CFD Online Logo CFD Online URL
www.cfd-online.com
[Sponsors]
Home > Forums > Mesh Generation & Pre-Processing Software > Pointwise & Gridgen

Joining Structured and unstructured domains

Register Blogs Community New Posts Updated Threads Search

Like Tree3Likes
  • 1 Post By saitej09
  • 1 Post By tcarrigan
  • 1 Post By dgarlisch

Reply
 
LinkBack Thread Tools Search this Thread Display Modes
Old   October 10, 2014, 10:05
Default Joining Structured and unstructured domains
  #1
New Member
 
Saiteja
Join Date: Aug 2014
Posts: 13
Rep Power: 11
saitej09 is on a distinguished road
Hi

I am new to pointwise and I was trying to use a hybrid mesh with structured layers near the wall. I have created both the domains but am unable to join them and when I am giving BC's, it is creating a problem as the interface between the is remaining unspecified. How should I join the domains?
I have attached the images so that the problem can be better understood.
Image1
Image2
Vino likes this.
saitej09 is offline   Reply With Quote

Old   October 10, 2014, 12:03
Default
  #2
Senior Member
 
Travis Carrigan
Join Date: Jul 2010
Location: Arlington, TX
Posts: 161
Rep Power: 15
tcarrigan is on a distinguished road
Hello Saiteja,

You cannot join a structured an unstructured domain in Pointwise. Currently, Pointwise does not support a "hybrid" domain. However, when you specify the boundary conditions in the CAE menu, simply do not specify a boundary condition at the interface between the structured and unstructured domain. By not specifying a boundary condition, the interface will be treated as just internal topology and a single block will be exported.

Regards,
Travis
__________________
Travis Carrigan
Manager, Business Development
Pointwise, Inc.
tcarrigan is offline   Reply With Quote

Old   October 10, 2014, 12:55
Default
  #3
New Member
 
Saiteja
Join Date: Aug 2014
Posts: 13
Rep Power: 11
saitej09 is on a distinguished road
Hello Travis,
Thanks for the prompt reply.I have exported it to SU2 format and when I run the config file, it is saying "The file does not have definition of Marker Unspecified " since I haven't specified any marker there. It is not taking it as an internal topology as you have specified!!
saitej09 is offline   Reply With Quote

Old   October 10, 2014, 17:50
Default
  #4
Senior Member
 
Travis Carrigan
Join Date: Jul 2010
Location: Arlington, TX
Posts: 161
Rep Power: 15
tcarrigan is on a distinguished road
Please post a picture of the CAE, Set Boundary Condition panel in Pointwise with all the BCs highlighted.
__________________
Travis Carrigan
Manager, Business Development
Pointwise, Inc.
tcarrigan is offline   Reply With Quote

Old   October 11, 2014, 00:45
Default
  #5
New Member
 
Saiteja
Join Date: Aug 2014
Posts: 13
Rep Power: 11
saitej09 is on a distinguished road
Hello

I have attached the images and also the pointwise .pw file.

Image1
Image2
Project_file
saitej09 is offline   Reply With Quote

Old   October 11, 2014, 19:16
Default
  #6
Senior Member
 
Travis Carrigan
Join Date: Jul 2010
Location: Arlington, TX
Posts: 161
Rep Power: 15
tcarrigan is on a distinguished road
It looks like there are duplicate domains at the interface. This means that they are in fact boundaries and not interior domains. You'll need to eliminate one set of domains at the interface.

In the end you should only have two structured domains at the interface. Because they will be shared by both the hex block and the prism block, they will be considered connections and no BCs should be applied.
haitham osman likes this.
__________________
Travis Carrigan
Manager, Business Development
Pointwise, Inc.
tcarrigan is offline   Reply With Quote

Old   September 19, 2016, 07:46
Default Dear Tcarrigan
  #7
Member
 
Join Date: Jul 2014
Posts: 31
Rep Power: 11
nba1942 is on a distinguished road
Quote:
Originally Posted by tcarrigan View Post
Hello Saiteja,

You cannot join a structured an unstructured domain in Pointwise. Currently, Pointwise does not support a "hybrid" domain. However, when you specify the boundary conditions in the CAE menu, simply do not specify a boundary condition at the interface between the structured and unstructured domain. By not specifying a boundary condition, the interface will be treated as just internal topology and a single block will be exported.

Regards,
Travis
Hi, I have read the message written by you few years ago.

You said pointwise would export the grid as a single block, if I simply do not specify a BC at the interface.

I made a hybrid grid(hex+tetra), which consists of two tetrahedral blocks and a tetrahedral block.

I found out that pointwise treats a block as a zone in a CGNS format.

According to your post, the exported grid supposed to have only a single zone.

But, the exported file has 3 zones in it.

And the problem is, if I sum numbers of points of each blocks, the result does not match to the number from "Grid - Cell Count - Total Points".

It seems to me that each zone(block) has duplicated points at an interface.

I check there are no duplicated domain at an interface.

------------------------------------------------------------------------------------------------------------------------------------------

My solver uses tecplot format.

So I read all the information from the exported cgns file, and rewrite it in tecplot format.

e.g.)

TITLE = "unstructured grid"
VARIABLE = "X", "y", "z"
ZONE NODES = 49740 , ELEMENTS = 46400 , DATAPACKING=POINT, ZONETYPE=FEBRICK
-1.2474613 1.52764907E-16 4.32800087E-18
-1.7662957 1.8516569 -1.13377544E-16
-1.2512790 1.48967407E-16 -6.96554333E-02
-1.2512790 1.57497452E-16 6.96554333E-02
-1.4101814 1.72691714E-16 -1.07852077E-32
-1.9290158 1.8516569 -1.17705538E-16
-1.4139991 1.68894214E-16 -6.96554333E-02
-1.4139991 1.77424259E-16 6.96554333E-02
.... <- x, y, z
10441 10442 10466 10465 11953 11954 11978 11977
10442 10443 10467 10466 11954 11955 11979 11978
10443 10444 10468 10467 11955 11956 11980 11979
10444 10445 10469 10468 11956 11957 11981 11980
10445 10446 10470 10469 11957 11958 11982 11981
10446 10447 10471 10470 11958 11959 11983 11982
10447 10448 10472 10471 11959 11960 11984 11983
10448 10449 10473 10472 11960 11961 11985 11984
10449 10450 10474 10473 11961 11962 11986 11985
10450 10451 10475 10474 11962 11963 11987 11986
10451 10452 10476 10475 11963
....... <- connectivity of cell

As you might have noticed, there should not be any duplicated point in each zone for my solver.

Is something wrong with my grid? Or if it is not, why does the sum of the number of points of each zone not match to the number of total points?

I am waiting for your reply.

Thank you.

Kim.
nba1942 is offline   Reply With Quote

Old   September 20, 2016, 11:34
Default
  #8
Senior Member
 
David Garlisch
Join Date: Jan 2013
Location: Fidelity Pointwise, Cadence Design Systems (Fort Worth, Texas Office)
Posts: 307
Rep Power: 14
dgarlisch is on a distinguished road
Please read this. It is very important.

CGNS is a multi-block format. Each block in PW is written as a single zone (Pointwise does not agglomerate blocks at export).

Each CGNS zone (block) has its own set of points. Hence, the points at block-to-block interfaces will be duplicated.

CGNS contains additional data nodes that define how the blocks connect to each other. A valid CGNS reader will be able to handle the interfaces and eliminate the duplicates.

You stated:
Quote:
My solver uses tecplot format.

So I read all the information from the exported cgns file, and rewrite it in tecplot format.
Why are you doing this? You can export your grid from Pointwise directly to a Tecplot file.
  1. Set the active solver to Tecplot using CAE, Select Solver....
  2. Assign all the BCs and VCs you need using CAE, Set Boundary Conditions... and CAE, Set Volume Conditions...
  3. Export the grid using File, Export, CAE....

Good luck.
nba1942 likes this.
dgarlisch is offline   Reply With Quote

Old   September 21, 2016, 10:17
Default
  #9
Member
 
Join Date: Jul 2014
Posts: 31
Rep Power: 11
nba1942 is on a distinguished road
I realized I revived very old thread. I apologize for that.

Thanks for your reply


I found the interface points list you mentioned from the grid file. I think I can eliminate duplicated points with this


My solver uses the data structure of tecplot format. But still, the solver needs to import cgns format which is one of the most widely-used formats. So preprocessing is needed.


Anyway thanks again. And next time I will open my own thread rather than reviving old threads!
nba1942 is offline   Reply With Quote

Old   September 23, 2016, 12:12
Default
  #10
Senior Member
 
David Garlisch
Join Date: Jan 2013
Location: Fidelity Pointwise, Cadence Design Systems (Fort Worth, Texas Office)
Posts: 307
Rep Power: 14
dgarlisch is on a distinguished road
It may help you to look at Pointwise CAE Plugins.

With a plugin, you would be able to export the data from Pointwise exactly how your solver needs it.

Good luck.
dgarlisch is offline   Reply With Quote

Reply


Posting Rules
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is On
HTML code is Off
Trackbacks are Off
Pingbacks are On
Refbacks are On


Similar Threads
Thread Thread Starter Forum Replies Last Post
Volume mesh for Cyclone separator Rajan Pointwise & Gridgen 17 May 4, 2014 05:38
How cell skew affects solution in structured vs unstructured grids lordvon Main CFD Forum 1 July 3, 2013 06:19
Structured or unstructured meshing of a STL file (surface) Milad06 Pointwise & Gridgen 4 April 25, 2013 11:55
Structured and Unstructured mesh Jingwei FLUENT 0 March 2, 2009 21:29
combining structured and unstructured grids Shane FLUENT 2 January 23, 2007 20:57


All times are GMT -4. The time now is 11:47.