# some question

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

 June 28, 2011, 05:07 some question #1 Senior Member   hamid Join Date: Nov 2010 Posts: 185 Rep Power: 6 Can I simulate two phases flow with VOF in steady state ? in grid independency study for UNSTEADY flow , how should I consider the time steps in every simulation, i mean i have grid1 with X amount of grids and grid2 with 3X and grid3 with 9X and i compare them, but should I consider different time steps as well?

 June 28, 2011, 06:59 #2 Senior Member     Svetlin Philipov Join Date: Mar 2009 Location: United Kingdom Posts: 176 Rep Power: 8 there is a kind of unwritten rule - time step is of a magnitude of that allows short phenomena modeling. it depends on the processes you are going to model and bigger mesh not always results in smaller timestep At the beginning try to use one and the same timestep. If not applicable - divide by 2 each time.

 June 28, 2011, 07:54 #3 Senior Member   hamid Join Date: Nov 2010 Posts: 185 Rep Power: 6 Hi Philipov Apparently denser mesh needs smaller time step (due to courant number), is it true? As I understood u mean I better to use the same time step for all three meshes , X,3X,9X, ? thank u Hamid

 June 28, 2011, 09:56 #4 Senior Member     Svetlin Philipov Join Date: Mar 2009 Location: United Kingdom Posts: 176 Rep Power: 8 As a starting point -Yes and if you do not obtain convergence reduce timestep by factor of 2 until you get convergence and proper results.

 June 30, 2011, 10:06 #5 Member   Adrien Lemoine Join Date: Jun 2011 Location: Paris,France Posts: 31 Rep Power: 6 Hi, if you have a problems due to the courant number (which means you are using an explicit VOF scheme) then you should switch to an Implicit VOF scheme whch gives good results and is easier to use. You should start with an time around 10-3 s. If you really want to stay with a explicit VOF scheme then you should calculate the time step with this formula : Time step < Size of the smallest cell / Speed of sound in your phase. Adrien __________________ ------------------------------------------- Plz add reputation if the answer is correct A cœur vaillant rien d'impossible ------------------------------------------

 June 30, 2011, 11:21 #6 Senior Member     Svetlin Philipov Join Date: Mar 2009 Location: United Kingdom Posts: 176 Rep Power: 8 I will a little disagree with the last comment as far as in air speed of sound is about 342m/s and if you have a cell size of 0.1m in X direction (for example) then you need dt of 3.0E-04 . From last post seems that the task is independent from velocity. Dont you think velocity of 1m/s; 10m/s and 100 m/s does not matter?!?!?!

 June 30, 2011, 16:38 #7 Member   Adrien Lemoine Join Date: Jun 2011 Location: Paris,France Posts: 31 Rep Power: 6 Hi Philipov, I forgot one thing in my last post : This formula will always make your scheme stable since the velocity will be always under the speed of sound. If you are 100% sure that the maxspeed will be for exemple 50 m/s (EVERYWHERE) then you can calculate the time step with : Time step < Size of the smallest cell / 50. The formula "Time step < Size of the smallest cell / Speed of sound in your phase" is always true but not always the quickest. Sincerly Adrien __________________ ------------------------------------------- Plz add reputation if the answer is correct A cœur vaillant rien d'impossible ------------------------------------------

July 1, 2011, 03:02
#8
Senior Member

Svetlin Philipov
Join Date: Mar 2009
Location: United Kingdom
Posts: 176
Rep Power: 8
Quote:
 Originally Posted by AdrienL Hi Philipov, I forgot one thing in my last post : This formula will always make your scheme stable since the velocity will be always under the speed of sound. If you are 100% sure that the maxspeed will be for exemple 50 m/s (EVERYWHERE) then you can calculate the time step with : Time step < Size of the smallest cell / 50. The formula "Time step < Size of the smallest cell / Speed of sound in your phase" is always true but not always the quickest. Sincerly Adrien
That's ok.... for me it is ok but sometimes people here are not so familiar and this explanation is good for them

 July 2, 2011, 16:28 #9 New Member   Join Date: May 2011 Posts: 3 Rep Power: 6 Hey, if you have problems with the Courant number. You could also try to decrease the under relaxation factors a little bit. That would cause a slower simulation but does not effect the speed so much like decreasing the time step.

 Thread Tools Display Modes Linear Mode

 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 OffTrackbacks are On Pingbacks are On Refbacks are On Forum Rules

 Similar Threads Thread Thread Starter Forum Replies Last Post tanven FLUENT 2 July 5, 2015 11:22 niklas OpenFOAM 2 July 31, 2013 16:03 universez OpenFOAM Running, Solving & CFD 0 January 12, 2010 21:31 Carlos Main CFD Forum 4 August 23, 2002 05:55 K.L.Huang CD-adapco 1 March 29, 2000 04:57

All times are GMT -4. The time now is 11:21.