Cast-in-place headed anchors in slab edge

[Gnique]

I have been working with a stair manufacturer for several years. One of the ways that stair stringers are attached at the concrete floor levels is to cast an angle with headed studs welded on the inside of an angle right at the intersection of the two legs of the angle. The studs are installed at 45 degrees to the angle legs. Under seismic loading the stringer(welded to the cast-in-place angle / stud assembly) puts the studs in tension. With big seismic loads, edge reinforcing is often required. The cross section would show the studs at the top corner of the slab edge projecting down into the slab at 45 degrees. The tension along the length of the studs can be resolved into two vectors: one parallel to the top of the slab and one perpendicular to the top of the slab. I was questioned by an engineer from the structural EOR firm that did the design of the building how I provided edge tension reinforcing for the vector that is perpendicular to the top of the slab. The load vector that is parallel to the top of the slab is easy; it just projects into the length of the slab. But the other way seems to require tension reinforcing in the direction of the slab thickness which is silly. It actually looks to me like edge reinforcing along the length of the slab is all that is needed but I can't argue that there is not a component that tends to break the slab out at the top. I would greatly appreciate any help or commentary about this issue. Thank you.

 

[RPMG]

Are you asking how to analyze/increase the concrete breakout strength of an embedded edge angle for shear-tension interaction?

 

[hokie66]

It seems to me that if the EOR is concerned about that, he should provide you with a detail to use. I assume he has designed the slab, and you are just connecting the stairs to it.

 

[dik]

would the tension, aka compression, not be provided by the angle bearing on the concrete 'landing' surface?

Dik

 

[hokie66]

dik,
Yes, but it sounds like the fellow is concerned about the building (or just the stair) jumping up and down.

 

[dik]

just tie it in with a rebar wired to the top of the headed stud, running horizontal across all studs.

 

[JStructsteel]

use some weldable bars to the angle that develop and give the tension capacity you need.

 

[KootK]

but I can't argue that there is not a component that tends to break the slab out at the top

You can actually. The trick here is to recognize that the studs are not, predominately, in tension under the action of lateral thrust. They are simply "lug" type things that transfer horizontal thrust from the stringers to the chunk of concrete at the edge of the slab that, hopefully, does not break off under the action of that horizontal (your edge reinforcing tends to this as you said). The lug action takes place over a few stud diameters from the angle corner and induces no serious vertical load component. This is similar, albeit not identical, to how studs in composite beams work.

If you have a stringer that is pinned at the bottom (rather than roller) you'll have a vertical load component to deal with coming out of the stringer. That's a different conversation however.

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.

 

[Gnique]

Thank you all for your input and help. I like the idea of arguing that the vertical (normal to the slab surface) component of the load in the stud doesn't even exist! But after having thought about this whole conundrum for over a week I have concluded that it actually DOES exist. My structural model is not as accurate as it should be. It must be remembered that stairs are sloped (7/11..about 50 degrees). The structural model that I use is to look at the run of stairs as if it were a length of ladder welded at the two legs to a support with a force applied to the free end. My analysis has always assumed that the force induced at the welded connection is purely shear (horizontal and parallel to the top of the slab). I now see that this is not accurate. There IS a vertical component at the connection that is generated by lateral seismic acceleration. The fact that the stairs resist this vertical load is not the result of reinforcement being designed and placed correctly. It looks like the vertical component of the load vector does put the slab into negative bending.

In a very seismically active place like San Francisco (SDS = 1.26), stair loading (using Ω = 2.5) at the landing can generate 35 kip loads. The vertical component of that is 25 kips (assuming that the stringer is a two-force member). With the slope of the stair at 50° and the stud at 45° the two members are roughly parallel. So now we can dispense with the stringer completely and apply a single force directly to the stud that is parallel with its length. That force DOES have a horizontal component and a vertical component. I have always neglected to look at the vertical component and dealt only with the horizontal component. The vertical component may well be the source of stair collapse in seismic events. It now seems to me that negative slab bending is how I have avoided stair collapse in the past. Negative slab bending can, of course, be analyzed but I am just the lowly "stair guy" and not the vaunted "BUILDING STRUCTURAL E! O! R!" so I am never privy to the structure of the various floor slabs.

What has occurred to me is that I could use the trim bars that are always installed at the slab edge. I could assume a single #4 bar, two inches from the top of the slab. Does any of this sound reasonable?

Thank you

 

[KootK]

Does any of this sound reasonable?

A sketch of your model would be helpful in ascertaining that. In high seismic applications, the stairs are often detailed to be pinned at the top and on horizontal rollers at the bottom. This allows one floor to drift laterally relative to the adjacent floors without the stringers acting as accidental diagonal bracing for the building. In this case, your original assumption of horizontal thrust only would be correct. You do have some vertical due to vertical ground acceleration but that's not going to be a big deal here. I'd start by looking at the detailing at the top and bottom of the stair to glean EOR intent. Better yet, share those details here.

I am just the lowly "stair guy" and not the vaunted "BUILDING STRUCTURAL E! O! R!"

For a couple of hours most days, I'm the "lowly precast guy". The work pays way better than any of my own EOR stuff but dealing with other EOR's is definitely the worst part of it. Nine times out ten, the "review" is done by some man-child EIT barely fit to change the batteries on my HP. Always wasting my time with goofy requests and attempting to hold my work to a ridiculous standard of care that I know perfectly well they don't apply to their own efforts. I'd know, I read their drawings. That's what I thought of when I read about your EOR riding you over this vertical component business. That guy's manager would probably slap him silly for wasting billable time if he was caught doing the same analysis he's asked you to do.

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.