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

Limitations on time the step for lid driven cavity flow

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

Like Tree1Likes
  • 1 Post By FMDenaro

Reply
 
LinkBack Thread Tools Search this Thread Display Modes
Old   October 21, 2016, 10:48
Default Limitations on time the step for lid driven cavity flow
  #1
New Member
 
Juan Camilo Acosta Arango
Join Date: Mar 2016
Posts: 5
Rep Power: 10
ja0335 is on a distinguished road
I'm following these slides

http://www3.nd.edu/~gtryggva/CFD-Cou...-Lecture-5.pdf

They formulate N-S equations usin stream, vorticity functions, I'm want to know what kind of analysis they do to conclude the time step restriction showed in the page 5. Its very important for me to know how it is done, thanks!

__________________
Juan Camilo Acosta Arango
ja0335 is offline   Reply With Quote

Old   October 21, 2016, 10:58
Default
  #2
Senior Member
 
Filippo Maria Denaro
Join Date: Jul 2010
Posts: 6,793
Rep Power: 71
FMDenaro has a spectacular aura aboutFMDenaro has a spectacular aura aboutFMDenaro has a spectacular aura about
Quote:
Originally Posted by ja0335 View Post
I'm following these slides

http://www3.nd.edu/~gtryggva/CFD-Cou...-Lecture-5.pdf

They formulate N-S equations usin stream, vorticity functions, I'm want to know what kind of analysis they do to conclude the time step restriction showed in the page 5. Its very important for me to know how it is done, thanks!



It stems from the numerical stability analysis for the explicit Forward Time Central Space (FTCS) applied on a 2D convection-diffusion equation. Note that the first constraint is exact only for the pure diffusive case (no convection). The method is unconditionally unstable for the pure convective case. In the mixed case the constraint involves a more complex functional law in the (cfl, Reh) plane. You can find details in any good CFD textbook.
FMDenaro is offline   Reply With Quote

Old   October 21, 2016, 11:11
Default
  #3
New Member
 
Juan Camilo Acosta Arango
Join Date: Mar 2016
Posts: 5
Rep Power: 10
ja0335 is on a distinguished road
Quote:
Originally Posted by FMDenaro View Post
It stems from the numerical stability analysis for the explicit Forward Time Central Space (FTCS) applied on a 2D convection-diffusion equation. Note that the first constraint is exact only for the pure diffusive case (no convection). The method is unconditionally unstable for the pure convective case. In the mixed case the constraint involves a more complex functional law in the (cfl, Reh) plane. You can find details in any good CFD textbook.
I'm curious about your answer, because as I know it is a Von Newman stability analysis, however this problem is not linear so the Von Newman analysis can not be applied here. I'm missing something?
__________________
Juan Camilo Acosta Arango
ja0335 is offline   Reply With Quote

Old   October 21, 2016, 11:17
Default
  #4
Senior Member
 
Filippo Maria Denaro
Join Date: Jul 2010
Posts: 6,793
Rep Power: 71
FMDenaro has a spectacular aura aboutFMDenaro has a spectacular aura aboutFMDenaro has a spectacular aura about
Quote:
Originally Posted by ja0335 View Post
I'm curious about your answer, because as I know it is a Von Newman stability analysis, however this problem is not linear so the Von Newman analysis can not be applied here. I'm missing something?
Yes. The Von Neumann analysis is applied for the linear case (as well as the matrix analysis you can perform by using the Gershgorin locus method). The constraint in those slides are valid for the linear equation. Usually, they are roughly used also for the non linear equation.
FMDenaro is offline   Reply With Quote

Old   October 21, 2016, 11:18
Default
  #5
Senior Member
 
Filippo Maria Denaro
Join Date: Jul 2010
Posts: 6,793
Rep Power: 71
FMDenaro has a spectacular aura aboutFMDenaro has a spectacular aura aboutFMDenaro has a spectacular aura about
P.S.:the second constraint must have the ratio to the mesh size h but, as I wrote, the case for the pure convective case is unstable. The key for a wiggles-free solution is to work with the Reh<2 condition.
FMDenaro is offline   Reply With Quote

Old   October 21, 2016, 11:20
Default
  #6
New Member
 
Juan Camilo Acosta Arango
Join Date: Mar 2016
Posts: 5
Rep Power: 10
ja0335 is on a distinguished road
Quote:
Originally Posted by FMDenaro View Post
P.S.:the second constraint must have the ratio to the mesh size h
sorry to be annoying, but do you know any paper or article where this specific analysis is done?
__________________
Juan Camilo Acosta Arango
ja0335 is offline   Reply With Quote

Old   October 21, 2016, 11:26
Default
  #7
Senior Member
 
sbaffini's Avatar
 
Paolo Lampitella
Join Date: Mar 2009
Location: Italy
Posts: 2,163
Blog Entries: 29
Rep Power: 39
sbaffini will become famous soon enoughsbaffini will become famous soon enough
Send a message via Skype™ to sbaffini
Fletcher: Computational Techniques for Fluid Dynamics, Vol. 2, pp. 335 cites a more general condition for the 2D equations. He actually cites this:

http://link.springer.com/book/10.100...-3-642-85952-6

pp. 183 as original source. If you don't trust them, I swear, I did it by myself (straightforward Neumann on the equivalent scalar equation) and it is correct .

It is (\Delta x = \Delta y):


\Delta t \le \frac{\Delta x^2}{4 \nu}

\Delta t \le \frac{4 \nu}{\left(|u|+|v|\right)^2}

Last edited by sbaffini; October 21, 2016 at 11:45. Reason: Added the actual formula from Fletcher
sbaffini is offline   Reply With Quote

Old   October 21, 2016, 11:57
Smile
  #8
New Member
 
Juan Camilo Acosta Arango
Join Date: Mar 2016
Posts: 5
Rep Power: 10
ja0335 is on a distinguished road
Quote:
Originally Posted by sbaffini View Post
Fletcher: Computational Techniques for Fluid Dynamics, Vol. 2, pp. 335 cites a more general condition for the 2D equations. He actually cites this:

http://link.springer.com/book/10.100...-3-642-85952-6

pp. 183 as original source. If you don't trust them, I swear, I did it by myself (straightforward Neumann on the equivalent scalar equation) and it is correct .

It is (\Delta x = \Delta y):


\Delta t \le \frac{\Delta x^2}{4 \nu}

\Delta t \le \frac{4 \nu}{\left(|u|+|v|\right)^2}
Thanks a lot!
__________________
Juan Camilo Acosta Arango
ja0335 is offline   Reply With Quote

Old   October 21, 2016, 12:05
Default
  #9
Senior Member
 
Filippo Maria Denaro
Join Date: Jul 2010
Posts: 6,793
Rep Power: 71
FMDenaro has a spectacular aura aboutFMDenaro has a spectacular aura aboutFMDenaro has a spectacular aura about
you can read any good textbook of both CFD and numerical anlysis...
as example, http://dl.iranidata.com/book/daneshg...idata.com).pdf
FMDenaro is offline   Reply With Quote

Old   October 21, 2016, 12:32
Default
  #10
New Member
 
Juan Camilo Acosta Arango
Join Date: Mar 2016
Posts: 5
Rep Power: 10
ja0335 is on a distinguished road
Quote:
Originally Posted by sbaffini View Post
Fletcher: Computational Techniques for Fluid Dynamics, Vol. 2, pp. 335 cites a more general condition for the 2D equations. He actually cites this:

http://link.springer.com/book/10.100...-3-642-85952-6

pp. 183 as original source. If you don't trust them, I swear, I did it by myself (straightforward Neumann on the equivalent scalar equation) and it is correct .

It is (\Delta x = \Delta y):


\Delta t \le \frac{\Delta x^2}{4 \nu}

\Delta t \le \frac{4 \nu}{\left(|u|+|v|\right)^2}
Sbafiny, I've checked the books you mentioned but again I don't see they do any analysis to arrive to the time step condition.

You mentioned something about "straightforward Neumann on the equivalent scalar equation", may you point me to some source about these formulation and the analysis done over it?
__________________
Juan Camilo Acosta Arango
ja0335 is offline   Reply With Quote

Old   October 21, 2016, 12:55
Default
  #11
Senior Member
 
sbaffini's Avatar
 
Paolo Lampitella
Join Date: Mar 2009
Location: Italy
Posts: 2,163
Blog Entries: 29
Rep Power: 39
sbaffini will become famous soon enoughsbaffini will become famous soon enough
Send a message via Skype™ to sbaffini
That's what i meant by all those "cites" and "swear". If you can handle the italian language between the 14 equations on 2 pages (i guess so), here it is attached my old analysis. As you can see, it is pretty straightforward Neumann analysis.
Attached Files
File Type: pdf stability.pdf (58.8 KB, 19 views)
sbaffini is offline   Reply With Quote

Old   October 21, 2016, 18:28
Default
  #12
Senior Member
 
Filippo Maria Denaro
Join Date: Jul 2010
Posts: 6,793
Rep Power: 71
FMDenaro has a spectacular aura aboutFMDenaro has a spectacular aura aboutFMDenaro has a spectacular aura about
I strongly suggest to plot the amplification factor for the value = 1 in the (Reh, cfl) plane.
FMDenaro is offline   Reply With Quote

Old   October 21, 2016, 19:13
Default
  #13
Senior Member
 
Filippo Maria Denaro
Join Date: Jul 2010
Posts: 6,793
Rep Power: 71
FMDenaro has a spectacular aura aboutFMDenaro has a spectacular aura aboutFMDenaro has a spectacular aura about
This issue is not as simple as it would appear...


https://estudogeral.sib.uc.pt/bitstr...20analysis.pdf
sbaffini likes this.

Last edited by FMDenaro; October 22, 2016 at 10:19.
FMDenaro is offline   Reply With Quote

Reply

Tags
time step size

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
Transient simulation not converging skabilan OpenFOAM Running, Solving & CFD 14 December 16, 2019 23:12
Stuck in a Rut- interDyMFoam! xoitx OpenFOAM Running, Solving & CFD 14 March 25, 2016 07:09
time step directories naming issue Andrea_85 OpenFOAM 3 April 3, 2014 08:38
dynamic Mesh is faster than MRF???? sharonyue OpenFOAM Running, Solving & CFD 14 August 26, 2013 07:47
Orifice Plate with a fully developed flow - Problems with convergence jonmec OpenFOAM Running, Solving & CFD 3 July 28, 2011 05:24


All times are GMT -4. The time now is 09:12.