# How to calculate optimum time step for VOF model in Fluent?

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

October 8, 2021, 03:25
How to calculate optimum time step for VOF model in Fluent?
#1
New Member

Join Date: Dec 2020
Posts: 23
Rep Power: 5
I am simulating 2D Sonic Nozzle & Under expanded jet impinging at the perpendicular wall,
with 0.2 mm water layer over the wall using the VOF model.
Distance from nozzle exit to the wall in 25mm
Area weighted velocity at exit is 320 m/s

I calculated time steps like following
Time step= 1/3*(25e-3/320)= 26e-6 or 26 micro-sec

Is the above calculation is right?

Details of flowfield in the attachments
Attached Files
 cfd.pdf (40.2 KB, 10 views)

 October 8, 2021, 04:58 #2 Senior Member   Lucky Join Date: Apr 2011 Location: Orlando, FL USA Posts: 5,680 Rep Power: 66 instead of 25e-03, you need to use your smallest cell dimension. If the water layer is 0.2mm, you're going to need cells on that order (preferable much smaller) to even resolve anything. You're looking at 2e-05 m.

 October 8, 2021, 10:25 Time Step Size #3 New Member   Join Date: Dec 2020 Posts: 23 Rep Power: 5 If i want to keep my global courant number 1, then from CFL= (V*Del t)/del x So, CFL=1, V=320 m/s , minimum cell size (near center of plate)=0.005e-3 m with this value, Del t= 1.5e-8 s.... So, should I calculate this way?

 October 8, 2021, 11:11 #4 Senior Member   Lucky Join Date: Apr 2011 Location: Orlando, FL USA Posts: 5,680 Rep Power: 66 Right... Mostly... That gets you the convective courant number. I noticed that you mentioned a sonic jet. So you might also need to consider the acoustic Courant number since the problem is compressible and add the sound velocity. The speed that goes into the acoustic courant number is V+c or V-c, with c being the sound speed That's still the Courant number btw. CFL is something else that people call it by mistake. You should correct your bad habits early. These are just for estimating the timestep size. You should do a calculation and then get the Courant number from the result and adjust it accordingly. A Courant number of 1 is not bad but a lot of problems will require it to be even smaller. In general you should also do a timestep size study if you haven't already. If you already know a Courant number of 1 is good enough then that's fine.

 October 9, 2021, 00:48 Time Step Size #5 New Member   Join Date: Dec 2020 Posts: 23 Rep Power: 5 Thanks a lot.

 Tags time step size, transient analisys, vof model