Stair Stringer Bolt Design Question

[zrck99]

I am putting together a set of calculations to justify the details/design that the a steel fabricator put together for a very typical office building stair set. They plan to connect stringers to walls with 1/2" diameter wedge bolts at 2'-0" o.c.

I've shown the detail in the attached pdf. My question is, won't gravity loads cause the bolt to rotate at the face of concrete because of the 3" flange width standoff? I assume people do this all the time so I'm just curious if anyone has seen a good justification for it. Some of the other guys in my office suggested adding 3" square metal shims in between the wall and inside face of the channel web. I agree this would add some stiffness but it still looks a little indirect to me...

Any thoughts?

Thanks

 

[r13]

Top flange has zero force. Bottom flange C = 500. Because upper corner is un-restrained, it can't take tension. You resolve the moment about the lower corner, solve the only unknow (bolt force in tension), then equate bolt force with the compression developed in the lower corner. Make sense?

But, I've revised/edit my previous comment. Please read again.

 

[r13]

Ok. Your system has a pin support at the bolt, and a roller support at the lower left corner. Thus the bolt will see both shear and tension, while the bottom flange will feel compression. Your plate and elements of the channel will develop internal force (bending) to resist the global moment. A FEM analysis can give you a better picture.

 

[zrck99]

retired13 -- Thanks for talking this through with me, I appreciate it. I think that we may already be on the same page but I'm just not communicating clearly. I clarified my sketch and show the bending checks in the plate just above the bolt and in the channel flange just adjacent to the web. As far as I can tell these are the only spots that I really need to be concerned with.

The way I see it, the outer 1000lb force will result in the 250lb couple force at the upper and lower flanges. The lower 250lb couple force will bear directly onto concrete through the flange tip. The upper 250lb couple force pulls at the top of the plate with 250lb. That 250lb force prys the bolt with 500lb of tension and bears with 250 lb of compression at the "roller" at the bottom of the plate. I know your moment arm will vary a little based on the bearing width but I think with the numbers I'm looking at this is pretty close.

Do you agree?

 

[Tomfh]

It’s not the same as the contractors detail. Instead of relying on bending the bolt (or friction of flange tips), the bolt is loaded in shear and load transferred back via the flanges. The flanges and web need to be checked for the bending and thE overall eccentricity needs to be resolved somewhere (a cross member?).

 

[human909]

I'm confused on what is trying to be achieved here.

Overall your connection design has numerous problems that other people have already posted. There are 100 ways in which you could secure a stringer to a concrete wall but this doesn't look an appropriate one and trying to clutz it until it works seems like you are avoiding fare simpler methods.

One simple way is to use bolted angles top and bottom of the stair stringer. Another is a continuous stringer with their ends supported.

How about you tell us what you are trying to achieve and what are your constraints (architecturally or structurally)

 

[skeletron]

Bolted angles to get your ledge.

The options could work theoretically, but I wouldn't want to bet on loading an anchor into a concrete/CMU wall in shear, tension, and bending as a method to support a point of egress.

 

[zrck99]

Tomfh -- What do you mean about the overall eccentricity being resolved? The eccentricity is resolved by first loading up the upper flange in tension and bottom flange in compression. The compression in the bottom flange transfers directly into the wall. The upper flange loads up the top of the plate via welds. The plate then acts as a cantilevered beam with the bolt acting as a pin and bottom of the plate acting as a roller.

human909 -- I am providing calculations for a set of fabrication drawings that are already considered 100% done. The original detail (not by me) is my detail without the internal plate so that there is an obvious hinge point at the wall unless you rely on pretensioning the bolts and shear friction. I'm trying to avoid changes as much as possible. If I were designing from the beginning, I agree that it would be much more straight forward to just span the C shapes out and do a more traditional end connection -- no arguments there. What numerous problems do you see with the attached connection?

 

[phamENG]

So is the stringer not capable of spanning and/or doesn't have connections top and bottom at the landings? Because if it can and the connections are there, it doesn't matter what these connections are doing. If you can justify that it works without these, then you don't have to make any changes. They can either pull the details from the drawings or put them in (whichever is cheaper) and the stringer will still support the loads.

If that doesn't work, your detail should be ok as long as the flanges take the load in bending.

 

[Ron]

JAE's detail is good. There are a couple of other considerations.

First, stringers are not typically connected to the side walls for two reasons....one, you don't need it when attached top and bottom and two, you transmit stair vibration (think noise) to the wall. Also, the other stringer will not be attached which will throw the stress distribution and bending off a bit.

I design stairs with no stringer attachment to walls...just the top and bottom supports.

 

[human909]

I am providing calculations for a set of fabrication drawings that are already considered 100% done. The original detail (not by me) is my detail without the internal plate so that there is an obvious hinge point at the wall unless you rely on pretensioning the bolts and shear friction. I'm trying to avoid changes as much as possible. If I were designing from the beginning, I agree that it would be much more straight forward to just span the C shapes out and do a more traditional end connection -- no arguments there. What numerous problems do you see with the attached connection?

The original detail seems highly complicated both structurally and for fabrication. I would suggest that trying to square that circle is more costly than it is worth. Cutting a plate a plat to size and full length welding along the channel is an expensive process. Other solutions such as angles above and below the channel would be simpler and cheaper both in terms of fabrication and calculative efficiency. Also using wedge bolts with an air gap is an exciting proposition. One that I think should require a bit more thought.

If these bolts are spaced appropriately using your design I'm not highly worried. But it still screams an unideal solution.

 

[Tomfh]

Tomfh -- What do you mean about the overall eccentricity being resolved

The bolt applies a shear force to the internal plate. The beam web (and shear centre) is some distance away. That eccentricity needs to be resolved somehow, otherwise the thing is trying to wrench itself off the wall.

 

[BridgeSmith]

I'm still a bit confused as to how the stair treads are attached to the stringer such that they don't provide rotational restraint to the stringers.

The addition of the 3" x 11" plate at the anchor locations seems like a straightforward, elegant solution to get the vertical load supported in shear only. If the moment due to the eccentricity is an issue, move the anchor up near the top of the channel to increase the distance between the force couple.

I also still wonder how the anchor protruding into the stair isn't a hazard to the people using the stairs. Instead of extending the anchor through the channel, why not put a handhole in the channel and secure the anchor to the 3" x 11" plate inside the channel?

Rod Smith, P.E., The artist formerly known as HotRod10

 

[Tmoose]

How flat and true is a (poured) concrete wall likely to be over the (unidentified) stringer length ?

I saw no mention of C-C spacing of the anchors along the channel/stringer, or whether the assembled stair or initially just the stringer will be bolted to the wall.

I anticipate a medium-to-big fight resulting when the anchors try to pull the stringer/channel against some sections of the wall.
Also plenty of gaps between the channel and wall at the upper flange, lower flange, or both, and some in the vicinity of the anchors.
By the time the inspector or EOR arrives the paintable caulk will have been applied, tooled and dry. In which case the variations of gap-osis of the assembly will indeterminate.

 

[r13]

zrck99,

First of all, your design maybe just work well, and all numbers could be checked out. But, from structural performance point of view, I just have too much concerns (hanging in my head) over a system relying upon so many factors to be successful. I made a 2D analysis, and the deflected shape is shown below for your information.

Analysis Parameters:
All member made of steel, width/depth = 1/1, member 7 (rod) = 0.25/0.25
Joint N6 - Fixed; Joint N5 - roller
Member 7 with moment release at joint N3
Lateral load = 0.1

 

[zrck99]

Thanks for all the comments. I'm going to just switch to spanning the channels out and doing a more typical connection at each end. I agree with most of the comments about how much of a pain it would be to add all the plates so it seems worthwhile to make the fabricator redraw with a more conventional layout.