Precast Hollow Core Plank Parallel to Wall 1

[RFreund]

What is your detail preference for hollow core plank spanning parallel to an exterior wall. Could be a basement wall or just exterior wall. Do you bear on the PC on the wall, or do you provide an angle with slotted holes? Any reason?

EIT

http://www.HowToEngineer.com

 

[bones206]

I think the clip angle is better, but with concrete anchors into the plank instead of a weld plate. As I understand it, for a precaster to provide weld plates in hollowcore can difficult, depending on the casting method they use. But providing a solid end core is easier, which allows you to use a higher strength concrete anchor.

You also don't have to bust out a notch in the top of the plank for a dowel bar. I think the clip angle is stronger and more robust than a dowel bar. Having a vertical dowel is generally a safety hazard that is better to avoid if possible. Field bending is not the biggest deal in the world, but can have variable results, especially when they are working on the edge of an elevated platform.

I would avoid bearing the plank on the wall since it can induce serviceability problems from differential deflection. Here's a snippet from the PCI Hollowcore Manual:

 

[r13]

A suggestion.

 

[JLNJ]

Look at all your forces and decide what is most critical. Are you transferring shear from a diaphragm into the wall? Is the wall an endwall which needs hold on the the slab for stability? How much live load deflection do you expect in the planks? Do you have a topping or not? All these things play into the design.

The slotted clip angle can work, but you will have some significant prying action if the only thing that stabilizes the wall is the fastener to the wall. I’m always concerned that the slot will bottom out before you want it to or the wall won’t be grouted where the fastener needs to be placed.

If the floor is stiff enough, I like to lap the plank onto the wall. I’m not big on field bending the rebar. If it’s masonry, I’ll run the verticals straight up through past the plank, relying on a sort of shear friction. If the endwall pullout forces are large you will need to think not only about how to hold the wall on the plank, but how the plank spans horizontally. Multiple planks may need to be engaged, especially if there is no topping.

 

[RFreund]

Excellent responses. Mostly because each of you picked a different side and offered your Pros/Cons.

Bones -> I agree with your concern. Does the PCI give any guidance on when to avoid a "rigid wall"? or just in general that it should be considered?

JLNJ -> what you say is "stiff enough", L/600? how are you figuring the deflection prior to fabrication? typically I would just rely on load tables rather than actually running deflection calculations. Interesting consideration on the hollow core spanning horizontally. Typically i just give the reactions and the PC engineer needs to figure that part out.

Thanks again!

EIT

http://www.HowToEngineer.com

 

[bones206]

Does the PCI give any guidance on when to avoid a "rigid wall"? or just in general that it should be considered?

I don't recall seeing much discussion about it besides that diagram. Might be more to be found in other PCI docs. The PCI hollowcore manual can be found with a google search, just FYI.

I don't have much personal experience with the lap joint causing issues. But in theory I can see where it could lead to some cracking issues in the topping slab or floor finishes. I think KootK is an expert on this stuff, so hopefully he will chime in. I'm still pretty much a hollowcore newbie. Pretty sure that my clip angle preference is due to something KootK mentioned in a past thread.

 

[JLNJ]

As far as the deflection goes, most of the time you don't get much. Most of it comes out under its own DL and when the topping it applied. After all the DL is on you grout under the planks. I'm thinking the DL deflection is usually around three times the LL deflection for topped planks. I might feel differently in a more industrial or storage type use where the LL is a greater percentage of the TL.


Say you have up to a 1/2" of live load instantaneous deflection. If the planks are 48" wide you have at most 1/2" at mid-span to accommodate with warping planks. For most offices, schools, and the like, the deflections are small and transient and there is a lot of load sharing between planks.


If I recall correctly, KootK likes the clip angles with the slots. The placing of the embeds and field welding thereto doesn't give him much pause. Having the entire system held together by an exposed steel clip might be an issue if you have a fire rating to maintain.

Lots of things to consider and more than one way to skin the cat. no matter how you detail it, some contractor will tell you he's never seen it done that way before!

 

[RFreund]

Thanks again for the responses. This atleast confirms that there is more than one way people are doing it.

some contractor will tell you he's never seen it done that way before!

Haha, very true.

EIT

http://www.HowToEngineer.com

 

[bones206]

I looked again for discussion in the PCI manual, and was able to find it this time:

While drag struts and boundary elements may have a vertical stiffness similar to that of the deck, the lateral force-resisting elements will usually have a significantly higher vertical stiffness. The connections to the lateral force-resisting elements will tend to be vertically rigid. While strength and toughness of such connections are certainly important, it is equally important to consider everyday performance of the structure. At rigid vertical elements, it may be desirable to allow hollow core slab camber growth or deflection to occur without distress at the connection.

Figure 4.6.4 shows potential damage at the first interior longitudinal joint when a vertically rigid connection is used. The potential for distress is dependent on the span and the actual applied loads. Short, lightly loaded spans may experience no problems. The effect of different vertical stiffnesses may be accounted for by:

• determining that distress will not affect the strength or performance of the system;
• locating vertically rigid connections near the hollow core slab supports where vertical movement is minimized; or
• providing allowance for vertical movement in the connection detail

 

[KootK]

@RFreund:

1) In your neck of the woods, I'd expect you to have a hard time doing anything other than the CIP version of the detail below without ruffling feathers.

2) Do consider whether or not allowing for differential movement at the slab wall joint is tolerable. Oftentimes, it's not and I'm inclined to consider that potential performance issue a more serious concern than the more arcane, if wholly valid, concerns about the impact of differential deflection on the slab itself.

3) In practice, 95% of hollow-core jobs are utterly riddled with differential deflection issues. Pretty much every re-entrant corner in the floor plan creates this problem. I used to worry about this a great deal but, with extensive experience with these products, I've relaxed considerably. My current thinking on this is:

a) These slabs tend to act like two way slabs to a far greater degree than we give them credit for owing to their torsional stiffness creating system response between individual units. For the most part, if I'd not get my undies in a bunch at the analogous two way slab condition, I usually let the plank condition slide too.

b) In dealing with hundreds of these conditions on hundreds of floor slabs, I've not once had a real world differential deflection issue bite me when using common industry detail.

c) In my opinion, it's impractical to rigorously address all potential differential deflection issues with these systems. Our pragmatic alternatives are a) keep calm and carry on or b) leave the design of hollowcore systems to other engineers who are, themselves, turning a blind eye to most of the differential deflection concerns.

 

[r13]

KootK,

Came back from long vacation? Seriously, is the #4 bar drilled in, or..?

 

[KootK]

Came back from long vacation?

I wish. I'm afraid that covid has caused me a number of difficulties that have me on a temporary, social media hiatus. And, if that weren't enough, I bought a new house last week. Energy and attention are in short supply. I'm just checking in here periodically to ensure that I haven't failed to help out with any "KootK Summons'".

Seriously, is the #4 bar drilled in, or..?

Yup, drilled in. I know, it's pretty dubious in terms of quality control and capacity. Spancrete has a technote on this connection which describes their testing and recommendations. Not that that make all concerns magically vanish. I hate to say it but my impression is that they're miles ahead of us in other parts of the world in terms of sophistication with these systems and, particularly, their connections. On the other hand, given the absence of problems, perhaps our level of rigor is just right for what we're doing... who's to say? Not KootK.

 

[r13]

Welcome back, and congratulate on the new house.