Unbraced Length for Twisting (Lt) - Cold Formed C

[Shanman_]

I am designing an 18'-0" wide structure that has two beams, spanning 18'-0", column to column. The beams are cold formed C-Purlins (with lips) but they have cold formed corrugated metal decking spanning between the beams with screws to their top flanges. There is no blocking between the beams.

It is my understanding that the decking provides a top flange restraint for the beam in positive bending as the decking fasteners help to prevent the top flange from rolling over due to torsional buckling. I have seen a publication on this before. In addition, 2012 AISI D6.1.1 has a section specifically referring to this, although the requirements appear stringent due to lack of testing.

My question is, aside from the provisions in the AISI mentioned above, what else can the engineer reasonably do to account for the restraint the fasteners provide through the top flange?

AISI EQ. C3.1.2.1-8 AND C3.1.2.1-9 are both used in determining the lateral torsional buckling strength of a member of open cross section. The first relates to Y-Axis bending and the other to Twisting of the section about its own axis.

I take the unbraced length used in equation 8 to be the unbraced length to lateral torsional buckling (Ly) of the compression flange, which happens to be the top flange in positive bending. In this case, I take my unbraced length to be my fastener spacing as opposed to my full span. I do see the other side of the coin and that perhaps I should be using the full-span, as the decking does not provide restraint for the member to bend about the Y-Axis like blocking does. This is not the most critical point for me, as the most drastic improvement to my design result comes from my next assumption.

I then take the unbraced twisting length (Lt) used in equation 9 to be fastener spacing, as the fastener can provide restraint through shear to resist the section's twisting about its own axis. This dramatically improves my results.

I would like to invite comments/opinions on the above.

Thanks.

 

[WARose]

My question is, aside from the provisions in the AISI mentioned above, what else can the engineer reasonably do to account for the restraint the fasteners provide through the top flange?

AISC's Steel manual (Appendix 6 of the 13th edition) is a good source for stability bracing. So is 'Guide to Stability Criteria for Metal Structures' (by Galambos).

AISI EQ. C3.1.2.1-8 AND C3.1.2.1-9 are both used in determining the lateral torsional buckling strength of a member of open cross section. The first relates to Y-Axis bending and the other to Twisting of the section about its own axis.

I take the unbraced length used in equation 8 to be the unbraced length to lateral torsional buckling (Ly) of the compression flange, which happens to be the top flange in positive bending. In this case, I take my unbraced length to be my fastener spacing as opposed to my full span. I do see the other side of the coin and that perhaps I should be using the full-span, as the decking does not provide restraint for the member to bend about the Y-Axis like blocking does. This is not the most critical point for me, as the most drastic improvement to my design result comes from my next assumption.

I then take the unbraced twisting length (Lt) used in equation 9 to be fastener spacing, as the fastener can provide restraint through shear to resist the section's twisting about its own axis. This dramatically improves my results.

I haven't looked at AISI in ages so I cannot address this specific issue.......however in my experience with most steel structures, it's easier to get lateral bracing than torsional bracing. (Bracing can be provided by either one.) Bracing requires both strength and stiffness. (AISC details the provisions for both.) My experience with deck is: it's hard to use it as a torsional brace because usually it is not flexuraly stiff enough to provide such restraint. If a deck is present, it's probably better to use it as a lateral restraint than a torsional one.

 

[XR250]

Best to use two lighter members back to back to make an "I" section. It gets rid of the twisting and makes the flange bracing easier.