CFD Online Logo CFD Online URL
www.cfd-online.com
[Sponsors]
Home > Forums > General Forums > Main CFD Forum

Use of k-epsilon and k-omega Models

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

Like Tree146Likes

Reply
 
LinkBack Thread Tools Search this Thread Display Modes
Old   February 8, 2017, 12:13
Default
  #21
New Member
 
Ricardo Ferreira
Join Date: May 2015
Posts: 16
Rep Power: 8
RLFerreira is on a distinguished road
Quote:
Originally Posted by fredicenci View Post
Dear RLFerreira,

Did you get any answer on that? I am trying the same settings to avoid walls functions with Y+ < 1 but the simulation explode...

Thanks
Dear Fredi,

In fact, simulations without wall functions you should keep y+ close unity. Try use "nutLowReWallFunction" in type of nut boundary condition in wall regions. In inlet you can use "calculated" type. Take especial attention to set wall BC in k/epsilon/omega Dict files. Use kLowReWallFunction, epsilonLowReWallFunction, omegaWallFunction, respectively, for wall BCs for k, epsilon and omega dict files. For inlet BC type i used turbulentIntensityKineticEnergyInlet, turbulentMixingLengthDissipationRateInlet, turbulentMixingLengthFrequencyInlet for k, epsilon and omega, respectively. If something is not clear let me know.

Cheers.
RLFerreira is offline   Reply With Quote

Old   February 8, 2017, 12:47
Default
  #22
Member
 
Fredi Cenci
Join Date: Dec 2016
Posts: 38
Rep Power: 6
fredicenci is on a distinguished road
Quote:
Originally Posted by RLFerreira View Post
Dear Fredi,

In fact, simulations without wall functions you should keep y+ close unity. Try use "nutLowReWallFunction" in type of nut boundary condition in wall regions. In inlet you can use "calculated" type. Take especial attention to set wall BC in k/epsilon/omega Dict files. Use kLowReWallFunction, epsilonLowReWallFunction, omegaWallFunction, respectively, for wall BCs for k, epsilon and omega dict files. For inlet BC type i used turbulentIntensityKineticEnergyInlet, turbulentMixingLengthDissipationRateInlet, turbulentMixingLengthFrequencyInlet for k, epsilon and omega, respectively. If something is not clear let me know.

Cheers.
Thanks for your reply,

So, on walls where I do NOT want to use wall functions (y+<1) I should set LowReWallFunction instead fixedValue?

best regards,
fredicenci is offline   Reply With Quote

Old   February 8, 2017, 14:43
Default
  #23
New Member
 
Ricardo Ferreira
Join Date: May 2015
Posts: 16
Rep Power: 8
RLFerreira is on a distinguished road
Quote:
Originally Posted by fredicenci View Post
Thanks for your reply,

So, on walls where I do NOT want to use wall functions (y+<1) I should set LowReWallFunction instead fixedValue?

best regards,
Yes. As far as I know, this type of BC allows integration up to wall. Besides, it can be used if you use wall functions too (it is automatic, you can use with or without wall functions). Therefore, who determines whether it will be or not integrated up to the wall is nut BC. Resuming, if you use a Low Reynolds Number Turb. Model to calculate nut you should use the same type of BC based on Low Reynolds for k and epsilon. Evidently, the models that use omega are already built in the concept of Low Reynolds models, so there is no Low Re in yours BC denominations (notice the name of BC: "omegaWallFunction").
RLFerreira is offline   Reply With Quote

Old   February 8, 2017, 15:27
Default
  #24
Member
 
Fredi Cenci
Join Date: Dec 2016
Posts: 38
Rep Power: 6
fredicenci is on a distinguished road
Quote:
Originally Posted by RLFerreira View Post
Yes. As far as I know, this type of BC allows integration up to wall. Besides, it can be used if you use wall functions too (it is automatic, you can use with or without wall functions). Therefore, who determines whether it will be or not integrated up to the wall is nut BC. Resuming, if you use a Low Reynolds Number Turb. Model to calculate nut you should use the same type of BC based on Low Reynolds for k and epsilon. Evidently, the models that use omega are already built in the concept of Low Reynolds models, so there is no Low Re in yours BC denominations (notice the name of BC: "omegaWallFunction").
I am simulating a turbulent flow around a 3-D cylinder using k-epsilon realizable , I've decided not to use wall functions so I've created a viscous layer and I was using fixedValue with Value uniform 1e-10 on the wall. It was exploding...

Now I changed to epsilonLowReWallFunction; kLowReWallFunction; and nutLowReWallFunction with values of the internalField.. seems like it is converging.. thanks a lot!

Best Regards,
fredicenci is offline   Reply With Quote

Old   May 9, 2017, 02:53
Default
  #25
New Member
 
AmolG
Join Date: Apr 2017
Posts: 4
Rep Power: 6
amolganjare@gmail.com is on a distinguished road
Dear William,
What did you get after using k-omega SST model with low Reynolds Number correction?

Regards!
-Amol
amolganjare@gmail.com is offline   Reply With Quote

Old   August 10, 2017, 10:35
Default
  #26
Senior Member
 
Vino
Join Date: Mar 2013
Posts: 130
Rep Power: 10
Vino is on a distinguished road
Dear fredicenci,

Are you able to solve k-epsilon or any other model with out wall function (y+<1). My simulation shoots up without wall function. thank you.
Vino is offline   Reply With Quote

Old   August 10, 2017, 12:36
Default
  #27
Member
 
Fredi Cenci
Join Date: Dec 2016
Posts: 38
Rep Power: 6
fredicenci is on a distinguished road
Yes. You can use lowReynolds wall functions.

Another alternative is to set the values of k,and omega (or epsilon) on the wall. In this case you should use fixedValue.

best,

Fredi Cenci
fredicenci is offline   Reply With Quote

Old   August 11, 2017, 16:41
Default
  #28
Senior Member
 
Vino
Join Date: Mar 2013
Posts: 130
Rep Power: 10
Vino is on a distinguished road
Dear fredicenci,

Thank you very much.
Vino is offline   Reply With Quote

Old   November 5, 2017, 23:40
Default
  #29
New Member
 
Harshal Akolekar
Join Date: Aug 2016
Location: Melbourne
Posts: 25
Rep Power: 7
hakolekar is on a distinguished road
Great! Great!
hakolekar is offline   Reply With Quote

Old   February 17, 2018, 12:25
Default selection of Turbulence model
  #30
New Member
 
Syed Waqar Mehdi
Join Date: Mar 2016
Posts: 5
Rep Power: 7
Waqar.Mehdi is on a distinguished road
Hello

I am currently doing a project regarding analysis on airfoil of Vertical axis wind turbine , I have already researched about many turbulence models like realizable K-epsilon , RNG K-Epsilon, spalart allmaras and K-Omega SST but which should i select and on what basis. Please help me about this
Waqar.Mehdi is offline   Reply With Quote

Old   February 17, 2018, 12:53
Default selection of Turbulence model
  #31
New Member
 
Syed Waqar Mehdi
Join Date: Mar 2016
Posts: 5
Rep Power: 7
Waqar.Mehdi is on a distinguished road
Hello

I am currently doing a project regarding analysis on airfoil of Vertical axis wind turbine , I have already researched about many turbulence models like realizable K-epsilon , RNG K-Epsilon, spalart allmaras and K-Omega SST but which should i select and on what basis. Please help me about this

regards

Waqar Mehdi

shahwaqar36@gmail.com
Waqar.Mehdi is offline   Reply With Quote

Old   September 9, 2018, 19:44
Default
  #32
Member
 
fede32's Avatar
 
Federico Agustín Caccia
Join Date: Jun 2015
Location: Buenos Aires, Argentina
Posts: 55
Rep Power: 8
fede32 is on a distinguished road
Quote:
Originally Posted by malv83 View Post
I just wanted to say that the RANS models mentioned in this post are indeed Fully turbulent models... I do not see that as a disadvantage, it is just their purpose. If you want to simulate transitional flows (laminar-to-turbulent), you need to use a transitional model. One if the most recent model is the k-omega-v2:

Maurin Lopez. D. K. Walters. “Prediction of transitional and fully turbulent free shear flows using an alternative to the laminar kinetic energy approach”. Journal of Turbulence.

If you want to simulate transient flows using a RANS model, I think this is the best model in the literature.
which model is appropiate to calculate head loss in a pipe with elbows, with Re=300000?
fede32 is offline   Reply With Quote

Old   November 6, 2018, 05:13
Default Menter's SST k-omega turbulent model
  #33
New Member
 
Nina Philipova
Join Date: Jul 2012
Posts: 11
Rep Power: 11
n.phililipova is on a distinguished road
Dear Users of ANSYS Fluent,
I have to apply Menter's SST k-omega turbulent model in Fluent for simulating wall bounded tortuous channel. I looked through Fluent User's Guide and Menter's publications:
Menter, F. R. (1994), "Two-Equation Eddy-Viscosity Turbulence Models for Engineering Applications", AIAA Journal, vol. 32, no 8. pp. 1598-1605
Menter, F, Kuntz, M., Laugtry, R., Ten years of industrial experience with SST Turbulence model ,Turbulence, Heat and Mass Transfer 4.
The cited constants in Fluent User's Guide are different from those in Menter' publications and the sites of NASA, Wikipedia.
Could some one help me which constant I have to use?
I will be grateful to you so much!
n.phililipova is offline   Reply With Quote

Old   November 6, 2018, 08:21
Default Menter's SST k-omega turbulent model
  #34
New Member
 
Nina Philipova
Join Date: Jul 2012
Posts: 11
Rep Power: 11
n.phililipova is on a distinguished road
Constants are the same. Fluent constants are reciprocal to the constants in Menter's publications
n.phililipova is offline   Reply With Quote

Old   December 16, 2018, 11:16
Default
  #35
New Member
 
Lee JP
Join Date: Dec 2018
Posts: 3
Rep Power: 4
ljpGBL is on a distinguished road
Quote:
Originally Posted by CapSizer View Post
... Life becomes a lot easier when you have some experimental data to provide some of that confidence.
As a student who has no conditions to experiment, I fully agree with you.
fede32 likes this.
ljpGBL is offline   Reply With Quote

Old   March 21, 2019, 05:56
Default
  #36
New Member
 
ahmed
Join Date: Jan 2016
Posts: 5
Rep Power: 7
midoronald is on a distinguished road
Quote:
Originally Posted by som87 View Post
Sorry to bump this old thread.
Just wanted to say, isnt DES (the ultimate mutant so far) aparent suitable solution?From my understanding, DES acts as:-
  • k-omega near wall,
  • k-epsilon just a lil after the BL
  • LES in fair enough distance from the wall
what my point is, the eddy viscosity model and LES model, both has their pros and cons.So since DES is the hybrid of these, it's probably time to focus on DES more than modifying the k-omega and k-epsilon models.(just my thought, I dont want to hurt any senior's feelings with their preference)
hello there,

As u referred to DES model, LES is quite computationally expensive compared to any RANS model. It depends on your problem and also depends on the degree of accuracy which you want to achieve. Even LES needs a turbulence model to be used. For some problem LES might take the solver few weeks to be solved, Actually few weeks definitely accepted compared to its fantastic results and its time saving compared to DNS solvers ofc. However, RANS are widely used as most of industrial proposes are not interested in the deep details of the flow fluctuations so they prefer to averaging the flow properties.
midoronald is offline   Reply With Quote

Old   March 21, 2019, 06:13
Default
  #37
New Member
 
ahmed
Join Date: Jan 2016
Posts: 5
Rep Power: 7
midoronald is on a distinguished road
Hello there,

With respect to ur question, it depends on your application and i highly recommend you to read Klaus A. Hoffmann, Computational Fluid Dynamics Vol. III from page 54 to 65, just 11 pages. In brief, I quick read a presentation before for a prof in some college, i really can't remember it, in which he recommended to use k-epsilon models for free-shear layer flows, external flows, small pressure gradient flow. While the k-omega model are better for internal flows and near wall behavior. SST k-omega merges the advantages of the two previous models. However the computational cost increases with using Spalart- Allmaras model, St. K-epsilon, RNG k-epsi, Realizable k-epsi, Standard k-omega, SST k-omega models, respectively.

Although it is recommended to use two or more models on the same application and compare their results so that u can be sure that the model which u chose is gonna work.
midoronald is offline   Reply With Quote

Old   June 24, 2019, 11:08
Default
  #38
New Member
 
abdo
Join Date: Apr 2018
Posts: 28
Rep Power: 4
khaledhmz is an unknown quantity at this point
Quote:
Originally Posted by vivekcfd View Post
I have heard so many times saying people SST is a better model than epsilon based models. This may be true if you are interested only in the region where the separation occurs especially for negative pressure gradients. If you are interested in the flow development after the attachment, lowRe k-eps or realizable k-epsilon model is definitely a better choice.

I have even seen performing realizable k-epsilon with low y+ treatment model performing better than SST in certain transient flows involving strong flow separation. To me it seems that is not right to say that SST is better than k-epsilon.

One must be careful in swirling and rotating flows, there k-epsilon model has certain difficulties.
--------------
hello brother I'am simulate the swirl flow and I have many problem..
could you help me?
khaledhmz is offline   Reply With Quote

Old   September 17, 2019, 04:05
Default k-w (wicox, BSL, SST) Vs. k-e
  #39
Member
 
Tomer
Join Date: Sep 2010
Location: Israel
Posts: 39
Blog Entries: 4
Rep Power: 13
Avr.Tomer is on a distinguished road
Here are some takes from my blog posts about the two families of turbulence models:

k-w:
https://cfdisraelblog.wordpress.com/...%89-sst-model/

k-e:
https://cfdisraelblog.wordpress.com/...odel%ef%81%a5/

And also the subject of wall-treatment, which is of most importance in it of itself:
https://cfdisraelblog.wordpress.com/...-ansys-fluent/
khaledhmz likes this.
Avr.Tomer is offline   Reply With Quote

Old   March 19, 2020, 19:27
Lightbulb
  #40
Member
 
Tomer
Join Date: Sep 2010
Location: Israel
Posts: 39
Blog Entries: 4
Rep Power: 13
Avr.Tomer is on a distinguished road
I love turbulence. Knowing how to model it in just the right fidelity is an art imho. An art I fell in love with 💕

The intention of this set of posts is to embark on a journey of connecting the dots between CFD and turbulence modeling with the phenomenological and practical concepts of engineering turbulence…

Even though it will by no means be all encompassing nor highly accurate I hope for the essence to be captured, and even more so communicated… ✨

https://allaboutcfdtomers.blog/2020/...-can-eat-menu/
Avr.Tomer is offline   Reply With Quote

Reply

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
Bounding epsilon or bounding omega Stylianos OpenFOAM 8 February 23, 2018 14:41
SimpleFoam k and epsilon bounded nedved OpenFOAM Running, Solving & CFD 16 March 4, 2017 09:30
Standard K Epsilon and Transitional SST models Robin FLUENT 1 January 17, 2010 19:06
Use of k omega turbulence model john_w OpenFOAM Running, Solving & CFD 2 September 22, 2009 06:15
SimpleFoam k and epsilon bounded nedved OpenFOAM Running, Solving & CFD 1 November 25, 2008 21:21


All times are GMT -4. The time now is 03:10.