Headed reinforement bars in compression

[bkal]

Hello,

I am considering using headed reinforcement bars to anchor bars which are subject to compression. ACI 318, Clause 12.6.3 explicitly rules out use of headed bars in compression, with an explanation that there are no data available to confirm their effectiveness. I am curious to understand what are the reasons for the headed bars not benefiting from the heads, what mechanism would prevent the heads from reducing the required anchorage length.

 

[KootK]

If the heads are near a free concrete edge, as they often are, the concrete behind the anchor may just pop out.

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.

 

[bkal]

Thanks KootK. I also thought this might be an issue. But, could this not be checked by using Appendix D (concrete breakout) or treating as a punching shear issue.

 

[KootK]

I was thinking the exact same thing regarding Appendix D. Such a method would be "unofficial" however and thus a bit scary in my opinion. Also, if you can make AppD work, you've probably already got enough distance to make regular compression development 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.

 

[Teguci]

Just talking, but consider - For compression steel in columns, the calculated compression stress in the steel, as a function of area, will be exceeded over time as the concrete creeps. For hooks or heads, as compared to a straight bar, I would imagine the bars delivering close to the full bar yield force to a small cone of concrete on the far side of the supporting element. Once the element spalls, that force redistributes to the shortened development length. With a straight bar, the transition of compression into the supporting element is more gradual and so would the local creep.

If you have a column subject to creep and you want to justify a head using appendix D, I'd say to at least design the cone against the full compression (max) yield strength of the bar.

 

[MrHershey]

For the concrete free edge what I've seen suggested in ACI 352 and elsewhere is to provide a closed tie or U bar (with legs developed) tight to the head. Similar to supplemental reinforcement allowed by Appendix D. Think that was mainly for ductility and joint integrity (as in a moment frame) though.

It's not that you can't put a head on bars being developed in compression, it's just that you can't count on it. Similar to a standard hook in compression, you'd just need to embed the standard compression development length as if the head/hook wasn't there. If you're not concerned with ductility then would think that even if you did break out the end you'd still have the bond over the compression development length to transfer the force from the bar.

 

[KootK]

I suspect that the sketches below are the ACI 352 ones the MrHershey mentioned. I agree that they would add confinement but disagree that they would help with the breakout cone behind the rebar head. None of the bars would actually intercept the breakout cone as would be required for supplemental AppD reinforcing.

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.

 

[MrHershey]

I had been thinking more along the lines of the second sketch except with the bars actually in physical contact (tight) with the heads. And, no, not exactly satisfying Appendix D. Similar in concept to Appendix D except instead of restraining the concrete in the failure cone you're restraining the head itself.

Would need your compression development anyways, would tend to argue that even if you don't work for breakout at the end that you still have that bond from the compression development length to transfer the load. You'd pop the end and then immediately grab hold in bond.

 

[KootK]

When you first mentioned the ties, I pictured something like what I've shown below. It sounds as though I had the right idea.

Originally, I thought the concept recursive. You start with a big bar that can't be developed in compression, add a headed anchor to compensate, then tie it all back with some smaller bars that need to be developed in tension. All tolled, it would just be the compression development of the whole assembly so you'd be back to square one. I still think that but, having sketched it, my impression is now that it amounts to an effective increase of the bond stress perimeter. My sketch below is a hypothetical case where not enough footing depth was available to compression develop some column bars.

Practically speaking, I wouldn't expect there to be many cases where you'd have enough space to exercise this concept to any real advantage.

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.

 

[Agent666]

Maybe I'm wrong but the compression development length for a straight length of bar has always been less than the tension development length for a headed bar every time I've ever calculated it. I'm working to a different code but it's based on the ACI code.

If the bars are developed before it gets to the headed part why is there a problem? .... Like I said maybe I'm missing the point here?

 

[bkal]

Thanks to everybody for a very interesting discussion. In my case (and I am afraid I cannot go into much details) there is a limited space to anchor the bars due to a very complex geometry, and the bars are predominantly loaded in compression. It seems that most people here are concerned with the concrete cone being pushed out by the bar head. I was wondering if there is any other mechanism which prevents the heads from being effective?

 

[Agent666]

Help yourself, what sized bars, and what depth do you have available?

I bet if we saw the exact constraints you might get a more focused answer (one that might help you instead of theoretical this and thats). We also might highlight other issues.

Weld a plate to the bar at the face if you are sure it's always in compression (reverse the head). Use this plate in bearing at the face of the joint, if there is joint...

 

[KootK]

For what it's worth:

1) I love the theoretical this and thats and believe that they are often the source of genuine understanding.

2) In my experience, development issues are usually just one symptom of a poorly proportioned joint. Exotic development solutions rarely resolve the underlying deficiency.

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.