Defining composites in ASYS

[Tudor Miron]

Does anybody have experience defining composites in ANYS? Laminates? Sandwich panels?
Any help will be highly appreciated.

Thank you
Ted

[Tudor Miron]

Does anybody have experience defining composites in ANYS? Laminates? Sandwich panels?
Any help will be highly appreciated.

Thank you
Ted

[MikeWaggoner at UW]

I think it was shell 89 or 92. If there's an alternative between one in the 80's and 90's, go with the one in the 90's. Try looking at the help file for shell

[Travis Garrison]

Ted, your question should really be:

"Does anybody have experience succesfully defining composites in ANYS?"

Don't expect fantastic results...I would encourage you to do lots of physical testing and make sure you can get in the neighborhood of a correct result before using ansys for safety equiv, or chassis stiffness predictions...

Best of luck to you though...

-Travis Garrison

[Tudor Miron]

Yep Travis,
I had to write it this way http://fsae.com/groupee_common/emoticons/icon_smile.gif

I saw a post from Kevin Hall that they had numerical results within 10% with experimental results. This is good enough for now (if one takes a safety margin) How ever I can't find him. Email in his profile is not working anymore.

Thanks
Ted

[gug]

keep in mind that Kev is probably talking about his fancy shmansy method of laying up. the final product using this method would be very consistant.

(edit: whoops, wrong Kev. was thinking of Kev from WA. sorry.)

what process/materials are you using? wet lay-up, prepreg, autoclave curing? depending on your process, IMHO ANSYS analysis could be pretty useless. any inconsistancies in the final product wouldnt be modeled. if i was in your position i would spend that time making test pieces and practising laying up.

but im also not particularly experienced in this area yet, just trying to learn.

how many other people out there have done numerical vs experimental analysis? i would be really interested if you could post the results and testing and construction methods.

[Kevin Hall]

We used standard ANSYS elements, and compared the results to those of a physical test with aluminum honeycomb panels, and they were VERY accurate.

[Travis Garrison]

Kevin,

By standard do you mean the shell elements Mike Waggoner refered to?

-Travis

[Kevin Hall]

The element of choice was Shell92. It worked very well. It took some learning to figure out, but I made a go of it

[Gumby]

If you are modeling carbon/honeycomb you have to keep in mind that the both the carbon fibre and the honeycomb are anisotropic materials. You can find equations for an equivalent model for honeycomb in books like 'Cellular Solids' which goes through everything from woods to honeycombs to foams. Another good book is 'Honeycomb Technology' again it has very similar information.

Just something to keep in mind, if you are planning on working with composites. Weave and honeycomb direction of your sandwich panel can give you more error than you are expecting between theoretical and experimental.

Adam Ballard
Ryerson University (http://www.ryerson.ca/~fsae)