Strengthening an Existing Hollowcore Plank Roof System

[KootK]

I've been engaged to look at a hollow core plank roof that needs to go from 20 PSF to 100 PSF. And probably some additional dead load too. I'd like to round up as many options as I can for strengthening. Here's what I've got so far, acknowledging that many of these options may not actually be viable for such a heavy increase in carrying capacity.

1) FRP on the bottom.

2) Steel strapping on the bottom.

3) External post-tensioning on the bottom.

4) More bonded topping on the top.

5) Parallel beams on the bottom, perhaps acting compositely.

6) A perpendicular beam line to break up the span.

7) An entirely new overbuilt roof structure shielding the plank from load.

8) Permanent sealing and pressurizing of the space below.

9) Very strong opposing magnets on the floor and ceiling below.

Any other ideas worthy of consideration? Come one, come all... Any comments on which of the above might actually be up to the task of a three to four fold load increase?





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.

 

[ajk1]

2) I would not do steel starpping on the bottom. Tried it 30 years ago on soffit a c.i.p post-tensioned slab and the adhesive of plate to concrete did not develop much bond.

1) Soffit FRP, is there not a limit to the percentage increase in capacity that CSA Standard will permit for FRP?

3) External p.t. may be problematic with developing a detail for end anchorage, etc. I stay away from p.t. wherever possible. Sounds very clever but it has its own set of issues.

4) I don't like bonded structural topping because I find that no matter how hard we try with surface prep, and with various bonding agents from cement paste to epoxy, that when we do bond tests there are some areas where the bond strength is low, sometimes very low. Perhaps if we could find a really conscientious contractor these issues could be overcome, but these days a "conscientious contractor" seems to be an oxymoron, at least among"low bidders". Bonded structural toppings are used, but they scare the hell out of me. Added to that concern is that if subject to freeze-thaw cycling (which maybe your case is not if it is below insulation), the bond line can eventually fail... I know of a parking structure repair where this happened (not one of our jobs, thank goodness).

7) An entirely new overbuilt structure seems the safest to me. Not elegant, but it is safe and predictable in its performance.

 

[jayrod12]

Based on some info I have here, for that depth and span, 4.1 kPa superimposed would be available. However that doesn't account for topping weight, nor topping contribution to capacity if it is a bonded topping.

so there may just be some more capacity than the 20psf design load.

Granted that's just one manufacturer's literature (LaFarge) and for the depth and span there would be no lower capacity than that. Take from that info what you will.

 

[TehMightyEngineer]

I found similar info in the PCI Design Handbook as Jayrod for typical 10" hollowcore with 4-1/2"Ø strands with a 2" composite topping. Looks like you might have more capacity unless the precaster used smaller strands than 1/2"Ø.

Professional Engineer (ME, NH, MA) Structural Engineer (IL)
American Concrete Industries

http://www.americanconcrete.com

 

[JAE]

Where it's permissible, I assume that this kind of thing is better done topside, right? Even with topping?

The way we had it done was on the bottom since the bottom was fully deteriorated and the top was still in good shape.

For your condition - yes, perhaps inserting rebar into the cores from the top might work.

Check out Eng-Tips Forum's Policies here:
faq731-376

 

[Ron]

KootK....your No. 10 will likely only get you another 20-40 psf LL and will add to the dead load of existing....I would vote "No".

 

[KootK]

KootK....your No. 10 will likely only get you another 20-40 psf LL and will add to the dead load of existing....I would vote "No".

Thanks for the input Ron. Why would adding a new precast overlay only get me 20-40 psf? Seems to me I could have it designed for any load I wish with my magic, EOR specifier wand.

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.

 

[Ron]

Anything other than 20 to 40 psf would be non-standard, so you would certainly need your Magic EOR Specifier Wand and a larger wallet!

 

[KootK]

Eh, I've ordered prestressed solid plank before. This is going to be an expensive adventure. I suspect that premium precast will get lost in the noise.

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.

 

[dik]

If you use #6, you should sawcut down a couple of inches on the beam centreline to make the HC slab behave more like a simple span condition. You may have to jack the slab to eliminate 'preflex' that may bind the sawblade. If the new moment from the new loading on the new span is approximately the same as the old moment on the old span, it should work. Also check new shear against old shear and length of new span is greater than 8' or 10' for bond on the cables.

Dik

 

[KootK]

Option 11) Build a network of trusses above the precast and then cut the precast planks in half and suspend them from the trusses.

Critical and/or constructive comments welcome. Probably need to give a bit more thought to truss bracing.

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.

 

[jayrod12]

Wait, why the second part?

cut the precast planks in half and suspend them from the trusses.

 

[KootK]

To cut the span of the planks in half and hopefully reduce bending and shear.

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.

 

[jayrod12]

Obviously. But why, if you're building a steel system, what's the need to improve the performance.

 

[KootK]

The steel system is overhead. As in 15-20 ft overhead. The precast is still the floor/patio surface. My sketches must need some work.

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.

 

[TehMightyEngineer]

Something, something, Hyatt Regency and stuff. It could work for sure but I would think it would require a lot of careful thought and detailing to ensure a durable hanger connection. Seems like if someone had a dance party or something up there and one of those hanger plates had broken or corroded through or whatever then you're going to have a bad day. I'd design it to be able to survive the failure of at least one of those hangers.

Also, unless you're hanging each hollowcore you'll want to look into shear transfer of forces between panels (PCI Manual for the Design of Hollow-Core Slabs covers this).

Overall I don't like it, it just seems like not the most efficient way to do it.

I know you looked at this earlier but how come you rejected #10, overbuilding with new precast panels? A 4' x 8" HC slab untopped gets you 110 PSF. Or if you needed to go solid slab you could do 10" thick untopped. These would probably be custom but not too much more than any stock slab panels.

Professional Engineer (ME, NH, MA) Structural Engineer (IL)
American Concrete Industries

http://www.americanconcrete.com

 

[jayrod12]

My sketches must need some work.

Never, except for maybe an elevation or two.

I'm not a fan of it from a durability sense either. Neat idea, minimal redundancy. Leaves me with the heebee jeebies.

 

[KootK]

Thanks for the feedback gentlemen. I agree, durability of the hanger connection is a concern. One somewhat mitigating factor is that my direct client is actually the building envelope firm engaged on the project. If nothing else, I should be able to negotiate some solid waterproofing details. I'll absolutely mention durability concerns in my feasibility report though. And the monstrous thermal break.

Part of my situation with this is that a few structural firms in town have already turned up their noses at this project, deeming it infeasible given the parameters involved. And they're probably right. The project team engaged me hoping that I could come up with some innovative, perhaps a little riskier than usual, solutions.

Something, something, Hyatt Regency and stuff.

How dare you summon the devil by name in reference to my work! Just kidding. The shock value was healthy for me.

I'd design it to be able to survive the failure of at least one of those hangers.

Also, unless you're hanging each hollowcore you'll want to look into shear transfer of forces between panels (PCI Manual for the Design of Hollow-Core Slabs covers this).

These comments lead me to believe that you may have missed something in my, clearly lacking, details. The hangers support a continuous, field assembled beam that, in turn, supports the precast continuously. It's still scary but, perhaps, a bit less scary than you're thinking. I never would have even thought to utilized plank to plank shear transfer. Even I'm not that much of an engi-cowboy.

I know you looked at this earlier but how come you rejected #10, overbuilding with new precast panels? A 4' x 8" HC slab untopped gets you 110 PSF. Or if you needed to go solid slab you could do 10" thick untopped. These would probably be custom but not too much more than any stock slab panels.

I'll take another look at this. Some issues that you may have thoughts on:

1) I need about 200 psf superimposed, factored over a 31' span.

2) There's a desire to keep the system depth increase to 6" or less. Granted, minds can be changed by enormous cost savings. The teams already looking at a steel overbuild option which would be much deeper.

3) I imagine that I'd need to put the new planks on bearing pads of some kind to avoid interaction with the existing planks. No big deal but more depth.

4) This thread is related to this thread: Link. You're helping out there too. So, with plank on plank, the shear would still have to travel through the existing planks unless I finagle a way to get the new planks right into the existing bearing wall. And I'd have to deal with in plane shear transfer, bearing, and all the other stuff that I mentioned in the other thread. Not necessarily a deal breaker but some noodling required for sure.

Never, except for maybe an elevation or two.

With all of the prior discussion of a steel overbuild, and me not showing the tops of the trusses, it boggles my mind how I didn't anticipate that this would be confusing. Doh! Anyhow, stay tuned for #12. Considerably more difficult to communicate in a few brief sketches.


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.

 

[TehMightyEngineer]

It's still scary but, perhaps, a bit less scary than you're thinking.

Ah, yes, I see what you're taking about now. I was imagining discrete support points, hence the Hyatt Regency "joke". Your system seems much less scary, though now I'm wondering how you're going to feasibly cut the HC slabs continuously to install said beam.

1) I need about 200 psf superimposed, factored over a 31' span.

2) There's a desire to keep the system depth increase to 6" or less.

Blarg, 6 inches and 200 PSF is no es bueno, deflection is through the roof (or should that be floor?)
I'd say you're looking at a minimum of a 10 inch HC slab to support that.

What happened to the 100 psf from your original post?

3) I imagine that I'd need to put the new planks on bearing pads of some kind to avoid interaction with the existing planks. No big deal but more depth.

Why? New slab will be stiffer, right?

4) This thread is related to this thread: Link.

Yeah, I think you'll have a shear problem with most systems. Noodling needed for sure, but not insurmountable.

Professional Engineer (ME, NH, MA) Structural Engineer (IL)
American Concrete Industries

http://www.americanconcrete.com

 

[KootK]

though now I'm wondering how you're going to feasibly cut the HC slabs continuously to install said beam.

1) Shore 'em.
2) Saw cut 'em.
3) Shove the WT up into the slot from the bottom.
4) Weld the top flange on.

Is that bad? I'll believe you if you say yes.

What happened to the 100 psf from your original post?

That was just the new occupancy load on the existing system, already designed for some stuff. And there have been some adds. New planks would need to support 30 psf roofing, 100 psf occupancy, 40 psf average snow drift, and 150 psf intensive green roof at 50% coverage. Easy peasy, right?

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.

 

[TehMightyEngineer]

Is that bad? I'll believe you if you say yes.

No, I just didn't know shoring was an option as you earlier said the lower story was occupied.

That was just the new occupancy load on the existing system, already designed for some stuff. And there have been some adds. New planks would need to support 30 psf roofing, 100 psf occupancy, 40 psf average snow drift, and 150 psf intensive green roof at 50% coverage. Easy peasy, right?

Oh my. I see now why the others engineers bailed. Well, I still think precast prestressed or cast-in-place PT slab perhaps using the existing HC as formwork will be your best bet and getting them the system they want with the least compromises.

You probably answered this earlier but are we allowed to do stuff below the existing hollow core slab?

Professional Engineer (ME, NH, MA) Structural Engineer (IL)
American Concrete Industries

http://www.americanconcrete.com