Time stepping method

luca007 · February 15, 2017, 06:21

Hello Everyone please what's the différence between the time stepping method fixed and variable ?

KevinZ09 · February 15, 2017, 07:11

You can look up the details in the Fluent manual as to how the variable time step is precisely used/calculated, but as you can understand, for fixed time stepping the timestep is a fixed number as specified by you, while for the variable time stepping method it's variable (just as with the adaptive time stepping methond) and the value then depends sort of on the local CFL number of the interface.

luca007 · February 15, 2017, 08:08

thanks for you answer in my case the velocity is 1 and the mesh size is 0.005 so the delta t is 0.007 so i want to simulat the flow for 24 seconde in reality please how i can do that ?

KevinZ09 · February 15, 2017, 08:11

Well, if your dt = 0.007s and you want your simulation to run for 24 seconds physical time, then run it for 24/0.007 = 3429 timesteps.

luca007 · February 15, 2017, 08:19

ok this is for fixed method ?

KevinZ09 · February 15, 2017, 08:20

Yes

Sent from my GT-I9505 using CFD Online Forum mobile app

luca007 · February 15, 2017, 08:42

ok please what's better fixed method or variable method

LuckyTran · February 15, 2017, 09:14

Adaptive time-stepping allows Fluent to globally and sometimes locally adjust the time-step to satisfy Courant number criteria. You end up with solutions are odd times though and for post-processing that can be a pain. However, if you don't need constant time-step size between intervals, such as in transient problems, then you can consider it. If you want to do something like fft at any point, you realize fft algorithm only works if time-step is constant.

I recommend to just start with fixed time stepping method to get your feet wet. It's simple, straightforward, and you don't wonder what is going on under the hood. After you are familiar with Fluent, then you can look into adaptive time-steps. To be honest, I've never used adaptive time-stepping, but I can see how useful it might be to some. It's extremely useful in problems where you have large disparity between flow velocities in different regions.

luca007 · February 15, 2017, 10:42

hello i'm using the fixed time steps but the results is not good please what's the wrong

February 15, 2017, 06:21	Time stepping method	#1
luca007 Member kaouachi anouar Join Date: Jul 2016 Posts: 66 Rep Power: 9	Hello Everyone please what's the différence between the time stepping method fixed and variable ?

February 15, 2017, 08:08	time stepping method	#3
luca007 Member kaouachi anouar Join Date: Jul 2016 Posts: 66 Rep Power: 9	thanks for you answer in my case the velocity is 1 and the mesh size is 0.005 so the delta t is 0.007 so i want to simulat the flow for 24 seconde in reality please how i can do that ?

Thread Tools	Search this Thread
Show Printable Version Email this Page	Search this Thread: Advanced Search
Display Modes
Linear Mode Switch to Hybrid Mode Switch to Threaded Mode

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
[Other] Contribution a new utility: refine wall layer mesh based on yPlus field	lakeat	OpenFOAM Community Contributions	58	December 23, 2021 02:36
Extrusion with OpenFoam problem No. Iterations 0	Lord Kelvin	OpenFOAM Running, Solving & CFD	8	March 28, 2016 11:08
[solidMechanics] solidMechanics gear contact in rotation	nlc	OpenFOAM CC Toolkits for Fluid-Structure Interaction	3	January 11, 2015 06:41
Star cd es-ice solver error	ernarasimman	STAR-CD	2	September 12, 2014 00:01
Moving mesh	Niklas Wikstrom (Wikstrom)	OpenFOAM Running, Solving & CFD	122	June 15, 2014 06:20

February 15, 2017, 07:11		#2
KevinZ09 Senior Member Kevin Join Date: Dec 2016 Posts: 138 Rep Power: 9	You can look up the details in the Fluent manual as to how the variable time step is precisely used/calculated, but as you can understand, for fixed time stepping the timestep is a fixed number as specified by you, while for the variable time stepping method it's variable (just as with the adaptive time stepping methond) and the value then depends sort of on the local CFL number of the interface.

February 15, 2017, 08:11		#4
KevinZ09 Senior Member Kevin Join Date: Dec 2016 Posts: 138 Rep Power: 9	Well, if your dt = 0.007s and you want your simulation to run for 24 seconds physical time, then run it for 24/0.007 = 3429 timesteps.

February 15, 2017, 08:19		#5
luca007 Member kaouachi anouar Join Date: Jul 2016 Posts: 66 Rep Power: 9	ok this is for fixed method ?

February 15, 2017, 08:20		#6
KevinZ09 Senior Member Kevin Join Date: Dec 2016 Posts: 138 Rep Power: 9	Yes Sent from my GT-I9505 using CFD Online Forum mobile app

February 15, 2017, 08:42		#7
luca007 Member kaouachi anouar Join Date: Jul 2016 Posts: 66 Rep Power: 9	ok please what's better fixed method or variable method

February 15, 2017, 09:14		#8
LuckyTran Senior Member Lucky Join Date: Apr 2011 Location: Orlando, FL USA Posts: 5,667 Rep Power: 65	Adaptive time-stepping allows Fluent to globally and sometimes locally adjust the time-step to satisfy Courant number criteria. You end up with solutions are odd times though and for post-processing that can be a pain. However, if you don't need constant time-step size between intervals, such as in transient problems, then you can consider it. If you want to do something like fft at any point, you realize fft algorithm only works if time-step is constant. I recommend to just start with fixed time stepping method to get your feet wet. It's simple, straightforward, and you don't wonder what is going on under the hood. After you are familiar with Fluent, then you can look into adaptive time-steps. To be honest, I've never used adaptive time-stepping, but I can see how useful it might be to some. It's extremely useful in problems where you have large disparity between flow velocities in different regions.