You might try sending your model into Tech Support. You don't even say what type of connection you're talking about. So, it's really hard to answer a question like this in detail when the only information given is generalities.

JoshPlum,

I appreciate the suggestion, however I am using a Demo copy to evaluate the software for personal use and this was one of the items that I was looking at (for a Girder-Beam Single Angle Connection). That is the reason I didn't send any of this information to RISA as I figured someone may just have a better understanding of evaluating out of plane (tension) forces on the weld, regardless of using RISA or not.

First off, you're talking at a single angle connection welded to the girder, correct? And, you're talking about putting it in tension.

I would recommend against that. At least for any significant amount of tension. This connection is intended to be a shear connection. There is another thread on eng-tips where this issue was discussed (though in a painfully long, round-about way). The result is that the next version (version 5.0) of RISAConnection is going to add some restrictions on how much tension is actually allowed.

Currently, when tension is detected by the program, the weld configuration automatically changes (otherwise, it wouldn't be stable). The method used in the program switches to an Elastic Method similar to that described on pages 8-12 through 8-14 of the AISC manual. The demand on the weld due to the out of plane loading (referred to as ro) is based on the elastic theory of welds. So, you just calculate section modulus and such and come up with a weld demand. I don't know if Blodgett has and example like this or not. But, that's probably the best reference that i know of for the elastic theory of welds.

Again, if you sent your question into tech support with a model then I'm sure you could get a more complete response with actual numbers and such.