TXEng-USA:
In terms of the screw stripping, it will be stronger if the screw bisects the/a glue joint. The glue joint will improve the long fiber resistance to stripping by a small amount, actually increasing the circumferential area of the fibers which will need to be sheared for stripping. Actually, some epoxy glue in the pilot holes of screw improves their pull-out strength too. I don’t think you get any special credit for the end grain reduction coef. because you are screwing into the end of a GlueLam member, it is still basically the end grain of solid sawn lumber. Longer, larger dia. screws would give improved pull-out values. You could use inserts which have a course outer thread and a machine screw interior thread. They are threaded/driven into a hole in the wood and then a machine screw replaces the wood screw. You could take .75 or 1" round bar stock about 3"- long and tap machine screw holes perpendicular to its long axis and 1.75" apart. Then, you drill a hole for the .75 or 1" round in the underside of the tread, and two clearance holes in the end grain for machine screws. I doubt that any of the above is covered by the NDS, but some fastener suppliers may have some ideas or testing data on some possibilities. Does this embedded standoff detail with screws at 1.75" o/c give a canti. moment which can handle 200# lateral load at the handrail?

dhengr: There are (2) embedded standoffs per tread so essentially 1 standoff per linear ft of glass guardrail. I am using a couple between a standoff at an upper tread and the standoff at the tread below to resist the overturning moment. so, in short, yes I believe that if I can get the screws to work in withdrawal then it can handle the 200lb point load at the top rail. Based on the geometry, I calculate that the maximum withdrawal reaction at each standoff is 560 lb with 90 lbs of shear. I have uploaded an elevation view to further illustrate the configuration.

Drew Dudley, P.E.
www.dudleyengineering.com

• http://files.engineering.com/getfile.aspx?folder=86f40fcd-8b29-4dbe-9f81-5e

Hi Drew,
I agree with you about not liking to screw into the end grain, but the wood code does allow it like you said with the reduction factor of 0.75. So I think you are right to just calculate the withdrawal and shear on the screws and find a screw that works. Personally I would pick something long and skinny, I just checked 1/4” x5” embed lag screw and it works, but I didn’t check the edge distances and bolt separation in NDS so don’t forget to check those. Maybe you already use this, but the AWC bolt calculator is one of my favorite websites: http://www.awc.org/codes-standards/calculators-sof...

I have a question for you though - how do you verify the 5/8” glass? Do you have an allowable tensile capacity you use for bending? And any local punching calculation for the screw connections? I’d be worried the glass will break before the screws.

--Luke

How are you going to be able to develop the required moment in the connection with only a 2" thick tread?

Mike McCann, PE, SE (WA)

Luke: The glass is monolithic fully tempered glass with a modulus of rupture of 24,000 psi. The IBC requires that you use a safety factor of 4, so the allowable bending stress is 6,000 psi. I am using elastic beam theory equation to derive the stress demand. To account for the concentrated supports in lieu of continuous support I have applied a amplification factor which for my configuration comes out to 1.24.

Mike: The moment is developed in a couple between the standoffs in the treads above and below. The treads are approximately 6 3/4" vertically part which is the moment arm.

Drew Dudley, P.E.
www.dudleyengineering.com

TXEng,

Like Mike, I think you have a problem developing the connection that way. Using the moment arm as the riser dimension doesn't work. Your connections follow the slope of the stair, so each connection has to work on its own.

I've done a bit of this kind of thing. One recommendation that I'd offer is not to fasten two panels to the same tread unless you're providing some play in the connections somewhere. You want to avoid forcing the glass to adopt the curvature of the stair system as that will invariably result in the glass seeing in plane flexural stresses owing to it's large in plane stiffness relative to the stair system.

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.

Thanks KootK, I will make this change. With respect to hokie and Mike's comments on the moment arm, hokie I think I see your point about using the riser dimension, see attached picture. Would you agree in the approximately 3.4" moment arm?

Drew Dudley, P.E.
www.dudleyengineering.com

• http://files.engineering.com/getfile.aspx?folder=ca21d9bd-0efb-4b0b-bcb4-a0

I would agree. You also get to take your load acting at lever arm of somewhat less than 42" as well.

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.

Drew,
I agree with that dimension, and agree with KootK about fastening two panels on the same tread.