Rafter unbraced length problem

[RobertEIT]

We got a difficult issue with our current design vs. an existing roof. Following is the senario:

I designed a flat roof structure for a workshop in a 40psf ground snow area, lower portion in attached pdf file shows part of the roof design plan view. In my design I would add one bracing member at middle of the 10JWE12 cold formed steel rafter to reduce the unbraced length of the rafter to half(12') based on required flexural strength calculation. But the contractor says it is un-necessary to add this bracing member and he showed me an existing workshop roof which is near exactly the same rafter span except it does not has the bracing member.

Acoording to my calculation with unbraced length =48" will definitely not be able to meet the required strength, unless the designer of that existing roof considers the top felt and a thin steel roof sheet have the bracing function to the 10JWE12 cold formed steel rafter. But to me, these soft and flexural material can not be relied on as an bracing member. On the other hand, this existing roof has already stand there for over 20 years without any problem according to the contractor.

What are you guys thoughts and suggestions?

Thanks a lot.

 

[RobertEIT]

Sorry, the third paragraph should be corrected as:
"Acoording to my calculation with unbraced length =24', the rafter will definitely not be able to meet the required strength, ....."

 

[ash060]

You can only trust a contractor as far as you can throw him

 

[msquared48]

In Light Gage steel building design, the rafters are frequently laterally stabilized using 28 or 29 gage steel panels. I am leary of their structural strength, especially if you walk on them wrong, but the system does seem to work.

What is the gage and depth of the steel sheeting? If the decking is 1.5" deep, 26 gage material, it should be OK, depending on the attachment to the joists. Screw the felt!

Mike McCann
MMC Engineering

 

[JAE]

Don't forget wind uplift on the rafters. If you rely only on top flange decking/roofing, your unbraced length may be limited for gravity loading but would be 24 feet under uplift.

 

[hokie66]

The bracing or bridging member is necessary, and you should just tell the builder to put it in. As JAE says, don't forget wind uplift. I would probably have used two rows of bridging.

 

[DaveAtkins]

When I design cold formed steel roof framing, I put in bridging @ 4'-0" oc. It does not need to be full depth bridging; it can be channel bridging running through the openings in the rafters. And you can assume the rafters are fully braced by the roof deck, if the deck is steel deck, plywood or OSB. If you have a standing seam roof, then I agree, the rafters would need the kind of bridging you are proposing.

DaveAtkins

 

[RobertEIT]

Checked the existing roof without bracing member, its roof sheet is 28g, decking is 1.5" deep. decking rib at 12" spacing, every 12" has a screw go through felt to attach to 10JWE12 cold formed steel rafter. the screw looks like 1/4" dia. Would it be rigid enough to consider the rafter as full braceed?

 

[hokie66]

Again, the top flange is braced, but not the bottom. Don't ignore uplift.

 

[geoffdale]

I very rarely see no bridging on this type of thing. I second every else.

 

[statiker]

this problem of lateral torsional buckling may be solved
by proofing that the rotational stiffness of the deck is
higher than the required rotational stiffness to
avoid lateral torional buckling.

the result will be that you need no bracing.

 

[hokie66]

statiker,

Your statement is fundamentally incorrect. Whether the bottom flange requires bracing is dependent on the member itself, not on the deck stiffness.

 

[youngstructural]

Actually Statiker's position has some validity, but is much more safely applied to steel portal frame rafters with stiffeners and pulins running continuously across the top flange.

Hokie is right to take issue with your approach Statiker, as you may be effectively restraining the CFS rafters agains LTB, but you will now have another failure mode governing, likely local buckling.

I posted the HERA research findings in an earlier thread, with a fairly detailed explanation. That said, the gist of the idea is that if your purlins are greater than half the depth of your rafter and connected to cleats passing across the top of your section with two or more bolts and full depth stiffeners are provided each side of the web, you have effectively restrained the section. I would advocate than anyone wanting to pursue removing fly-bracing by this method get a paper on the topic to ensure they're confident with it.

All of that said, you're almost certainly not going to get away with it in CFS; It's just too unforgiving.

Regards,

YS

B.Eng (Carleton)
Working in New Zealand, thinking of my snow covered home...

 

[statiker]

i think we should differentiate between global and local
instabilities. global lateral tors. buckl. may be avoided by the deck itself and local buckling may be avoided by
maintaining the form of the cross-section for example by
local stiffeners. this is proven by experiments.

i think it is possible to analyse this correctly by solving the corresponding stability problem.

 

[hokie66]

In the case of the OP, the uplift load would be transmitted to the top flange by screws or spot welds through the sheeting. The sheeting would not be connected in any way which would resist torsion on the member.

 

[diarmud]

If you can't verify it, don't use it. Being there 20 years doesn't mean anything, it may have never been fully loaded. What was the code, then and now? Was it even designed to code then?

While you can appreciate the contractors comments and should treat him as an associate, the two of you are responsible for two different things.

He wants to get it built, he has no responsibility after it's up, you are responsible for a long time.

If you're truly uncertain, go with the bracing.

 

[youngstructural]

Actually I now agree with Hokie on the buckling issue... The screw fixing will be lightly semi-rigid at best, and certainly not the required moment fixing to permit segmenting the rafter length.

I actually second Diarmud's comment, with a sprinkle of Ash060's pessimisum. Add the bracing, sleep at night.

AND a contractor with twenty years of experience (or a structure which has been field tested for that period of time) does not prove performance for as long as you must, and remain responsible for...

Good luck, Please send an update post with the end result,

YS

B.Eng (Carleton)
Working in New Zealand, thinking of my snow covered home...

 

[RobertEIT]

Thanks for everybody's good comments.

Now I am going to tell the contractor that the existing roof might only meet the old code of 20 years ago, it does not meet current code requirement. So I will insist on putting these bracing member.

Thanks everybody again.