Lagrangian model

yorelchr · May 18, 2014, 05:51

Hello!!!

Do someone knows how to put a permeable condition for a phase when using the Lagrangian model?
In the phase condition under the boundary node, I only have "wall", no possibility to choose "phase impermeable".
Any help?

Thank you

me3840 · May 19, 2014, 12:04

You need to create a multiphase interaction in your physics continuum.

yorelchr · May 20, 2014, 04:19

Hi, thank you for your reply; I'll try that.
Do you think it is a good method for modeling particle air filters?

me3840 · May 20, 2014, 10:11

That depends on what you want out of the simulation.

yorelchr · May 20, 2014, 10:28

Hi me3840,

I need to evaluate the drop pressure and the efficiency in term of particle absorption. I am new to multiphase and I have tried a lot of things without big success for now.
If using Euler multiphase, I don't understand how to simulate porosity for the filter zone. VOF would impose same velocity for air and particles.
For the absorption of particles, a sink term is ok in the porous zone?

Thank you if you have any advice for me!

me3840 · May 20, 2014, 10:54

What is meant by particle absorption? Particles will only have the ability to be blocked by your filter; none will move through it.

yorelchr · May 20, 2014, 11:09

yes that's what I meant.
In fact I have 2 projetcs, 2 same kind of situations:

first project : filters with particle. Drop pressure and efficiency.
Drop pressure is ok with porous zone for the filter and just air. I am at the moment just trying to add particles, keeping in mind that later I have to add a source term to allow the filter to "keep" particles.

Second projcet : this time in water and oil, the "filter" has to keep a certain amount of oil. I think this could be done with VOF. I tried with only water and put a constant value as source term and it "kept" the good anount of water, that is why I though of doing this for the first project.

I really don't know what I should choose between Euler, Lagrange and VOF.

me3840 · May 20, 2014, 11:35

First project: Lagrangian is okay here. If the particles are solid, I would use DEM.

Second project: VOF is a hard sell here, depending on the particle sparsity. I would probably use Lagrangian and have the particles collect in a fluid film on the filter surface or something.

yorelchr · May 22, 2014, 05:14

Hello me3840,

thank you for your reply.
I am working on the first project using DEM as you suggested. But I think it is really slow and I will try just Lagrangian model since the particles do not have any interaction with each other.
- is it better to launch DEM or Lagrange from a solution with just air that is already converged or from the beginning?
- I've let all default values for a first try, it was ok, but I would like to be more precise; how do we choose Poisson Coeff, Young's modulus, Rolling Resistance etc?
- since I have porosity zone, better using a porous zone or porous baffle with these models?
- will I be able with one of these 2 models to model absorption or is it better in that case using some other multiphase model?

Thanks a lot for any new advice.

me3840 · May 22, 2014, 10:09

You can launch a single shot of particles with a steady approach, yes, but the usefulness of this depends on what you want out of the simulation; you won't get any feedback from the particles to the flow.

Those few values you specified there are material properties - they are not chosen. You need to input them depending on your material.

It likely won't matter, for this case they'll do the same thing.

You can force particles to not pass the porous interface whether you're using the porous baffle or a porous media.

yorelchr · May 22, 2014, 11:31

I have to model the pressure drop, so find the good coefficients for viscous and inertial resistence. This is ok now.
I also have to model the absorption, given an inlet mass flow of particles (all distributed on the inlet) and their characteristics (density, size...). The particles don't have any impact on the air flow since their density will be very low.
So I have to let some of the particles enter into the filter zone and stay inside: for thisshould I write a field function or add a source term at the filtre zone so it keeps the particles? (Some others should simply cross the filter). Without any source term, porosity just slows the particle but don't keep them or don't "block" them. Is that correct?

me3840 · May 22, 2014, 11:46

If you just want the porous media to slow the particles, I don't think you have to do anything with a source term, just ensure the interfaces are set correctly to allow particles to pass through them.

yorelchr · May 22, 2014, 11:53

No no, I want the filter to let some particles stay inside the filter zone.

yorelchr · May 28, 2014, 07:36

Hello,

could anyone explain a little about lagrangian model?
I tried particle tracks, injectors ...but I can't see anythong at all...what is it supposed to do? I've also done the turorial...no way...even trying the same conditions as in the tutorials, no way to "see " anything.
did I miss something?
I've tried injectors fromt the inlet, from grid distribution...in any way possible (per injector etc...)...

What is the residence time? maybe understanding this could help me.

me3840 · May 28, 2014, 09:18

To see the particles you need to add the particles (or particle tracks) to a displayer in a scene.

yorelchr · May 28, 2014, 09:49

ok for particle tracking, but in my situation, it is not interesting : I just see lines...
When running, is there a way to see what's hanppening to the discrete phase?? residuals for continuous zone remains the same (I'm starting from a previous simulation with just air, then added Lagrangian model and particles).
I would like to see, as in tutorials, particles coming from injectors that are crossing my domain; but nothing happens. A grid distribution appears and then stays at the same place...
I ask for particle velocity on a scalar view but nothing...
I've done many times exactly as in tuttorials, but I don't know, maybe in my situation I can't use Lagrande model.
And this residence time? what is it?

yorelchr · May 28, 2014, 09:59

in fact, whatever I try to visualize concerning particles, I have nothing, as if there were not

yorelchr · May 28, 2014, 10:06

do someone has a simple example of a model using Lagrange? if she/he could send me the summary report it would be really nice because I think I've tried everything I could since week without any result.

yorelchr · May 29, 2014, 03:40

hello,

still about Lagrangian model : I have some results but my track file remains empty whatever I do.
I've selected it in the phase model. I have saved the simulation, but it is empty.

When running, I have some views of the particles but then when asking in a report for velocity I have no values at all.
If i do a plot of the particle velocity, I just have some points at the beginning of the domain, then nothing.
So, I mean, I can't use any results about particles.

any help?

May 18, 2014, 05:51	Lagrangian model	#1
yorelchr Senior Member christine Join Date: Jul 2009 Location: europe Posts: 125 Rep Power: 16	Hello!!! Do someone knows how to put a permeable condition for a phase when using the Lagrangian model? In the phase condition under the boundary node, I only have "wall", no possibility to choose "phase impermeable". Any help? Thank you

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
Is it possible to model natural convection in a 2D horizontal model in fluent	caitoc	FLUENT	1	May 5, 2014 13:32
Error finding variable "THERMX"	sunilpatil	CFX	8	April 26, 2013 07:00
Problems bout CFD model of biomass gasification, Downdraft gasifier	wanglong	FLUENT	2	November 25, 2009 23:27
2 stage axial turbine model convergence issues	sherifkadry	CFX	2	September 7, 2009 20:51
Advanced Turbulence Modeling in Fluent, Realizable k-epsilon Model	Jonas Larsson	FLUENT	5	March 13, 2000 03:27

May 19, 2014, 12:04		#2
me3840 Senior Member Join Date: Nov 2010 Location: USA Posts: 1,232 Rep Power: 24	You need to create a multiphase interaction in your physics continuum.

May 20, 2014, 04:19		#3
yorelchr Senior Member christine Join Date: Jul 2009 Location: europe Posts: 125 Rep Power: 16	Hi, thank you for your reply; I'll try that. Do you think it is a good method for modeling particle air filters?

May 20, 2014, 10:11		#4
me3840 Senior Member Join Date: Nov 2010 Location: USA Posts: 1,232 Rep Power: 24	That depends on what you want out of the simulation.

May 20, 2014, 10:28		#5
yorelchr Senior Member christine Join Date: Jul 2009 Location: europe Posts: 125 Rep Power: 16	Hi me3840, I need to evaluate the drop pressure and the efficiency in term of particle absorption. I am new to multiphase and I have tried a lot of things without big success for now. If using Euler multiphase, I don't understand how to simulate porosity for the filter zone. VOF would impose same velocity for air and particles. For the absorption of particles, a sink term is ok in the porous zone? Thank you if you have any advice for me!

May 20, 2014, 10:54		#6
me3840 Senior Member Join Date: Nov 2010 Location: USA Posts: 1,232 Rep Power: 24	What is meant by particle absorption? Particles will only have the ability to be blocked by your filter; none will move through it.

May 20, 2014, 11:09		#7
yorelchr Senior Member christine Join Date: Jul 2009 Location: europe Posts: 125 Rep Power: 16	yes that's what I meant. In fact I have 2 projetcs, 2 same kind of situations: first project : filters with particle. Drop pressure and efficiency. Drop pressure is ok with porous zone for the filter and just air. I am at the moment just trying to add particles, keeping in mind that later I have to add a source term to allow the filter to "keep" particles. Second projcet : this time in water and oil, the "filter" has to keep a certain amount of oil. I think this could be done with VOF. I tried with only water and put a constant value as source term and it "kept" the good anount of water, that is why I though of doing this for the first project. I really don't know what I should choose between Euler, Lagrange and VOF.

May 20, 2014, 11:35		#8
me3840 Senior Member Join Date: Nov 2010 Location: USA Posts: 1,232 Rep Power: 24	First project: Lagrangian is okay here. If the particles are solid, I would use DEM. Second project: VOF is a hard sell here, depending on the particle sparsity. I would probably use Lagrangian and have the particles collect in a fluid film on the filter surface or something.

May 22, 2014, 05:14		#9
yorelchr Senior Member christine Join Date: Jul 2009 Location: europe Posts: 125 Rep Power: 16	Hello me3840, thank you for your reply. I am working on the first project using DEM as you suggested. But I think it is really slow and I will try just Lagrangian model since the particles do not have any interaction with each other. - is it better to launch DEM or Lagrange from a solution with just air that is already converged or from the beginning? - I've let all default values for a first try, it was ok, but I would like to be more precise; how do we choose Poisson Coeff, Young's modulus, Rolling Resistance etc? - since I have porosity zone, better using a porous zone or porous baffle with these models? - will I be able with one of these 2 models to model absorption or is it better in that case using some other multiphase model? Thanks a lot for any new advice.

May 22, 2014, 10:09		#10
me3840 Senior Member Join Date: Nov 2010 Location: USA Posts: 1,232 Rep Power: 24	You can launch a single shot of particles with a steady approach, yes, but the usefulness of this depends on what you want out of the simulation; you won't get any feedback from the particles to the flow. Those few values you specified there are material properties - they are not chosen. You need to input them depending on your material. It likely won't matter, for this case they'll do the same thing. You can force particles to not pass the porous interface whether you're using the porous baffle or a porous media.

May 22, 2014, 11:31		#11
yorelchr Senior Member christine Join Date: Jul 2009 Location: europe Posts: 125 Rep Power: 16	I have to model the pressure drop, so find the good coefficients for viscous and inertial resistence. This is ok now. I also have to model the absorption, given an inlet mass flow of particles (all distributed on the inlet) and their characteristics (density, size...). The particles don't have any impact on the air flow since their density will be very low. So I have to let some of the particles enter into the filter zone and stay inside: for thisshould I write a field function or add a source term at the filtre zone so it keeps the particles? (Some others should simply cross the filter). Without any source term, porosity just slows the particle but don't keep them or don't "block" them. Is that correct?

May 22, 2014, 11:46		#12
me3840 Senior Member Join Date: Nov 2010 Location: USA Posts: 1,232 Rep Power: 24	If you just want the porous media to slow the particles, I don't think you have to do anything with a source term, just ensure the interfaces are set correctly to allow particles to pass through them.

May 22, 2014, 11:53		#13
yorelchr Senior Member christine Join Date: Jul 2009 Location: europe Posts: 125 Rep Power: 16	No no, I want the filter to let some particles stay inside the filter zone.

May 28, 2014, 07:36		#14
yorelchr Senior Member christine Join Date: Jul 2009 Location: europe Posts: 125 Rep Power: 16	Hello, could anyone explain a little about lagrangian model? I tried particle tracks, injectors ...but I can't see anythong at all...what is it supposed to do? I've also done the turorial...no way...even trying the same conditions as in the tutorials, no way to "see " anything. did I miss something? I've tried injectors fromt the inlet, from grid distribution...in any way possible (per injector etc...)... What is the residence time? maybe understanding this could help me.

May 28, 2014, 09:18		#15
me3840 Senior Member Join Date: Nov 2010 Location: USA Posts: 1,232 Rep Power: 24	To see the particles you need to add the particles (or particle tracks) to a displayer in a scene.

May 28, 2014, 09:49		#16
yorelchr Senior Member christine Join Date: Jul 2009 Location: europe Posts: 125 Rep Power: 16	ok for particle tracking, but in my situation, it is not interesting : I just see lines... When running, is there a way to see what's hanppening to the discrete phase?? residuals for continuous zone remains the same (I'm starting from a previous simulation with just air, then added Lagrangian model and particles). I would like to see, as in tutorials, particles coming from injectors that are crossing my domain; but nothing happens. A grid distribution appears and then stays at the same place... I ask for particle velocity on a scalar view but nothing... I've done many times exactly as in tuttorials, but I don't know, maybe in my situation I can't use Lagrande model. And this residence time? what is it?

May 28, 2014, 09:59		#17
yorelchr Senior Member christine Join Date: Jul 2009 Location: europe Posts: 125 Rep Power: 16	in fact, whatever I try to visualize concerning particles, I have nothing, as if there were not

May 28, 2014, 10:06		#18
yorelchr Senior Member christine Join Date: Jul 2009 Location: europe Posts: 125 Rep Power: 16	do someone has a simple example of a model using Lagrange? if she/he could send me the summary report it would be really nice because I think I've tried everything I could since week without any result.

May 29, 2014, 03:40		#19
yorelchr Senior Member christine Join Date: Jul 2009 Location: europe Posts: 125 Rep Power: 16	hello, still about Lagrangian model : I have some results but my track file remains empty whatever I do. I've selected it in the phase model. I have saved the simulation, but it is empty. When running, I have some views of the particles but then when asking in a report for velocity I have no values at all. If i do a plot of the particle velocity, I just have some points at the beginning of the domain, then nothing. So, I mean, I can't use any results about particles. any help?