External anchorage of reinforcement - is this feasible?

[BridgeEngineer21]

I have a situation that I can't seem to find any examples of in practice. During staged construction, a slab is loaded right to the edge, utilizing the primary longitudinal reinforcement. These bars will eventually be coupled and continue into the next stage of the slab (figure 1). But temporarily, they need to be anchored somehow. We've proposed various solutions like hairpin bars and additional headed bars, but the reinforcement density is quite dense, so for ease of constructability the contractor has asked about this alternate solution to anchor the bars externally, to a steel plate at the face of the slab (figure 2).

In principle it is very similar to headed bar anchors, but I guess the capacity of those is based on manufacturer testing that would include things like the effects of bond between head and concrete and confinement of adjacent concrete, which aren't present here. In theory, it seems possible and easy to just verify the bearing capacity of the concrete behind the plate, as well as the required torque on the nuts to ensure the plate is tight against the face of concrete. But I'm a bit unsure about using a solution which there is seemingly no precedent for, at least not in any code or reference about rebar anchorage I can find. It's more comparable to post-tensioning - but we're not using post-tensioning bars here.

Has anyone seen something like this before? And if not, can anyone think of any concerns or reasons this wouldn't work that I'm maybe not identifying?

Figure 1:


Figure 2:

 

[jhnblgr]

You do bends when you don’t have enough length to develop the rebar into the concrete when transitioning from one pour to the next. Like footings that the heel or toe when the length is inadequate in length.

 

[BridgeEngineer21]

Yes, that's one example of a situation where you need a hook to anchor. So take that situation then. Instead of a hook, do you see a reason it couldn't be anchored externally? Since this is a temporary condition, ignore deterioration of the external plate.



Sorry for the terrible sketch on my phone with my index finger, but I hope you get the idea.

 

[jhnblgr]

But is you ld req < l available? Development length basically is the amount of bond the concrete needs for that full section to develop enough tension in the steel to yield. Also there are reduction factors based area steel excess. How long is the cantilever and what is the calculated ld?

 

[BridgeEngineer21]

Ld required is 1000mm. Ld available is 0mm.

 

[BridgeEngineer21]

But again, as I said, if the situation i described is getting you hung up then just forget about it. I really just want to discuss the general question of external anchorage. Not why or whether the external anchorage is needed, but if it's feasible at all.

 

[jhnblgr]

So you’re saying you no cantilever than. What I am hung up on is the need isn’t there. But what you want is essentially post tensioning the concrete, can it be done sure but you’d be wasting money and the contractors time for nothing. The edge will have shear but no real significant moment, by the time the moment developed the steel will have developed sufficiently to resist the moment. Your ld is measured for the interior beam to the end of the bar, if l=0 than you have no cantilever, if l = 3’ you have 3’ ld

 

[BridgeEngineer21]

It's definitely not a simple cantilever, and the reinforcement demand can't be easily determined based on rules of thumb or hand calcs. It's a two-way slab with complex geometry and openings, experiencing large axial forces and bending in both directions. There is a significant demand for reinforcement at the face of the slab, determined by an FE model for both ULS and SLS design checks (including service level stresses and crack widths). The contractor is the one who initiated the request to anchor externally instead of using hooked bars (due to reinforcement density it will be difficult for them to construct). If I need to I can tell them no, the external plate is not possible, you need to use hooks.

I had the thought that it's basically like PT, but I can't use actual high strength PT bars, or apply any tensioning outside of some nominal torque in the nuts. These are just regular reinforcement bars that still need to be coupled eventually.

But as long as it can be ensured the plate stays in contact with the face of concrete, I can't think of a reason it's not doable. Seems that is the consensus here too.

 

[jhnblgr]

can you post a sketch of the slab and the result of the FEM? We do bridges like this all the time with 32 Kip loads without issue and only 9.5” thick. And for that steel plate to work the concrete would have to crack likely.

 

[BridgeEngineer21]

Sorry, I'm not going to post anything more detailed due to sensitivities of the project. All I can say is the loading and geometry is complex and probably not what you're picturing. It's not really comparable to the behavior you'd see on a typical bridge deck with staged construction.

There will of course be cracking in this area as the bar is in tension, but what type of cracking are you envisioning in order for the steel plate to work? Do you mean something like the bar would first "try" to anchor itself with bond stress, and "fail" because there's not sufficient length available, before the force can propagate to the external plate?

 

[jhnblgr]

https://drlatifee.weebly.com/uploads/4/8/3/2/48322241/bond_development_lengths_and_hook-7-2-2016v1.pdf

Cantilever Development Length | PDF | Beam (Structure) | Column

This document discusses bond and development length in reinforced concrete structures. It explains that reinforcing bars need to be embedded a certain length into concrete to develop sufficient bond stress that prevents the bars from being pulled out when the structure is loaded. This minimum...

www.scribd.com

Google Image Result for https://res.cloudinary.com/engineering-com/image/upload/v1641315822/tips/C01_bkwnty.png

 

[jhnblgr]

For the steel plate to take load the bond must break, if the bond breaks the concrete will crack. You will already be in failure regardless of what you do.

 

[BridgeEngineer21]

Yea, that was my thought. The bond fails, so the plate picks up the tensile load and transfers it back in compression on the side face of the slab, providing the necessary anchorage.

 

[HTURKAK]

Thanks HTURKAK, but the coupler is not the problem. We know the couplers we're going to use. The question is, can a rebar be anchored by being bolted to a plate bearing externally on the surface of the concrete (rather than traditional methods of bonding over a straight anchorage length, or bearing internally on concrete via hooks, legs, or headed bars)?

The bar needs to be anchored before the future slab (in dotted lines) is cast and there is anything to couple the bar into.

Your approach is complicating the situation and i have never seen such an example. My suggestion would be , prop the first slab and use conventional threaded couplers .

 

[BridgeEngineer21]

Your approach is complicating the situation and i have never seen such an example. My suggestion would be , prop the first slab and use conventional threaded couplers .

Agreed. Propping the slab is also an option under consideration. But this anchorage idea is not my approach, it's something the contractor is asking to do. So I should only say no if I have a good reason it won't work.

 

[HTURKAK]

Agreed. Propping the slab is also an option under consideration. But this anchorage idea is not my approach, it's something the contractor is asking to do. So I should only say no if I have a good reason it won't work.

you did not provide a lot of detail here. I understand your reason but it is hard to figure out the full picture.
Assume the use of temporary external plate is decided and the first stage bars preloaded with turning nut method. Lets imagine the deflected shape of the first stage slab. The second stage will not be continuous for the dead loading of the first stage. ( It has already experienced elastic preliminary deflection).
That is why i would go with propping.

 

[JAE]

I think the plate certainly anchors the rebar at its end.

But your contention that a hook would also work isn't entirely correct as a hook needs a small embedment length in front of its bend to meet ACI development criteria right?
So if you had room for a hook, it wouldn't be at 100% anchorage until the ldh was met - 8 x diameter or 6 inches - inward from the end of the hook.