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

Non equilibruim wall function

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

Reply
 
LinkBack Thread Tools Search this Thread Display Modes
Old   September 19, 2006, 13:58
Default Non equilibruim wall function
  #1
sam
Guest
 
Posts: n/a
1. What is difference between non equilibriuim and standard wall fuction.

2 . Whent to use both.

3. What advantage and disadvantage both have
  Reply With Quote

Old   September 20, 2006, 09:58
Default Re: Non equilibruim wall function
  #2
Andrew Hayes
Guest
 
Posts: n/a
I will take a shot at answering this; I am assuming you are referring to non-equilibrium in terms of temperature. TNE model assumes there is a difference in temperature between the fluid and the solid. The wall function of the TE model is a temperature gradient in the fluid from the wall. The TNE model accounts for temperature gradients in both the solid and the fluid. I am speaking of porous media/packed beds/matrix heat exchanger stuff. The TE model is a lot easier to solve b/c it only involves the one equation. The TNE model has two equations, thus a little more number crunching. The TNE model is used on small length scales. Authors to look into; Whitaker, Vafai, and Jiang. They all do a lot of work with porous media and the thermal non-equilibrium stuff.
  Reply With Quote

Old   September 20, 2006, 11:28
Default Re: Non equilibruim wall function
  #3
Iain Barton
Guest
 
Posts: n/a
Hello

This question is to do with a post I put up recently, regarding k-epsilon wall boundary conditions. I didn't use the terms equilibrium and non-equilibrium, but I can clarify that now...

Non-equilibrium is the first approach I describe, basically k is solved at the wall but you modify the production and dissipation terms. Epsilon is found using an algebraic formula using the node normal distance and shear value.

Equilibrium is the second approach I described, for me, it is brute-force approach, I would guess it is likely to be more stable. With this approach, take your law of the wall, find your tangential velocity adjacent to the wall, calculate your frictional velocity (i.e. this gives you shear value) and then use this algebraicly to find values for k and epsilon.

Similar arguments would apply for other turbulence models, k-omega, k-g, k-f, etc.

Since the law of wall, is not "correct" near points of separation, curved sufaces, etc. The Non-equilibrium model assumably is better, and authors on the web report it must be apply if you are doing heat-transfer calculations (otherwise your local nusselt number would be incorrect).

My impression that the Launder-Sharma started the game off with non-equilibrium modelling, but others subsequently prefer to use the equilibrium model probably because it is easier to code?

Iain

ps I have not seen any papers which have made direct comparisons, but I have not looked that hard, I would be interested in reading some, anyone any suggestions, pdfs, off-prints, etc.?

  Reply With Quote

Old   September 20, 2006, 11:59
Default Re: Non equilibruim wall function
  #4
sam
Guest
 
Posts: n/a
thank Barton. you have got the point which i want to have answer.

My question is that in which cases we have to turn on the no equilibrium wall fuction in turbulence model
  Reply With Quote

Old   September 21, 2006, 03:56
Default Re: Non equilibrium wall function
  #5
Iain Barton
Guest
 
Posts: n/a
Regarding which one to use.

Obviously, non-equilibrium is in principle better, but I believe the main point is that it MUST be used for heat transfer calculations...

If you used the equilibrium model then you are making assumptions about the boundary growth (you are enforcing law of wall on the adjacent nodes), and since the heat transfer is dependent on the gradient of this profile, it therefore means your heat transfer computations become in a sense simulation independent!

The other problem is at points of separation, using the equilibium model means that you will predict close to zero frictional velocity (the tangential velocity will be close to zero) at this location. For the flow this means the turbulent viscosity becomes zero, which means the heat transfer will drop at points of separation (there is no turbulent transfer of heat). But of course you would actually expect heat transfer to be at a local maximum at these points!

Iain

ps I repeat that the equilibrium model seems to be growing in popularity in research papers over the years, I don't know why, but I suspect it could be the following reasons:- 1. easier to implement/code 2. more stable/quicker to converge 3. while your answer maybe wrong, at least it wont be crazy 4. if you are doing a boundary layer simulations it probably gives better results (note that research papers tend to do "idealistic" flow problems).

I am very interested to read papers, pdfs, etc. that have done back-to-back comparisons using the same code.

  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
Compile problem ivanyao OpenFOAM Running, Solving & CFD 1 October 12, 2012 10:31
channelFoam for a 3D pipe AlmostSurelyRob OpenFOAM 3 June 24, 2011 14:06
OpenFOAM static build on Cray XT5 asaijo OpenFOAM Installation 9 April 6, 2011 13:21
Problem with rhoSimpleFoam matteo_gautero OpenFOAM Running, Solving & CFD 0 February 28, 2008 07:51
[blockMesh] Axisymmetrical mesh Rasmus Gjesing (Gjesing) OpenFOAM Meshing & Mesh Conversion 10 April 2, 2007 15:00


All times are GMT -4. The time now is 05:11.