[CFD] What Wall Functions Do I Need for Turbulent Kinetic Energy?
HTML-код
- Опубликовано: 3 фев 2019
- [CFD] What Wall Functions Do I Need for Turbulent Kinetic Energy?
An introduction to the wall functions that are used to capture the near wall behaviour of the turbulent kinetic energy (k) in all mainstream CFD codes (OpenFOAM, Fluent, CFX, Star). The following topics are covered:
1) 2:11 How do we modify the production and dissipation of k in the wall adjacent cell?
2) 13:33 Why do some CFD codes specify dk/dn at the wall?
3) 22:12 Why do some CFD codes specify k directly at the cell centroid?
#wallfunctions #tke #kwallfunction
=============================================
Some useful references:
=============================================
1) ANSYS FLUENT User Manual
12.10.2 Standard Wall Functions
www.afs.enea.it/project/neptun...
2) G. Kalitzin, G. Medic, G. Iaccarino and P. Durbin
Near-wall behaviour of RANS turbulene models and implications for wall functions
Journal of Computational Physics 204 (2005) 265-291
citeseerx.ist.psu.edu/viewdoc/...
3) Moser, Kim & Mansour, Direct numerical simulation of turbulent channel flow up to Re_tau = 590. Physics of Fluids, 11(4), 1999.
cfd.spbstu.ru/agarbaruk/doc/1...
4) The complete set of statistical DNS data for turbulent channel flow:
turbulence.ices.utexas.edu/MKM...
5) J. Bredberg. On the wall boundary condition for turbulence models. Internal report 00/4, Chalmers University of Technology, Sweden, 2000.
www.tfd.chalmers.se/~lada/post...
============================================
Want to learn more?
============================================
Grab a copy of my CFD Fundamentals Course (for beginners):
www.udemy.com/course/computat...
Learn how to write your own CFD code in MATLAB and python (for intermediates):
dr-aidan-wimshurst-s-school.t...
Learn how I draw my figures and diagrams in Inkscape (for everyone):
dr-aidan-wimshurst-s-school.t...
============================================
Did you like the video?
============================================
Download the lecture slides from my website:
www.fluidmechanics101.com/pag...
Buy me a coffee to say thanks:
www.buymeacoffee.com/NKEZrhvg0
Support the channel on Patreon (and get useful extras for your CFD studies):
/ fluidmechanics101
============================================
Donations:
============================================
1) PayPal
www.paypal.me/fluidmechanics101
==================================
Disclaimer
==================================
The methods, algorithms, equations, formulae, diagrams and explanations in this talk are for educational and demonstrative purposes only. They should never be used to analyse, design, accredit or validate real scientific / engineering / mathematical structures and flow systems. For such applications, appropriate trained, qualified and accredited (SQEP) engineers / scientists should be consulted along with the appropriate documentation, procedures and engineering standards. Furthermore, the information contained within this talk has not been verified, peer reviewed or checked in any way and is likely to contain several errors. It is therefore not appropriate to use this talk itself (or any of the algorithms, equations, formulae, diagrams and explanations contained within this talk) as an academic or technical reference. The reader should consult the original references and follow the verification and validation processes adopted by your company / institution when carrying out engineering calculations and analyses. Fluid Mechanics 101 and Dr. Aidan Wimshurst are not accountable or liable in any form for the use or misuse of the information contained in this talk beyond the specific educational and demonstrative purposes for which it was intended. 
Наука
![[CFD] How are Wall Functions Modified for Roughness?](http://i.ytimg.com/vi/vYbRUmVTmGM/mqdefault.jpg)
![[CFD] How are Wall Functions Modified for Roughness?](/img/tr.png)
![[CFD] The Transition SST (gamma - Re_theta) model](http://i.ytimg.com/vi/5htknS9uVEk/mqdefault.jpg)
![[CFD] The Spalart-Allmaras Turbulence Model](/img/1.gif)
I’m totally referencing you in my dissertation XD
God bless you
Me too
Thanks a lot, another excellent explanation.
awesome videos! very grateful
What a coincidence, I just now wanted to learn it, and you published it today! Thanks Bing to bring me here.
Amazing talk!
Thank you!
As usual great
Good explained, congratulations!
Thanks Jorge!
Hello,
1) Thank you for the videos, They have been very informative.
2) After watching most of the wall function videos, researches usually fit one or two equations based on what regime they are in. Why do researches not just take it a step further and fit the experimental data via a B-Spline curve? I do see an issue if a few hundred segments of a B-Spline are used but I do not see much of a code speed difference between utilizing 10 or fewer segments vs 2 or 3 equations that utilize non-standard operations (sqrt, exp, log). Utilizing a spline curve would also achieve an estimation to the experimental data that is in C^4.
Again, thank you.
Hi Matt, i have been thinking about this for a long time. I think my next video on enhanced wall functions should be able to explain it. It seems that the reason to keep the wall functions in this form is so that they can be extended for compressible flows. But yes, for an incompressible flow, there doesnt seem to be any reason why a B-spline or polynomial couldnt be used. I completely agree.
That is a fantastic series of videos! Beside that, can you also have a flew talk on the wall treatment on rough boundaries?
Certainly :) rough wall treatment is quite interesting. The main modification to the codes is the log-law function is translated up and down the y axis. But dont worry, i will explain all in the video!
@@fluidmechanics101 That's so great! I have been stuck on this for a long time!
Hi @Aidan, I have one doubt. It will be very kind of u to put some light on it. At 20.44 minute in the video, the k* is non-dimensional TKE at the corresponding y*. Then why is it directly imposed on wall, i.e., y=0, as the TKE at y=0 must be zero. Isn't k_w should be k_p = k* sq(friction velocity)?
Means the data fitted curve of k* should be used for the value of TKE at the first cell center.
Hope u get some time of ur busy schedule to answer this.
But what to do after getting k*? How do I get the k_p value? And then how to get the P_k? I don't get the next steps. This is different than the epsilon wall function where epsilon_p was directly calculated.
notations are a little bit confusing it will be very helpful for us if you explicitly mention them which letter stands for which quantity, rest the video was very comprehensive
thanks a lot
Thanks for this great video Aidan. For y+ > 11.25, I can see how "Pk" and "eps" for the grid adjacent to the wall are balanced. But I don't get it for y+ < 11.25. I saw in your other video that we use a function for "eps" when y+
Hi,
Pk and eps are not balanced in the wall adjacent cells if we are in viscous sub layer. In the viscous sub layer, dissipation is far greater than the prouduction. However as we move away from the wall, the strength of dissipation decreases and it increases for production. As we reach in log layer, both production and dissipation are nearly equal.
Doesn’t this suffer the same problem as U+? In that you cannot assume the experiment data as accurate for many cases?
Hello, after all these videos on Wall Functions, I wonder about this buffer layer.
Why codes do not add an intermediate treatment for this layer, approached by some polynomial function or whatever, as long as it fits the DNS data? It would simply add a case for 5 < y+ < 30. Doesn't seem like an issue for me, but there must be a reason I suppose ^^.
I completely agree! It would make sense for the velocity wall function. I think it is a bit more difficult for turbulent kinetic energy as we switch over from a zero gradient condition to a fixed value condition. I think this is quite difficult to do!
Did you make a video/class about the term G (turbulent kinetic energy production field) at the wall ("epsilonWallFunction" in openfoam)? Thank for sharing, your content is really nice to watch!
Yep, i have a video for epsilon but dont have one for G yet 👍
@@fluidmechanics101 Thank for your quick reply! I try to implement a new G formulation for an ABL in openfoam.
Hi Aidan, I wonder if its possible for cell based soft like fluent to obtain non-zero value of TKE at the wall(no slip) because i always obtain a little bit of more than 0?
Is it about the fixedvalue=0 or zerogradient?
Great channel btw!
Yes, at the cell centroid fluent will calculate a non-zero value of TKE. It is fixed value = 0 on the cell face when y+< 11.25 and zero gradient on the cell face when y+ > 11.25. Always remember boundary conditions are applied on cell faces and not at cell centroids! Hope this helps
@@fluidmechanics101 Thanks for the response! Knowing that I am a bit confused why after simulating a very simple fan with moving reference frame I get a non zero TKE at the fan blades and zero at the hub? (Fluent. y+ < 11.25 on blades) Maybe I did some wrong BC's.
How i ensure that my CFD mesh in ansys fluent can grab least 80% of total turbulent kinetic energy?are there any options are available in ansys fluent?
If you check out my videos on LES, there is a good explanation in there 😊
So summarize. In the wall k should be define as in slide 19/24 (kw) or as zeroGradient. But what about internal field value?
Hi Krystian, yes this wall function is a bit more confusing than the others. The modern approach is to set zero gradient for k at the wall and then modify the generation of k in the cell. The CFD code will then compute the appropriate value at the cell centroid. Sorry if this is a tricky one, it took me ages to understand it!
@@fluidmechanics101
Good, because I have this ages !
Can I ask also about patch conditions. If we define k and epsilon or omega on the patch conditions (inlet&outlet) the numbers in the calculations will be corrected or they have significant impact on the calculations or it the depends on the code which are using to calculations?
Thank you for answer.
Hi,
I am currently working on a multiphase flow problem in ANSYS and I have certain doubts regarding the same. Will you be comfortable in sharing your email address, so that I can discuss the issues with you?
fluidmechanics101@gmail.com