twoPhaseEulerFoam and LES
 

Hello all,

anyone tried implementing LES in twoPhaseEulerFoam solver?


awaiting for replies.. :)

Regards
Raghavendra

Hi,

yes, I worked on that during my PhD. I essentially used Smagorynsky type models for the gas phase to simulated gas-particle flows. The work was done with OF 1.4.0, and does not compile with more recent versions of OpenFOAM. I did not port it to OpenFOAM 1.5.x for lack of time (I'm working on a different topic right now).

What I can suggest to you is to reuse the single phase classes for LES, and adapt them at the multiphase case, which is essentially what I did.

Best regards,

Hi,

I actually started modifying the single phase incompressible LES-models for use with twoPhaseEulerFoam yesterday evening. Based on Albertos' example. Didn't get that far with it yet though.

I'd be interested in including some sort of model to account for modulation of turbulence due to particles in them. (gas-solid flows)

Regards,
Juho

Hi Alberto and Juho,

Thanks for your replies..Its good to know about LES implementation in twoPhaseEulerFoam.

Are there any references about LES in mutiphase flows which I can go through before working on it?

Is it possible to share the LES implementation for twoPhaseEulerFoam (in OF-1.4) and I can continue from there to implement the same in OF-1.5 ..?


Regards
Raghavendra

Hi Raghavendr,

I have no problem to share the code, but I don't really know how useful it can be for your purposes, because starting from the single-phase LES classes in OpenFOAM and adapting them to the multiphase case is not a long work (1-2 days should be enough if you know OF a bit). You mainly need to rename the classes, make some changes to the submodels according to the literature (for gas-particle flows there are many paper, tell me if you need references) and make some minor change to the twoPhaseEulerFoam to replace the turbulent viscosity with the SGS viscosity.

An Juho said, it would be necessary to account for the effect of particles on the sub-grid phenomena description (See for example the very interesting work of Prof. Simonin on this purpose), and also include the effect of sub-grid phenomena on the dispersed phase (drag correction). This was not done in my code.


Best,

Hi Alberto,

Yes, it seems to be quite simple and just to change the way nutb is calculated to implement LES..
but
I am very much new to OF and I haven't tried modifying the source code. Till now (past 2months) I was just using the available solvers :)

If its not a problem is it OK to share the LES code you implemented for twoPhaseEulerFoam?..
are there any references available with you regarding implementation of LES in twoPhaseEulerFoam for liquid gas flows?..
(Similar to Phd thesis of Henrik Rusche what you mentioned earlier for K-epsilon model and twoPhaseEulerFoam)

Hi, I don't see any contact information in your profile where I can send the code.

Best,

ohh!!

u can send it to me through

raagh77@gmail.com

in the old forum my mail id was visible in myy profile..
don't know why its not showing in the new forum!

Check you email. :-)

About your address, I have no idea. Also mine should be visible, and it should be possible to download a vCard, but the forum is so bloated I didn't find it anywhere! :-P

Hi Alberto,

Yes, I received your mail..
Thanks :)

I made changes in my profile (which I made in the old forum also) so now its fine :)
thanks for notifyiong me and also for the code


Regards
Raghavendra

Hi Alberto and Raghavendra,

I am playing around with twoPhaseEulerFoam as well and I am interested in using LES with that solver. Could one of you please send me the code as well? It would help me a lot!

Many thanks in advance!

Regards,

Gijsbert

gijsbert.wierink@gmail.com

Edit: Thanks Raghavendra, I received the file. Now for some homework to get it to compile nicely on OpenFOAM-1.5.

It seems Raghavendra sent it to you without even asking me...

Just to clarify, that code is NOT released publicly, so I would like to know who is using it.

Hi Alberto,

Yes, Raghavendra has sent me the code. I understand you want to know what goes on with it and where, and it's not so nice if other people throw your stuff around. I am sure no harm was intended. As for me, I am developing a model to simulate three-phase stirred tank reactors (continuous liquid with dispersed bubbles and solids) for my PhD. The issue with models in literature is that most of them simulate a (passive) scalar field on top of just a single phase. Another big thing is that commonly the k-epsilon model is used, so that turbulence is isotropic. For bubble-particle interaction in my case the anisotropy, in my opinion, is significant. Hence my interest in a LES two-phase solver. I have understood that your PhD thesis is on CFD of two-phase flows. Would it be possible to obtain your thesis as a pdf from somewhere, or should I perhaps ask your former university about it? (I sent you a message about this already, sorry if you get it twice ...). Your thesis would be very useful and of course, if I use your solver I will refer to your work.

For now I'm having some compilation trouble with twoPhaseLESEulerFoam, probably due to a difference in versions of OpenFOAM. I'll let you know if I come up with something interesting :). And please inform me if you would just not like me to use your solver the way it is. If you are interested I can send you a more detailed mail about what I do and my progress, in any case your feedback is very welcome.

Have a great day! :)

Best regards,

Gijsbert

@ Alberto
 

I had forwarded LEStwophaseEulerfoam code to him! (the same mail, without asking you)
am sorry I should have asked you before as you are the sole owner of the code..

:confused:

Regards
Raghavendra

Hello,

no problem. I simply wanted to clarify the code is not published. The reasons are many:

a) it is not up to date
b) some parts are not tested (see interstitial fluid effects, for example)
c) there are some quite rough edges (instabilities when the granular temperature goes to alphaMax, limitations on SGS models (no account for particle influence, no SGS model for gas-liquid flow)).
d) finally, I would like to keep track of what goes on with the code to eventually put all the efforts together and avoid repeating work done by others ;-)

Just an idea: why don't we keep in touch about the progress you have? I should work soon on the basic solver again to make it more stable.

Best,
Alberto

Yes,
thats a good idea..
(is that through Forums we can be in touch? or anyother means)

Hopefully we can see twoPhaseLESEulerFoam in the next version of OF :)

Hi Raghavendra,

to keep in touch email is OK I think.

Best,

hello,

I am new in OpenFOAM and I am starting my work on LES multiphase with EE approach. I see you guys are already working on that topic and probably have a lot of experience in that field. I was wondering Alberto if it would be possible to have a look on your code? I read it is not checked yet but I am sure it would help me to understand how I can make some changes inside the source terms in OpenFOAM. Thank you in advance.

Peter

Hi Peter,

the main problem is that the code is outdated. I have written it using OF 1.3/1.4, and I did not adapt it to the new versions of OpenFOAM(r).

However if you drop me an email I can send it to you.

P.S. Juho Peltola used my code as a starting point to write his own, and he used OF 1.5. Maybe you want to contact him too ;-)

Best,

Advancement in twoPhaseEulerLESFoam
 

Hi Alberto, Raagh, Juho, and others,

I am working on a particle cloud settling problem.
I wonder whether dynMixSmagorinsky can be added.
The single phase problem has been well addressed.
I would like to know if any advancement has been done and be happy to be the first user.

PS: Alberto, may I have a look at your code?

Thanks in advance.

Roro

How dense (average volume fraction)?

Yes, it can be added very easily re-using the work done for single-phase flows by OpenFOAM developers.

I can send you my code, but it wasn't ported to a recent release of OpenFOAM. You can either email me, or, if you want a newer version, get in touch with Juho.

Best,

twoPhaseLESEulerFoam
 

Hello Alberto and all particle settling Foamers,

I am quite interested in twoPhaseLESEulerFoam, too. I have so far used a modified interDyMFoam solver to simulate freeface fluid structure interactions of shallow landslides with flexible protection barriers in my PhD thesis, but since this is done i would like to work on erosion around piles in hyperconcentrated turbulent flows, as they occur during flood events. Could you imagine simulating particles in a turbulent Herschel-Bulkley rheology flow using twoPhaseEulerFoam? Please let me know if I can contribute.

Regards,

Albrecht
(Albrecht v. Boetticher, albrecht.vonboetticher[at]wsl.ch, Swiss Federal Institute for Forest, Snow and Landscape Research WSL, Züricherstrasse 111, CH-8903 Birmensdorf)

Hi,

the solution algorithm should not have extreme difficulties to deal with non-Newtonian problems.

In OpenFOAM 2.0.0 twoPhaseEulerFoam implements the PIMPLE algorithm, which makes it more stable. If you use the kinetic theory of granular flows, the treatment of the particle pressure is still explicit, and this makes the solution less stable in dense regions.

You'll have to add the non-Newtonian model, but there should be no difficulty in managing large values of the viscosity (it's similar to what happens in a granular flow in the frictional regime, and there the algorithm, once the particle pressure is treated implicitly, works fine).

Best,

Hi Alberto,

thanks for these advices, the mixture of LES with non-Newtonian models causes discussions from a physical point of view, but the implementation works fine in interFoam and I think I'll run some simple experiments with a flume for validation. I will start writing the proposal in fall so that the project could start in next summer. But instead of dealing with the particles in twoPhaseEulerFOAM I think I will use InterFOAM and link it with an external discrete finite element code that traces the particles, and allows for complex bindings between the particles which makes it possible to simulate fishing nets as well as the foundation of piles in eroding granular material with different grain sizes. I thought about feeding the particle position and velocity to InterFOAM, together with a particle radius. Then finding all cells that ly within the particle radius and then work through the resulting actio and reactio forces, applied to the flow and to the particle. I will post the first trial tests when ready...

Good luck. It seems a very complicated problem. :-)

News on twoPhaseEulerFoam and LES
 

Are there news on twoPhaseEulerFoam and LES?

Who has worked that out and how difficult is this?
Will this combination be publicly available in the future?

Hi Gerhard,

if with LES you mean a Smagorinsky-type model for the continuous phase, the implementation is easy, and can be based on what is already available into OpenFOAM.

Hi, all,

I am trying to run twoPhaseEulerFoam with one of the RSTM.
I am wondering if anyone has done this before or any instruction on how to modify the code?

thank you in advance,

yy

You have to replace the k-epsilon implementation with the RSTM implementation for multiphase flows. The code does not rely on the turbulence models in the library, since you have to include the volume fraction.

Best,

hi all
I am new in Open FOAM and I am starting my work on LES of an evaporating two-phaseflow with LE approach.I built my code for the liquid phase.and i want to know if it is possible to find in OF for the gas phase a LES code solves the filtered compressible Navier-Stokes equations and if i can add my source term in the code and integrate the gas solver with my liquid solver.

mohamed

Hi, OpenFOAM has compressible LES solvers. You might want to take a look at rhoPimpleFoam.

thanks alberto for your help , what about the integration between LES gas solver and my liquid solver ?

I do not know enough about your solver to comment on this. Is your solver written in OpenFOAM? What kind of coupling do you need?

Alberto

my solver written in fortran 90 and there is no problem to written in c++ .
the coupling (i need u, v,T,roh from LES solver & export the conductive heat flux to it at each time step )

This should not be too difficult: you can simply create an interface between the two codes. If you want to do things more cleanly, you probably should take a look at the Lagrangian solver already present in OpenFOAM, which includes heat and mass transfer too (see the combustion / diesel) solvers.

thanks alberto for your help :)

Thank you very much for your reply, Alberto.

I am trying the standard kepsilon model for now but the simulation is very unstable. I have to limit my Courant Number below 0.2 otherwise it will diverge.
I am using limitedLinearV 1 for my convective terms and I am wondering if anyone has similar experience or any idea to improve it?

Thank you.

hi alberto ,
I start to look at rhoPimpleFoam solver , I need your help to find some tutorial about governing equations and how to run the solver, Thanks for your help.

mohamed

Hi,

the governing equations for rhoPimpleFoam are the continuity, momentum and energy equation of a compressible fluid. You can refer to Ferziger and Peric and Hirsch books, for example.

For the tutorials, a starting point are the examples here

~/OpenFOAM/OpenFOAM-2.1.x/tutorials/compressible/rhoPimpleFoam/

If you have questions, just ask on the forum :-)