Retaining Wall - Flexural Reinforcement from Stem Into Footing 21

[CWEngineer]

I am trying to get some clarification regarding the flexural reinforcement of the stem of a retaining wall into the footing.

Does the flexural reinforcement in the stem of a wall, need to be developed into the toe, such as show in Figure 1 of the attached document. Or is providing a standard hook (12db), sufficient, such as that show on Figure 2 of the attached document? If providing a standard hook is sufficient, can the hook be turned towards the heel?

Thanks in advance

 

[CELinOttawa]

Aren't Appendix D and Anned D specific to post-installation?

I don't think they really justify anything about cast in place bar development. They are an entirely different animal and more about how drilling into set and cured concrete behaves...

 

[TehMightyEngineer]

Appendix D covers both cast-in dowels and post-installed. You're correct (at least in my opinion) that it's a little apples and oranges, but my point was that if concrete failure wasn't an issue then why wouldn't we just be calculating the "development length" for a dowel and calling it a day? Perhaps a little too apples and oranges but hopefully you get what I was alluding to.

Professional and Structural Engineer (ME, NH, MA)
American Concrete Industries

http://www.americanconcrete.com

 

[KootK]

What do I object to? Posts like 30 Sep 16 00:18. It is exactly that type of non-collegial and argumentative post that had me so sick of Eng-Tips I didn’t log in for the better part of a year.

Bah! That was just a little snarky humor. If Tom had to take the day off to cry in his bear and re-watch Bridgette Jones, he should let me now. I'll send a basket.

I well remember the thread the precipitated your exit: Link. And I felt badly about it once I realized that you were gone.

Despite your claiming the contrary, I believe that the truth is that you do not enjoy vigorous debates when they are with me. And I don't think that "why" really matters. Our styles are just incongruous I guess. Different strokes for different folks. What I don't understand is why you keep engaging me in these vigorous debates when they clearly make you miserable.

I was very happy to see you back here again after your long absence. And, in threads where my level of investment is much less, I've made a concerted effort to give you some space so that we can avoid the kind of arguments that led to your prolonged absence. To some extent, it will take two however. You'll need to stop "poking the bear" as it were.

So here's what I'm asking... begging of you. When you start getting frustrated, disengage. I tend not to get frustrated with these kinds of discourse so it's very difficult for me to tell when enough is enough. Perhaps I'm a high functioning autistic.


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.

 

[KootK]

Developed bar does not equal concrete sufficient to avoid failure; it's simply a bond strength check. Otherwise why would appendix D even exist?

TME is spot on with this. It's also an oddly contentious issue and has been debated at length here: Link. In fact, I do believe that TME may have stolen my favorite, snarky illustration of the concept from there.

Whenever I pitch the concept, folks jump all over me saying that it's too ridiculous to be relevant. I believe that the juxtaposition of its ridiculousness with its technical correctness is what makes the sketch so salient. It is a properly developed bar by all measures. And it's tension capacity is zilch due to concrete breakout issues.

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.

 

[TehMightyEngineer]

cry in his bear

I well remember the thread the precipitated your exit: Link.

Hmmmm, I seem to be unintentionally dragging this thread towards that thread by discussing the definition of development length. I'll leave that debate for another thread lest we make this thread any more contentious.

I do believe that TME may have stolen my favorite, snarky illustration of the concept from there.

Guilty as charged, it was a good example.

Professional and Structural Engineer (ME, NH, MA)
American Concrete Industries

http://www.americanconcrete.com

 

[CELinOttawa]

And I find it most ironic, and funny, that I see the two threads as having now become nearly interchangeable... Only this time you've gotten all the stars, and last time it was more even keel.

I have become quite annoyed at the way that Eng-Tips becomes a "I'll vote my view up" game instead of a "Hey, that was genuinely helpful; I'll give them a star to thank them for their effort", which is what I think was/is intended.

As to the issues with Development Length: I have been, and as best as I can tell always will be, a strong believer in the reliability of a developed bar to pass the subject stress into the surrounding concrete. After this has occurred, you must then resolve the stress elsewhere. That's good detailing practice, and effectively what people do (without the thinking that I believe should go into our work) when they insist on developing a bar which has already been developed (ie: Pass a "tension" around a corner and anchor elsewhere).

Look: Everything we do it based on a gross number of assumptions. Which way you design your reinforced concrete *should* vary based on how you choose to handle the design. I suppose I am a "Bernoulli Region" thinker, and you have clearly become an avowed Strut-and-Tie proponent. I prefer modified compression stress field in my work, so I don't really think about STM very often anymore, other than for pile caps and inverted corbels carrying an existing structure on a new exterior pile.

And you're right. That thread basically left me entirely sickened with this site, like this one has nearly now managed to do. I wish I had the spare time to run an Ansys Mechanical simulation on this question, but I'm just too busy. My only hope is that you realize that I believe you're right about the assumptions required for your work, that they don't always apply to how someone else is choosing to do a design, and further that NEITHER ONE is a true reflection of what is going on, but a practical model with necessary simplifications to enable us to get some stuff done in the real world.

 

[CELinOttawa]

Also: Canada, and even more so the USA, have become a "There is one right way to design X". That's cook-book thinking and leads to crappy, overly conservative designs and chokes out innovation, and particularly to a poor quality of University graduate from the Engineering schools. I have worked with new grad interns from universities in five countries so far, and we don't compare well against the skills coming out of the NZ universities.

I hope and pray for the day where Canada adopts a genuinely Performance-Based approach and the codes loosen up and allow for practical thinking again. That and I would welcome Authorities Having Jurisdiction requiring the submission of calculations with permits. The fees are so low sometimes that I find it hard to believe anyone is doing more than simply drawing lines, labeling best-bet beam sizes from a half-checked model and submitting the work.

If it isn't already obvious from the discussion (above), I do very little work with A23 and have not become reacquainted with the finer points of detailing of Reinforced Concrete according to the Canadian Code, despite being back designing in Canada. I simply don't do reinforced concrete outside of my very specialized position with the Federal Government, where we use a custom code for high strain rate and impulsive loads. I'll let you fill in the blanks on that one, but suffice it to say that the real world behaviour of concrete far outweighs the belts-and-suspenders while reading a cookbook approach that our code demands.

I hope that my next job for design in reinforced concrete in Canada is a tilt-up job. I would love a chance to bring some of the practical detailing innovations from New Zealand (which would satisfy the A23 code) to local practice, and I really need such a job to refamiliarize myself with the Canadian code. Knowing NZ 3101 like the back of my hand and being functionally familiar with the codes for Hong Kong, Vietnam, and French Polynesia (specifically New Caledonia) isn't really what I need for local practice. I just haven't been willing to get to studying the new code until I have a job that needs it; Only time will tell if I get the chance.

In the meantime, I think I'm out.

 

[KootK]

While I'm anxious to get to "done" as well, I feel that I've come up with one more salient argument that hasn't yet seen the light of day.

Below, I've included a blurb from MacGregor's text which references a related ACI clause. Essentially, it's a fundamental statement of what reinforced concrete is. Not just concrete mind you. REINFORCED concrete. The fundamental principle being that, in reinforced concrete, flexural tension is passed from reinforcing bar to reinforcing bar without utilizing concrete in tension other than, perhaps very locally at lap splices. It's an important distinction because, as has been pointed out, concrete does have some tension capacity that can be mobilized in certain cases (app D etc).

Here's a summary of my thinking with this:

1) I went with an L-shaped retaining wall because getting rid of the heel simplifies things in a helpful way.

2) If all that matters is development, then I would argue that cases 2 & 3 are effectively identical.

3) Case 1 would pass the RC concrete litmus test indicated in the blurb at section A-A.

4) Neither case 2 nor case 3 would pass the RC concrete litmus test indicated in the blurb at Section A-A.

5) While cases 2&3 would have some capacity as a result of concrete tensile stress, that capacity would be greatly reduced.

6) I've never encountered a design method that relied on concrete in tension at the moment joint.

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.

 

[KootK]

The fees are so low sometimes that I find it hard to believe anyone is doing more than simply drawing lines, labeling best-bet beam sizes from a half-checked model and submitting the work.

Based on my time in engineering management, this is absolutely true. It was great fun to manage a stable of poor bastards who didn't have the budget available to do tight technical work but then had to deal with a KootK review at the end of the line.

And your statement below is a big part of why. Frankly, I'm ashamed of Canada on this front. What did we think would happen in a competitive environment sans sheep dog? First, folks get sloppy in order to make more money. Second, folks stop having the fee available to do anything but be sloppy. Not. Impressed. With selves.

That and I would welcome Authorities Having Jurisdiction requiring the submission of calculations with permits.

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.

 

[TehMightyEngineer]

I feel that I've come up with one more salient argument that hasn't yet seen the light of day.

I'm glad you did; that post summarizes in a very concise manner exactly why I think you're right. If people can't agree on what you've shown there, then it's probably best to just agree to disagree and move on.

Professional and Structural Engineer (ME, NH, MA)
American Concrete Industries

http://www.americanconcrete.com

 

[KootK]

So here's the Nilsson stuff which I'll let folks interpret for themselves. Really no new revelations beyond what has already been covered except, perhaps, the importance of the diagonal bars to crack control. There was no testing specifically for retaining walls having only standard hooks at the bottom of the stem bars. Reinforcing was 10M. Stems were 200 wide where as I estimated 250.

I won't post the entire article because I harbor some sporadic and selective respect for intellectual property rights. That said, if anyone has any specific questions, I'm happy to review the article and post whatever answers I can.

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.

 

[sandman21]

For what its worth, U70 is what I typical see in CA and what we use almost all the time. Just wish they would only use real units, in, ft. etc.

 

[Tomfh]

Developed bar does not equal concrete sufficient to avoid failure; it's simply a bond strength check.

TME is spot on with this.

The fundamental principle being that, in reinforced concrete, flexural tension is passed from reinforcing bar to reinforcing bar without utilizing concrete in tension other than, perhaps very locally at lap splices.

Fundamentally incorrect. Stress development IS concrete tension. Specifically it is perpendicular concrete tension along the lengths of the bar. It's not simply some magical "bond" between the bar and concrete which is unrelated to the concrete tensile-rupture. Read the Gilbert link KootK posted about. As gilbert puts it - "bar anchorage is concrete tension". That's what it is.

A straight bar embedded in slab or tube of concrete (TME and KootK sketch) is irrelevant to our problem because the applied actions and failure plane are perpendicular the anchorage tensile forces meaning you can forget about them.

In our cases with hooks into a heel the anchorage tensile forces align with other tensile forces, and thus it's now critical (hence the poor joint efficiency). The concrete is preloaded with tension and hence the bars cannot develop. They can no longer induce the tensile forces needed to develop the bar because the concrete tensile capacity is almost used up already in those areas. As CELinO points out, developement is much worse in tensile zone. It is much harder to develop the bar in these tensile zones. In the case with the hook turning into the toe there is great confinement so anchorage is much better.

This idea that development is merely a bond force which is unrelated to rupturing of the surrounding concrete is ridiculous. It is simply ridiculous to call bars which are tearing up concrete as being "developed bars". A developed bar is one that DOES NOT rip up the concrete before it yields. It is a bar that does not overwhelm the concrete's tensile capacity prior to it yielding.

 

[KootK]

Read the Gilbert link KootK posted about. As Gilbert puts it - "bar anchorage is concrete tension". That's what it is.

Gilbert's just talking about the perpendicular splitting forces that develop along bars as they anchor, slice, and develop. That's something that I've acknowledged previously and simply isn't part of the primary structural action.

flexural tension is passed from reinforcing bar to reinforcing bar without utilizing concrete in tension other than, perhaps very locally at lap splices.

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.

 

[KootK]

It is simply ridiculous to call bars which are tearing up concrete as being "developed bars". A developed bar is one that DOES NOT rip up the concrete before it yields. It is a bar that does not overwhelm the concrete's tensile capacity prior to it yielding.

Using ACI appendix D, I ran an analytical experiment. I put it together quickly so, no doubt, there's a mistake in there someplace. I'm sure that someone will point it out for me in short order. For now, here are the details:

1) Went with #8 bars at 12" o/c as a representative retaining wall case.

2) Calculated the tension demand along the bars in kip/ft.

3) Calculated the resisting concrete breakout capacity, also in kip/ft.

4) The capacity in this instance is less than half of the demand.

5) I used phi = 0.9 on the bars. To rupture a bar, phi = 1.25 is probably more apt.

6) It plays out better for smaller bars and worse for larger.

If you do this same exercise for a single bar, all by it's lonesome, you usually will have enough breakout capacity. That's not because development guarantees breakout capacity however. Rather, it's just how the numbers pan out. When edge or group conditions apply, development often will not guarantee adequate breakout capacity, as seems to be the case in this instance.

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.

 

[Tomfh]

Gilbert's just talking about the perpendicular splitting forces that develop along bars as they anchor, slice, and develop.

That's something that I've acknowledged previously and simply isn't part of the primary structural action.

I don't agree you've acknowledged it because you are characterizing the tension as "just a perpendicular splitting force" and are saying it's has nothing to with overall failure.


The perpendicular tension is sometimes related to primary failure (i.e. global forces in the concrete), hence it being more difficult to develop a bar through a tension field. Hence the hook in detail (a) being far less developed than the hook in your "abominable" detail. In detail (a) the perpendicular splitting forces are pushing against concrete that's already wanting to pop out, and the joint fails early. In the abominable detail the hook wraps back into the confined core, where the concrete isn't already at breaking point.

Your rod-snapping and frustrum-pull-out examples do not have global tension forces aligned with the bar perpendicular splitting forces, plus the geometry ensures the concrete wont split along the bar, so it's cheating to cite these as proof that splitting along a bar is entirely unrelated to so-called "primary failure".

 

[KootK]

The perpendicular tension is sometimes related to primary failure (i.e. global forces in the concrete), hence it being more difficult to develop a bar through a tension field. Hence the hook in detail (a) being far less developed than the hook in your "abominable" detail. In detail (a) the perpendicular splitting forces are pushing against concrete that's already wanting to pop out, and the joint fails early. In the abominable detail the hook wraps back into the confined core, where the concrete isn't already at breaking point.

I'm having a hard time parsing out what you're trying to get at here. How about a sketch or two?

Your rod-snapping and frustrum-pull-out examples do not have global tension forces aligned with the bar perpendicular splitting forces, plus the geometry ensures the concrete wont split along the bar, so it's cheating to cite these as proof that splitting along a bar is entirely unrelated to so-called "primary failure".

Cheating huh? I thought that one of your arguments was that developed bars can't be pulled out of concrete by way of concrete breakout. I've show an analytical example of that assumption being inaccurate. Seems perfectly valid to me.

This idea that development is merely a bond force which is unrelated to rupturing of the surrounding concrete is ridiculous. It is simply ridiculous to call bars which are tearing up concrete as being "developed bars". A developed bar is one that DOES NOT rip up the concrete before it yields. It is a bar that does not overwhelm the concrete's tensile capacity prior to it yielding.

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.

 

[Tomfh]

I'm having a hard time parsing out what you're trying to get at here. How about a sketch or two?

What I'm getting at is that anchorage tensile forces aren't just "perpendicular splitting forces" that are unrelated to overall failure. Yes in most occasions these perpendicular splitting forces do not interact critically with the concrete tension forces (e.g. your go to example above), but in many occasions they do, and failure occurs. As CELinO puts it, the anchorage needs to grab onto something. If that something (concrete tension capacity) is already used up then you have problems. here are a couple of examples. one a regular beam, and the second our joint cases.

 

[SteelPE]

This thread has been going on for 10 months and makes my head hurt.

 

[CELinOttawa]

SteelPE gets a star for the first non-partisan post in months... And because HYPOCRISY!!!! Hahahaha.... And manic tiredness of child-rearing. *sigh*