Time marching in incomprerssible flow

amazdeh · May 25, 2011, 13:47

There was a discussion a while back about how CFX treats unsteady incompressible Navier-Stokes equations.
The question was that as the Time marching methods cannot be applied directly to incompressible flows because the governing equations are not hyperbolic so whether CFX uses an artificial compressibility method in order to treat the inviscid fluxes using the traditional characteristics-based schemes.
I have searched CFX manual I had no luck finding any info regarding time marching for incompressible flow
I wonder if u can shed some light on this, I am interested in mathematics because if we know how this is being dealt with CFX then we can make better judgment about our results
Thanks
Al

ghorrocks · May 25, 2011, 19:04

CFX uses backward eular scheme for the transient term, either first or second order. You can march this out in time just fine. CFX does not use artificial compressibility.

This is all described in the theory documentation.

amazdeh · May 26, 2011, 10:07

GHORROCKS,
I know u from your previous posts, that u have a vast knowledge of the tool-so please help me understand this

u are saying that the solution is not done differently in case of compressible and incompressible

how the time marching is applied on a continiuity equation that has only space variables-

also if u can please point me to the some reference that CFX uses for its solver

in other posts, I saw that u mentioned pressure and velocity are fully coupled in CFX is that why there is no need for artificail compressibility-

If so please tell me how pressure and velocity are coupled for the incompressible case-thanks

ghorrocks · May 26, 2011, 17:22

Obviously there are differences when compressibility is activated, but that is just to allow the density to vary. See the doco for details.

Everything you are asking is discussed in the CFX Theory manual.

amazdeh · June 24, 2011, 15:20

Quote:

Originally Posted by ghorrocks

Obviously there are differences when compressibility is activated, but that is just to allow the density to vary. See the doco for details.

Everything you are asking is discussed in the CFX Theory manual.

Hi Glen,

I have spent a month reviewing literature and cfx manual along with fluent documentation-

Finally I know some basics from all of these

I can tell that CFX uses pressure-based coupled solvers but I can not find much about the math on any of these manual-fluent offers more info

Now I like to go deeper and would like to know some kind of clasical ref where CFX was developed based upon -

to be honest with in my search I came up with so many articles on this topic but I like to get closer to the one that CFX uses

this is very critical for me since my dissertation will not go forward if I can not answer question on CFX solver theory

ghorrocks · June 25, 2011, 06:53

The CFX solver is not as well documented as Fluent. Also, as it is a little unusual it is hard to find information about how it works. The documentation is pretty good and the references it links to are important too. Also the CFX solver is based on one developed for an old CFD code called TASCFlow. You might find more information on documentation for tascflow.

May 25, 2011, 13:47	Time marching in incomprerssible flow	#1
amazdeh New Member al mazdeh Join Date: Mar 2009 Posts: 16 Rep Power: 17	There was a discussion a while back about how CFX treats unsteady incompressible Navier-Stokes equations. The question was that as the Time marching methods cannot be applied directly to incompressible flows because the governing equations are not hyperbolic so whether CFX uses an artificial compressibility method in order to treat the inviscid fluxes using the traditional characteristics-based schemes. I have searched CFX manual I had no luck finding any info regarding time marching for incompressible flow I wonder if u can shed some light on this, I am interested in mathematics because if we know how this is being dealt with CFX then we can make better judgment about our results Thanks Al

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
Physical Reason for stability of Implicit Schemes?	radhakrishnan	Main CFD Forum	26	October 3, 2023 22:05
Transient simulation not converging	skabilan	OpenFOAM Running, Solving & CFD	14	December 16, 2019 23:12
Extrusion with OpenFoam problem No. Iterations 0	Lord Kelvin	OpenFOAM Running, Solving & CFD	8	March 28, 2016 11:08
flow time plot units	alex	FLUENT	0	July 2, 2007 13:38
AMG versus ICCG	msrinath80	OpenFOAM Running, Solving & CFD	2	November 7, 2006 15:15

May 25, 2011, 19:04		#2
ghorrocks Super Moderator Glenn Horrocks Join Date: Mar 2009 Location: Sydney, Australia Posts: 17,703 Rep Power: 143	CFX uses backward eular scheme for the transient term, either first or second order. You can march this out in time just fine. CFX does not use artificial compressibility. This is all described in the theory documentation.

May 26, 2011, 10:07		#3
amazdeh New Member al mazdeh Join Date: Mar 2009 Posts: 16 Rep Power: 17	GHORROCKS, I know u from your previous posts, that u have a vast knowledge of the tool-so please help me understand this u are saying that the solution is not done differently in case of compressible and incompressible how the time marching is applied on a continiuity equation that has only space variables- also if u can please point me to the some reference that CFX uses for its solver in other posts, I saw that u mentioned pressure and velocity are fully coupled in CFX is that why there is no need for artificail compressibility- If so please tell me how pressure and velocity are coupled for the incompressible case-thanks

May 26, 2011, 17:22		#4
ghorrocks Super Moderator Glenn Horrocks Join Date: Mar 2009 Location: Sydney, Australia Posts: 17,703 Rep Power: 143	Obviously there are differences when compressibility is activated, but that is just to allow the density to vary. See the doco for details. Everything you are asking is discussed in the CFX Theory manual.

June 25, 2011, 06:53		#6
ghorrocks Super Moderator Glenn Horrocks Join Date: Mar 2009 Location: Sydney, Australia Posts: 17,703 Rep Power: 143	The CFX solver is not as well documented as Fluent. Also, as it is a little unusual it is hard to find information about how it works. The documentation is pretty good and the references it links to are important too. Also the CFX solver is based on one developed for an old CFD code called TASCFlow. You might find more information on documentation for tascflow.