splice/joint analysis using FE model with shell elements alone

[rptr]

Hi,

I am working on a splice analysis, which is not modeled explicitly in the FEM. All I have are a couple of CQUAD elements, which look like continuous skin. I overlapped the geometry on to the FEM in order to identify the location of the splice and only for those elements extracted the running loads from the shells. Now, what is the best way to analyze the rivets (shear, bending, tension etc.) for this kind of modeling?
Given:
1. The running loads in shell elements Nxx, Nyy etc in, lbs/in.
2. The actual fastener pitch p in, in.
3. No. of fasteners n
4. Element lengths le in, in.
5. There is no intent to model the fasteners by one dimensional elements or MPCs
Any suggestions would be greatly appreciated!

Thanks

 

[GrandpaDave]

Just reading your question above it seems to me you need to collect the lineloads from your global loads model(s).
The nodal forces, which you appear to be talking about, have no meaning here. Toss them out. Then conduct a classical
hand stress analysis (See E. Bruhn, D. Peery, M. Niu, etc. If the joint is a riveted shear splice there should not be
any significant tension or bending in the rivets. The allowables are extracted from tables in MIL-HDBK-5 (MMPDS?)which account
for any pretensions or bending in the rivets. G-pa Dave

 

[rb1957]

using the model in-plane forces (shear flows, in-plane tension) are straight forward; bending moments are a little questionable.

there are plenty of texts that address compliance models (for calculating the load transferred on different rivet rows.

be careful about fatigue design (the splice joint should have lower design allowables than continuous structure).

Quando Omni Flunkus Moritati