CFD Online Discussion Forums

CFD Online Discussion Forums (https://www.cfd-online.com/Forums/)
-   FLUENT (https://www.cfd-online.com/Forums/fluent/)
-   -   Eulerian lagrangian model (https://www.cfd-online.com/Forums/fluent/94404-eulerian-lagrangian-model.html)

shk12345 November 15, 2011 01:23

Eulerian lagrangian model
 
Hi
Can anyone help me in telling me how can i set up a eulerian lagrangian model in fluent.
Thanks in advance
SHK

raj.cfd November 16, 2011 20:16

You have to use Discrete phase models available in Fluent . The continuous phase is solved like an continuum and the secondary phase/discrete phase is solved through lagrangian method.

shk12345 November 17, 2011 13:20

Cfd
 
Kindly tell me the steps to put a eulerian langrangian model in fluent.
I am not been able to interlink the secondary phase of the DPM model with the secondary phase of the eulerian model
kindly tell me how to do that
Thanks
SHK

mali28 November 21, 2011 10:15

Quote:

Originally Posted by shk12345 (Post 332548)
Kindly tell me the steps to put a eulerian langrangian model in fluent.
I am not been able to interlink the secondary phase of the DPM model with the secondary phase of the eulerian model
kindly tell me how to do that
Thanks
SHK

Hi,

After loading the mesh file, go to the models. Select your turbulence model then select Discrete Phase. If you want the particles to exchange momentum with the continuous plahse than select the option "interaction with continuous phase". Now click the injections button. Then click create button. Add your particle properties and the way particles are injected. You may have these via spray cone, single stream etc...
Specify the position of injector (x, y and z) in the same window. Also specify the number of particle streams.

That should be it.

You can specify the physical properties of the injected particles in the Materials menu.

shk12345 November 22, 2011 08:01

Help
 
Hello
Thanks for the reply
I have one clarification. While using the Eulerian-Langrangian Model is it so that we have to on the eulerian model from define>model>multiphase>eulerian and then on the dpm model as discussed above
or

Dpm itself takes care of the eulerian and Lagrangian framework of the calculation.

and
Is that possible to use both eulerian and dpm model at the same time in fluent if yes how??

Thanks
SHK

mali28 November 22, 2011 14:48

Quote:

Originally Posted by shk12345 (Post 333077)
Hello
Thanks for the reply
I have one clarification. While using the Eulerian-Langrangian Model is it so that we have to on the eulerian model from define>model>multiphase>eulerian and then on the dpm model as discussed above
or

Dpm itself takes care of the eulerian and Lagrangian framework of the calculation.

and
Is that possible to use both eulerian and dpm model at the same time in fluent if yes how??

Thanks
SHK

You dont have to select multiphase if you are tracking the particles using Lagrangian approach. You will need it if you want the second phase to be solved using eulerian approach.

shk12345 November 23, 2011 01:09

Help
 
Thanks for the reply but my question remains the same??
Can we use both both DPM and multiphase eulerian model at the same time.If yes how??

Thanks
Shk

mali28 November 23, 2011 05:27

Quote:

Originally Posted by shk12345 (Post 333204)
Thanks for the reply but my question remains the same??
Can we use both both DPM and multiphase eulerian model at the same time.If yes how??

Thanks
Shk

I dont know whether you can use both multiphase eulerian and DPM at the same time.

mit.R November 29, 2013 03:12

hi mali

is discrete phase (particles), turbulence always?

masoudeh March 24, 2015 23:19

Quote:

Originally Posted by shk12345 (Post 333204)
Thanks for the reply but my question remains the same??
Can we use both both DPM and multiphase eulerian model at the same time.If yes how??

Thanks
Shk


Hi, I have the same question, any one can help me please

`e` March 25, 2015 01:16

Have you tried using the multiphase Eulerian model and then enabling the DPM as well to see for yourself? I can't think of any reason that DPM couldn't be used with the multiphase model; especially if you're only concerned with one-way coupling.

masoudeh March 25, 2015 21:39

Quote:

Originally Posted by `e` (Post 538068)
Have you tried using the multiphase Eulerian model and then enabling the DPM as well to see for yourself? I can't think of any reason that DPM couldn't be used with the multiphase model; especially if you're only concerned with one-way coupling.

Thanks for your attention. Let me ask my question clearly: Is it necessary to have injection in lagragian model? Actually, I did multiphase Elurian with DPM without injection, Can I consider it as lagragian?!

By the way, I used two way coupling, what is the difference it makes compared to one way coupling?

Thanks

masoudeh March 25, 2015 21:44

Quote:

Originally Posted by `e` (Post 538068)
Have you tried using the multiphase Eulerian model and then enabling the DPM as well to see for yourself? I can't think of any reason that DPM couldn't be used with the multiphase model; especially if you're only concerned with one-way coupling.

Yes, I did Eulerian and DPM together and I got good results, but I was wandering without having injection can I consider it as lagragian!
I have problem when I add injection! Actually i am interested to have injection so that i can see particle track. can you help me please

`e` March 25, 2015 21:55

What do you mean by "injection"? If you don't specify an injection for the DPM, then you're not using the Lagrangian model (or DPM) at all.

One-way coupling is where the particles are influenced by the fluid; for example, the particles would tend toward the fluid velocity due to drag (and the speed of which is a function of the particle relaxation time). Two-way coupling also includes the influence of the particles on the fluid. Lastly, four-way coupling also considers the particle-particle interactions.

masoudeh March 25, 2015 22:08

Quote:

Originally Posted by `e` (Post 538263)
What do you mean by "injection"? If you don't specify an injection for the DPM, then you're not using the Lagrangian model (or DPM) at all.

One-way coupling is where the particles are influenced by the fluid; for example, the particles would tend toward the fluid velocity due to drag (and the speed of which is a function of the particle relaxation time). Two-way coupling also includes the influence of the particles on the fluid. Lastly, four-way coupling also considers the particle-particle interactions.

I tried to specify injections but it did not converge so I changed initial condition and finally it converged, but still the path lines are not reasonable. however the contours and other results are reasonable, can you tell me what is the problem and how can i improve it! should i increase the number of injections (now i have 3 single injections, i could not specify group injection as it was not be identified by the software and it gave me error)

Thanks,

`e` March 25, 2015 22:20

What didn't converge when you specified the injections? What initial conditions, mass flow rates etc have you specified for your injection streams?

masoudeh March 25, 2015 22:25

Quote:

Originally Posted by `e` (Post 538265)
What didn't converge when you specified the injections? What initial conditions, mass flow rates etc have you specified for your injection streams?

the software gave me the error and stop it. i specify the mass flow rate , velocity temperature ( I assume these parameters the same as the fluid) and diameter of the particle for each single injection

`e` March 25, 2015 22:31

Specifically: did the solver crash/stop during the DPM calculations or the continuous phase iterations? What error was displayed?

masoudeh March 25, 2015 22:38

Quote:

Originally Posted by `e` (Post 538267)
Specifically: did the solver crash/stop during the DPM calculations or the continuous phase iterations? What error was displayed?

after specific number of iteration it stopped, anyway i changed initial condition such as momentum pressure volume fraction and it could be helpful and it gave me the results. i just have problem with path lines which are not in the same pattern i expected! how can i work on that

`e` March 25, 2015 22:52

You can specify how many DPM iterations the solver does until it stops (the default is 500). If the particle stream hasn't escaped the domain within this number of iterations then this stream would be incomplete (finishes partway through the domain).

What path lines are you generating (show us a figure) and what are you expecting (show us another figure)?

masoudeh March 25, 2015 23:18

1 Attachment(s)
Quote:

Originally Posted by `e` (Post 538272)
You can specify how many DPM iterations the solver does until it stops (the default is 500). If the particle stream hasn't escaped the domain within this number of iterations then this stream would be incomplete (finishes partway through the domain).

What path lines are you generating (show us a figure) and what are you expecting (show us another figure)?

this is the path line i got. all the dreams have the same color i want different colors corresponds to velocity.
Attachment 38104
for particle track the streams are incomplete! i will also provide fig for this

`e` March 25, 2015 23:25

Where is the mesh (display the outline only to avoid cluttering) to see the overall shape of the domain? What path lines are you expecting?

You've just quoted my post on how to solve the "incomplete" particle tracking issue, please read carefully...

masoudeh March 26, 2015 10:42

3 Attachment(s)
Quote:

Originally Posted by `e` (Post 538278)
Where is the mesh (display the outline only to avoid cluttering) to see the overall shape of the domain? What path lines are you expecting?

You've just quoted my post on how to solve the "incomplete" particle tracking issue, please read carefully...

Hi e

I think i do not understand your question clearly, but the max number of steps for tracking is 500..
this is the path line i expect to seeAttachment 38110
and this my particle track which i said it is incompleteAttachment 38111

and this is the path line i got with outline mesh ( i am not sure if u were asking me this!)Attachment 38112

`e` March 26, 2015 20:17

If the particles are incomplete at the end of the simulation then that would indicate they require more steps until they leave the domain.

Your task: change the maximum number of steps for tracking from 500 to say 50,000.

masoudeh March 26, 2015 21:16

Quote:

Originally Posted by `e` (Post 538460)
If the particles are incomplete at the end of the simulation then that would indicate they require more steps until they leave the domain.

Your task: change the maximum number of steps for tracking from 500 to say 50,000.

Thanks a lot for your attention.
Would you please tell me if i just use eulerian approach can it consider the brownian motion?

masoudeh March 26, 2015 22:43

Quote:

Originally Posted by `e` (Post 538460)
If the particles are incomplete at the end of the simulation then that would indicate they require more steps until they leave the domain.

Your task: change the maximum number of steps for tracking from 500 to say 50,000.

I have a message" temperature limited to.." what does it mean? What should I do, does it affect my results? I got results but they not exactly what i expected.

`e` March 27, 2015 00:24

Tackle one issue at a time: have you tried a tracking simulation after increasing the maximum number of steps for tracking to remove the "incomplete" particle problem?

masoudeh March 27, 2015 11:10

Quote:

Originally Posted by `e` (Post 538478)
Tackle one issue at a time: have you tried a tracking simulation after increasing the maximum number of steps for tracking to remove the "incomplete" particle problem?

It works :) thanks

masoudeh March 27, 2015 21:40

Quote:

Originally Posted by `e` (Post 538478)
Tackle one issue at a time: have you tried a tracking simulation after increasing the maximum number of steps for tracking to remove the "incomplete" particle problem?

this problem solved, thanks.
I got the message that the temperature is limited to..and when i remove the limit form temperature the outlet temperature the software gave me was out of range. Can you help me how to solve it?

`e` March 28, 2015 02:45

Have a read of the Fluent FAQ section on convergence issues and we'll go from there.

masoudeh March 30, 2015 00:13

Quote:

Originally Posted by `e` (Post 538649)
Have a read of the Fluent FAQ section on convergence issues and we'll go from there.

Hi e

I read this part it said the momentum and pressure should be 0.6 and 0.1. I changed mine (pressure 1 and momentum 0.7) to these numbers it becomes worse than before, it gave me the reverse flow message and then increasing sweep and then it said no divergence. these errors came in the first iterations!!
it also said to set k and eps to 0.4 which i could not find out where should i set these.

masoudeh March 30, 2015 00:25

Quote:

Originally Posted by `e` (Post 538649)
Have a read of the Fluent FAQ section on convergence issues and we'll go from there.

I tried to follow the instruction to remove this temperature limited error but it did not work. 1. I do not know how i can solve this error " temperature is limited ..."

2. does eulerian consider interaction between particles? or just eulerian with lagragian could consider this interaction?

Thanks

`e` March 30, 2015 01:12

1. When did this temperature limited error arise, did the simulation previously work and then did you change something?

2. The one- or two-way coupling between the discrete phase (Lagrangian) and continuous phase (Eulerian) dictates the interaction between particles, or lack thereof. Both of these options are available in the DPM of Fluent. Have a read of the "Modeling Discrete Phase" chapter of the user's guide.

masoudeh March 30, 2015 14:48

Quote:

Originally Posted by `e` (Post 538911)
1. When did this temperature limited error arise, did the simulation previously work and then did you change something?

2. The one- or two-way coupling between the discrete phase (Lagrangian) and continuous phase (Eulerian) dictates the interaction between particles, or lack thereof. Both of these options are available in the DPM of Fluent. Have a read of the "Modeling Discrete Phase" chapter of the user's guide.

1. Yes it works fine when i used just eulerian model (DPM with no injection). when I added injection the temperature limited error arise at iteration around#20. When you said to follow the fluent FAQ i decreased momentum and increase pressure to the number in the guid (0.7 and1) and then it gave me this error"reverse flow" and it said no converge!!

2. As it gave me good results with eulerian i preferred to work with eulerian but i was wondering if particles interaction (brownian motion) could be considered in eulerian model?! I know i can use two way coupling but need to be sure eulerian consider solid particles interactions. I read it but still need to know. please give me your opinion.

Thanks

`e` March 30, 2015 17:39

1. Terminology: you're not using DPM if there are no injections. Ensure you're not exceeding the limits of DPM (particle volume fraction etc).

2. I'm familiar with the DPM but not the Euler-Euler approach of modelling particles in a fluid. If your simulations are working in the Eulerian reference frame, then that's great, but you'll need to check the user's manual (or theory guide) for checking how it treats diffusion / Brownian motion.

masoudeh March 31, 2015 17:48

Quote:

Originally Posted by `e` (Post 539097)
1. Terminology: you're not using DPM if there are no injections. Ensure you're not exceeding the limits of DPM (particle volume fraction etc).

2. I'm familiar with the DPM but not the Euler-Euler approach of modelling particles in a fluid. If your simulations are working in the Eulerian reference frame, then that's great, but you'll need to check the user's manual (or theory guide) for checking how it treats diffusion / Brownian motion.

Ok thanks.
Would you please tell me why i got this "temperature limit " error and how can i fix it?

`e` March 31, 2015 19:30

Quote:

Originally Posted by `e` (Post 539097)
Ensure you're not exceeding the limits of DPM (particle volume fraction etc).

Try injecting a small single particle stream of negligible mass flow rate.

masoudeh March 31, 2015 20:42

Quote:

Originally Posted by `e` (Post 539339)
Try injecting a small single particle stream of negligible mass flow rate.

i have done it with 3 single injections (of solid particles) and as my flow is laminar the mass flow rate is very low!

masoudeh March 31, 2015 22:29

Quote:

Originally Posted by `e` (Post 539339)
Try injecting a small single particle stream of negligible mass flow rate.

now i changed the type of injection from single to cone and put 3 particles stream, fortunately it gives me good result and i did not have any "temperature limite" error. but, my problem is when i track particles i just see 1 stream!!
1. i put 3 particles for the cone , i expect i see 3 streams. why i see just 1 stream?
2. i need to see the location of each particle. i mean i want to find out particles motion. can u tell me how should i set particle track to display it for me? i could not find any option in particle track that display the location of each of these 3 particles!

`e` March 31, 2015 23:20

Have a read of the user's guide and tutorials on visualising particle streams; all of your questions are answered there in step by step detail.


All times are GMT -4. The time now is 11:52.