🎉Community Raffle - Win $25

An exclusive raffle opportunity for active members like you! Complete your profile, answer questions and get your first accepted badge to enter the raffle.

Join and Win

Connect offset Surfaces

Alex_Eberhard

Hello everyone,

I'm struggling to properly connect offset surfaces. I've separated the solid into components of different thicknesses, extracted the midsurfaces and connected the elements with rigids (see attached picture).

FE-Analysis shows different results compared to a solid model and stress jumps at the connection points.

Can anyone suggest me a better idea, how to connect these surfaces?

Thanks!

<?xml version="1.0" encoding="UTF-8"?>

Find more posts tagged with

English

OptiStruct

Accepted answers

All comments

Rahul Rajan_21763

Could you please share the screenshot of actual geometry?

Alex_Eberhard

Here's what the solid looks like.

<?xml version="1.0" encoding="UTF-8"?>

QuyNguyenDai

Sometime, life is easier with solid meshing /emoticons/default_tongue.png' srcset='/emoticons/tongue@2x.png 2x' title=':P' width='20' />

Alex_Eberhard

Sometime, life is easier with solid meshing /emoticons/default_tongue.png' srcset='/emoticons/tongue@2x.png 2x' title=':P' width='20' />

Haha yes, totally agree.

But unfortunately I must use shell elements ...

Rahul Rajan_21763

Any specific reason for using shell? You could also try RSSCON – Shell-to-Solid Element Connector.

Would be good if you could share the hm file.

Alex_Eberhard

Hello Sir,

the reason I have to use shell elements is because I have to size optimize the element thicknesses of this model (for a university project).

Attached you can find the files (solid and hm shell model).

Thanks in advance for your help!

Unable to find an attachment - read this blog

Rahul Rajan_21763

I think SPC is creating a problem.You should constrain all dof for SPC and use RBE2 rather constraining edges directly.

Alex_Eberhard

Hello,

thank your for the tip, unfortunately this delivers the same results after trying it out.

I think the problem is, that rigids insert too much stiffness into the model.

I've tried out different approaches and came to the conclusion, that using several rbe3 elements distributed symmetricaly at the connection points deliver the best results for my problem. This is not perfect, but good enough for my issue.

The model is a bit stiffer than the solid model. The deviaton value for displacement is 2% and for the vM-stress only 0,03%.

Maybe this can help someone who has the same problem.

Nevertheless, thanks for your help!

Rahul Rajan_21763

Thanks for sharing.