Steel moment frame retrofit for school

[Enable]

This is an extension of this thread where I was called in to see if we could fast-track some lintels above the windows. Obviously that did not occur and we are now fashioning a system to remove the glass block in its entirety and in-fill with CFS.

Framing System of Work Area
0. Single story school (approx 3.7m or 12 feet) with pinned columns spaced at 5.3m (17 feet)
1. Steel W150x30(W6x20) columns have their strong-axis oriented perpendicular to windows / wall plane
2. Steel angle 150x150x6.4 (6x6x1/4) connecting week axis of columns at top. Seems to be riveted into adjacent 2x member behind. Seems like the 2x member collects shear from the roof diaphragm and distributes to the angle which then puts it into the weak axis of the columns
3. Roof decking is old style planks on angled direction across joists
4. Roof joists run parallel with the exterior wall. What those joists frame into is unconfirmed.

Exterior View

General Framing Scheme

Angle Connection

Problems I have
1. I don't see a true lateral system on the exterior walls and I'm fairly confident the roof diaphragm would be considered flexible (angled plank on 2x12 members) so the load is going to the exterior framing system (based on trib area) and into the ground...somehow. Right now I'm thinking there is just such a mass with the glass-block infill (it's almost the entire top half of the column) that inadvertently it's helping resist the lateral forces. So by taking it out I would making these columns cantilevers (more or less since but of course the angle would help some) which will increase deflection in the weak-axis considerably
2. Whatever solution we come up with cannot induce too much of a different load path else things start moving / cracking since all existing finishes are more or less staying
3. I'm hoping someone is just going to tell me I have missed the obvious here and DUH there is a system as is or that the CMU walls are likely hogging everything so not to worry

Solutions (??) I have
1. Create a weak-axis moment frame at the top of the columns to recreate the lateral stiffness of the block as much as I can
2. Place a pin-connected HSS member lower down above the window to pickup the CFS infill just above the new window location
3. My preliminary connection thoughts are attached. I'm rather concerned about inducing bi-directional moment capacity at my top connection. Ideally I want to avoid doing that and have a moment frame in the weak axis of the W-column only...don't know how possible that is. If you look at my sketch I think on the moment connection I have with a top plate / angle underneath might work well if I shorten the plate to within the confines of the HSS considerably (e.g. maybe a 3" wide plate but thickeer)
4. Key is to also think of fitup. We need to provide quite a bit of tolerance for the installer / erector given things will be totally baffed / out of plumb / whatever and they cannot be expected to line up holes to within 1/16" here

Misc Questions
1. I was going to place CFS infill on top of the HSS and use some powder actuated fasteners to hold it down. But this will puncture the HSS. How worried do I need to be about creating these holes given that the HSS is on an exterior wall and capped both sides? Any concerns about moisture / water infiltration / etc (assume there is a good water barrier on the exterior sheathing at least initially)?

Load information
1. Equivalent inplane lateral wind point load at top of moment frame of 10kN (2.3 kips)
2. Moment demand (assuming a moment frame) at top connection in weak axis of column of approx 6kN*m (4.4 kips*ft)
3. Vertical loads on HSS framing members (self-dead load only for the most part. Some infill CFS but nothing to get riled up about)

 

[phamENG]

Off the cuff...what if you make the CFS infill a shear wall of sorts? You'll have vertical and axial reactions at the connections of the top angle and bottom HSS to transfer the boundary shear but you won't have to worry about a moment connection.

I wouldn't worry about water getting in there - that's a common connection detail. If flashing/moisture barriers aren't installed or maintained, there will be lots more problems than just this. Though water-filled HSS in freezing temperatures makes for some pretty impressive failures.

 

[Greenalleycat]

Just to make sure I've understood you correctly:
In the direction of concern there are two full-height masonry walls up the guts of the building (that presumably have loads of capacity)
However, you do not have a rigid roof diaphragm, so therefore tributary width assumption of a flexible diaphragm would put loads on the outside walls
These outside walls are basically 100% glazing with, at best, some collector beams (angles) framing into the sides of columns
There's potential that these columns could be turned into weak-axis moment frames to help resist the wind, and that is what you're asking for thoughts on?

If this is correct, here are some thoughts:

1. The building is very long and not very wide for the wind direction of concern. Could you rip out the windows at the top end of the building (where your wind arrows are drawn) and bang in a large, stiff moment/braced frame in this bay? Design this to take all the load on that side. There's not really wind from the other long end due to the adjacent building at the bottom of the page

2. Alternatively, those shear walls look pretty massive. Your diaphragm in this direction is really deep so probably stiff AF
Can you put a number to a beam model that has two internal pinned supports (walls) and overhangs either side (out to the glazing wall lines) to show the deflection under wind
I would think that the diaphragm would be so stiff that this model would show very little real-world deflection in the ceiling so you could just bang all the load through the masonry walls
Frankly, this seems like the true load path

3. If you want proof of structural glass... the Ministry of Education in NZ tested a whole bunch of prefab classrooms in NZ that were built through the 1900s
Most of them had no identifiable load path, same as yours
I think they found the buildings were something like 3x stronger with the windows shut lol

Edit with another thought:

4. Right now it looks as if you have a wall that is half-height framing, top half glazing.
Your columns sit between glazing so basically are supported by wall to their half height then unsupported to roof level.
This effectively makes them a two-support cantilever column (support at ground level and sill trimmer height)
Can you put some numbers to that?
The trib width for loads looks low, and it looks like you have a bunch of columns so this could work?

Edit 2:

No idea what literature you have access to for assessing such buildings, but here are a few relevant references from here that could help you with putting a number to capacities of diaphragms etc

https://www.building.govt.nz/assets...re/seismic-assessment/c9-timber-buildings.pdf

And have a look at Section 5 of this report lol

https://assets.education.govt.nz/pu...e/FinalReportOnStructuralTestingCarterton.pdf

 

[kipfoot]

My take is that you've probably already spent more time considering the lateral load system of this building wing than the original designer did.
I think I would:
1) assume that the interior concrete block walls were intended to be sufficient for the one story building in its long axis; and
2) place the new cold formed studs in a way that replaces the capacity of the glass block, perhaps with light gauge diagonal straps.

re: your pinned connection, If I want more weak axis capacity in the connection, I'll detail the connection with a C on one side or the other in lieu of the plate.

 

[phamENG]

I missed the full height CMU walls. Those are likely the primary shear walls. Wood may be a flexible diaphragm, but it's capable of cantilevers. I agree with kipfoot's approach.

 

[Enable]

Thanks team. No doubt the CMU walls have oodles of capacity. So the consensus seems to be given span to depth ratio the deck will cantilever and deliver most of the shear to the CMU with the angles picking up incidental loading? I can buy that and is kind of what I wanted to hear.

I didn't think (still dont) that I could create a stiff enough shear wall out of CFS to match the glass block, which is why I got somewhat concerned about actual paths since the deflection would increase mightily if the exterior walls were actually taking their tributary share of the load. But for incidental loads we should have no problem.

Kipfoot: thanks for the c-channel idea

phamENG: thanks for confirming about the PAF. What's the typical standard spacing you'd use for nominal loads something like 1 every 8-12" c/c staggered?

Greenalleycat: You had it right, and thanks for those links. The structural glazing is hilarious! I also did checks as cantilevers and it's a no-go. The deflection is some 20mm. Load is small but the columns are so freaking unstiffen in the weak axis direction.

 

[Greenalleycat]

Sweet as. If the column are that flexible then I'd bet my cat that the diaphragm stiffness will be >>>>> by comparison, so a rigid diaphragm assumption would be very valid for this model
So, chuck all the load through the central masonry walls, stop worrying about the global lateral system, and start focusing in on all the connections and details in the end wall to masonry wall load path

Edit: drew my < and > signs the wrong way woops. Diaphragm >>>> Columns / outside wall line

 

[phamENG]

What's the typical standard spacing you'd use for nominal loads something like 1 every 12" c/c staggered?

It's been a while since I've had to do that detail, but that sounds about right. That'll give you about 140plf of shear capacity for thin light gauge depending on the manufacturer. So you may be able to space it out to 16" if that's your stud spacing.

 

[KootK]

Yeah, I'm guessing this is the Power-I Formation.