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

On the time step size.

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

Reply
 
LinkBack Thread Tools Display Modes
Old   March 31, 2005, 21:49
Default On the time step size.
  #1
mini
Guest
 
Posts: n/a
Hello everyone: During the unsteady computation, can the increased time step size affect the results and postprocessing? If I set the time step size 1e-05 initially and 1e-04 finally , are the results reasonable?
  Reply With Quote

Old   April 1, 2005, 02:35
Default Re: On the time step size.
  #2
debs
Guest
 
Posts: n/a
What is your numerical Scheme ? If the scheme is implicit the difference should not be so much . In an explicit scheme the difference can be considerable even for simple problems.

  Reply With Quote

Old   April 1, 2005, 07:42
Default Re: On the time step size.
  #3
ap
Guest
 
Posts: n/a
Right. Also consider the physical time scale of the system you're modeling. You should use a time step smaller than it.

Best regards, ap
  Reply With Quote

Old   April 3, 2005, 06:46
Default Re: On the time step size.
  #4
khalid sultan
Guest
 
Posts: n/a
Hi mini;

The time step - according to my knowledge - is dependent on: 1- the cell length. 2- the {|u+a|} term which express the maximum speed with which the information can travel. 3-and the CFL. If the cell has del(x) != del(y), you can take the smaller to compute the time step, although there are some other tricks that could be used.

Having this said, the time step should be calculated and not given. Of course if you give a little bit smaller time step than that computed, your solver will work. this is equvalent to decreasing the CFL, but you will increase; meanninglessly, your executaion time. The same arguement apply -to some extent- for a little bit increase.

Hoping it will help. cheers ;-)

  Reply With Quote

Old   April 3, 2005, 09:27
Default Re: On the time step size.
  #5
ap
Guest
 
Posts: n/a
This is true if you use an explicit method, while if you use a fully implicit one, the method is stable for each value of the time step.

However, this doesn't mean you can use a time step as big as you like. Obviously you still have the physical constraints on your dt to obtain an accurate solution.

ap
  Reply With Quote

Old   April 3, 2005, 20:29
Default Re: On the time step size.
  #6
mini
Guest
 
Posts: n/a
implicit scheme, thanks.
  Reply With Quote

Old   April 3, 2005, 20:32
Default Re: On the time step size.
  #7
mini
Guest
 
Posts: n/a
Yes. Thanks.
  Reply With Quote

Reply

Thread Tools
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 On
Pingbacks are On
Refbacks are On


Similar Threads
Thread Thread Starter Forum Replies Last Post
critical error during installation of openfoam Fabio88 OpenFOAM Installation 21 June 2, 2010 03:01
Modeling in micron scale using icoFoam m9819348 OpenFOAM Running, Solving & CFD 7 October 27, 2007 00:36
Transient simulation not converging skabilan OpenFOAM Running, Solving & CFD 12 September 17, 2007 17:48
natural convection - time step size co2 FLUENT 7 June 3, 2004 21:56
how to increase time step size? co2 FLUENT 6 May 17, 2004 07:25


All times are GMT -4. The time now is 17:11.