Modelling a Thin Solid

[HVSmith]

Hello all

I'm a new user to ANSYS (and this forum), and a physicist by trade, so apologies in advance for any blatant engineering/ANSYS mistakes being made...

It is my task to create a thermal model. I have been having problems trying to create a mesh over the object I must model. The object is very thin in comparison with it's area. It is rectangular in shape, 110mm x 40mm, yet only 1 micron in thickness.

Initially I attempted this in Workbench, but have since been advised to move to ANSYS "Classic" and I have read and worked through many tutorials for both, but am still unable to find any solution.

The reading I have done indicates that maybe a shell-model is the way to go, specifically Shell 131 for the thermal study. But when I try to mesh using this it states that a volume is not found. However, I understand that shell-models are primarily 2D and the thickness of the object is important to my studies, so really a 3D model is what I require!

So, if anyone is able to share any wisdom about how I may start to go about modelling such an object successfully I would be very grateful!

Apologies if the answer is trivial, or if I have missed out any important details that may be required...

Thank you in advance, for any knowledge you are able to impart.

HVSmith

 

[Drej]

This is a 2D planar element and therefore require areas to be present for a mesh to be generated. Strange to see that error message therefore. The element sounds applicable to your needs given your requirements ie a highly slender structure, with the SHELL131 giving you a layered through-thickness thermal capability not present in a standard thermal shell.

To be honest, I wouldn't want to go down the route of using solids for this structure.


------------
See faq569-1083 for details on how to make best use of Eng-Tips.com

 

[HVSmith]

OK, thanks for your reply. I have since read about the qualities of Shell57 and it is with this I am progressing at present. Setting the thickness as a real constant and defining the rectangular area as before. I hope this is still valid? Or is it a mistake to switch from Shell131 to Shell57?

(As I say, I am incredibly new to this and results are required by next week(!) and despite doing much reading of the help files and various tutorials I still find much of the content incomprehensible - so thank you in advance for your continued help!!!)

HVSmith

 

[Drej]

SHELL57 is the same as SHELL131 without the layered capability (you can 'convert' a 131 to a 57 by setting KEYOPTs in the model if you need to). If the layered capability - ie through-thickness - is important then you need to include in your analysis, else simply consider using solids. For the structure you describe, however, I would imagine that shells is the appropriate way to go, and that 57 would be suitable for your needs.

Hope it all goes well.


------------
See faq569-1083 for details on how to make best use of Eng-Tips.com

 

[flash3780]

Workbench/Mechanical should be able to handle this with no problems. There are tons of great (and new) meshing tools available. The trick is, you need to let Ansys know if you intend for the model to be 2D or 3D. You also may need to slice up your geometry and glue it back together (see Multibody Parts) to get it to mesh properly (that applies to both Mechanical and APDL). Often the meshing software isn't able to deal with larger, more complex geometries as well as it is able to deal with groups of simpler geometries.