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!

MPC_RBE2 vs. FASTENERS SIMULATION

Status
Not open for further replies.

051174

New member
Jun 23, 2006
20
Can MPC RBE2 with 6 d.o.f. be used correctly to simulate bolt in FEM instead of bar and other things.....The mpc shape i'm talking is to create from a 3d model an indipendent node at one face and all dipendent connected in an umbrella shape...the other face of the same solid the other node and the other solid nodes (hole nodes all around instead )....
| | /|o
| |/ |
| | |
| /| |
|/_|o |
|\ | |
| \| |
| \ |
| |\ |
| | \|o ?????
 
Replies continue below

Recommended for you

Remember the RBE2 MPC will be infinitely stiff and will create local stress/strain concentrations. Take these into account when you interpet your results.

Jxc
 
Following jxc's line of thought; if the stress at the hole is the important thing for you to calculate (that is, your goal is the accurate computation of the stress field at the hole), then these MPC RBE2 if they are indeed 'infinitely stiff' (I can't vouch for that since I don't use this software), you will not be able to use them and still reach a goal of computing accurate stress fields at the holes (if that's your goal). If however all you are interested in is getting the load transfer correct, you might be able to use these MPC RBE2. BTW, what FEA software is this?
 
I recommend a combination of beam and rbe2 elements.
I'm using a beam to model the bolt, and an RBE connection of the bolt to the shell elements.
this way the connection is stiff, but the bolt has the real material properties.

for extremely precise calculations, model in 3D bolt and body, but I think it's overkill.
 
I recommend NOT using an RBE2 and use instead a CBUSH (pref zero length). You then have the freedom to calculate equivalent fastener stiffness which will result in more realistic load transfer.

See also post about CBUSH rotational stiffness (dated today).
 
Hello,

You can't use a rigid body element (infinitely stiff) to model a bolt (finite stifness). The load path will be wrong.

IF you don't (or can't) use continuum elements, you can define a rigid body on both solids and linked each master node of the two rigid bodies by a bush element (to simulate a prestressed bolt).

An easy way to choose among the different proposals would be to compare 3D model and "bush" models especially if you often simulate the same bolt connections.

Regards,

Torpen
 
Status
Not open for further replies.

Part and Inventory Search

Sponsor