Time Increment Significance in Steady state FEM CFD solvers?

Abhya · October 14, 2015, 04:26

Hi,
I'm using AcuSolve which is a FEM based solver

In AcuSolve steady state simulations are solved as pseudo transient.
Recently I have noticed that the 'Initial Time increment' in steady state problems affects the convergence behavior of a problem I'm dealing with largely

I want to know what exactly is this Inital time increment (for steady state escp. what role it plays), as currently my understanding says that the time increment shall not affect the steady state convergence in anyway

Also, is the CFX too a FEM based Solver like AcuSolve?
(I don't understand what "CFX is an element based Finite Volume Method" means, I know what a FVM solver is)

mprinkey · October 14, 2015, 20:07

I don't know anything about the solver you mention, but it is not uncommon for steady state solvers (or even unsteady solvers) to use pseudo-transient terms as a relaxation method to reach a converged solution. Many (most?) density-based CFD codes use pseudo-timestepping in some form. The pseudo-transient approach is useful, because it works with any nonlinear system, even if the solution has discontinuities, etc. You can always introduce a pseudo-time variable and march the solution forward in pseudo-time...and as pseudo-time approaches infinity, the transient term will vanish and you will be left with the solution of the original system. But, just like real time stepping, pseudo-timestepping is subject to CFL stability limits. You want to choose a large timestep to get convergence, but not so large that the iterations are unstable.

October 14, 2015, 04:26	Time Increment Significance in Steady state FEM CFD solvers?	#1
Abhya Member Join Date: Jul 2012 Posts: 48 Rep Power: 13	Hi, I'm using AcuSolve which is a FEM based solver In AcuSolve steady state simulations are solved as pseudo transient. Recently I have noticed that the 'Initial Time increment' in steady state problems affects the convergence behavior of a problem I'm dealing with largely I want to know what exactly is this Inital time increment (for steady state escp. what role it plays), as currently my understanding says that the time increment shall not affect the steady state convergence in anyway Also, is the CFX too a FEM based Solver like AcuSolve? (I don't understand what "CFX is an element based Finite Volume Method" means, I know what a FVM solver is) AlexRonto likes this.

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
High Courant Number @ icoFoam	Artex85	OpenFOAM Running, Solving & CFD	11	February 16, 2017 13:40
same geometry,structured and unstructured mesh,different behaviour.	sharonyue	OpenFOAM Running, Solving & CFD	13	January 2, 2013 22:40
time step in steady state	naveen_010	Main CFD Forum	3	February 3, 2011 19:49
time step and iterations in steady state problem using transient solver	jing113cn	FLUENT	2	January 15, 2010 03:18
Where do we go from here? CFD in 2001	John C. Chien	Main CFD Forum	36	January 24, 2001 21:10

October 14, 2015, 20:07		#2
mprinkey Senior Member Michael Prinkey Join Date: Mar 2009 Location: Pittsburgh PA Posts: 363 Rep Power: 25	I don't know anything about the solver you mention, but it is not uncommon for steady state solvers (or even unsteady solvers) to use pseudo-transient terms as a relaxation method to reach a converged solution. Many (most?) density-based CFD codes use pseudo-timestepping in some form. The pseudo-transient approach is useful, because it works with any nonlinear system, even if the solution has discontinuities, etc. You can always introduce a pseudo-time variable and march the solution forward in pseudo-time...and as pseudo-time approaches infinity, the transient term will vanish and you will be left with the solution of the original system. But, just like real time stepping, pseudo-timestepping is subject to CFL stability limits. You want to choose a large timestep to get convergence, but not so large that the iterations are unstable.