Topping slab in hollow core plank as diaphragm? 1

[lutein]

I am working on a 5 story hollow core plank on CMU wall projects in Florida, ie no seismic. As permitted in PCA hollow core design manual, I managed to get grout filled core to work as diaphragm action lateral load distribution. However, I have heard concerns of reliability of grout filled cells as diaphragm agent since they may crack. I also do not need the topping slab as composite action for Gravity load. The owner will like to use gypcrete and subfloor and leveler for floor finishes. This will greatly reduce the footing design.

I was wondering what's your experience in hollow core plank with no topping slab on a multi-story building. Have you always spec topping slab on hollow core plank?

Please help....thanks

 

[Ron]

In my opinion, the addition of a topping slab helps to better distribute forces throughout the structural system. Leaving off the topping slab concentrates forces at the top of the supporting wall, which would contribute to the cracking concern you have.

 

[KootK]

Firstly, I share the concerns over cracking and agree with Ron that a topped system would surely perform better. But the real question here is whether or not an untopped system would perform good enough. It may.

Untopped precast has been used extensively in my area in underground parking facilities where it is topped with asphalt and performs well. This isn't quite your scenario but it's something.

Cracking may not be as big a concern as it would at first seem. Within reason, concrete retains its shear capacity after cracking. This should be all the more true here where it's really a permutation of shear friction at work between planks. I'd want grout that was non-shrink or even slightly expansive like Sika212. Check with the precaster to determine what makes sense.

One thing to pay very close attention to is the need for diaphragm tension chords running perpendicular to your planks. I believe that the untopped diaphragm testing was done with masonry bond beams as the chords and dowels extending from the bond beams into grouted plank cores. You'll need something similar to prevent large tension cracks from opening up between planks and truly compromising your shear capacity. And chord strain is important here too, not just chord strength.

Diaphragm chord continuity is also an issue as well. If the perimeter of your building -- and the planks -- jogs in and out, you'll need to give careful consideration to how you'll maintain your diaphragm chords around the corners. It can be done but it's not always easy.

Diaphragm issues aside, differential plank movement may well crack your gypcrete extensively. If your flooring is carpet and pad, no problem. If it's stone tile, not so much. You'll want to educate your client on this and carefully manage expectations.



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.

 

[lutein]

Ron and KootK,
thank you for the valuable input, greatly appreciate it.

Another alternate to a diaphragm topping slab is physical plank to plank connection with shear ties or embed plates. Totally agree on the chord strength and continuity. Found a good article abstract from FEMA P751 showing calculation of diaphragm strength for untopped precast floor. Based on more research, it appears that untopped precast floor system is permitted in SDC A - C buildings.

I have been trying to find CTA technical bulletin 80b3, but have no luck.

I was wondering what kind of gypcrete thickness would be appropriate for this kind of application.

 

[KootK]

1.5" min. Thicker at the ends of the plank to accommodate camber.

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.

 

[KootK]

This is likely similar to the CTC doc: Link.

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.

 

[Brad805]

I suggest you speak to your hollow core suppliers before specifying connection plates. This is not very common as it really complicates their production. Plates tend to get in the way of the extruder. It is pretty easy to add connection plates to double tees since they are not extruded, and that is what is common in parking garages. The company we work for has done un-topped precast floor elements, but they are small and not a hollow core producer. It is very easy for them to add connection plates. Personally, I would add a topping. I am not a big fan of gypcrete, and I do not know how well it will perform on hollowcore over its lifespan. The topping will also help to damp the system in case you have any long clear spans or little mass above.

By the way, Piekko has a neat beam system that works well with hollowcores in case you are interested.

 

[lutein]

Agree, I am all for topping slab, no doubt it's the best practice for a reliable diaphragm, I was just wondering if untopped hollow core plank has been used for multi story structure.

I have also found a project done locally where light weight 2" non structural topping is used.

 

[Ron]

lutein....I have not seen a multi-story hollow core/masonry project in Florida without a topping slab, so you might want to consider the standard of care for such design. The roof section sometimes does not have a topping except for lightweight insulating concrete.

Like Brad805, I'm not a fan of Gypcrete.

 

[lutein]

Ron: I appreciate your concern, and I am myself as well. Just want to clarify: you mentioned "I have not seen a multi-story hollow core/masonry project in Florida without a topping slab..." this 'topping slab' that you referenced, do you mean by structural topping slab for diaphragm or just lightweight concrete topping slab (not gypcrete) for subfloor?

here are several options I have seen, and Option 1 is what I have always used:

Option 1: Structural topping slab (f'c=3000psi) reinforced with wire mesh, serves as lateral diaphragm, and subfloor at the same time. So, the CMU walls above will bear on top of this topping slab per standard PCI detail.

Option 2: Non-Structural topping slab (f'c=3000psi), lightweight concrete, for subfloor use only. CMU walls above will bear on top plank per standard PCI detail. Topping slab comes later.

Option 3: Possibly the lowest cost. Gypcrete on plank.

 

[KootK]

@lutein: in your original post, you mentioned that the gypcrete topping would significantly reduce foundation demands. The impact can't really be all that much can it?

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.

 

[lutein]

Well, if we go with 0.75" gypcrete, we save about 18psf per floor, and for 5 stories it's 90 psf, and that's significant.
0.75" Gypcrete = 7 psf
8" bare plank = 54 psf (from Old Castle catalog)
8" plank + 2"topping = 79 psf (from Old Castle catalog)



For buildings in SDC A-C, either method is permitted by code and there are published literatures, including PCI hollow core manual, with step/step calculations for untopped diaphragm strength calculations. I just don't want to do something stupid that no one out there has done before; if most projects in the south have always been plank + toppings, I certainly want to do the same thing. However, if it's the other way around, then I might consider untopped floor.

 

[KootK]

If you compared factored loads including a 40 psf LL, the difference is only about 15%. And if you include the considerable dead weight of your block walls, the impact would be considerably less still. Lastly, 3/4" gyp on a cambered plank system is a little optimistic unless you're dealing with very short spans.

Ron has given you the Florida perspective and the general consensus here seems to be that topping is the way to go. Before you step up and do something a little off reservation, but numerically defensible, I think that it's important to ensure that there are legitimate, significant savings to be had. Hence my comment regarding foundation savings.

Do we know the difference in installed cost between the topping and the gypcrete, irrespective of the foundations? If I were the owner of this facility, I'd want some meaningful savings to be demonstrated before I'd consent to having my floors topped with crusty toothpaste.

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.

 

[lutein]

Kootk: Agreed, you have a good point about the labor to pour topping versus gypcrete. Very insightful discussion and greatly appreciate the opinion. I guess I need to contact the contractor for some pricing exercise.

 

[KootK]

I'm designing a couple of 14 story buildings that will have plywood diaphragms. No typo there. I'll be submitting the concept on this site for critique, just as you have done. And I'll be asking to see the cost comparison relative to concrete floors before I'm sold on the concept.

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.

 

[Brad805]

There is no question there is a labor savings to pour gypcrete and it is quite significant for this one building component. I have watched crews pour a complete floor in hours (wood floors in motels) and it is a cheaper trade because pretty much any laborer can do it since it is self levelling. You have a few laborers filling a pump (not much bigger than a grout pump) with gypsum somewhere at grade and a guy running around with a hose on the floor; whereas, you tend to see the same concrete trades and finishers when it comes to toppings (in our areas anyway). We've done up to four stories that way. I know we can make concrete self leveling too, but our trades always seem to cringe at the cost of super P. Short sited I know, but that is what I see the guys paying for the mud think.

Now is the cost savings of the gypcrete v. concrete topping significant on the basis of the overall project, that I doubt, but developers always disagree when they see the option that says they can save $x on the overall project.

This will come back to your comfort level. If your loads are that low and the spans are reasonable to avoid vibration problems without the added weight, I suspect you can make the numbers work just like Koot's example shows. I would bet you can drag your chord forces into the walls or concrete frames using a variety of connectors. Relying on the grout bond between the panels is the part where I would have the most difficulty. I suppose one could add a frame for redundancy, but then you are probably right back at the cost of a topping. I still do not like the serviceability aspect of gypcrete, but that is a discussion for you and your client. That discussion I suspect will be short unless they care about something other than cost. Be certain of your position before you go down that path.

Assuming your hollow cores are in the range of 24 - 40ft, I bet your pre-camber will be in the range of 0.5" - 1.25" when they are first delivered. Will the mech/elec trades be allowed to install their piping or conduit in the topping? That can be a good savings as well if the architect and timing permit. I would allow for something greater than 3/4" for the topping as well.

As for the footing cost savings, that is minor in the grande scheme of things. Unless you are minimum wall reinforcing levels, you could spend a bit more time detailing the wall reinforcement and you would save more on the basis of area alone. Now if you have problematic soils, I totally understand.

 

[Ron]

lutein.....Option 1

 

[lutein]

Ron: I would really like to go that way.
Brad805: What kind of gypcrete thickness you've used in the 4-stories? 3/4" or 1" or 2"?

Thank you all for input. It would have been an easy decision if the code were to forbid the method.
Calculated a maximum ultimate diaphragm shear of 0.8 k/ft, which is extremely low. It will be a tough debate.

 

[Brad805]

1.5" is common for wood framing (1 extra bottom plate for screeding). With precast + topping situations 1.5" - 3" is what I see more commonly.