'm using O2+H2 and the bed is now a porous zone instead of a granular eulerian zone and has sites of 'ga' (gallium) on it }

mkmk.JPG

in this simulation, 'ga' is the site species and 'gah3' is the solid species. As seen in the picture, it reacts with h2 to produce gah3. And the reaction is set in the porous zone only. { This is simply adsorption of h2 gas from the feed mixture, by preferential adsorption of h2 gas. The outlet mixture will have a lower h2 concentration }

But after some time the 'ga' at the site should decrease in concentration as the time marches ahead. And since the 'ga' is getting lowered, the other reactant h2 now must increase at the outlet. { This phenomenon is not getting in place there }

The image below shows the o2 (oxygen) component getting increased due to the preferential adsorption of h2 (hydrogen) but this o2 must decrease after a certain time when all the site species is consumed or gets covered with gah3 ( product of the surface reaction). But this doesn't happen!

mkmkmk.JPG

Please help me with this.

Thanks
Anshs is offline   Reply With Quote

Old   May 11, 2020, 11:58
Default Consumption Rate
  #6
Senior Member
 
vinerm's Avatar
 
Vinerm
Join Date: Jun 2009
Location: Nederland
Posts: 2,946
Blog Entries: 1
Rep Power: 36
vinerm will become famous soon enough
Have you calculated how much time should it take for almost complete consumption of the site species?
__________________
Regards,
Vinerm

PM to be used if and only if you do not want something to be shared publicly. PM is considered to be of the least priority.
vinerm is offline   Reply With Quote

Old   May 11, 2020, 12:57
Default
  #7
Member
 
Anshuman Sinha
Join Date: Oct 2018
Posts: 70
Rep Power: 8
Anshs is on a distinguished road
Quote:
Originally Posted by vinerm View Post
Have you calculated how much time should it take for almost complete consumption of the site species?
Not really, I was just looking for things to work first, and then I would have implemented the actual numbers!

I can increase the reaction rate of the reaction and check whether the site species is getting consumed or not?
Anshs is offline   Reply With Quote

Old   May 11, 2020, 13:04
Default Consumption Rate
  #8
Senior Member
 
vinerm's Avatar
 
Vinerm
Join Date: Jun 2009
Location: Nederland
Posts: 2,946
Blog Entries: 1
Rep Power: 36
vinerm will become famous soon enough
Yes, artificially increased rate is a good idea.
Anshs likes this.
__________________
Regards,
Vinerm

PM to be used if and only if you do not want something to be shared publicly. PM is considered to be of the least priority.
vinerm is offline   Reply With Quote

Old   May 11, 2020, 15:35
Default
  #9
Member
 
Anshuman Sinha
Join Date: Oct 2018
Posts: 70
Rep Power: 8
Anshs is on a distinguished road
Quote:
Originally Posted by vinerm View Post
Yes, artificially increased rate is a good idea.
I tried to increase the reaction rate but no change was observed.

The solutions appear to reach a steady-state and even after increasing the time step size to 5 sec ( earlier I was using 0.001 sec) no divergence is observed and the solution remains the same.

I also checked for the results of surface coverage of site species at different timestamps but it remains constant throughout to value 1.0 everywhere in the reactor zone.

What may I do now?
Anshs is offline   Reply With Quote

Old   May 12, 2020, 08:22
Default Site species density
  #10
Senior Member
 
vinerm's Avatar
 
Vinerm
Join Date: Jun 2009
Location: Nederland
Posts: 2,946
Blog Entries: 1
Rep Power: 36
vinerm will become famous soon enough
I suppose the number you have is quite large for site density. What's the area of the surface where you have reaction?
__________________
Regards,
Vinerm

PM to be used if and only if you do not want something to be shared publicly. PM is considered to be of the least priority.
vinerm is offline   Reply With Quote

Old   May 12, 2020, 11:47
Default
  #11
Member
 
Anshuman Sinha
Join Date: Oct 2018
Posts: 70
Rep Power: 8
Anshs is on a distinguished road
Quote:
Originally Posted by vinerm View Post
Yes, artificially increased rate is a good idea.
Quote:
Originally Posted by vinerm View Post
I suppose the number you have is quite large for site density. What's the area of the surface where you have reaction?

I have decreased the site density to 5e-10 kgmol/m2. which was earlier 0.005 , but still the solutions have come to standstill, I'm posting results after 400 seconds. Please also look at the residual plot.

check.jpg

residual plot

resi.JPG
Anshs is offline   Reply With Quote

Reply

Tags
eulerian-granular, heterogeneous reaction

Thread Tools Search this Thread
Search this Thread:

Advanced Search
Display Modes

Posting Rules
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is On
HTML code is Off
Trackbacks are Off
Pingbacks are On
Refbacks are On


Similar Threads
Thread Thread Starter Forum Replies Last Post
UDF for Automatic Solution Initialization for previous case data file gartz89 Fluent UDF and Scheme Programming 6 March 30, 2020 08:38
Steady-state solution as IC for transient simulation pardoa OpenFOAM Running, Solving & CFD 3 July 24, 2019 07:49
Solution diverges giof STAR-CD 4 November 5, 2015 13:04
porosity in packed bed with eulerian multiphase Sree FLUENT 1 April 15, 2015 07:17
Questions about gas-solid fluidized bed simulation Sinferm FLUENT 1 January 31, 2007 15:54


All times are GMT -4. The time now is 09:05.