Guardrail load paths 1

[joaex]

Hi, EIT here. My interests in college were mostly focused on steel design, and unfortunately I'm not in that line of work now, but I try to keep learning more about it so I'll be better-equipped when it comes time to move on. I often try to look at how things are put together "in the wild" to get a better idea of general practices and a better handle on nuances of design. One thing I've seen a few times that has me a little confused is guardrail connections striking me as a little underdesigned. I've read that C and MC sections have very little torsional strength/stiffness. But I've seen guardrail sections (on stairs especially) that connect the posts to the channel stringer with two bolts through the flange, oriented longitudinally, or to a small clip (4 or 5 inches in height, say) in the center of the beam web. Using the design LL as a 200 lb point load applied at the top of rail, it strikes me that, say, 1.6*200lb*4ft=1.28k*ft is a lot of torsion to apply to these sections. The "two bolts through the flange" approach sometimes looks like it can't have a resisting moment arm of much greater than half an inch or so, which would mean almost 31k of uplift, and the "clip on the web" approach on a flimsy web just gives me the impression that it would deform a lot, locally. And either way, the channel section itself eventually has to deal with the torque, and I haven't tracked down a design procedure to give me the torsional stress/deflection for a C-shape (which I gather is complex) but treating the section as rectangular but making some educated-guess equivalent rectangular sections gives me the impression that the sections are borderline at best in handling it. And that's to say nothing of torsional deflection, and over a stair stringer with 12' of rise, say, I imagine the deflection (assuming the treads don't somehow restrain it) would be large.

Anyway, I've been lurking here and thought I'd ask this, because it's been nagging at me for a little while. It seems to me that the lateral live-load from guardrails should very much be a controlling factor in elevated platform/stair design, but I don't really see it talked about explicitly all that much. Is there a reason for that? Or am I way off here?

 

[wannabeSE]

I will bite my tongue and not saying anything about under-designed (not designed) hand and guard rails on stairs.

Stair stringers may be braced by the treads. AISC has a Design Guide for torsion

https://www.aisc.org/store/p-1585-d...nalysis-of-structural-steel-members-see.aspx.

NAAMM has some design free information on rails and stairs:

http://www.naamm.org/amp/amp_technical_literature.aspx

 

[KootK]

You're spot on joaex. The only thing that I'll add to Wannabe's comment is that deflections aren't usually considered important. I've seen some approved testing for highrise condo balcony railings that deflected 18" at the top under specified loading and are deemed to have passed.

I like to debate structural engineering theory -- a lot. If I challenge you on something, know that I'm doing so because I respect your opinion enough to either change it or adopt it.

 

[Jerehmy]

Stair treads are not rated to brace the stair stringers.

"Jerehmy,
Thanks for contacting me. Neither Ohio Gratings nor the other members
of NAAMM Metal Bar Grating Division rate the grating stair treads as
lateral braces for the stringers. Other structural members should be
used for this purpose.

Please let me know if I can be of additional assistance.

Kenneth P. Apperson, PE
Director of Engineering and Product Development
Ohio Gratings, Inc.
Office 330-479-4256
Cell 330-418-5624"

 

[Once20036]

Without any specific geometry being mentioned - it may be worth considering that the rails actually span horizontally (ish) between the returns at the top and bottom of the stair run, rather than spanning between the vertical posts.
Typically at the top and bottom you'll have a U-turn which will be significantly more rigid than the channels and connections that you're describing above.
NAAMM has information regarding distribution of load when you have a number of posts - that is to say - when you apply 200 lb at the top of a post, it will begin to deflect horizontally. As it deflects horizontally, a portion of the lateral load will be shared with adjacent posts. From memory - I think intermediate rails can be designed for 60% of the code specified load. I don't believe that NAAMM does anything to account for rigidity when you have perpendicular rails at the ends.

All this being said - I've seen a lot of guardrails that just don't work on paper, but seem to have served for many years without issue.

 

[SteelPE]

I have designed a bunch of stairs in my day (50-100). 90% of my posts are welded to the top flange of the stingers. Stairs and guards are often not designed by the EOR as they require the fabricator to provided calculations that prove the stairs work. The torsional capacity of the channel with regards to the guards is suspect however, I believe that neglecting the torsional resistance of the stringer (or not checking it) is something that has been generally accepted by the engineering community.

I haven't read this in about 5 years but here is an article from MSC:

http://msc.aisc.org/globalassets/modern-steel/archives/2009/02/2009v02_holding_on.pdf

Jerehmy is correct with his post GRATING does not brace the stair stringers. Headers that frame perpendicular to the stringers or tube stringers should be used when GRATING is used. However, I would argue that the Z profile used in most concrete stringers would help brace the stair stringers.

 

[Jerehmy]

joaex I wish I fully read your post before. I design stair stringers pretty regularly as I work with a manufacturer. They typically use Ohio Grating stair treads with two little bolts to attach them and a noseguard, so we ingore it for bracing. We typically run some angle cross bracing along the bottom flange.

Also, AISC Design Guide 9 deals with torsional capacity of channels.

 

[Jerehmy]

sorry, I meant fabricator, not manufacturer.

 

[Ingenuity]

I've seen some approved testing for highrise condo balcony railings that deflected 18" at the top under specified loading and are deemed to have passed.

ASTM E985 Standard Specification for Permanent Metal Railing Systems and Rails for Buildings requires a horizontal load of 200 lbf with a maximum deflection of h/12 of metal handrail systems...not sure if this includes balcony handrails?

We once were asked to load test a handrail system to a series of balconies of a multistory building in Guam that consisted of PVC tubing!

 

[XR250]

If you put web stiffeners in the channel at each post, it cuts down on deflection dramatically as most of the deflection is from the top flange bending.