yield line analysis for web connection in direct tension

huangyhg

yield line analysis for web connection in direct tension
hi to all,
i have stumbled upon an article on aisc by richard h. kapp (engineering journal 2nd quarter 1974). my question is related to the applications of the yield line analysis for local bending in column webs. in what situations does one consider the column flanges to be fixed/pinned? in other derivations that i have seen (for example one by mr. muir,
like many analyses in our profession, the consideration of whether to apply pin-pin or fixed edge restraints to this application of johanssen comes down to engineering judgement.  i've heard arguments that the web should be treated fixed-fixed due to the welds to the flanges (but what about flange rotation in this case?), as well as arguments about applying a more conservative pin-pinned approach (but then what about a well latterally braced situation?
my honest opinion:  if your forces do not cause much rotation of your "support" when applied at a reasonable eccentricity (say 1/4 flange width), then consider your supports to be reasonably fix-fixed.  otherwise go pin-pin.
note:  i've seen a break-down by class of sections (ie: class 1 pin-pin, class 3 fix-fix, class 2 by judgement with a rotation suggestion), but i don't buy it... this really always needs to be considered as part of how stable and stiff your supports should be.
hope that helps,
cheers,
ys
b.eng (carleton), p.eng (ontario), mipenz (structural-new zealand)
working in canada, and missing my adoptive new zealand family... at least i brought the little kiwi with me!
my practice has always been to assume fixed edges for hss members and pinned edges for wide flange shapes. this is because the flanges of a wide flange can rotate, as has been pointed out.
i think you may find others in the literature who also suppoort this practice.