CLT roof with opening

[greznik91]

Hi,

I'm designing a CLT roof - span in main direction is 6,20 m.
CLT slab is 200 mm thick and it has opening in the middle of the span 1,0 x 1,0 m.
I made FEM analysis and the strip model analysis. I figured that the bending capacity is OK but
I don't know how to approach huge spike of stresses at the edges of the opening?
Would you reinforce them? What do you suggest (steel angles?) and how to calculate it?

Thank you for reply.

Regards.

 

[MotorCity]

Do the stresses around the opening exceed the allowable stresses or are they just too high for your comfort level?

If they do not exceed allowable stresses, I'd leave it alone. Otherwise, I may look at adding some flat steel bars or angles below to assist with load sharing.
No FEA needed, just consider the steel acting as a beam and transferring the reactions to the adjacent panel strips.

 

[jayrod12]

Also, if you tighten your mesh, how does that affect the concentrations you're seeing. If the tighter mesh causes tighter areas of concentrations, then it could be a modelling anomaly rather than a real life issue.

 

[phamENG]

Is it appropriate to model a CLT floor like that? I know there is load sharing between adjacent panels, but is it that robust that it truly behaves as a monolithic slab?

Are you doing two layers of CLT panels, or are you trying to look at each layer of a single panel?

 

[jayrod12]

TO be fair, his 2.75m wide panel would potentially be a single panel. The ones I had on my project ranged from 2m to 3m wide. Essentially as long as it's shippable they can build them. So he may be trying to approximate the orthagonal layers.

 

[phamENG]

Got it. I guess that's just a hair over 9ft. Doable with oversized permits here. If that's the case, I'd probably leave the internal performance of the panel up to the manufacturer.

 

[jayrod12]

...I'd probably leave the internal performance of the panel up to the manufacturer.

This is a good idea, however a word of caution to some. Ensure that your specifications during tender indicate that the supplier is responsible for final detailed design of the planks. On my first CLT project we had assumed that CLT was a proprietary product and as such would be designed by the supplier for whatever loads and openings we showed in the drawings.

Turns out that most of the manufacturers, although capable of providing that design assist, bid projects as if their product is a commodity lumber type product and unless mandated by the project documentation do not include the design and detailing of the planks around openings etc. They expect the EOR to do all of that.

So in my situation, we ended up working out a fee adjustment with the supplier for them to provide the design assist as well instead of me frantically having to figure out how to deal with openings, etc.

 

[phamENG]

Thanks, jayrod. Good to know. My only CLT exposure was for a rigging plan and connection design. To my knowledge, CLT has been used on exactly one project in my region. One panelizing company in the area looked into offering CLT elevator and stair shafts, but they seem to have abandoned it.

 

[jayrod12]

There's only been one actually built so far locally. Just so happened to be one of mine.

We've got two more upcoming on the books in the office, one has started piling, other is still in design stages.

Where I am, steel is still really expensive since Covid, so heavy timber buildings are coming in cost competitive.

 

[KootK]

Similar to motorcity, I'd approach this almost identically to how precast holliwcore plank openings are handled. Most likely, that means a header beam. If anything, I suspect that clt sidelap connections are worse than holliwcore for out of plane shear. No FEM on this for me unless realistically captures joint behavior.

 

[jayrod12]

If anything, I suspect that clt sidelap connections are worse than holliwcore for out of plane shear.

This depends on the joint detail. For their ideal OSB spline sidelap connection I would agree, however for an overlap or keyed joint, I'd say they perform far better in terms of out of plane shear than the grout friction we account for in hollow-core.

And maybe better isn't the right term, more predictable would be a better description of the overlap performance.

 

[KootK]

If the opening can be less than one plank width, then you can often hang the interrupted plank from its neighbors with z_clips or the like. This probably implies the ability to strategically locate the opening.

 

[KootK]

A grouted keyway is good for like 700 psi. A lapped CLT connection encourages delamination and/or local tension perpendicular to grain. I'll take the groutred keyway all day long, even with the qc issues.

 

[jayrod12]

Ballsy, I've only ever used 0.2 MPa which is drastically less than 700 PSI. That's also why I changed the term from better, to more predictable.

Either way, back to the question at hand,

I'd be surprised if a single plank worked for this size opening upon further examination unless the panel thickness is drastically oversized for strength elsewhere. Potentially if the opening was split in half for two planks, there's a possibility it works out, but there is no moment capacity across the joint between planks. So that would need to be treated as a pin connection with the capacity of the joint kept in mind for load transfer. I'd likely be ignoring out of plane shear transfer as well and just look at the weak axis plies cantilevering to pick up the areas affected by the opening.

Essentially I'd be looking at the remaining unaffected width of the panels as the only effective portions of the panels for overall load support. The areas affected by the opening would just be transferring their regular design loads back to the un-cut primary plies.

 

[KootK]

70 psi. Extra zero there. Still, that beats the heck out of what is essentially a wooden corbel. You could screw reinforce it I suppose but that would be a whole totta screws.

 

[greznik91]

Thank you for your opinions/suggestions.

I don't have an issue with bending.
If there was no opening the bearing capacity of the slab is around 50%.
I'm not really fond of high shear stresses at edges of the opening since I'm not really sure how to consider them for design...
What do you think about reinforcing it with steel angles (150 x 150 x 10 mm)?

 

[jayrod12]

I'd likely be doing what KootK indicated with just a bunch of inclined screws to reinforce the corners. But you're the one that needs to be confident in the design. I don't see what the two angles running left right are doing for you? Also, to run your angles past the opening either requires you to remove the flange of the angle (then what was the point of using an angle section) or cutting a flange slot in the entire panel (to what benefit?).

I'd be looking at the weak axis bending capacity of the panel to span the opening. Like the attached sketch. And if you're worried about the primary plies peeling from the bottom of the panel at the corner of the opening, I'd be looking at some inclined screws to tie the bottom ply to the upper plies.

 

[KootK]

I totally misunderstood the direction of span here. Would your transverse reinforcing not interrupt your primary bending capacity? I feel as though your in pretty good shape here without the reinforcing. I imagine you are governed by rolling shear?

 

[jayrod12]

If the span were the other direction, I'd be even less concerned.