[GiDlist] GiDlist Digest, Vol 30, Issue 2

[He Jinzhi]

Thank you very much. I got it .








At 2013-01-11 19:00:02,gidlist-request at listas.cimne.upc.edu wrote:
Send GiDlist mailing list submissions to
gidlist at listas.cimne.upc.edu

To subscribe or unsubscribe via the World Wide Web, visit
http://listas.cimne.upc.edu/cgi-bin/mailman/listinfo/gidlist
or, via email, send a message with subject or body 'help' to
gidlist-request at listas.cimne.upc.edu

You can reach the person managing the list at
gidlist-owner at listas.cimne.upc.edu

When replying, please edit your Subject line so it is more specific
than "Re: Contents of GiDlist digest..."


Today's Topics:

1. Re: about cartesian grid generation (Enrique Escolano)


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

Message: 1
Date: Thu, 10 Jan 2013 13:00:42 +0100
From: "Enrique Escolano" escolano at cimne.upc.edu
Subject: Re: [GiDlist] about cartesian grid generation
To: gidlist at listas.cimne.upc.edu
Message-ID: 4DF7F4BC3A5A466885A0468D8536A997 at mediador
Content-Type: text/plain; charset="gb2312"

Hi Jinzhi,

Now in GiD there are two options to generate the cartesian volume grid: with spacing uniform or variable. (can set it in Utilities-Preferences... Meshing)

1- Grid spacing: Uniform

All cell sizes are equal for all cell in all directions. The cell size will be exactly the one specified by the user as general mesh size.
The grid will have n x m x l cells of this exact size, the number of cells on each direction will be greater or equal to the bounding box of the model, and the corner of the grid will be set centering the bounding box of the model inside the grid, then typically the grid mesh will be a little bigger that the real model

2- Grid spacing: Variable

This is the option that you are using.

The cell sizes could be non-uniform along each direction, and also diffenent in x, y and z directions.
the bounding box of each volume is considered separatelly, and its planes are forced to be grid thicks. The number of cells between is interval depends on the specific size assigned to the entity (if not zero) or the general size. If automatic correct sizes is set (usual) then grid thicks are modified to force a size ratio smaller that the one set in preferences (3.0 default) and the distance between thicks bigger that a minimum (1e-8 default).

In your case, with probably only one volume, the corners or the grid are set by the bounding box of your volume, to obtain a grid mesh of this total size and not a little bigger, and the number of divisions is set as an integer number that best approximate your required size. This is the reason to obtain a grid cell size only similar to the user required, and not exactly equal.

If do you want exactly your cell size=0.0375, set the 'grid spacing to uniform' (but notice that the grid mesh bounding box could be a little bigger that the model one to enclose them)

Regards

Enrique Escolano
----- Original Message -----
From: He Jinzhi
To: gidlist at listas.cimne.upc.edu
Sent: Thursday, January 10, 2013 2:12 AM
Subject: [GiDlist] about cartesian grid generation


Hello,
I have a question about the cartesian grid generation of GID. I want to generate the cartesian grid of object. When i set the Mesh Cartesian Data, Generate Mesh, and export the GID mesh data, I find the interval exported is different with i set before.


For example, I set the cartesian volume mesh size as 0.0375(same sizes), but i find the interval of the exported mesh data may be 0.037037037?Likewise , i set the value as 200, but the result may be 198.7578 or 196.226415, or something approximated to 200. i want to ask the reason.Thanks!


regards.
Jinzhi He





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


_______________________________________________
GiDlist mailing list
GiDlist at listas.cimne.upc.edu
http://listas.cimne.upc.edu/cgi-bin/mailman/listinfo/gidlist
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://listas.cimne.upc.edu/pipermail/gidlist/attachments/20130110/84ad6666/attachment.html

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

_______________________________________________
GiDlist mailing list
GiDlist at listas.cimne.upc.edu
http://listas.cimne.upc.edu/cgi-bin/mailman/listinfo/gidlist


End of GiDlist Digest, Vol 30, Issue 2
**************************************
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://listas.cimne.upc.edu/pipermail/gidlist/attachments/20130114/0fcf8dce/attachment.htm