[GiDlist] Semistructured volume

Moderator: GiD Team

Isabel Gil

[GiDlist] Semistructured volume

Post by Isabel Gil »

Hello!

Thank-you. I could finally find it.

However, when I try to "Generate mesh", the program, both GID version
8.1.2b in Windows and GID version 8.1.1b in Linux, breaks down, the
message is, in Linux, "Segmentation fault".
Could it be due to both of them are beta versions?

Thanks again for your time
Isa

Isabel Gil wrote:
Hi!

Thanks for your reply. I answer through your e-mail:

Enrique Escolano wrote:
I think that we didn't understand your meaning of "irregularities in
the mesh"
I think that you don't want to say "elements with bad angles or
sizes", and what you want
is that all extruded nodes have exactly the same XY proyection. Is not?

Yes, you are right. I meant just what you have described.
Structured or semi-structured meshes will not generate aligned-like
nodes in the extrusion direction, the algorithm is not a simple
translation, is more general (lateral surfaces can be curved, and
both tops must have the same number of boundaries, but can have
different shape)

Last 8.1.x beta versions have an "special option" to try to force
this alignement if possible, by setting the GiD variable
AlignSemiStructuredNodes to 1
escape escape escape utilities variables AlignSemiStructuredNodes 1
(but your 8.0.8 official version doesn't has this option)

Then, I have just downloaded Gid8.1.2b-win-x32.exe, which is a beta
version but I am not able to find the GiD variable
AlignSemiStructuredNodes and to put it to 1.
Could you say to me where it is exactly?
Thanks in advance and best regards
Isa
ftp://gid.cimne.upc.es/pub/gid_adds/Windows/GiD8.1.2b-win-x32.exe
Next week we are releasing a new beta version 8.1.3 that also has a
new unstructured quadrilateral meshing algorithm,
Obtained quadrilaterals will have much better shape quality. Can set
this mesher with the variable:
escape escape escape utilities variables QuadrilateralMesher 1

Regards

Enrique Escolano

----- Original Message ----- From: "Isabel Gil" misabel at litec.csic.es
To: gidlist at gatxan.cimne.upc.edu
Sent: Friday, April 27, 2007 6:12 PM
Subject: Re: [GiDlist] Semistructured volume



Hello!

First of all, thanks for your reply.

Your are right with the so different size of mesh elements I am using,
but if you, in the previous project I sent you, put "RENDER-normal"
and
put on and off the "layer capa2_a", you can see that different lines
appear (you can see this effect at first sight on the left at the
bottom
with only the "layer capa1_a" on and putting the "layer capa2_a" on and
off, with the view of XY plane).
Do you mean that these irregularities in the mesh is due to my bad size
election?

Again, thank-you very much in advance.

Best regards!
Isa

Miguel Pasenau wrote:

Hi,

If you mean that the elements are quite distorted, i think it's
because you
have assigned element size of '2' for some volumes (layers
'capa1_a' and
'capa1_b' with a height of 0.002) and '4' for others (layers
'capa2_a' and
'capa2_b' with a height of 0.002). Some details of the surfaces
conforming
the volumes is just 0.001. You can see it at 'ig_volume_size_1.png'.

And, for instance, at the layer 'capa1_b', there are surfaces with
element
size assigned of 0.0059 (and the width of the surface is just
0.002) and
just beside them surfaces with element size of 0.000325, 15 times
less, as
you can see in the image 'ig_surface_size_1.png', resulting in the
mesh
shown in 'ig_mesh_zoom_1.png'.
Best regards,

Miguel

-----Mensaje original-----
De: gidlist-admin at gatxan.cimne.upc.edu
[mailto:gidlist-admin at gatxan.cimne.upc.edu] En nombre de Isabel Gil
Enviado el: jueves, 26 de abril de 2007 11:56
Para: GID_FORO
Asunto: [GiDlist] Semistructured volume

Hello GiD'users!

I have a little problem when I want to create a mesh.
I am working with GID 8.0.8 in Windows.


I attach my project because I don't get to make a good semistructured
volume with structured mesh
in z-direction. You can see the irregularities of the mesh if you use
the XY-plane view and the rotate trackball tool.

I don't know what I am doing wrong. I think I followed correctly the
advice that previous e-mails
gave me but perhaps I am forgetting something.

I would be really grateful if you could cast a glance at it.

Thanks in advance and best regards
Isa



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



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



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



_______________________________________________
GiDlist mailing list
GiDlist at gid.cimne.upc.es
http://gid.cimne.upc.es/mailman/listinfo/gidlist


_______________________________________________
GiDlist mailing list
GiDlist at gid.cimne.upc.es
http://gid.cimne.upc.es/mailman/listinfo/gidlist


_______________________________________________
GiDlist mailing list
GiDlist at gid.cimne.upc.es
http://gid.cimne.upc.es/mailman/listinfo/gidlist
User avatar
escolano
Posts: 1918
Joined: Sun Sep 05, 1982 10:51 pm

[GiDlist] Semistructured volume

Post by escolano »

GiD 8.1.2 is crashing because you are using the new quadrilateral mesher

You must set the old quadrilateral mesher (value=0)
escape escape escape utilities variables QuadrilateralMesher 0
And activate the align nodes option (value=1)
escape escape escape utilities variables AlignSemiStructuredNodes 1

As I said we are releasing a new 8.1.3 beta version in few days, with quadrilateral mesher fixed bugs.

Regards

Enrique

----- Original Message -----
From: "Isabel Gil" misabel at litec.csic.es
To: gidlist at gatxan.cimne.upc.edu
Sent: Wednesday, May 02, 2007 2:29 PM
Subject: Re: [GiDlist] Semistructured volume


Hello!

Thank-you. I could finally find it.

However, when I try to "Generate mesh", the program, both GID version
8.1.2b in Windows and GID version 8.1.1b in Linux, breaks down, the
message is, in Linux, "Segmentation fault".
Could it be due to both of them are beta versions?

Thanks again for your time
Isa

Isabel Gil wrote:
Hi!

Thanks for your reply. I answer through your e-mail:

Enrique Escolano wrote:
I think that we didn't understand your meaning of "irregularities in
the mesh"
I think that you don't want to say "elements with bad angles or
sizes", and what you want
is that all extruded nodes have exactly the same XY proyection. Is not?

Yes, you are right. I meant just what you have described.
Structured or semi-structured meshes will not generate aligned-like
nodes in the extrusion direction, the algorithm is not a simple
translation, is more general (lateral surfaces can be curved, and
both tops must have the same number of boundaries, but can have
different shape)

Last 8.1.x beta versions have an "special option" to try to force
this alignement if possible, by setting the GiD variable
AlignSemiStructuredNodes to 1
escape escape escape utilities variables AlignSemiStructuredNodes 1
(but your 8.0.8 official version doesn't has this option)

Then, I have just downloaded Gid8.1.2b-win-x32.exe, which is a beta
version but I am not able to find the GiD variable
AlignSemiStructuredNodes and to put it to 1.
Could you say to me where it is exactly?
Thanks in advance and best regards
Isa
ftp://gid.cimne.upc.es/pub/gid_adds/Windows/GiD8.1.2b-win-x32.exe
Next week we are releasing a new beta version 8.1.3 that also has a
new unstructured quadrilateral meshing algorithm,
Obtained quadrilaterals will have much better shape quality. Can set
this mesher with the variable:
escape escape escape utilities variables QuadrilateralMesher 1

Regards

Enrique Escolano

----- Original Message ----- From: "Isabel Gil" misabel at litec.csic.es
To: gidlist at gatxan.cimne.upc.edu
Sent: Friday, April 27, 2007 6:12 PM
Subject: Re: [GiDlist] Semistructured volume



Hello!

First of all, thanks for your reply.

Your are right with the so different size of mesh elements I am using,
but if you, in the previous project I sent you, put "RENDER-normal"
and
put on and off the "layer capa2_a", you can see that different lines
appear (you can see this effect at first sight on the left at the
bottom
with only the "layer capa1_a" on and putting the "layer capa2_a" on and
off, with the view of XY plane).
Do you mean that these irregularities in the mesh is due to my bad size
election?

Again, thank-you very much in advance.

Best regards!
Isa

Miguel Pasenau wrote:

Hi,

If you mean that the elements are quite distorted, i think it's
because you
have assigned element size of '2' for some volumes (layers
'capa1_a' and
'capa1_b' with a height of 0.002) and '4' for others (layers
'capa2_a' and
'capa2_b' with a height of 0.002). Some details of the surfaces
conforming
the volumes is just 0.001. You can see it at 'ig_volume_size_1.png'.

And, for instance, at the layer 'capa1_b', there are surfaces with
element
size assigned of 0.0059 (and the width of the surface is just
0.002) and
just beside them surfaces with element size of 0.000325, 15 times
less, as
you can see in the image 'ig_surface_size_1.png', resulting in the
mesh
shown in 'ig_mesh_zoom_1.png'.
Best regards,

Miguel

-----Mensaje original-----
De: gidlist-admin at gatxan.cimne.upc.edu
[mailto:gidlist-admin at gatxan.cimne.upc.edu] En nombre de Isabel Gil
Enviado el: jueves, 26 de abril de 2007 11:56
Para: GID_FORO
Asunto: [GiDlist] Semistructured volume

Hello GiD'users!

I have a little problem when I want to create a mesh.
I am working with GID 8.0.8 in Windows.


I attach my project because I don't get to make a good semistructured
volume with structured mesh
in z-direction. You can see the irregularities of the mesh if you use
the XY-plane view and the rotate trackball tool.

I don't know what I am doing wrong. I think I followed correctly the
advice that previous e-mails
gave me but perhaps I am forgetting something.

I would be really grateful if you could cast a glance at it.

Thanks in advance and best regards
Isa



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



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



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



_______________________________________________
GiDlist mailing list
GiDlist at gid.cimne.upc.es
http://gid.cimne.upc.es/mailman/listinfo/gidlist


_______________________________________________
GiDlist mailing list
GiDlist at gid.cimne.upc.es
http://gid.cimne.upc.es/mailman/listinfo/gidlist


_______________________________________________
GiDlist mailing list
GiDlist at gid.cimne.upc.es
http://gid.cimne.upc.es/mailman/listinfo/gidlist

_______________________________________________
GiDlist mailing list
GiDlist at gid.cimne.upc.es
http://gid.cimne.upc.es/mailman/listinfo/gidlist
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://listas.cimne.upc.edu/pipermail/gidlist/attachments/20070502/810f77d1/attachment.htm
Isabel Gil

[GiDlist] Semistructured volume

Post by Isabel Gil »

Hi again,

I have tried to put in both, GiD8.1.2 in Windows and in GiD8.1.1b in
Linux escape escape escape utilities variables QuadrilateralMesher 0
but I have obtained the following error message: "Unknown variable name
QuadrilateralMesher".
Perhaps, in these versions, the variable was called in another way¿?

Thanks in advance
Isa


Enrique Escolano wrote:
GiD 8.1.2 is crashing because you are using the new quadrilateral mesher

You must set the old quadrilateral mesher (value=0)
escape escape escape utilities variables QuadrilateralMesher 0
And activate the align nodes option (value=1)
escape escape escape utilities variables AlignSemiStructuredNodes 1

As I said we are releasing a new 8.1.3 beta version in few days, with
quadrilateral mesher fixed bugs.

Regards

Enrique

----- Original Message -----
From: "Isabel Gil" misabel at litec.csic.es mailto:misabel at litec.csic.es
To: gidlist at gatxan.cimne.upc.edu mailto:gidlist at gatxan.cimne.upc.edu
Sent: Wednesday, May 02, 2007 2:29 PM
Subject: Re: [GiDlist] Semistructured volume

Hello!

Thank-you. I could finally find it.

However, when I try to "Generate mesh", the program, both GID version
8.1.2b in Windows and GID version 8.1.1b in Linux, breaks down, the
message is, in Linux, "Segmentation fault".
Could it be due to both of them are beta versions?

Thanks again for your time
Isa

Isabel Gil wrote:
Hi!

Thanks for your reply. I answer through your e-mail:

Enrique Escolano wrote:
I think that we didn't understand your meaning of
"irregularities in
the mesh"
I think that you don't want to say "elements with bad angles or
sizes", and what you want
is that all extruded nodes have exactly the same XY proyection.
Is not?

Yes, you are right. I meant just what you have described.
Structured or semi-structured meshes will not generate aligned-like
nodes in the extrusion direction, the algorithm is not a simple
translation, is more general (lateral surfaces can be curved, and
both tops must have the same number of boundaries, but can have
different shape)

Last 8.1.x beta versions have an "special option" to try to force
this alignement if possible, by setting the GiD variable
AlignSemiStructuredNodes to 1
escape escape escape utilities variables AlignSemiStructuredNodes 1
(but your 8.0.8 official version doesn't has this option)

Then, I have just downloaded Gid8.1.2b-win-x32.exe, which is a beta
version but I am not able to find the GiD variable
AlignSemiStructuredNodes and to put it to 1.
Could you say to me where it is exactly?
Thanks in advance and best regards
Isa
ftp://gid.cimne.upc.es/pub/gid_adds/Windows/GiD8.1.2b-win-x32.exe
Next week we are releasing a new beta version 8.1.3 that also has a
new unstructured quadrilateral meshing algorithm,
Obtained quadrilaterals will have much better shape quality. Can set
this mesher with the variable:
escape escape escape utilities variables QuadrilateralMesher 1

Regards

Enrique Escolano

----- Original Message ----- From: "Isabel Gil"
misabel at litec.csic.es mailto:misabel at litec.csic.es
To: gidlist at gatxan.cimne.upc.edu
mailto:gidlist at gatxan.cimne.upc.edu
Sent: Friday, April 27, 2007 6:12 PM
Subject: Re: [GiDlist] Semistructured volume



Hello!

First of all, thanks for your reply.

Your are right with the so different size of mesh elements I am
using,
but if you, in the previous project I sent you, put
"RENDER-normal"
and
put on and off the "layer capa2_a", you can see that different lines
appear (you can see this effect at first sight on the left at the
bottom
with only the "layer capa1_a" on and putting the "layer capa2_a"
on and
off, with the view of XY plane).
Do you mean that these irregularities in the mesh is due to my
bad size
election?

Again, thank-you very much in advance.

Best regards!
Isa

Miguel Pasenau wrote:

Hi,

If you mean that the elements are quite distorted, i think it's
because you
have assigned element size of '2' for some volumes (layers
'capa1_a' and
'capa1_b' with a height of 0.002) and '4' for others (layers
'capa2_a' and
'capa2_b' with a height of 0.002). Some details of the surfaces
conforming
the volumes is just 0.001. You can see it at
'ig_volume_size_1.png'.

And, for instance, at the layer 'capa1_b', there are surfaces with
element
size assigned of 0.0059 (and the width of the surface is just
0.002) and
just beside them surfaces with element size of 0.000325, 15 times
less, as
you can see in the image 'ig_surface_size_1.png', resulting in the
mesh
shown in 'ig_mesh_zoom_1.png'.
Best regards,

Miguel

-----Mensaje original-----
De: gidlist-admin at gatxan.cimne.upc.edu
mailto:gidlist-admin at gatxan.cimne.upc.edu
[mailto:gidlist-admin at gatxan.cimne.upc.edu] En nombre de Isabel Gil
Enviado el: jueves, 26 de abril de 2007 11:56
Para: GID_FORO
Asunto: [GiDlist] Semistructured volume

Hello GiD'users!

I have a little problem when I want to create a mesh.
I am working with GID 8.0.8 in Windows.


I attach my project because I don't get to make a good
semistructured
volume with structured mesh
in z-direction. You can see the irregularities of the mesh if
you use
the XY-plane view and the rotate trackball tool.

I don't know what I am doing wrong. I think I followed
correctly the
advice that previous e-mails
gave me but perhaps I am forgetting something.

I would be really grateful if you could cast a glance at it.

Thanks in advance and best regards
Isa




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




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




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



_______________________________________________
GiDlist mailing list
GiDlist at gid.cimne.upc.es mailto:GiDlist at gid.cimne.upc.es
http://gid.cimne.upc.es/mailman/listinfo/gidlist


_______________________________________________
GiDlist mailing list
GiDlist at gid.cimne.upc.es mailto:GiDlist at gid.cimne.upc.es
http://gid.cimne.upc.es/mailman/listinfo/gidlist


_______________________________________________
GiDlist mailing list
GiDlist at gid.cimne.upc.es mailto:GiDlist at gid.cimne.upc.es
http://gid.cimne.upc.es/mailman/listinfo/gidlist

_______________________________________________
GiDlist mailing list
GiDlist at gid.cimne.upc.es mailto:GiDlist at gid.cimne.upc.es
http://gid.cimne.upc.es/mailman/listinfo/gidlist
User avatar
abelcs
Posts: 146
Joined: Mon Oct 13, 2014 9:14 am

[GiDlist] Semistructured volume

Post by abelcs »

Dear Isabel,

The name of the variable is TypeOfQuadMesher, not QuadrilateralMesher.

Regards,

ABEL

__________________________________________________________

Abel Coll Sans

CIMNE - International Center for Numerical Methods in Engineering

Office C2 - C1Building - Campus Nord UPC

Gran Capità s/n, 08034 Barcelona

Tfn: 34 - 93 401 74 03 Fax: 34 - 93 401 65 17

abelcs at cimne.upc.edu

_________________________________________________________


-----Mensaje original-----
De: gidlist-admin at gatxan.cimne.upc.edu
[mailto:gidlist-admin at gatxan.cimne.upc.edu] En nombre de Isabel Gil
Enviado el: jueves, 03 de mayo de 2007 13:52
Para: gidlist at gatxan.cimne.upc.edu
Asunto: Re: [GiDlist] Semistructured volume

Hi again,

I have tried to put in both, GiD8.1.2 in Windows and in GiD8.1.1b in
Linux escape escape escape utilities variables QuadrilateralMesher 0
but I have obtained the following error message: "Unknown variable name
QuadrilateralMesher".
Perhaps, in these versions, the variable was called in another way¿?

Thanks in advance
Isa


Enrique Escolano wrote:
GiD 8.1.2 is crashing because you are using the new quadrilateral mesher

You must set the old quadrilateral mesher (value=0)
escape escape escape utilities variables QuadrilateralMesher 0
And activate the align nodes option (value=1)
escape escape escape utilities variables AlignSemiStructuredNodes 1

As I said we are releasing a new 8.1.3 beta version in few days, with
quadrilateral mesher fixed bugs.

Regards

Enrique

----- Original Message -----
From: "Isabel Gil" misabel at litec.csic.es mailto:misabel at litec.csic.es
To: gidlist at gatxan.cimne.upc.edu mailto:gidlist at gatxan.cimne.upc.edu
Sent: Wednesday, May 02, 2007 2:29 PM
Subject: Re: [GiDlist] Semistructured volume

Hello!

Thank-you. I could finally find it.

However, when I try to "Generate mesh", the program, both GID version
8.1.2b in Windows and GID version 8.1.1b in Linux, breaks down, the
message is, in Linux, "Segmentation fault".
Could it be due to both of them are beta versions?

Thanks again for your time
Isa

Isabel Gil wrote:
Hi!

Thanks for your reply. I answer through your e-mail:

Enrique Escolano wrote:
I think that we didn't understand your meaning of
"irregularities in
the mesh"
I think that you don't want to say "elements with bad angles or
sizes", and what you want
is that all extruded nodes have exactly the same XY proyection.
Is not?

Yes, you are right. I meant just what you have described.
Structured or semi-structured meshes will not generate aligned-like
nodes in the extrusion direction, the algorithm is not a simple
translation, is more general (lateral surfaces can be curved, and
both tops must have the same number of boundaries, but can have
different shape)

Last 8.1.x beta versions have an "special option" to try to force
this alignement if possible, by setting the GiD variable
AlignSemiStructuredNodes to 1
escape escape escape utilities variables AlignSemiStructuredNodes 1
(but your 8.0.8 official version doesn't has this option)

Then, I have just downloaded Gid8.1.2b-win-x32.exe, which is a beta
version but I am not able to find the GiD variable
AlignSemiStructuredNodes and to put it to 1.
Could you say to me where it is exactly?
Thanks in advance and best regards
Isa
ftp://gid.cimne.upc.es/pub/gid_adds/Windows/GiD8.1.2b-win-x32.exe
Next week we are releasing a new beta version 8.1.3 that also has a
new unstructured quadrilateral meshing algorithm,
Obtained quadrilaterals will have much better shape quality. Can set
this mesher with the variable:
escape escape escape utilities variables QuadrilateralMesher 1

Regards

Enrique Escolano

----- Original Message ----- From: "Isabel Gil"
misabel at litec.csic.es mailto:misabel at litec.csic.es
To: gidlist at gatxan.cimne.upc.edu
mailto:gidlist at gatxan.cimne.upc.edu
Sent: Friday, April 27, 2007 6:12 PM
Subject: Re: [GiDlist] Semistructured volume



Hello!

First of all, thanks for your reply.

Your are right with the so different size of mesh elements I am
using,
but if you, in the previous project I sent you, put
"RENDER-normal"
and
put on and off the "layer capa2_a", you can see that different lines
appear (you can see this effect at first sight on the left at the
bottom
with only the "layer capa1_a" on and putting the "layer capa2_a"
on and
off, with the view of XY plane).
Do you mean that these irregularities in the mesh is due to my
bad size
election?

Again, thank-you very much in advance.

Best regards!
Isa

Miguel Pasenau wrote:

Hi,

If you mean that the elements are quite distorted, i think it's
because you
have assigned element size of '2' for some volumes (layers
'capa1_a' and
'capa1_b' with a height of 0.002) and '4' for others (layers
'capa2_a' and
'capa2_b' with a height of 0.002). Some details of the surfaces
conforming
the volumes is just 0.001. You can see it at
'ig_volume_size_1.png'.

And, for instance, at the layer 'capa1_b', there are surfaces with
element
size assigned of 0.0059 (and the width of the surface is just
0.002) and
just beside them surfaces with element size of 0.000325, 15 times
less, as
you can see in the image 'ig_surface_size_1.png', resulting in the
mesh
shown in 'ig_mesh_zoom_1.png'.
Best regards,

Miguel

-----Mensaje original-----
De: gidlist-admin at gatxan.cimne.upc.edu
mailto:gidlist-admin at gatxan.cimne.upc.edu
[mailto:gidlist-admin at gatxan.cimne.upc.edu] En nombre de Isabel Gil
Enviado el: jueves, 26 de abril de 2007 11:56
Para: GID_FORO
Asunto: [GiDlist] Semistructured volume

Hello GiD'users!

I have a little problem when I want to create a mesh.
I am working with GID 8.0.8 in Windows.


I attach my project because I don't get to make a good
semistructured
volume with structured mesh
in z-direction. You can see the irregularities of the mesh if
you use
the XY-plane view and the rotate trackball tool.

I don't know what I am doing wrong. I think I followed
correctly the
advice that previous e-mails
gave me but perhaps I am forgetting something.

I would be really grateful if you could cast a glance at it.

Thanks in advance and best regards
Isa




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




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




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



_______________________________________________
GiDlist mailing list
GiDlist at gid.cimne.upc.es mailto:GiDlist at gid.cimne.upc.es
http://gid.cimne.upc.es/mailman/listinfo/gidlist


_______________________________________________
GiDlist mailing list
GiDlist at gid.cimne.upc.es mailto:GiDlist at gid.cimne.upc.es
http://gid.cimne.upc.es/mailman/listinfo/gidlist


_______________________________________________
GiDlist mailing list
GiDlist at gid.cimne.upc.es mailto:GiDlist at gid.cimne.upc.es
http://gid.cimne.upc.es/mailman/listinfo/gidlist

_______________________________________________
GiDlist mailing list
GiDlist at gid.cimne.upc.es mailto:GiDlist at gid.cimne.upc.es
http://gid.cimne.upc.es/mailman/listinfo/gidlist

_______________________________________________
GiDlist mailing list
GiDlist at gid.cimne.upc.es
http://gid.cimne.upc.es/mailman/listinfo/gidlist
User avatar
escolano
Posts: 1918
Joined: Sun Sep 05, 1982 10:51 pm

[GiDlist] Semistructured volume

Post by escolano »

Hello Isabel,

Now is available the last 8.1.3b GiD beta version for Windows.
http://www.gidhome.com/download/do07.html

With this version your model can be also meshed with the new quadrilateral mesher.

Enrique
----- Original Message -----
From: "Abel Coll" abelcs at cimne.upc.edu
To: gidlist at gatxan.cimne.upc.edu
Sent: Friday, May 04, 2007 9:41 AM
Subject: RE: [GiDlist] Semistructured volume


Dear Isabel,

The name of the variable is TypeOfQuadMesher, not QuadrilateralMesher.

Regards,

ABEL

__________________________________________________________

Abel Coll Sans

CIMNE - International Center for Numerical Methods in Engineering

Office C2 - C1Building - Campus Nord UPC

Gran Capità s/n, 08034 Barcelona

Tfn: 34 - 93 401 74 03 Fax: 34 - 93 401 65 17

abelcs at cimne.upc.edu

_________________________________________________________


-----Mensaje original-----
De: gidlist-admin at gatxan.cimne.upc.edu
[mailto:gidlist-admin at gatxan.cimne.upc.edu] En nombre de Isabel Gil
Enviado el: jueves, 03 de mayo de 2007 13:52
Para: gidlist at gatxan.cimne.upc.edu
Asunto: Re: [GiDlist] Semistructured volume

Hi again,

I have tried to put in both, GiD8.1.2 in Windows and in GiD8.1.1b in
Linux escape escape escape utilities variables QuadrilateralMesher 0
but I have obtained the following error message: "Unknown variable name
QuadrilateralMesher".
Perhaps, in these versions, the variable was called in another way¿?

Thanks in advance
Isa


Enrique Escolano wrote:
GiD 8.1.2 is crashing because you are using the new quadrilateral mesher

You must set the old quadrilateral mesher (value=0)
escape escape escape utilities variables QuadrilateralMesher 0
And activate the align nodes option (value=1)
escape escape escape utilities variables AlignSemiStructuredNodes 1

As I said we are releasing a new 8.1.3 beta version in few days, with
quadrilateral mesher fixed bugs.

Regards

Enrique

----- Original Message -----
From: "Isabel Gil" misabel at litec.csic.es mailto:misabel at litec.csic.es
To: gidlist at gatxan.cimne.upc.edu mailto:gidlist at gatxan.cimne.upc.edu
Sent: Wednesday, May 02, 2007 2:29 PM
Subject: Re: [GiDlist] Semistructured volume

Hello!

Thank-you. I could finally find it.

However, when I try to "Generate mesh", the program, both GID version
8.1.2b in Windows and GID version 8.1.1b in Linux, breaks down, the
message is, in Linux, "Segmentation fault".
Could it be due to both of them are beta versions?

Thanks again for your time
Isa

Isabel Gil wrote:
Hi!

Thanks for your reply. I answer through your e-mail:

Enrique Escolano wrote:
I think that we didn't understand your meaning of
"irregularities in
the mesh"
I think that you don't want to say "elements with bad angles or
sizes", and what you want
is that all extruded nodes have exactly the same XY proyection.
Is not?

Yes, you are right. I meant just what you have described.
Structured or semi-structured meshes will not generate aligned-like
nodes in the extrusion direction, the algorithm is not a simple
translation, is more general (lateral surfaces can be curved, and
both tops must have the same number of boundaries, but can have
different shape)

Last 8.1.x beta versions have an "special option" to try to force
this alignement if possible, by setting the GiD variable
AlignSemiStructuredNodes to 1
escape escape escape utilities variables AlignSemiStructuredNodes 1
(but your 8.0.8 official version doesn't has this option)

Then, I have just downloaded Gid8.1.2b-win-x32.exe, which is a beta
version but I am not able to find the GiD variable
AlignSemiStructuredNodes and to put it to 1.
Could you say to me where it is exactly?
Thanks in advance and best regards
Isa
ftp://gid.cimne.upc.es/pub/gid_adds/Windows/GiD8.1.2b-win-x32.exe
Next week we are releasing a new beta version 8.1.3 that also has a
new unstructured quadrilateral meshing algorithm,
Obtained quadrilaterals will have much better shape quality. Can set
this mesher with the variable:
escape escape escape utilities variables QuadrilateralMesher 1

Regards

Enrique Escolano

----- Original Message ----- From: "Isabel Gil"
misabel at litec.csic.es mailto:misabel at litec.csic.es
To: gidlist at gatxan.cimne.upc.edu
mailto:gidlist at gatxan.cimne.upc.edu
Sent: Friday, April 27, 2007 6:12 PM
Subject: Re: [GiDlist] Semistructured volume



Hello!

First of all, thanks for your reply.

Your are right with the so different size of mesh elements I am
using,
but if you, in the previous project I sent you, put
"RENDER-normal"
and
put on and off the "layer capa2_a", you can see that different lines
appear (you can see this effect at first sight on the left at the
bottom
with only the "layer capa1_a" on and putting the "layer capa2_a"
on and
off, with the view of XY plane).
Do you mean that these irregularities in the mesh is due to my
bad size
election?

Again, thank-you very much in advance.

Best regards!
Isa

Miguel Pasenau wrote:

Hi,

If you mean that the elements are quite distorted, i think it's
because you
have assigned element size of '2' for some volumes (layers
'capa1_a' and
'capa1_b' with a height of 0.002) and '4' for others (layers
'capa2_a' and
'capa2_b' with a height of 0.002). Some details of the surfaces
conforming
the volumes is just 0.001. You can see it at
'ig_volume_size_1.png'.

And, for instance, at the layer 'capa1_b', there are surfaces with
element
size assigned of 0.0059 (and the width of the surface is just
0.002) and
just beside them surfaces with element size of 0.000325, 15 times
less, as
you can see in the image 'ig_surface_size_1.png', resulting in the
mesh
shown in 'ig_mesh_zoom_1.png'.
Best regards,

Miguel

-----Mensaje original-----
De: gidlist-admin at gatxan.cimne.upc.edu
mailto:gidlist-admin at gatxan.cimne.upc.edu
[mailto:gidlist-admin at gatxan.cimne.upc.edu] En nombre de Isabel Gil
Enviado el: jueves, 26 de abril de 2007 11:56
Para: GID_FORO
Asunto: [GiDlist] Semistructured volume

Hello GiD'users!

I have a little problem when I want to create a mesh.
I am working with GID 8.0.8 in Windows.


I attach my project because I don't get to make a good
semistructured
volume with structured mesh
in z-direction. You can see the irregularities of the mesh if
you use
the XY-plane view and the rotate trackball tool.

I don't know what I am doing wrong. I think I followed
correctly the
advice that previous e-mails
gave me but perhaps I am forgetting something.

I would be really grateful if you could cast a glance at it.

Thanks in advance and best regards
Isa




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




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




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



_______________________________________________
GiDlist mailing list
GiDlist at gid.cimne.upc.es mailto:GiDlist at gid.cimne.upc.es
http://gid.cimne.upc.es/mailman/listinfo/gidlist


_______________________________________________
GiDlist mailing list
GiDlist at gid.cimne.upc.es mailto:GiDlist at gid.cimne.upc.es
http://gid.cimne.upc.es/mailman/listinfo/gidlist


_______________________________________________
GiDlist mailing list
GiDlist at gid.cimne.upc.es mailto:GiDlist at gid.cimne.upc.es
http://gid.cimne.upc.es/mailman/listinfo/gidlist

_______________________________________________
GiDlist mailing list
GiDlist at gid.cimne.upc.es mailto:GiDlist at gid.cimne.upc.es
http://gid.cimne.upc.es/mailman/listinfo/gidlist

_______________________________________________
GiDlist mailing list
GiDlist at gid.cimne.upc.es
http://gid.cimne.upc.es/mailman/listinfo/gidlist


_______________________________________________
GiDlist mailing list
GiDlist at gid.cimne.upc.es
http://gid.cimne.upc.es/mailman/listinfo/gidlist
Isabel Gil

[GiDlist] Semistructured volume & concentrate elements

Post by Isabel Gil »

Dear GiD Team,

Thanks for your help, I mean the use of TypeOfQuadMesher and
AlignSemiStructuredNodes. Now I can create a mesh just I wanted.

However, if I want to concentrate elements in the structured direction
(z-direction), I obtain the same XY projection in all extruded nodes but
if I put plane XZ or plane YZ I don't obtain the same projection over
those planes (if I don't concentrate elements everything is right on
each plane).

Could you give me any ideas?

Thanks in advance
Isa

Abel Coll wrote:
Dear Isabel,

The name of the variable is TypeOfQuadMesher, not QuadrilateralMesher.

Regards,

ABEL

__________________________________________________________

Abel Coll Sans

CIMNE - International Center for Numerical Methods in Engineering

Office C2 - C1Building - Campus Nord UPC

Gran Capità s/n, 08034 Barcelona

Tfn: 34 - 93 401 74 03 Fax: 34 - 93 401 65 17

abelcs at cimne.upc.edu

_________________________________________________________
is that all extruded nodes have exactly the same XY proyection.

-----Mensaje original-----
De: gidlist-admin at gatxan.cimne.upc.edu
[mailto:gidlist-admin at gatxan.cimne.upc.edu] En nombre de Isabel Gil
Enviado el: jueves, 03 de mayo de 2007 13:52
Para: gidlist at gatxan.cimne.upc.edu
Asunto: Re: [GiDlist] Semistructured volume

Hi again,

I have tried to put in both, GiD8.1.2 in Windows and in GiD8.1.1b in
Linux escape escape escape utilities variables QuadrilateralMesher 0
but I have obtained the following error message: "Unknown variable name
QuadrilateralMesher".
Perhaps, in these versions, the variable was called in another way¿?

Thanks in advance
Isa


Enrique Escolano wrote:

GiD 8.1.2 is crashing because you are using the new quadrilateral mesher

You must set the old quadrilateral mesher (value=0)
escape escape escape utilities variables QuadrilateralMesher 0
And activate the align nodes option (value=1)
escape escape escape utilities variables AlignSemiStructuredNodes 1

As I said we are releasing a new 8.1.3 beta version in few days, with
quadrilateral mesher fixed bugs.

Regards

Enrique

----- Original Message -----
From: "Isabel Gil" misabel at litec.csic.es mailto:misabel at litec.csic.es
To: gidlist at gatxan.cimne.upc.edu mailto:gidlist at gatxan.cimne.upc.edu
Sent: Wednesday, May 02, 2007 2:29 PM
Subject: Re: [GiDlist] Semistructured volume


Hello!

Thank-you. I could finally find it.

However, when I try to "Generate mesh", the program, both GID version
8.1.2b in Windows and GID version 8.1.1b in Linux, breaks down, the
message is, in Linux, "Segmentation fault".
Could it be due to both of them are beta versions?

Thanks again for your time
Isa

Isabel Gil wrote:

Hi!

Thanks for your reply. I answer through your e-mail:

Enrique Escolano wrote:

I think that we didn't understand your meaning of

"irregularities in

the mesh"
I think that you don't want to say "elements with bad angles or
sizes", and what you want
is that all extruded nodes have exactly the same XY proyection.

Is not?



Yes, you are right. I meant just what you have described.

Structured or semi-structured meshes will not generate aligned-like
nodes in the extrusion direction, the algorithm is not a simple
translation, is more general (lateral surfaces can be curved, and
both tops must have the same number of boundaries, but can have
different shape)

Last 8.1.x beta versions have an "special option" to try to force
this alignement if possible, by setting the GiD variable
AlignSemiStructuredNodes to 1
escape escape escape utilities variables AlignSemiStructuredNodes 1
(but your 8.0.8 official version doesn't has this option)


Then, I have just downloaded Gid8.1.2b-win-x32.exe, which is a beta
version but I am not able to find the GiD variable
AlignSemiStructuredNodes and to put it to 1.
Could you say to me where it is exactly?
Thanks in advance and best regards
Isa
ftp://gid.cimne.upc.es/pub/gid_adds/Windows/GiD8.1.2b-win-x32.exe

Next week we are releasing a new beta version 8.1.3 that also has a
new unstructured quadrilateral meshing algorithm,
Obtained quadrilaterals will have much better shape quality. Can set
this mesher with the variable:
escape escape escape utilities variables QuadrilateralMesher 1

Regards

Enrique Escolano

----- Original Message ----- From: "Isabel Gil"

misabel at litec.csic.es mailto:misabel at litec.csic.es

To: gidlist at gatxan.cimne.upc.edu

mailto:gidlist at gatxan.cimne.upc.edu

Sent: Friday, April 27, 2007 6:12 PM
Subject: Re: [GiDlist] Semistructured volume




Hello!

First of all, thanks for your reply.

Your are right with the so different size of mesh elements I am

using,

but if you, in the previous project I sent you, put

"RENDER-normal"

and
put on and off the "layer capa2_a", you can see that different lines
appear (you can see this effect at first sight on the left at the
bottom
with only the "layer capa1_a" on and putting the "layer capa2_a"

on and

off, with the view of XY plane).
Do you mean that these irregularities in the mesh is due to my

bad size

election?

Again, thank-you very much in advance.

Best regards!
Isa

Miguel Pasenau wrote:


Hi,

If you mean that the elements are quite distorted, i think it's
because you
have assigned element size of '2' for some volumes (layers
'capa1_a' and
'capa1_b' with a height of 0.002) and '4' for others (layers
'capa2_a' and
'capa2_b' with a height of 0.002). Some details of the surfaces
conforming
the volumes is just 0.001. You can see it at

'ig_volume_size_1.png'.

And, for instance, at the layer 'capa1_b', there are surfaces with
element
size assigned of 0.0059 (and the width of the surface is just
0.002) and
just beside them surfaces with element size of 0.000325, 15 times
less, as
you can see in the image 'ig_surface_size_1.png', resulting in the
mesh
shown in 'ig_mesh_zoom_1.png'.
Best regards,

Miguel

-----Mensaje original-----
De: gidlist-admin at gatxan.cimne.upc.edu

mailto:gidlist-admin at gatxan.cimne.upc.edu

[mailto:gidlist-admin at gatxan.cimne.upc.edu] En nombre de Isabel Gil
Enviado el: jueves, 26 de abril de 2007 11:56
Para: GID_FORO
Asunto: [GiDlist] Semistructured volume

Hello GiD'users!

I have a little problem when I want to create a mesh.
I am working with GID 8.0.8 in Windows.


I attach my project because I don't get to make a good

semistructured

volume with structured mesh
in z-direction. You can see the irregularities of the mesh if

you use

the XY-plane view and the rotate trackball tool.

I don't know what I am doing wrong. I think I followed

correctly the

advice that previous e-mails
gave me but perhaps I am forgetting something.

I would be really grateful if you could cast a glance at it.

Thanks in advance and best regards
Isa





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




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




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



_______________________________________________
GiDlist mailing list
GiDlist at gid.cimne.upc.es mailto:GiDlist at gid.cimne.upc.es
http://gid.cimne.upc.es/mailman/listinfo/gidlist


_______________________________________________
GiDlist mailing list
GiDlist at gid.cimne.upc.es mailto:GiDlist at gid.cimne.upc.es
http://gid.cimne.upc.es/mailman/listinfo/gidlist


_______________________________________________
GiDlist mailing list
GiDlist at gid.cimne.upc.es mailto:GiDlist at gid.cimne.upc.es
http://gid.cimne.upc.es/mailman/listinfo/gidlist

_______________________________________________
GiDlist mailing list
GiDlist at gid.cimne.upc.es mailto:GiDlist at gid.cimne.upc.es
http://gid.cimne.upc.es/mailman/listinfo/gidlist


_______________________________________________
GiDlist mailing list
GiDlist at gid.cimne.upc.es
http://gid.cimne.upc.es/mailman/listinfo/gidlist


_______________________________________________
GiDlist mailing list
GiDlist at gid.cimne.upc.es
http://gid.cimne.upc.es/mailman/listinfo/gidlist

Post Reply