Population Balance - Example Projects

mnolan93 · January 19, 2016, 10:22

Hi guys,

I seem to be having strange divergence issues when activating nucleation and growth phenomena with population balance modelling. I was wondering if anyone has example projects they would be able to send me, where these phenomena work? Just to see if i can narrow down where the problem is.

Regards

CFDYourself · January 19, 2016, 11:46

Hi MNolan,
I'm unable to give you a definitive answer about this as I haven't got practical experience with the PDF model. but I might be able to give you a few pointers.

Firstly nucleation & growth themselves can have aspects in common with stiff chemistry problems; it's a possibility that the problem itself is just a pain to solve. Have you tried using more conservative solver settings?

What about the ansys literature ? you should be able to get the "Ansys fluent Population Balance Module Manual" from Ansys if you don't already have it. They should also have a tutorial where it's implimented. Try looking for one called "tut-16-dqmom" titled "Tutorial: Modeling Liquid Reactions in CIJR Using the Eulerian PDF transport (DQMOM-IEM) model" as a starter

mnolan93 · January 19, 2016, 11:51

Thanks for your reply.

What do you mean by conservative solver settings? Furthermore, as for the tutorial, I have already followed the crystallisation tutorial in the guide line by line, however I am still getting the same divergence error

CFDYourself · January 21, 2016, 14:19

"conservative solver settings" generally means lower underrelaxation factors (but can mean other things aswell). Typically stiff chemistry can be solved by setting extremely low urfs, and slowly ramping them up, but historically stiff chemistry problems (assuming stiff chemistry is the problem) are prone to difficulties solving.

In the past I have had divergence problems due to errors in the way I've actually set up the phases & species for models. As I said I don't have experience with the population balance model so I can't really help, but my understanding is it's similar to a Euler-Euler model, with the separate 'bins' of sizes acting like separate Eulerian dispersed phases so It could be similar ?

Ansys also has an option called "stiff chemistry solver" (which has helped me in other work) but I'm not sure if this'll be available in the population balance model ?
it is accessed by going to models-species-reactions-volumetric(tickbox)-stiffchemistrysolver(tickbox).

Sorry to hear the tutorial isn't working, have you tried contacting ansys support ?

mnolan93 · January 21, 2016, 15:50

Hi again,

Thanks again for your reply. How you described my model is exactly the case. I'm not sure if it's to do with the phases or bins though, because the PB works fine until I turn on the nucleation and growth phenomena.

The tutorial I actually tried was a step by step guide to PB setup in the fluent user guide. The tutorial you quoted I can't actually find online. Do you by any chance have a URL?

Kind regards

CFDYourself · January 22, 2016, 08:34

You need to get access to Ansys' support service in order to get access to their tutorials etc.

They've changed the policy since I have last used it, but they have a "customer portal" and a "student portal" now - this will give you access to all of their tutorials.

You might also be able to phone or email the nearest branch and ask them to send through the specific tutorial, but long-term it's much better if you get access to the portal.

January 19, 2016, 10:22	Population Balance - Example Projects	#1
mnolan93 Member Martin Nolan Join Date: Dec 2014 Posts: 33 Rep Power: 11	Hi guys, I seem to be having strange divergence issues when activating nucleation and growth phenomena with population balance modelling. I was wondering if anyone has example projects they would be able to send me, where these phenomena work? Just to see if i can narrow down where the problem is. Regards

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
Coupling RPI wall boiling model with population balance model in Fluent	softice2006	Fluent Multiphase	1	April 19, 2023 02:09
coupling population balance with reaction	mehdi kamyabi	FLUENT	7	December 5, 2015 01:14
Problems when using Population balance model with Boiling model	luliqiang	FLUENT	0	May 28, 2013 05:22
Population balance for airlift reactor (FLUENT)	amira	Main CFD Forum	1	October 31, 2009 07:54
Population Balance Theory	student	Main CFD Forum	2	August 10, 2008 04:12

January 19, 2016, 11:46		#2
CFDYourself Member Jim Join Date: May 2015 Posts: 47 Rep Power: 10	Hi MNolan, I'm unable to give you a definitive answer about this as I haven't got practical experience with the PDF model. but I might be able to give you a few pointers. Firstly nucleation & growth themselves can have aspects in common with stiff chemistry problems; it's a possibility that the problem itself is just a pain to solve. Have you tried using more conservative solver settings? What about the ansys literature ? you should be able to get the "Ansys fluent Population Balance Module Manual" from Ansys if you don't already have it. They should also have a tutorial where it's implimented. Try looking for one called "tut-16-dqmom" titled "Tutorial: Modeling Liquid Reactions in CIJR Using the Eulerian PDF transport (DQMOM-IEM) model" as a starter

January 19, 2016, 11:51		#3
mnolan93 Member Martin Nolan Join Date: Dec 2014 Posts: 33 Rep Power: 11	Thanks for your reply. What do you mean by conservative solver settings? Furthermore, as for the tutorial, I have already followed the crystallisation tutorial in the guide line by line, however I am still getting the same divergence error

January 21, 2016, 14:19		#4
CFDYourself Member Jim Join Date: May 2015 Posts: 47 Rep Power: 10	"conservative solver settings" generally means lower underrelaxation factors (but can mean other things aswell). Typically stiff chemistry can be solved by setting extremely low urfs, and slowly ramping them up, but historically stiff chemistry problems (assuming stiff chemistry is the problem) are prone to difficulties solving. In the past I have had divergence problems due to errors in the way I've actually set up the phases & species for models. As I said I don't have experience with the population balance model so I can't really help, but my understanding is it's similar to a Euler-Euler model, with the separate 'bins' of sizes acting like separate Eulerian dispersed phases so It could be similar ? Ansys also has an option called "stiff chemistry solver" (which has helped me in other work) but I'm not sure if this'll be available in the population balance model ? it is accessed by going to models-species-reactions-volumetric(tickbox)-stiffchemistrysolver(tickbox). Sorry to hear the tutorial isn't working, have you tried contacting ansys support ?

January 21, 2016, 15:50		#5
mnolan93 Member Martin Nolan Join Date: Dec 2014 Posts: 33 Rep Power: 11	Hi again, Thanks again for your reply. How you described my model is exactly the case. I'm not sure if it's to do with the phases or bins though, because the PB works fine until I turn on the nucleation and growth phenomena. The tutorial I actually tried was a step by step guide to PB setup in the fluent user guide. The tutorial you quoted I can't actually find online. Do you by any chance have a URL? Kind regards

January 22, 2016, 08:34		#6
CFDYourself Member Jim Join Date: May 2015 Posts: 47 Rep Power: 10	You need to get access to Ansys' support service in order to get access to their tutorials etc. They've changed the policy since I have last used it, but they have a "customer portal" and a "student portal" now - this will give you access to all of their tutorials. You might also be able to phone or email the nearest branch and ask them to send through the specific tutorial, but long-term it's much better if you get access to the portal.