You are on the right track.

Make sure your single angle satisfied the E5 requirements such as being connected through the same leg at each end with two or more bolts, etc.

Compute the modified L/r in E5 for flexural buckling. Compute Fcr per E3 and/or E4.

Determine the local buckling classification using Table B4.1a. That will decide if you need to deal with E7.

IIRC, the AISC Manual Companion Vol 1 has an example calculation.

Thank you @271828. I guess that's the way to do it. I'm doing an internship and updating a calculation sheet for single angles under compression and other forces. This had me struggling because my boss told me that. in practice, single angles in compression must be checked for flexural buckling (Section E3 of AISC 360-10) with the minor principal axis "V".

This translates in calculating the elastic stress Fe with (Kv·Lv/rv) right? But if I do it as we discussed above, first calculating the modified Lc/r ... i would never use the "minor axis" properties.

As far as I understand, when you compute the modified Lc/r on Section E5 of AISC 360-10 , it only depends on L/rx , where rx is the radius of gyration about the geometric axis parallel to the connected leg, and L is the length of member between work points at truss chord centerlines.

So If i go back to chapter E3, the Fe stress would be computed without anything related to the minor axis "V".. Is it okay? Also: When reading the Section E7 (local buckling) of AISC 360-10, it says:


" Fe = elastic buckling stress, calculated using Equations E3-4 and E4-4 for doubly symmetric members, Equations E3-4 and E4-5 for singly symmetric mem- bers, and Equation E4-6 for unsymmetric members, except for single angles with b/t ≤ 20, where Fe is calculated using Equation E3-4. "

I understand that for E7 I should use the Fe calculated with the equation E3-4, but using the modified Lc/r right? My problem is on Section E4 (torsional-flexural buckling). I should calculate Fe with the provided equations, but then im confused again about if i should use the modified Lc/r or not, and where to include it there...

I will check the example that you told me, I think i've alredy seen it , also i checked the SAP2000 manual, and i haven't seen them using modified Lc/r to calculate Fe ( i think )

Thanks for the help, it is hugely appreciated !

There are two approaches:

Rigorous General Approach. Compute Pn using E3, E4, and E7 if slender per Table B4.1a. Buckling modes are flexural buckling (z-axis controls for typical cases) and flexural-torsional buckling. Compute Mn also. For an angle with typical connections through one leg, Pu is at an eccentricity relative to the centroid so it causes a moment also. This moment goes into the overall bending moment diagram and Mu is established. Take Pu, phiPn, Mu, and phiMn over to the interaction equations in Chapter H for evaluation. Vol 1 of the v15 Manual Companion example E.14 covers this.

Simplified Approach in E5. See the Commentary for some discussion. The purpose is to cut out some of the steps listed above. This method only applies when the five scope limitations in E5 are met. This is a prescriptive procedure. Typically, you can get out of considering flexural-torsional buckling. Check flexural buckling using the effective slenderness ratio, L/ra, not L/rz. E7 kicks in if either leg is slender. No need to consider eccentricity.