Lap splices of column bars in concrete columns with column faces set back > 75 mm (3)

[ajk1]

If the column face of a column above a floor is set back > 75 mm (3") from the column face of the column immediately below, and vertical dowels are used to lap with the vertical bars in the column, these dowels will be separated from the vertical bars in the column below by > 75 mm (3").
► Questions:
1. Is there any Code requirement on how close the bars in a compression lap splice must be, in order for the Code expressions for compression lap lengths to be valid?
2. If so, what happens if that limit is exceeded? How long a lap length would you use in that case?

I ask this questions as part of my review and revising/updating of all our typical details, and not in reference to any specific project.
The Standard to which I refer is CSA A23.3-2014, but I would be interested in what other Standards such as, but not limited to, ACI-318 and the Australian Standard.

 

[KootK]

In my opinion, the offset lap splice business is for planar elements in the plane of those elements. This, for the reasons shown below. I wouldn't dream of relying on this in a concrete column that's likely to creep and transfer a disproportionate amount of its compression to the rebar. Below, I've also shown what I consider the two sage ways to handle the offset.

This is effectively the same reason why we never transition 12" CMU retaining walls to 8" part way up.

 

[KootK]

Just gonna anticipate where this is headed. I suspect this is related to HotRod's comments: Link

 

[ajk1]

What is the source paper of this study?
Was there any follow-up discussion by others in a Journal?

 

[KootK]

There's a link to the source paper provided in my last post ajk1.

 

[ajk1]

Thanks.

 

[KootK]

@ajk1: For what it's worth, I actually agree with your take on the "for flexural members" thing. While I don't have complete faith in code writers -- particularly Canadian code writers -- my default stance is to take what is written as having been purposeful. And my read on it, from a semantic perspective, is the same as yours. The question then becomes why should the offset lapping only apply for flexural members. The code doesn't make that clear in my opinion but, as usual, I have a theory. I think that it's got to do with the creep in axially loaded members that I mentioned above.

Two things that are generally true of predominately flexural members:

1) At service loads, bar stresses will be low.

2) While not always mandatory, standard detailing practice usually has splices placed where peak bar tensions are not.

With a predominately axially loaded concrete column, both of these nifty features go to hell in a hand-basket. There are no low demand locations for axial load so #2 is shot. And creep will draw substantial compression to the rebar so #1 is not reliable either.

It's kind of a weird thing in codified concrete design in that we don't really seem to have faith in lap splices, even with transverse reinforcement, when the demand for them is high and ever present. This shows up conspicuously in the clause related to tension members that prohibits the use of lap splicing altogether, even with ties/STM.

I get that almost every member will have some flexure in it but I still feel that there are members out there that are predominately axial and therefore may be worthy of differential treatment. To paraphrase a past comment from Hokie66 that has stuck with me: high-rise columns don't care about moments. The bars in such columns do in fact have an important role to play in compression resistance. It's only low-rise and and upper story columns where the rebar is playing an important role in resisting tension.

 

[saikee119]

I think the need of rebar splicing in tension is different from compression.

Let's say if we can use some high-tech means, like high power water jet of laser beam, to clean cut a column in the middle and stack them up as before. The two-section can take axial load nearly as much as before as long as the two sections are restrained laterally without any displacement. The rebar bearing on each other can take same level of stress as before as each bar is well confined. So is the concrete. By severing the column into two halves the stack-up column effectively has no rebar splicing.

In the reinforced design concrete is assumed no resistance against tension so in the tensile zone below the neutral axis we have just the rebar and no concrete area to contribute tension to balance the applied load. In reality we can allow the concrete section to crack and the member will work as long as the rebar below and above the cracking plane is bonded successfully to concrete with adequate development length.

If a column has to resist a modest bending moment then it will have a neutral axis with tension and compression zones with some rebar in small tension.

 

[ajk1]

@ajk1: For what it's worth, I actually agree with your take on the "for flexural members" thing.

I agree completely with you.

I get that almost every member will have some flexure in it but I still feel that there are members out there that are predominately axial and therefore may be worthy of differential treatment. To paraphrase a past comment from Hokie66 that has stuck with me: high-rise columns don't care about moments. The bars in such columns do in fact have an important role to play in compression resistance. It's only low-rise and and upper story columns where the rebar is playing an important role in resisting tension.

On this too I agree completely with you. There are many (most?) columns where the rebar takes a very significant part of the axial load.

I find Hotrod10's reasoning round-about and unpersuasive. Saikee119 has totally lost me.

I suppose the only way to be sure is to write to CSA and ask them (that will likely take a long time to get an answer back) or to call someone I know on the CSA A23.3 technical committee. Frankly, I think the issue arises most probably because of failure to use the best wording in the clause. I suspect that it applies to compressive splices in columns, but they screwed up the wording. Maybe they meant "compressive members and flexural members, but not tension members".

 

[ajk1]

I spoke to one of the academics on the CSA A23.3 Technical Committee and he says that CSA A23.3 Clause 12.14.2.3 does not apply to columns except if the column rebar is in tension. He says lap splices are different in tension than compression, and the bar separation specified by the Standard for contact tension splices do not apply to compression splices. So it appears my original reading of the Standard was correct.