Continue to Site

Eng-Tips is the largest engineering community on the Internet

Intelligent Work Forums for Engineering Professionals

  • Congratulations cowski on being selected by the Eng-Tips community for having the most helpful posts in the forums last week. Way to Go!

Joint modeling

Status
Not open for further replies.

pylandjw

Petroleum
Mar 30, 2007
2
Hi all,

We are constructing an FEA model of a structural steel frame. The members are custom built from steel plate. They are bolted together with custom built flanges also made of steel plate.

We have constructed a shell model in ANSYS using shell181 elements. We are having difficulty modeling the connections, and I am looking for suggestions on how to treat them.

Our first approach was to model both halves of the flange and connect the plates together with beam elements. There was no contact between the plates. This method produces very high stresses in the flanges and the nearby members. By bonding the plates together, the stresses decrease by 3 or four times, and this result matches more closely to simple hand calcs. But we will also need to get the loads on the bolts in the joint for a bolted joint type of calculation such as from Shigley and Mischke.

Any suggestions would be appreciated.
 
Replies continue below

Recommended for you

Check this thread for more information about bolted joints. thread727-169039
 
As far I understood you have used beam elements to connnect 2 flanges which means these beam elements represents bolts, is this true? If so, what kind of elements did you take to connect these beam elements? I would suggest to use rigid elements, and I hope you have understood what I mean. If not,assume you have a bunch of lotus flowers and keep all them in a row. Centre of each flower is connected by a rigid element the rest are beam elements which connects to the cylinder surface of the flange hole. This should give you good results..

good luck..
Tobias
 
Status
Not open for further replies.

Part and Inventory Search

Sponsor