Swirl ratio failure in remeshing in CFX

a_cucen · May 4, 2012, 11:27

Dear all,

I want to run an internal combustion engine simulation with CFX. I have connected CFX and ICEM-CFD. I have made some analysis so far but after most of the remeshing sections in ICEM-CFD, swirl ratio changes. I have attached a pictures shows this situation. I have tried different mesh sizes, but result is the same. How can I fix this situation?

Regards

ghorrocks · May 5, 2012, 08:12

The fix is easy - use the IC engine modelling wizard in workbench

It uses Fluent as the solver as it has much more IC engine stuff in it that CFX does.

I have no idea what your remeshing problem is. Save a results file before and after a remesh and see if you can spot the problem.

a_cucen · May 5, 2012, 08:29

I have found an extension package that you can use ICE very effectively but this failure happened. The tutorials work fine but when I prepare my simulation like tutorial, it causes this error.

ghorrocks · May 5, 2012, 08:31

Have you looked at the flow field immediately before and after a remesh? Do they look the same?

a_cucen · May 5, 2012, 11:05

Yes they look like the same. This failure happens in averadge of pressure in domain. Swirl ratio was one of an example.

ghorrocks · May 6, 2012, 07:46

If they look the same then why is the swirl ratio different?

Is the thing which is not carried over smoothly the pressure? That could affect swirl ratio through the density.

a_cucen · May 6, 2012, 08:53

Yes after remesh phase pressure could not been carried over smoothly. It changes just like swirl ratio, so physics of this case fails I think. and It s not depend on mesh sizes or time step sizes. I have not found a solution yet.

ghorrocks · May 6, 2012, 21:11

So the problem is the pressure is not being carried over smoothly in remeshes, and the swirl ratio change is just a symptom of that.

a_cucen · May 7, 2012, 04:02

I don' t know. And I haven' t found any solution yet. Really lost in here. Do you have an idea about this situation?

ghorrocks · May 7, 2012, 07:59

Other than saying the problem seems to be in mapping the pressure over, no. You will have to do some investigation and try to track it down. Contact ANSYS support if you need to.

a_cucen · May 7, 2012, 15:11

Dear Glenn,

Maybe I can change cfx to Fluent but IC engine modelling wizard in workbench is only modelling cold flow, not combustion. How can I model combustion in this wizard?

ghorrocks · May 7, 2012, 18:13

I have never used the wizard so do not know, but I am sure you will be able to include combustion in the model. Fluent has many combustion models.

Shahr · March 1, 2013, 05:13

I am sorry that I resurrect this thread. I think I have same problem with pressure field after remeshing. My simulation is not an ICE one though.

My question has been answered previously in here. But I don't expect to have same problem after every single remeshing step.

The velocity field is nicely transferred to the new mesh; The pressure field however has weird mismatching behaviour. Has anybody found a solution for the problem?

ghorrocks · March 3, 2013, 06:08

The thread you quote is talking about initial conditions. This thread is very different - it is about mapping the ersults over to a new mesh and continuing a simulation. This appears to be some problem with the interpolator.

Shahr · March 4, 2013, 12:11

Thank you for reminding, Mr. Horrocks. I am aware that my previous question was about initial conditions, but if I run the simulation long enough, the pressure will be fixed after some time steps. However, every time remeshing takes place, interpolation (remapping) of pressure field is done wrong, and the same behaviour repeats. This, I believe is similar to the problem mentioned in this thread. You can see this plot from my monitors:

ghorrocks · March 4, 2013, 17:28

You will have to do a bit of research your self on this one. Have a look at the pressure field immediately before and after a remesh. Do you see any differences?

March 4, 2013, 12:11		#15
Shahr New Member Join Date: Feb 2013 Posts: 21 Rep Power: 13	Thank you for reminding, Mr. Horrocks. I am aware that my previous question was about initial conditions, but if I run the simulation long enough, the pressure will be fixed after some time steps. However, every time remeshing takes place, interpolation (remapping) of pressure field is done wrong, and the same behaviour repeats. This, I believe is similar to the problem mentioned in this thread. You can see this plot from my monitors: Last edited by Shahr; March 4, 2013 at 12:32.

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
Quantifying the swirl in CFX, Swirl Number/Swirl Angle	tauqirnawaz	CFX	23	July 24, 2016 23:28
CFX Wind turbine mesh trouble and solver failure	tallliam	CFX	9	April 29, 2012 23:55
CFX - Solidworks Flow, Impeller comparison	aerospace_guy_	CFX	8	March 20, 2012 23:06
Pros and Cons for CFX, CFdesign, COMSOL	Val	Main CFD Forum	3	June 10, 2011 03:20
CFX 4.4 installation problem	Pandu Sattvika	CFX	1	December 1, 2001 05:07

May 5, 2012, 08:12		#2
ghorrocks Super Moderator Glenn Horrocks Join Date: Mar 2009 Location: Sydney, Australia Posts: 17,852 Rep Power: 144	The fix is easy - use the IC engine modelling wizard in workbench It uses Fluent as the solver as it has much more IC engine stuff in it that CFX does. I have no idea what your remeshing problem is. Save a results file before and after a remesh and see if you can spot the problem.

May 5, 2012, 08:29		#3
a_cucen Member Ahmet Join Date: Mar 2011 Location: Ankara/ Turkey Posts: 37 Rep Power: 15	I have found an extension package that you can use ICE very effectively but this failure happened. The tutorials work fine but when I prepare my simulation like tutorial, it causes this error.

May 5, 2012, 08:31		#4
ghorrocks Super Moderator Glenn Horrocks Join Date: Mar 2009 Location: Sydney, Australia Posts: 17,852 Rep Power: 144	Have you looked at the flow field immediately before and after a remesh? Do they look the same?

May 5, 2012, 11:05		#5
a_cucen Member Ahmet Join Date: Mar 2011 Location: Ankara/ Turkey Posts: 37 Rep Power: 15	Yes they look like the same. This failure happens in averadge of pressure in domain. Swirl ratio was one of an example.

May 6, 2012, 07:46		#6
ghorrocks Super Moderator Glenn Horrocks Join Date: Mar 2009 Location: Sydney, Australia Posts: 17,852 Rep Power: 144	If they look the same then why is the swirl ratio different? Is the thing which is not carried over smoothly the pressure? That could affect swirl ratio through the density.

May 6, 2012, 08:53		#7
a_cucen Member Ahmet Join Date: Mar 2011 Location: Ankara/ Turkey Posts: 37 Rep Power: 15	Yes after remesh phase pressure could not been carried over smoothly. It changes just like swirl ratio, so physics of this case fails I think. and It s not depend on mesh sizes or time step sizes. I have not found a solution yet.

May 6, 2012, 21:11		#8
ghorrocks Super Moderator Glenn Horrocks Join Date: Mar 2009 Location: Sydney, Australia Posts: 17,852 Rep Power: 144	So the problem is the pressure is not being carried over smoothly in remeshes, and the swirl ratio change is just a symptom of that.

May 7, 2012, 04:02		#9
a_cucen Member Ahmet Join Date: Mar 2011 Location: Ankara/ Turkey Posts: 37 Rep Power: 15	I don' t know. And I haven' t found any solution yet. Really lost in here. Do you have an idea about this situation?

May 7, 2012, 07:59		#10
ghorrocks Super Moderator Glenn Horrocks Join Date: Mar 2009 Location: Sydney, Australia Posts: 17,852 Rep Power: 144	Other than saying the problem seems to be in mapping the pressure over, no. You will have to do some investigation and try to track it down. Contact ANSYS support if you need to.

May 7, 2012, 15:11		#11
a_cucen Member Ahmet Join Date: Mar 2011 Location: Ankara/ Turkey Posts: 37 Rep Power: 15	Dear Glenn, Maybe I can change cfx to Fluent but IC engine modelling wizard in workbench is only modelling cold flow, not combustion. How can I model combustion in this wizard?

May 7, 2012, 18:13		#12
ghorrocks Super Moderator Glenn Horrocks Join Date: Mar 2009 Location: Sydney, Australia Posts: 17,852 Rep Power: 144	I have never used the wizard so do not know, but I am sure you will be able to include combustion in the model. Fluent has many combustion models.

March 1, 2013, 05:13		#13
Shahr New Member Join Date: Feb 2013 Posts: 21 Rep Power: 13	I am sorry that I resurrect this thread. I think I have same problem with pressure field after remeshing. My simulation is not an ICE one though. My question has been answered previously in here. But I don't expect to have same problem after every single remeshing step. The velocity field is nicely transferred to the new mesh; The pressure field however has weird mismatching behaviour. Has anybody found a solution for the problem?

March 3, 2013, 06:08		#14
ghorrocks Super Moderator Glenn Horrocks Join Date: Mar 2009 Location: Sydney, Australia Posts: 17,852 Rep Power: 144	The thread you quote is talking about initial conditions. This thread is very different - it is about mapping the ersults over to a new mesh and continuing a simulation. This appears to be some problem with the interpolator.

March 4, 2013, 17:28		#16
ghorrocks Super Moderator Glenn Horrocks Join Date: Mar 2009 Location: Sydney, Australia Posts: 17,852 Rep Power: 144	You will have to do a bit of research your self on this one. Have a look at the pressure field immediately before and after a remesh. Do you see any differences?