Hello everyone,
I was wandering which is the effect of the command:
-Sel by normal (select by a direction)
in the normal menu once the normal are displayed.
A wish:
For BEM and other applications the orientation of surface and line
normals is fundamental, and it is almost always necessary to control
and reorient them in a model. It would be than definitely handy to
have a dialog window (similar to the layer one) opening at the
command "Draw Normals", where the user could manage the
normals and have always the result displayed without needing to re-
issue the command "Draw Normals" again.
I hope this suggestion could be useful for the GiD interface
development.
Best regards
Andrea
Andrea Calaon
Ashurst Lodge, Ashurst
Southampton, Hampshire
SO40 7AA, UK
Tel. office: +44 23 80 293 223
Fax office: +44 23 80 292 853
Homepage: http://mailserver1.wessex.ac.uk/~acalaon
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://listas.cimne.upc.edu/pipermail/gidlist/attachments/20050622/3d86f677/attachment.htm
[GiDlist] Question above normals
Moderator: GiD Team
[GiDlist] Question above normals
Hello Andrea.
You are right, control of normals is interesting, and the current mode to use this tools is old and not very intuitive.
It must be enhanced adding some window or better menus.
Sel by normal is interesting for example, if you have a shape that can be projected in a direction (useful in sheet forming, etc)
to select only the entities with normal pointing approximately an user direction (in fact all with positive dot-product)
For example, if you have this surfaces in the ZY plane, you can swap only the surfaces with normal pointing to -Z axis (0,0,-1)
use: Utilities-Draw normals-Surfaces
then you select all surfaces (elements in mesh view)
and you filter this selection with
SelByNormal and enter a direction, for example: 0 0 -1
then only the entities with normal pointing in the same half-space of this direction remain selected
and you can use AllSwapped to swap the orientation of the yellow surfaces
Best wishes
Enrique
----- Original Message -----
From: Andrea Calaon
To: gidlist at gatxan.cimne.upc.edu
Sent: Wednesday, June 22, 2005 1:22 PM
Subject: [GiDlist] Question above normals
Hello everyone,
I was wandering which is the effect of the command:
-Sel by normal (select by a direction)
in the normal menu once the normal are displayed.
A wish:
For BEM and other applications the orientation of surface and line normals is fundamental, and it is almost always necessary to control and reorient them in a model. It would be than definitely handy to have a dialog window (similar to the layer one) opening at the command "Draw Normals", where the user could manage the normals and have always the result displayed without needing to re- issue the command "Draw Normals" again.
I hope this suggestion could be useful for the GiD interface development.
Best regards
Andrea
Andrea Calaon
Ashurst Lodge, Ashurst
Southampton, Hampshire
SO40 7AA, UK
Tel. office: +44 23 80 293 223
Fax office: +44 23 80 292 853
Homepage: http://mailserver1.wessex.ac.uk/~acalaon
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://listas.cimne.upc.edu/pipermail/gidlist/attachments/20050622/38dbeba0/attachment.htm
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: image/gif
Size: 5650 bytes
Desc: not available
Url : http://listas.cimne.upc.edu/pipermail/gidlist/attachments/20050622/38dbeba0/attachment.gif
You are right, control of normals is interesting, and the current mode to use this tools is old and not very intuitive.
It must be enhanced adding some window or better menus.
Sel by normal is interesting for example, if you have a shape that can be projected in a direction (useful in sheet forming, etc)
to select only the entities with normal pointing approximately an user direction (in fact all with positive dot-product)
For example, if you have this surfaces in the ZY plane, you can swap only the surfaces with normal pointing to -Z axis (0,0,-1)
use: Utilities-Draw normals-Surfaces
then you select all surfaces (elements in mesh view)
and you filter this selection with
SelByNormal and enter a direction, for example: 0 0 -1
then only the entities with normal pointing in the same half-space of this direction remain selected
and you can use AllSwapped to swap the orientation of the yellow surfaces
Best wishes
Enrique
----- Original Message -----
From: Andrea Calaon
To: gidlist at gatxan.cimne.upc.edu
Sent: Wednesday, June 22, 2005 1:22 PM
Subject: [GiDlist] Question above normals
Hello everyone,
I was wandering which is the effect of the command:
-Sel by normal (select by a direction)
in the normal menu once the normal are displayed.
A wish:
For BEM and other applications the orientation of surface and line normals is fundamental, and it is almost always necessary to control and reorient them in a model. It would be than definitely handy to have a dialog window (similar to the layer one) opening at the command "Draw Normals", where the user could manage the normals and have always the result displayed without needing to re- issue the command "Draw Normals" again.
I hope this suggestion could be useful for the GiD interface development.
Best regards
Andrea
Andrea Calaon
Ashurst Lodge, Ashurst
Southampton, Hampshire
SO40 7AA, UK
Tel. office: +44 23 80 293 223
Fax office: +44 23 80 292 853
Homepage: http://mailserver1.wessex.ac.uk/~acalaon
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://listas.cimne.upc.edu/pipermail/gidlist/attachments/20050622/38dbeba0/attachment.htm
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: image/gif
Size: 5650 bytes
Desc: not available
Url : http://listas.cimne.upc.edu/pipermail/gidlist/attachments/20050622/38dbeba0/attachment.gif
-
Andrea Calaon
[GiDlist] Question above normals
Thank you Enrique!
In cases with many surfaces or elements the command is very
handy.
In BEM, as you know, the most important information about
normals is their ORIENTATION in relation to the geometrical
entities of which their support is the boundary, like:
- (2D) line normals: oriented outwards or inwards with respect to
surface/s "owing" them?
- (3D) surface normals: oriented outwards or inwards with respect
to the volume/s "owing" them?
Three years ago I programmed something which allows extracting
this critical information from the GiD VOLUME properties, but only
provided the user uses some precautions (it makes use of the info
DIFF1ST and SAME1ST). It works but is not as automatic as it
could.
When creating a single Volume from a series of surfaces GiD
reorients them in order to point inwards, therefore I guess it
makes use of the information I was mentioning. But than, with the
presence of other volumes sharing some surfaces with the first
one, part of the information is “lost” or anyway becomes no more
retrievable. From the VOLUME list it is in fact possible to know only
the RELATIVE orientation of surfaces with respect to other surfaces
around a volume, but not their ABSOLUTE orientation with respect
to every single volume they are part of.
I don't know if my "analysis" makes any sense in the scheme GiD
uses for these things. If it does, do you think in some future GiD
betas or releases it will be possible to have access to this
information, so critical for some applications?
Sorry for the boring and long mail.
Best regards,
Andrea
On 22 Jun 2005 at 16:34, Enrique Escolano wrote:
Hello Andrea.
You are right, control of normals is interesting, and
thecurrentmode to use this tools is old and
not very intuitive.
It must be enhanced adding some window or better menus.
Sel by normal is interesting for example, if you have a shape
that can be projected in a direction
(useful in sheet forming, etc)
to select only the entities with normal pointing approximately an
userdirection (in fact all with
positive dot-product)
For example, if you have this surfaces in the ZY plane, you can
swap only the surfaces with
normal pointing to -Z axis (0,0,-1)
use: Utilities-Draw normals-Surfaces
then you selectall surfaces (elements in mesh view)
and you filter this selection with
SelByNormal and enter a direction, for example: 0 0 -1
then only the entities with normal pointing in the same half-
space of this direction remain selected
andyou can use AllSwapped to swap the orientation of the yellow
surfaces
Best wishes
Enrique
----- Original Message -----
From: Andrea Calaon
To: gidlist at gatxan.cimne.upc.edu
Sent: Wednesday, June 22, 2005 1:22 PM
Subject: [GiDlist] Question above normals
Andrea Calaon
*****************************************************
WESSEX INSTITUTE on the Web - http://www.wessex.ac.uk
SEE 2005 Call for Papers at: http://www.wessex.ac.uk/conferences/2005/index.html
WIT PRESS on the Web - visit http://www.witpress.com
to download individual Wessex Institute conference papers from our WIT eLIBRARY
and browse our BOOKSTORE. Save 10% and ensure fast delivery when you order our books on-line!
BEASY on the Web: http://www.beasy.com
*****************************************************
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://listas.cimne.upc.edu/pipermail/gidlist/attachments/20050623/82ae9c24/attachment.htm
In cases with many surfaces or elements the command is very
handy.
In BEM, as you know, the most important information about
normals is their ORIENTATION in relation to the geometrical
entities of which their support is the boundary, like:
- (2D) line normals: oriented outwards or inwards with respect to
surface/s "owing" them?
- (3D) surface normals: oriented outwards or inwards with respect
to the volume/s "owing" them?
Three years ago I programmed something which allows extracting
this critical information from the GiD VOLUME properties, but only
provided the user uses some precautions (it makes use of the info
DIFF1ST and SAME1ST). It works but is not as automatic as it
could.
When creating a single Volume from a series of surfaces GiD
reorients them in order to point inwards, therefore I guess it
makes use of the information I was mentioning. But than, with the
presence of other volumes sharing some surfaces with the first
one, part of the information is “lost” or anyway becomes no more
retrievable. From the VOLUME list it is in fact possible to know only
the RELATIVE orientation of surfaces with respect to other surfaces
around a volume, but not their ABSOLUTE orientation with respect
to every single volume they are part of.
I don't know if my "analysis" makes any sense in the scheme GiD
uses for these things. If it does, do you think in some future GiD
betas or releases it will be possible to have access to this
information, so critical for some applications?
Sorry for the boring and long mail.
Best regards,
Andrea
On 22 Jun 2005 at 16:34, Enrique Escolano wrote:
Hello Andrea.
You are right, control of normals is interesting, and
thecurrentmode to use this tools is old and
not very intuitive.
It must be enhanced adding some window or better menus.
Sel by normal is interesting for example, if you have a shape
that can be projected in a direction
(useful in sheet forming, etc)
to select only the entities with normal pointing approximately an
userdirection (in fact all with
positive dot-product)
For example, if you have this surfaces in the ZY plane, you can
swap only the surfaces with
normal pointing to -Z axis (0,0,-1)
use: Utilities-Draw normals-Surfaces
then you selectall surfaces (elements in mesh view)
and you filter this selection with
SelByNormal and enter a direction, for example: 0 0 -1
then only the entities with normal pointing in the same half-
space of this direction remain selected
andyou can use AllSwapped to swap the orientation of the yellow
surfaces
Best wishes
Enrique
----- Original Message -----
From: Andrea Calaon
To: gidlist at gatxan.cimne.upc.edu
Sent: Wednesday, June 22, 2005 1:22 PM
Subject: [GiDlist] Question above normals
Andrea Calaon
*****************************************************
WESSEX INSTITUTE on the Web - http://www.wessex.ac.uk
SEE 2005 Call for Papers at: http://www.wessex.ac.uk/conferences/2005/index.html
WIT PRESS on the Web - visit http://www.witpress.com
to download individual Wessex Institute conference papers from our WIT eLIBRARY
and browse our BOOKSTORE. Save 10% and ensure fast delivery when you order our books on-line!
BEASY on the Web: http://www.beasy.com
*****************************************************
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://listas.cimne.upc.edu/pipermail/gidlist/attachments/20050623/82ae9c24/attachment.htm
[GiDlist] Question above normals
Like BEM, GiD also needs to know what is inside and what is outside to generate the mesh.
A volume has a list of surfaces and their senses:
If the surface normal points inside, its sense must be SAME1ST, else must be DIFF1ST
(usually when you create a volume the surface normals are swapped in order to have all SAME1ST, but if the surface is shared by two volumes
it must be SAME1ST for one volume and DIFF1ST for the other)
Do you have access to this information from Tcl with the command GiD_Info list_entities volumes id_number
The surfaces also have a similar topology, its lines are oriented: a SAME1ST line has the surface interior turning to the left
What information do you need that it is unavailable now?
Enrique
----- Original Message -----
From: Andrea Calaon
To: gidlist at gatxan.cimne.upc.edu
Sent: Thursday, June 23, 2005 11:18 AM
Subject: Re: [GiDlist] Question above normals
Thank you Enrique!
In cases with many surfaces or elements the command is very handy.
In BEM, as you know, the most important information about normals is their ORIENTATION in relation to the geometrical entities of which their support is the boundary, like:
- (2D) line normals: oriented outwards or inwards with respect to surface/s "owing" them?
- (3D) surface normals: oriented outwards or inwards with respect to the volume/s "owing" them?
Three years ago I programmed something which allows extracting this critical information from the GiD VOLUME properties, but only provided the user uses some precautions (it makes use of the info DIFF1ST and SAME1ST). It works but is not as automatic as it could.
When creating a single Volume from a series of surfaces GiD reorients them in order to point inwards, therefore I guess it makes use of the information I was mentioning. But than, with the presence of other volumes sharing some surfaces with the first one, part of the information is "lost" or anyway becomes no more retrievable. From the VOLUME list it is in fact possible to know only the RELATIVE orientation of surfaces with respect to other surfaces around a volume, but not their ABSOLUTE orientation with respect to every single volume they are part of.
I don't know if my "analysis" makes any sense in the scheme GiD uses for these things. If it does, do you think in some future GiD betas or releases it will be possible to have access to this information, so critical for some applications?
Sorry for the boring and long mail.
Best regards,
Andrea
On 22 Jun 2005 at 16:34, Enrique Escolano wrote:
Hello Andrea.
You are right, control of normals is interesting, and thecurrentmode to use this tools is old and
not very intuitive.
It must be enhanced adding some window or better menus.
Sel by normal is interesting for example, if you have a shape that can be projected in a direction
(useful in sheet forming, etc)
to select only the entities with normal pointing approximately an userdirection (in fact all with
positive dot-product)
For example, if you have this surfaces in the ZY plane, you can swap only the surfaces with
normal pointing to -Z axis (0,0,-1)
use: Utilities-Draw normals-Surfaces
then you selectall surfaces (elements in mesh view)
and you filter this selection with
SelByNormal and enter a direction, for example: 0 0 -1
then only the entities with normal pointing in the same half- space of this direction remain selected
andyou can use AllSwapped to swap the orientation of the yellow surfaces
Best wishes
Enrique
----- Original Message -----
From: Andrea Calaon
To: gidlist at gatxan.cimne.upc.edu
Sent: Wednesday, June 22, 2005 1:22 PM
Subject: [GiDlist] Question above normals
Andrea Calaon
*****************************************************
WESSEX INSTITUTE on the Web - http://www.wessex.ac.uk
SEE 2005 Call for Papers at: http://www.wessex.ac.uk/conferences/2005/index.html
WIT PRESS on the Web - visit http://www.witpress.com
to download individual Wessex Institute conference papers from our WIT eLIBRARY
and browse our BOOKSTORE. Save 10% and ensure fast delivery when you order our books on-line!
BEASY on the Web: http://www.beasy.com
*****************************************************
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://listas.cimne.upc.edu/pipermail/gidlist/attachments/20050629/f1ca9096/attachment.htm
A volume has a list of surfaces and their senses:
If the surface normal points inside, its sense must be SAME1ST, else must be DIFF1ST
(usually when you create a volume the surface normals are swapped in order to have all SAME1ST, but if the surface is shared by two volumes
it must be SAME1ST for one volume and DIFF1ST for the other)
Do you have access to this information from Tcl with the command GiD_Info list_entities volumes id_number
The surfaces also have a similar topology, its lines are oriented: a SAME1ST line has the surface interior turning to the left
What information do you need that it is unavailable now?
Enrique
----- Original Message -----
From: Andrea Calaon
To: gidlist at gatxan.cimne.upc.edu
Sent: Thursday, June 23, 2005 11:18 AM
Subject: Re: [GiDlist] Question above normals
Thank you Enrique!
In cases with many surfaces or elements the command is very handy.
In BEM, as you know, the most important information about normals is their ORIENTATION in relation to the geometrical entities of which their support is the boundary, like:
- (2D) line normals: oriented outwards or inwards with respect to surface/s "owing" them?
- (3D) surface normals: oriented outwards or inwards with respect to the volume/s "owing" them?
Three years ago I programmed something which allows extracting this critical information from the GiD VOLUME properties, but only provided the user uses some precautions (it makes use of the info DIFF1ST and SAME1ST). It works but is not as automatic as it could.
When creating a single Volume from a series of surfaces GiD reorients them in order to point inwards, therefore I guess it makes use of the information I was mentioning. But than, with the presence of other volumes sharing some surfaces with the first one, part of the information is "lost" or anyway becomes no more retrievable. From the VOLUME list it is in fact possible to know only the RELATIVE orientation of surfaces with respect to other surfaces around a volume, but not their ABSOLUTE orientation with respect to every single volume they are part of.
I don't know if my "analysis" makes any sense in the scheme GiD uses for these things. If it does, do you think in some future GiD betas or releases it will be possible to have access to this information, so critical for some applications?
Sorry for the boring and long mail.
Best regards,
Andrea
On 22 Jun 2005 at 16:34, Enrique Escolano wrote:
Hello Andrea.
You are right, control of normals is interesting, and thecurrentmode to use this tools is old and
not very intuitive.
It must be enhanced adding some window or better menus.
Sel by normal is interesting for example, if you have a shape that can be projected in a direction
(useful in sheet forming, etc)
to select only the entities with normal pointing approximately an userdirection (in fact all with
positive dot-product)
For example, if you have this surfaces in the ZY plane, you can swap only the surfaces with
normal pointing to -Z axis (0,0,-1)
use: Utilities-Draw normals-Surfaces
then you selectall surfaces (elements in mesh view)
and you filter this selection with
SelByNormal and enter a direction, for example: 0 0 -1
then only the entities with normal pointing in the same half- space of this direction remain selected
andyou can use AllSwapped to swap the orientation of the yellow surfaces
Best wishes
Enrique
----- Original Message -----
From: Andrea Calaon
To: gidlist at gatxan.cimne.upc.edu
Sent: Wednesday, June 22, 2005 1:22 PM
Subject: [GiDlist] Question above normals
Andrea Calaon
*****************************************************
WESSEX INSTITUTE on the Web - http://www.wessex.ac.uk
SEE 2005 Call for Papers at: http://www.wessex.ac.uk/conferences/2005/index.html
WIT PRESS on the Web - visit http://www.witpress.com
to download individual Wessex Institute conference papers from our WIT eLIBRARY
and browse our BOOKSTORE. Save 10% and ensure fast delivery when you order our books on-line!
BEASY on the Web: http://www.beasy.com
*****************************************************
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://listas.cimne.upc.edu/pipermail/gidlist/attachments/20050629/f1ca9096/attachment.htm
-
Andrea Calaon
[GiDlist] Question above normals
Hello Enrique,
thanks for the answer and the explanation.
I think I eventually understood how it works: if a surface is
labelled as SAME1ST its normal is ALWAYS pointing inward,
no matter how the volumes have been generated, otherwise
it is pointing outward. All info is then provided and there isn't
any missing bit. Sorry for having been fussy about this.
I was getting the info regarding SAME1ST and DIFF1ST from
the Tcl command:
[.central.s info list_entities volumes 1:[NumVolumes]]
Best regards
Andrea
On 29 Jun 2005 at 22:08, Enrique Escolano wrote:
Like BEM, GiD also needs to know what is inside and what is outside to generate the mesh.
A volume has a list of surfaces andtheir senses:
If the surface normal points inside, itssensemust be SAME1ST, else must be DIFF1ST
(usually when you create a volume the surface normals are swapped in order to have all
SAME1ST, but if the surface is shared by two volumes
it must be SAME1ST for one volume and DIFF1ST for the other)
Do you have access to this informationfrom Tcl with the command GiD_Info list_entities volumes
id_number
The surfaces also have a similar topology, its lines are oriented: a SAME1ST line has the surface
interior turning to theleft
What information do you need that it is unavailable now?
Enrique
Andrea Calaon
*****************************************************
WESSEX INSTITUTE on the Web - http://www.wessex.ac.uk
SEE 2005 Call for Papers at: http://www.wessex.ac.uk/conferences/2005/index.html
WIT PRESS on the Web - visit http://www.witpress.com
to download individual Wessex Institute conference papers from our WIT eLIBRARY
and browse our BOOKSTORE. Save 10% and ensure fast delivery when you order our books on-line!
BEASY on the Web: http://www.beasy.com
*****************************************************
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://listas.cimne.upc.edu/pipermail/gidlist/attachments/20050630/68c6f9df/attachment.htm
thanks for the answer and the explanation.
I think I eventually understood how it works: if a surface is
labelled as SAME1ST its normal is ALWAYS pointing inward,
no matter how the volumes have been generated, otherwise
it is pointing outward. All info is then provided and there isn't
any missing bit. Sorry for having been fussy about this.
I was getting the info regarding SAME1ST and DIFF1ST from
the Tcl command:
[.central.s info list_entities volumes 1:[NumVolumes]]
Best regards
Andrea
On 29 Jun 2005 at 22:08, Enrique Escolano wrote:
Like BEM, GiD also needs to know what is inside and what is outside to generate the mesh.
A volume has a list of surfaces andtheir senses:
If the surface normal points inside, itssensemust be SAME1ST, else must be DIFF1ST
(usually when you create a volume the surface normals are swapped in order to have all
SAME1ST, but if the surface is shared by two volumes
it must be SAME1ST for one volume and DIFF1ST for the other)
Do you have access to this informationfrom Tcl with the command GiD_Info list_entities volumes
id_number
The surfaces also have a similar topology, its lines are oriented: a SAME1ST line has the surface
interior turning to theleft
What information do you need that it is unavailable now?
Enrique
Andrea Calaon
*****************************************************
WESSEX INSTITUTE on the Web - http://www.wessex.ac.uk
SEE 2005 Call for Papers at: http://www.wessex.ac.uk/conferences/2005/index.html
WIT PRESS on the Web - visit http://www.witpress.com
to download individual Wessex Institute conference papers from our WIT eLIBRARY
and browse our BOOKSTORE. Save 10% and ensure fast delivery when you order our books on-line!
BEASY on the Web: http://www.beasy.com
*****************************************************
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://listas.cimne.upc.edu/pipermail/gidlist/attachments/20050630/68c6f9df/attachment.htm