A discussion which might related to recent roof collapse

[chrislaope]

I have recently done some research and analysis on metal building rigid frame under snow load. I found there are 2 issues which I would like to bring here to discuss with you guys:

1. I understand that AISC-13th does not explicitely require to check the I-Beam web buckling or crippling strength under compression at suspended beam portion. It only request to check web buckling or crippling strength under compression at bearing location. However, to me, I think it is an item which needs to be checked, because to guarantee the flexural strength of the whole beam under applied load, the web of the I-Beam at every cross section has to be able to resist applied local vertical load without local buckling or crippling.

2. I understand that it is a common practice (at least in U.S.) that for most built-up I-Beam, only single sided weld is requested as is shown in Figure 1.a. in attached document. To me, with single sided weld, the web can only considered as pinned at both ends when checking its compression buckling capacity, only when double sided weld, then the web can be considered as fixed at both ends when checking its compression buckling capacity under local vertical load (Figure 1.b.).

If above 2 issues are true, then I'll bet that there are many metal buildings which can not pass that local I-Beam web buckling check.

I am not sure how much it related to the the recent roof collaps case, that is why I would like to bring this topic here to discuss.

Thanks.

 

[msquared48]

Regarding #2, I can understand why that is done in the shops with automated welding machines with the beam laying inclined. However, I believe in staggered welds of equal length on both sides to minimize differential stresses due to the welds.

Mike McCann
MMC Engineering
Motto: KISS
Motivation: Don't ask

 

[gobsmacked]

A collapse of falsework at Loddon bridge, UK (date 1970's)was put down to incorrect assessment of the effective height of a rolled I-beam web under a concentrated load. The designer assumed that the filleted ends of the web meant that it had encastre conditions at both ends, but failed to recognise that the top flange could drift relative to the bottom flange, and rotate a bit too, as I recall. That incident lead to a change in the UK code of practice at that time.