AISC Specification for Structural Steel Buildings - 2016 - Chapter E 2

[penpe]

W6X15 column has a slender element: the web. That sends me to E7: members with slender elements. Second paragraph reads: Nominal compressive strength shall be the lowest value based on the applicable limit states of flexural buckling, torsional buckling, and flexural-torsional buckling in interaction with local buckling. But where to check those values, other than E4: Torsional and Flexural-Torsional Buckling of Single angles and members WITHOUT slender elements; and E3: Flexural buckling of members WITHOUT slender elements ?

 

[JoshPlumSE]

You use those sections, but only to determine Fcr. Then you come back into E7 to get Aeffective. And, your capacity becomes Fcr*Ae instead of Fcr*Ag like it would have been in the non-slender sections.

 

[penpe]

Thanks, Josh. But E4 sends me back to equations E3-2 and E3-3 to determine Fcr again. Seems pretty convoluted - makes me think I'm missing something.

 

[JoshPlumSE]

That does feel a little convoluted, but it's correct.

1) Use E3 by itself to get an F_cr for pure Euler Buckling. Call this F_cr_euler
2) If you have a WT or Double angle (which are subject to FTB) then use E4 to calculate an Fe for flex torsional Buckling. Use this Fe in the equations for E3 to come up with an F_cr for the FTB limit state. Call it F_cr_ftb
3) Take the controlling case from items 1 and 2 back into E7 and use it to get your effective area and the final capacity.

Note: Technically item 2 is required for all shapes, but it really only become an issue for singly symmetric shapes like WT's or LL's or unsymmetric shapes like single angles. So, it's almost always safe to ignore it for wide flanges like you have. The only time it comes into play (in theory) is for torsional buckling of wide flanges. But, I don't believe this limit state has ever been reproduced in laboratory testing with a standard WF. I believe the commentary in some of the older codes says essentially the same thing.

 

[Hokie93]

Determine F_e as indicated in Section E7 (see definition of F_e). That is, use Eq. (E3-4) for flexural buckling and Eq. (4-4) for torsional buckling. Flexural-torsional buckling does not apply to a doubly-symmetric shape. Take the lesser of the values from Eqs. (E3-4) and (E4-4) and plug into the appropriate equation from Eqs. (E7-2) and (E7-3) to get F_cr. Determine Q (really Q_a, in your case since the web is a stiffened element) and then determine P_n from Eq. (E7-1).

 

[JoshPlumSE]

Hokie -

I haven't reviewed it in detail, but at first glance, your method appears to be correct for the 2010 spec / 14th edition.

PenPe's question, was related to the 2016 spec / 15th edition, where the procedure no longer uses Q factors and takes some getting used to.

 

[Hokie93]

Josh; that is correct. My response was based on AISC 360-10, not AISC 360-16. I missed that the OP was inquiring about the 2016 Specification.

 

[penpe]

Thanks, guys! Your input helps me to have more confidence in my results. The W6x15 turns out to be more than adequate. Probably could go smaller/lighter, but W6 is good for practicality in this application.