Rotation in Cold-Formed Steel Moment Connections

[cldea8]

BACKGROUND

Our firm does a considerable amount of cold-formed steel design. We are beginning to see more and more architects specify what we call "ribbon windows". These are long narrow windows that run the length of the wall. The headers and sills are not supported by king studs. The studs above and below the windows are either by-pass (where they continue from sill above to header below) or are cantilevered off the concrete slabs. My question concerns the cantilevered framing and proprietary moment resisting connections in conjunction with serviceability requirements.

THE PROBLEM

The Steel Network and Dietrich both manufacture moment resisting clips for these conditions. However, they are published with a foot note stating the values are based on a 0.02 radian serviceability limit. We requested test data and found that indeed several of the listed capacities correspond to this rotation. This allows too much deflection in almost any circumstance! (L/165 compared to my required L/660)

Design software from the Steel Network does not include this rotation in deflection calculations. It appears as though people are ignoring this. Generally, when we model deflection of fixed-fixed beams(in steel) we don't account for rotation allowed by the connection (if so we can oversize the member or develop the full section at the connection). Unlike a fixed-fixed span (where you can account for additional rotation by increasing the stiffness of the member), serviceability of a cantilevered stud is entirely based on the allowable rotation of the connection at the base.

QUESTIONS

What are the accepted rotational requirements in cold-formed to consider a connection rigid?

If it is considered rigid by code does that allow me to neglect the rotation induced at the connection (for overall deflection at the top of the cantilever)?

How are people detailing this (if not using this clip)?

Isn't neglecting connection rotation on any cantilever the wrong thing to do(w/ respect to serviceability)?

Any insight would be greatly appreciated. We have no options as far as changing framing, kickers, or steel embeds.

Thanks,

cldea8

 

[jike]

My preference as EOR would be to use the bypass studs not to cantilever the studs off of the slab. When I have had to cantilever a stud wall off the slab, I would use a structural angle or channel at 48" oc and infill between with studs.

I agree that you need to take into account the connection rotation.

Could you call the engineer (or architect) and explain the problem and see what he wants to do? I know that I am asking alot!

 

[cldea8]

jike,

Thanks for the response. I agree with the bypass studs. We do a lot of EOR jobs and the work we do on the contractor side definitely makes us more attentive to such situations.

During our early discussions w/ the architect/engineer (AE firm) about this problem there was a strong push toward using the clip "..that is what we typically see on all of our details of this nature..." They shot down an original detail that used welded straps (to the pour stop) and anchor bolts.

I think we are going to have to stand our ground...they may agree to design some cantilevered steel framing.

Just an FYI - I spoke with a design engineer for Dietrich (as you know they do full design on projects). He said that to his knowledge, they have never specified their moment resisting clip. They always insist on steel framing. It almost seems irresponsible for companies to offer such ineffective products.

 

[UcfSE]

The company can offer what they want. It is up to us as engineers and EORs to determine suitability.

I agree with your concerns and would insist on using some other framing method besides cantilevering from a clip. I would provide at least two supports for the studs.

You can brace back to a beam at top and bottom and set up a stud that is simply-supported with an overhang. I have also used kickers back to the red iron. I think that's fairly common.

Technically rigid is zero deformation. With moment capacity, like you get with this clip, you can take that into account as though it's "rigid" for strength but you still need to account for the additional deflection in your system due to the clip, or support, rotation.

 

[cldea8]

UcfSE,
Thanks for the response. As you know cold-formed steel design is almost always deflection controlled. In most states it is not required to be designed by an engineer. Therefore, architects with cursory knowledge of structures are specifying this clip. I also found, in my many phone calls to other firms designing cold-formed, that many engineers have been oblivious to this small footnote. Again, I agree w/ you 100%...shame on them!

However, my search for information left be sour. The Steel Network's first email to me stated that the clip qualified as a rigid clip per the allowable rotation for rigid connections in the AISI 2001. I later found this to be incorrect. The code does not make such qualifications.

They wanted to dismiss the concern as quickly as possible. The Steel Network shouldn't advertise a "Rigid" clip if it doesn't allow a cantilevered stud to meet the most minimum deflection requirements (l/180). They finally aggreed this was indeed a problem. If Dietrich design group has never even specified their own moment resisting clip (and they design thousands of projects), why do they offer it?

I am sure there are applications for the clips, however, I think their liscenced engineers and designers carry somewhat of a responsibility to at least show what values in the table correspond to this service limit. A new clip needs to be designed that attaches to the flanges of the stud.

As far as options for kicking back, the framing is under the window and is not 'by-pass framing'. It rests on the concrete slab and spans from slab to sill (just below the window). We can't kick back inside the hospital rooms.

After some discussion I think we are going to get them to accept a change order to install cantilevered steel framing every 4-8' inside the wall cavity.

cldea8

 

[cldea8]

oops...their LICENSED engineers should carry...