[GiDlist] problems meshing section of building envelope

GiDList Archive 2013 - 2005
1

Hi all;

I’m interested in modelling the heat-transfer characteristics of several
building envelope designs. I’m attempting to use Elmer to do this, with GiD
for the mesh.

However, I’m having some serious problems creating a mesh. There are two
elements of the construction (wall studs and z-girts, which are used to hold
up insulation) that have very small cross-sections in directions orthogonal
to one another and these seem to be the source of most of my problems.

I’ve attempted to do an unstructured mesh of the entire assembly with a very
small element size (1) with no success (ie, I get messages telling me the
mesh creation failed). I’ve also tried just setting the sizes of potentially
problematic elements of the construction (ie, the studs and z-girts) to this
extremely small size and used more moderate sizes for the rest of the
assembly, and this also hasn’t worked. Finally, I’ve attempted to use a
structured orthogonal mesh for the whole assembly, which seems as if it
would work quite well because there are no non-orthogonal angles in the
entire assembly, but a number of the volumes seem to resist being meshed in
this fashion, with the “Centered Structured Mesh” dialogue popping up and
non-orthogonal mesh resulting (and failing).

So, I basically don’t know what to do. I’m wondering if anyone has modelled
anything similar to this, or could offer any hints regarding the approach I
should take.

Many thanks,

Josh

–
Josh Kjenner, P.Eng., LEED AP
Manasc Isaac Architects Ltd.
10225 100 Avenue :: Edmonton, AB T5J 0A1
780.429.3977
-------------- next part --------------
An HTML attachment was scrubbed…
URL: http://listas.cimne.upc.edu/pipermail/gidlist/attachments/20100119/76a1b896/attachment.htm

2

As you said, very thin parts are problematic to be meshed and could generate a lot of elements,
probably meshing structuredly of semi-structuredly some parts is a good option.

You can split the problematic parts to isolate the problem: ‘divide and conquer’

Send us the model to have a look (if you want directly to my e-mail instead the whole GiDList)

Regards

Enrique Escolano
GiD Team

----- Original Message -----
From: Josh K
To: gidlist at gatxan.cimne.upc.edu
Sent: Tuesday, January 19, 2010 1:08 AM
Subject: [GiDlist] problems meshing section of building envelope


Hi all;

I’m interested in modelling the heat-transfer characteristics of several building envelope designs. I’m attempting to use Elmer to do this, with GiD for the mesh.

However, I’m having some serious problems creating a mesh. There are two elements of the construction (wall studs and z-girts, which are used to hold up insulation) that have very small cross-sections in directions orthogonal to one another and these seem to be the source of most of my problems.

I’ve attempted to do an unstructured mesh of the entire assembly with a very small element size (1) with no success (ie, I get messages telling me the mesh creation failed). I’ve also tried just setting the sizes of potentially problematic elements of the construction (ie, the studs and z-girts) to this extremely small size and used more moderate sizes for the rest of the assembly, and this also hasn’t worked. Finally, I’ve attempted to use a structured orthogonal mesh for the whole assembly, which seems as if it would work quite well because there are no non-orthogonal angles in the entire assembly, but a number of the volumes seem to resist being meshed in this fashion, with the “Centered Structured Mesh” dialogue popping up and non-orthogonal mesh resulting (and failing).

So, I basically don’t know what to do. I’m wondering if anyone has modelled anything similar to this, or could offer any hints regarding the approach I should take.

Many thanks,

Josh

–
Josh Kjenner, P.Eng., LEED AP
Manasc Isaac Architects Ltd.
10225 100 Avenue :: Edmonton, AB T5J 0A1
780.429.3977
-------------- next part --------------
An HTML attachment was scrubbed…
URL: http://listas.cimne.upc.edu/pipermail/gidlist/attachments/20100119/2f7a6c65/attachment.htm