Long story short, I have a very simple model that consists of a single angle with a compressive force on it. The member is a 6x6x9/16. I am unable to derive the same unity as STAAD when using AISC 360-05 (360-10 gives same answer). I am getting 2.97 and STAAD is getting 2.43. Axial load is 71.7 kips.
I applied the ASD for the METHOD and TRACK 2 to see how STAAD was arriving at the 2.43 unity. The output showed that the only code check with a unity ratio was for "Check for Compression," which was expected, and that unity was 2.43. After digging into the output more, it became apparent the difference between my calcs and model was the KL/r ratio. Looking at E5(a) and E5(b) in AISC, I was (and still am) convinced that my situation falls under E5(a) since this is an "individual member." Given that the L/r ratio is greater than 80, I used equation E5-2 to get a KL/r of 202, or 200 due it being the max. However, STAAD's KL/r is listed as 181.19 in the compression check. I was finally able to find that STAAD was actually using equation E5-4, which is in E5(b).
My question is why did STAAD use section E5(b)? Is it a default? It seems like E5(a) would be the default if there was one. Am I missing something? Let me know what you guys think.
Answer Verified By: Tim Bordelon
Same question, different program. It seems RAM Elements either uses E5(b) which is unconservative (planar box unchecked, default) or uses KL/rz which is too conservative (planar box checked), is there a way to have it apply E5(a), when designing an individual single angle member.
I came across this post when I had a similar problem as Tim. I was comparing STAAD's results to Table 4-11 in AISC 14th ed. and found STAAD to be unconservative, at least compared to the table. I got the same value as the table doing it by hand. Like Tim, I recognized that KL/r was the culprit and Sye's reply confirmed what is going on.
I can't find the parameter to change how it computes KL/r. OK, I'm razzing you a little, I'm pretty sure it's not there. If it is let me know because it is really needed.
I really don't agree with using the least conservative approach. Someone could do moment releases instead of using the TRUSS command and STAAD would miss that it should be using E5(a). Also, an angle could not meet the criteria for using this section (compression through same leg, attached by welding or at least two bolts, no intermediate transverse loads) but there is no way to tell STAAD not to use this section.
Speaking of it not meeting the criteria, I found AISC a bit confusing on this point. It says to use Chapter H, but Chapter H doesn't tell you how to compute KL/r. I'm assuming they mean just use the minor axis radius of gyration to compute KL/r (assuming it's not braced about the weak axis) but that's not abundantly clear to me.
One final point, and I've probably said this ad nauseam, but us users really need to know what is happening behind the curtain. It would be nice if decisions like this that the program is making were documented.