Metal Stud Cladding System 1

[davidfi]

I am starting work on a new 3-story office building in Calif with braced frames in one direction and moment frames in the other direction. The floors will be lightweight concrete over metal deck.

Before I get too far in the design, I want to make sure I have a good handle on attaching the metal stud cladding (finished with cement plaster) to the exterior of the building. The metal studs will be balloon framed from the slab up to the roof and clipped to each floor. After reading ASCE 13.5.3 regarding in-plane drift, it sounds like I can:

(a) Keep the metal stud cladding independent of the primary structural steel with the use of drift clips.

OR

(b) Connect the metal stud cladding laterally to the primary structural steel and have the metal stud cladding "go for the ride".

I have the following questions:

1) If I do option (a), can I use the exterior Dens Glass sheathing to resist the in-plane self-weight loads of the metal stud cladding system.
2) If I do option (a), what do you do at the corners where the perpendicular wall is rigidly attached for out-of-plane loads?
3) If I do option (b), I am concerned that the drift of the moment frames will rip the cement plaster cladding apart. Is this a valid concern?

Thanks for any advice!

 

[XR250]

I have wondered this myself. I am not convinced you are going to get hundreds or even thousands of "drift clips" to cooperate in a seismic event. And, like you said, what do you do at the corners?
Maybe you are better off just repairing the stucco when needed.

 

[DETstru]

Facade attachment is often overlooked. Good on you for thinking about it now.

I would not balloon frame non-structural facades in high seismic zones unless you are doing spandrel systems where you can take the drift at the strip window head. Your building drift at the top will be a lot.

Typically here's what I do for non-spandrel systems:
Bottom of facade wall (top of slab): Vertical, in-plane, and out-of-plane connection
Top of facade wall (bottom of slab): Out-of-plane connection. Drift connection for in plane drift of story. Deflection track connection for vertical deflection from floor above.
Corners: Often ignored, especially with lightweight/flexible facades. Usually trying to accommodate drift in the corners is more trouble than it's worth. That being said, if you have heavy, non-flexible systems, it may be worth the trouble.

Here's a good article:

http://www.fwcse.com/wp-content/themes/FWC/Publications/Exterior.pdf

There are a lot of great products to handle drift, examples:

https://www.steelnetwork.com/Product/DriftConnection

If you provide a sketch of your elevation I'd be happy to comment further on it.

 

[davidfi]

Thanks for the responses!

I had read that presentation from FWCSE earlier today; now it is making more sense.

So if I understand you correctly, the reason for using a floor to floor system in high seismic regions is because your drift at the roof will be too high, correct? If I design my moment frame up to the code limit of 2.5% drift, I would have 13.5" at the roof of the 3 story bldg (15 ft stories). So it sounds like I can use a drift track at each story to limit the drift to 4.5", correct?

Do you see an issue using balloon frame studs on the side of my building with braced frames? I imagine the drift will be quite small.

Unfortunately I do not have any exterior elevations from the architect yet. They are just trying to pin down exactly where to locate the columns.

Thanks!

 

[DETstru]

Correct, you don't want to take 13" of drift at the top floor. That's just not feasible for typical metal stud systems.

Study the elevation carefully. Architects don't think about drift (not the ones I work with anyway...).
Look for areas where your drift plane (the line that separates what is moves laterally with the floor below vs. the floor above) and see where it jogs. If it's not entirely horizontal (has vertical jogs) that means those vertical joints will need a gap to take up the drift. Caulk only compresses 50% or so, which makes these joints get huge. I can post a sketch of what I mean if you would like.

 

[davidfi]

The architect is also asking if we can locate the moment frame beams/columns in the plane of the exterior curtain wall. If we were to do this, this is how I see it done:

1) Install the drift track to the bottom of the beam rather than to the bottom of the slab.
2) Keep a gap between the column and the curtain wall of at least 4.5".
3) Span horizontally over the column with flat metal studs.

Does this sound appropriate?

Thanks,

 

[DETstru]

Typical architects...
This is doable but will be a lot of work.

1) Yes, you can do this. Your track might not line up with the beam so you'll need to make sure the track can be properly anchored. Don't forget about the protected zone on your moment frames. No welding/fasteners.
Don't forget that this track needs to provide vertical deflection and in-plane drift tolerance (i.e. nested track or specific drift track product).
2) Yes. The architect will hate this. Joints over +/-3" need very special detailing and often can't just be caulked. Don't forget about combressibility. For example, if you have 1" of deflection and you're using 50% compressible material, you need a 2" joint.
3) Yes but you can't attach to the column since the column moves laterally in the plane of the wall. Also your columns are probably wider than your beams. Don't forget you have to support facade at the face of the beams as well.

 

[XR250]

I have done alot of buildings with vertical deflection (none with drift). I have seen job pix where they just sheathed right across the joints. Have not heard of any problems yet......

 

[DETstru]

XR250,
Agreed but the deflection joint is really for worst case scenarios. You don't typically have full live load on your slab so you won't typically see full deflection. Even with that, every beam/slab is almost always somewhat over designed so you don't see the L/360 that you may be designing for regardless.

Regarding drift, you'd only see this in a severe wind event or a seismic event. I agree that there may be other factors at play here but I think the idea is to provide the ability to drift to whatever degree we can calculate it.


davidfi, one more thing. If you haven't done so already, I recommend pushing back on the architect regarding the frame being in the plane of the wall. It never works out well.

 

[wannabeSE]

AISC's Design Guide 22, Façade Attachments to Steel-Framed Buildings, is a good resource.

https://www.aisc.org/Design-Guide-22-Facade-Attachments-to-Steel-Framed-Buildings#.Wa2bjrpFw2w

I think the architect will have problems detailing the building envelope if the metal stud framing is in the same plane as the moment frames. The joint on each side of each column will be challenging.

In actual practice, I have not seen a drift joint at the top of a window. How is the top of the window flashed - a caulked joint? If the story drift is real low, they might be able to accommodate the movement between glass and the window frame.

It is common practice to limit the perimeter beam live load deflection to L/600 or less so it is easier to detail the exterior. Often times, these beams are part of a frame or collectors, so live load deflection may not govern the design.

 

[davidfi]

Thanks for the feedback!

I didn't like the idea of putting the curtain wall in the plane of the moment frames and you all seem to agree. I will push back on this.

Thanks for the AISC Design Guide link. I had not seen that.

Thanks for all of the help!

 

[Ron]

A few other thoughts...

Since your cladding will include cement plaster, pay attention to deflection. Your code requirement is L/360, usually considered only in the long bending dimension. Keep in mind that for CFS studs, you will get a lot of "stiffness absorption" at windows and other openings. This creates a bi-directional bending issue with rigid coverings that do not serve as membranes. The result... increased cracking, water intrusion, and CFS deterioration.

 

[XR250]

Ron,
Are you saying the window jamb studs are generally stiffer or noodlier than the common studs? In my design experience, they are usually the same or less stiff.

 

[KootK]

My understanding is that the jamb studs create a point of extra stiffness and you wind up with a bit of two way bending in the wall panels. I think that the jamb studs might calc out noodlier if they are assumed to see their full design load coming in through the sills etc. I don't think that happens in practice 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.

 

[XR250]

My understanding is that the jamb studs create a point of extra stiffness and you wind up with a bit of two way bending in the wall panels. I think that the jamb studs might calc out noodlier if they are assumed to see their full design load coming in through the sills etc. I don't think that happens in practice however.

Has anyone in metal stud history ever taken this into account? When I design jambs, I usually just use tributary area but I decrease the pressure accordingly due to it being more area than a common stud (maybe I should not do that anymore). I don't think i have ever done a building that had real stucco so I can still sleep well at night. Thanks Ron for spoiling the party!

 

[KootK]

Has anyone in metal stud history ever taken this into account?

I believe that it was done in Roman times but the technology was lost. But no, I haven't attempted it. For me, it crosses into the area of things where prediction accuracy, of lack there of, makes prediction effort a poor investment. If one were to attempt to address this explicitly, I'd think that the answer would be tighter limits on field studs rather than intentionally more flexible jamb studs.

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.