Lets say you have a large unreinforced slab (infinite thickness) and you embed a rebar vertically ld deep. Do you think the bar will yield before the concrete breaks? Appendix D cone shear calc is only for headed stud or hook. I think the rebar will yield before the concrete breaks but my coworker thinks otherwise. Isn't that what development length is? It is the distance the rebar need to develop full strength? What do you guys think?
Tek-Tips is the largest IT community on the Internet today!
Members share and learn making Tek-Tips Forums the best source of peer-reviewed technical information on the Internet!
-
Congratulations TugboatEng on being selected by the Eng-Tips community for having the most helpful posts in the forums last week. Way to Go!
Development length 8
- Thread starter precast78
- Start date
CELinOttawa
Structural
Okay, so now you're talking about something that most engineers can relate more closely to when discussing rock anchors and that type of application where different zones along the anchor are expressly calculated.
I think what you're saying is that you expect there to be a length over which the bond fails and the bar pulls out cleanly, then an area where the bond holds and the cone develops above this.
You're right to think this happens, but wrong to think that Ld doesn't address it. Development length gives you the full length upon/over which the bar gets developed, regardless of which of the multiple failure modes actually occur. As you get shorter you'd have failures that would range from strain hardening overtaking the strength of the theoretical "cone" and having a very flat inverted volcano pop-out all the way down to failures of the bond and a straight "pop out" of the bar.
Some of the very early work on reinforced concrete did indeed involve such lengths being calculated, particularly early French and German codes. This simply isn't the case anymore...
When we get PASS the regular bar development into the idea of where the stresses in the bar go we get to Kootk's varied concerns... Until that point the bar plain old holds and yields.
I think what you're saying is that you expect there to be a length over which the bond fails and the bar pulls out cleanly, then an area where the bond holds and the cone develops above this.
You're right to think this happens, but wrong to think that Ld doesn't address it. Development length gives you the full length upon/over which the bar gets developed, regardless of which of the multiple failure modes actually occur. As you get shorter you'd have failures that would range from strain hardening overtaking the strength of the theoretical "cone" and having a very flat inverted volcano pop-out all the way down to failures of the bond and a straight "pop out" of the bar.
Some of the very early work on reinforced concrete did indeed involve such lengths being calculated, particularly early French and German codes. This simply isn't the case anymore...
When we get PASS the regular bar development into the idea of where the stresses in the bar go we get to Kootk's varied concerns... Until that point the bar plain old holds and yields.
In broad terms, you need two things to ensure that the theoretical rebar won't pull away from the theoretical concrete before it yields:
#1) The bar should not slide out from the concrete in which it is embedded by way of bond stress failure / local concrete splitting.
#2) A chunk of concrete should not separate from the main body of the parent material and come away with the rebar.
So far, this ought not be fodder for debate. It's just statics. So, analytically, how do we make sure that #1 and #2 don't happen? Like this:
#1) We embed our bar by a development length and;
#2) We employ some calculation procedure to assess concrete breakout.
So, in my opinion, bar development is necessary, but not sufficient, to ensure that our hypothetical reinforcing bar can be stressed to yield before initiating a concrete breakout failure. This makes sense to me for a couple of reasons:
1) If development were sufficient on it's own, then it would be sufficient for all cases where the code definition is met, including situations with small edge distances etc. And we all seem to agree that concrete breakout will be an issue where edge distances are an issue. For what it's worth, my understanding is that
2) As shown in the test setup sketches that I posted above, the testing for Ld involves a concrete strut that the bond stress forces can push back against. As TXstructural pointed out, development length was never intended for situations where the bond stresses would push against a free edge.
I too have long been frustrated with the lack of an Appendix D provision for cast in rebar used as anchorage devices. The ACI 318 provisions on post installed adhesive anchors are coming tantalizingly close. In Eligenhausen's book, which was the primary source for the 318 App D provisions, he explicitly differentiates between situations where concrete breakout applies and where development length is the name of the game (snipped below).
For what it's worth, my understanding of Eligenhausen's work is that bond and concrete breakout are both incorporated into the Appendix D. As such, it's debatable whether or not it's even appropriate to be using the term "development length" in the context of our problem in this thread. In this situation, bond and breakout are inextricably intertwined. This is consistent with Precast78's intuition that a 35deg cone emanating from the bar end might not be "fair".
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.
#1) The bar should not slide out from the concrete in which it is embedded by way of bond stress failure / local concrete splitting.
#2) A chunk of concrete should not separate from the main body of the parent material and come away with the rebar.
So far, this ought not be fodder for debate. It's just statics. So, analytically, how do we make sure that #1 and #2 don't happen? Like this:
#1) We embed our bar by a development length and;
#2) We employ some calculation procedure to assess concrete breakout.
So, in my opinion, bar development is necessary, but not sufficient, to ensure that our hypothetical reinforcing bar can be stressed to yield before initiating a concrete breakout failure. This makes sense to me for a couple of reasons:
1) If development were sufficient on it's own, then it would be sufficient for all cases where the code definition is met, including situations with small edge distances etc. And we all seem to agree that concrete breakout will be an issue where edge distances are an issue. For what it's worth, my understanding is that
2) As shown in the test setup sketches that I posted above, the testing for Ld involves a concrete strut that the bond stress forces can push back against. As TXstructural pointed out, development length was never intended for situations where the bond stresses would push against a free edge.
I too have long been frustrated with the lack of an Appendix D provision for cast in rebar used as anchorage devices. The ACI 318 provisions on post installed adhesive anchors are coming tantalizingly close. In Eligenhausen's book, which was the primary source for the 318 App D provisions, he explicitly differentiates between situations where concrete breakout applies and where development length is the name of the game (snipped below).
For what it's worth, my understanding of Eligenhausen's work is that bond and concrete breakout are both incorporated into the Appendix D. As such, it's debatable whether or not it's even appropriate to be using the term "development length" in the context of our problem in this thread. In this situation, bond and breakout are inextricably intertwined. This is consistent with Precast78's intuition that a 35deg cone emanating from the bar end might not be "fair".
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.
CELinOttawa
Structural
Kootk: That simply isn't true. You're overcomplicating the issue. Where you have full development length, without cover issues, you will be able to transfer the stress into the surrounding concrete. Whether the concrete can handle the loading is the next question, not part of the problem.
Precast78 said:
That was the original question CEL. How do you propose that we address it without considering whether or not the concrete breaks? OP even referenced App D by name in the problem statement.
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.
CELinOttawa
Structural
The OP mentioned that monstrosity in excluding it only. No reference to any content was made.
I think we may need to agree to disagree on this one. The alternative is to keep running around in circles as the conversation appears to be endless so long as you keep redefining what we should be discussing.
I do have to say that the two examples you give from Eligehausen (no 'n', a very common mistake) are obvious to the point of being moot. Of course development length is insufficient to anchor a bar at the far end of a beam; where would the stress go? Strut and Tie, as is implied by the stress field, have very particular requirements for good reason. As for the lapped and spaced bars, no code that I know allows a row of laps like this. And who would ever do this? Tie the bars together and deal with this as a lap splice...
Nor am I beyond the academic interest. I worked in seismic anchor development, and had the privilege of working with the good Dr-Prof-Ing Eligehausen when he acted as an expert witness for us in a patent suit. I just don't think you're right in your thinking.perhaps it is a simple matter of terminology, who knows?
I think we may need to agree to disagree on this one. The alternative is to keep running around in circles as the conversation appears to be endless so long as you keep redefining what we should be discussing.
I do have to say that the two examples you give from Eligehausen (no 'n', a very common mistake) are obvious to the point of being moot. Of course development length is insufficient to anchor a bar at the far end of a beam; where would the stress go? Strut and Tie, as is implied by the stress field, have very particular requirements for good reason. As for the lapped and spaced bars, no code that I know allows a row of laps like this. And who would ever do this? Tie the bars together and deal with this as a lap splice...
Nor am I beyond the academic interest. I worked in seismic anchor development, and had the privilege of working with the good Dr-Prof-Ing Eligehausen when he acted as an expert witness for us in a patent suit. I just don't think you're right in your thinking.perhaps it is a simple matter of terminology, who knows?
- Status
Similar threads
