different CFX Pre and Post mesh region

mactech001 · April 9, 2010, 07:09

Hi all,

i was performing a CHT analysis on a simple rectangular model of about 30m long with a rectangular hollow hole where fluid flows.

In the Pre, mesh region for solid only consist of 1 body mesh region. But when i check the results in Post, mesh region for the solid became 3 parts.

Why has the mesh region in Post divided into 3 parts please?

Look forward to hear of any comments.

ghorrocks · April 9, 2010, 08:34

CFX-Pre does this when the mesh has a non-contiguous section.

mactech001 · April 9, 2010, 08:54

Hi again Glenn, thanks for your reply.

Quote:

Originally Posted by ghorrocks

CFX-Pre does this when the mesh has a non-contiguous section.

Is a non-contiguous section desirable please? if not, how should i get rid of this section? in meshing?

ghorrocks · April 9, 2010, 09:26

Yes, you need to remesh.

mactech001 · April 10, 2010, 05:57

Hi Glenn, thanks for your reply.

is there any parameter that i can check before solving so that i can know i need to remesh the model beforehand please?

if the solid in Post has mesh regions divided into 3 parts, is the solid the only part i need to remesh?

ghorrocks · April 10, 2010, 07:32

If you meshed in WB then did you make the body a multi-body part? If you forget to do this you will get non-contiguous meshes.

Also you can join non-contiguous meshes with GGI interfaces. But only do this if it is not possible to make the mesh contiguous.

mactech001 · April 11, 2010, 00:32

Hi Glenn,

Quote:

Originally Posted by ghorrocks

If you meshed in WB then did you make the body a multi-body part? If you forget to do this you will get non-contiguous meshes.

Also you can join non-contiguous meshes with GGI interfaces. But only do this if it is not possible to make the mesh contiguous.

1) i've attached a picture 'Pre_PostMshRegProb.jpg". The Mesh Region in Post is shown to have the body divided into 4 parts, namely B1510, B1510 1, B1510 2, B1510 3. This is non-contiguous?

in WB and reality, this part is just ONE single part, not composed by many parts.

2) What causes non-contiguous section please?

3) I've used GGI for the interface between fluid and solid, but even the fluid body is divided into 4 parts, namely B1590, B1590 1, B1590 2, B1590 3. Can non-contiguous sections influence my results, even if the solver has shown convergence?

ghorrocks · April 11, 2010, 07:54

As I said, usually these non-contiguous meshes are a meshing error and a common cause of this error is a domain consisting of several parts which have not been joined into a multi-body part in DesignModeller. This is all covered in the tutorials so have a look there for further details.

Unless you have put GGI interfaces in the non-contiguous interfaces then you will have spurious walls in the middle of your domain. These are very bad and will mean your simulation is totally wrong. Have a look in CFD-Post at the flows near these regions and see what I mean.

mactech001 · April 11, 2010, 21:58

Dear Glenn,

thanks again for your reply.

i went back to check on the Heating Coil example which i followed. I discovered that, the FluidZone Mesh Region is non-contiguous as well, please see attached(non-contiguous-mesh.jpg).

Quote:

As I said, usually these non-contiguous meshes are a meshing error and a common cause of this error is a domain consisting of several parts which have not been joined into a multi-body part in DesignModeller. This is all covered in the tutorials so have a look there for further details

Would you mind directing me to the tutorial you were referring to above please?

Furthermore, you mentioned that causes of non-contiguous section is because several parts have not been joined into a multi-body part in DM. Attached in 'parts into multibody.jpg' is anothe model (not the Heating coil example), i've 4 parts of 50bodies, does it mean i have to put all 50 Bodies into ONE part please? This however, will make defining materials to the bodies less convenient in Simulation.

Look forward to your comments again.

ghorrocks · April 11, 2010, 22:08

I can't explain graphical stuff like this over a forum. Make sure you do the tutorial examples on this sort of stuff. Also be aware that there are lots of good examples on multi body meshes on the CFX Community page. You will just have to look at the mesh regions and determine whether there is a problem or not. Sometimes a mesh can be split into several regions, but if they are joined by contiguous meshes they can be combined together in CFX-Pre.

April 9, 2010, 07:09	different CFX Pre and Post mesh region	#1
mactech001 Senior Member Join Date: Nov 2009 Posts: 125 Rep Power: 17	Hi all, i was performing a CHT analysis on a simple rectangular model of about 30m long with a rectangular hollow hole where fluid flows. In the Pre, mesh region for solid only consist of 1 body mesh region. But when i check the results in Post, mesh region for the solid became 3 parts. Why has the mesh region in Post divided into 3 parts please? Look forward to hear of any comments. __________________ Thank you for your kind attention. Kind regards, mactech001 Currently using: ANSYS v13

April 10, 2010, 05:57		#5
mactech001 Senior Member Join Date: Nov 2009 Posts: 125 Rep Power: 17	Hi Glenn, thanks for your reply. is there any parameter that i can check before solving so that i can know i need to remesh the model beforehand please? if the solid in Post has mesh regions divided into 3 parts, is the solid the only part i need to remesh? __________________ Thank you for your kind attention. Kind regards, mactech001 Currently using: ANSYS v13

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
ATTENTION! Reliability problems in CFX 5.7	Joseph	CFX	14	April 20, 2010 16:45
No display on CFX Pre and Post	Dipl.-Ing. Jayadi Lukito	CFX	0	May 23, 2007 10:30
(CFX POST) Rotational Speed	ARJUN	CFX	9	July 13, 2006 21:38
CFX 10 -pre and -post problams on Linux	Henry Liu	CFX	3	June 22, 2006 07:57
CFX 5.7.1 PRE and solver won't start	daniel	CFX	1	January 20, 2006 11:09

April 9, 2010, 08:34		#2
ghorrocks Super Moderator Glenn Horrocks Join Date: Mar 2009 Location: Sydney, Australia Posts: 17,854 Rep Power: 144	CFX-Pre does this when the mesh has a non-contiguous section.

April 9, 2010, 09:26		#4
ghorrocks Super Moderator Glenn Horrocks Join Date: Mar 2009 Location: Sydney, Australia Posts: 17,854 Rep Power: 144	Yes, you need to remesh.

April 10, 2010, 07:32		#6
ghorrocks Super Moderator Glenn Horrocks Join Date: Mar 2009 Location: Sydney, Australia Posts: 17,854 Rep Power: 144	If you meshed in WB then did you make the body a multi-body part? If you forget to do this you will get non-contiguous meshes. Also you can join non-contiguous meshes with GGI interfaces. But only do this if it is not possible to make the mesh contiguous.

April 11, 2010, 07:54		#8
ghorrocks Super Moderator Glenn Horrocks Join Date: Mar 2009 Location: Sydney, Australia Posts: 17,854 Rep Power: 144	As I said, usually these non-contiguous meshes are a meshing error and a common cause of this error is a domain consisting of several parts which have not been joined into a multi-body part in DesignModeller. This is all covered in the tutorials so have a look there for further details. Unless you have put GGI interfaces in the non-contiguous interfaces then you will have spurious walls in the middle of your domain. These are very bad and will mean your simulation is totally wrong. Have a look in CFD-Post at the flows near these regions and see what I mean.

April 11, 2010, 22:08		#10
ghorrocks Super Moderator Glenn Horrocks Join Date: Mar 2009 Location: Sydney, Australia Posts: 17,854 Rep Power: 144	I can't explain graphical stuff like this over a forum. Make sure you do the tutorial examples on this sort of stuff. Also be aware that there are lots of good examples on multi body meshes on the CFX Community page. You will just have to look at the mesh regions and determine whether there is a problem or not. Sometimes a mesh can be split into several regions, but if they are joined by contiguous meshes they can be combined together in CFX-Pre.