Quote (RabitPete)

What would be an effect of adding longitudinal reinforcement to the stem wall, may be make a section of it into a beam? Would it help to distribute the load from the anchors over a longer footing area, so steel which is farther away than 1/2Hef from the anchors would also contribute?

1) I feel that your logic is sound with that and that it could be made to work with the right attention to detail. Even without the beam detailing, it's pretty common to assume a 1:1 load spread from your base plate down to the footing for bending.

2) It's strictly more correct to turn your dowel hooks the other way but, for the proportions that you're showing, I doubt it would make a difference and your arrangement is more constructible.

For me this would be a little too exotic relative to simpler solutions but it's doable. You're basically turning the joint into a reinforced beam/column joint rather than breakout style anchorage. You'd need a pretty good footing depth to make the detailing workable.

I think your sketch shows the cross section of the stem wall. If I am not mistaken, You need to consider the available/effective concrete area bounded in both direction. The example below uses h_ef = 10". Didn't ACI address this situation? Link

Quote (r13)

Didn't ACI address this situation?

They did, and I did account for the effective area on all sides. And without reinforcement, φNcbg is roughly 50% of the required capacity

Quote (KootK)

for the proportions that you're showing, I doubt it would make a difference and your arrangement is more constructible

It is not to scale. Stem wall is 14" thick and footer is 12". I might be able to upgrade it to 14", but that is all. It is going right next to the existing structure and digging a deeper footing would undercut foundation

The wall is 20ft long (10ft on each side of the column), so if we could only distribute that moment over a 20ft... Doubt it can be designed rigid enough

Quote:

My issue is that bottom footing is not deep enough to resist the breakout either.

I guess I am not following what is going on. Are you saying you cannot develop the rebar shown because the footing is too shallow.....so you are checking it by Appendix D (or whatever section it is now in 318) and it isn't working there either?

In that case, you have no choice but to utilize the vertical reinforcement to carry the tension. You need to ensure there is adequate development length above the intercept of the assumed break plane (for hooked bar), lengthen the anchor bolts as required. I think ACI DG 1 does have example for it. For the base slab, ACI does not address breakout cone, but with control in development length. You need to maintain the development length for bars in tension. If Ld is inadequate, you can add concrete thickness (most effective), or use smaller bars to reduce the development length requirement (note the minimum requirement you can't violate though).

Quote (r13)

ACI does not address breakout cone, but with control in development length

Technically speaking, for a single bar it makes sense, breakout does not control. But let's consider 4 hooked bars (#4) closely spaced (on 4in square). Each one is fully developed and hooks oriented in 4 directions (+). Would you really get 4 times the capacity? I would think that break out failure will occur shortly after breakout capacity of a single bar is exceeded.

I've investigated that and come to the same conclusion:

1) Generously spaced bars: development is usually enough/.

2) Gaggles of bars: development is often not enough.

When you design spread footing, is there any suggestion to check the breakout from base pad? I couldn't recall any, but the requirement on development length. Same logic here. Similar concept as develop the rebars of concrete walls in a T joint. Sketch be low is to make sure we are on the same topic - develop rebars in tension. Note the different breakout mechanisms of rebar and headed stud/bend smooth dowel in concrete.

Agreed r13. That is what was behind my question above. Wasn't sure why we were talking breakout with rebar development (two different animals).

Kootk has made some interesting points on this over the years (if you do a thread search).

WARose,

I am puzzled as well.

"Interesting points" = it shall be my life's work to shed light on the fact that development <> anchorage.

A couple of the more salient threads, the first of which would surely have seen violence visited upon me if it were a real world pub discussion:

Development by precast78

Anchor Bolt / Pedestal / Foundation Connection by azcats

1) Failure mechanism of reinforced concrete and round smooth stud/dowel in tension. 2) ACI do address influence of rebar spacing and clear cover in development length calculation.

ACI table 25.4.2.2 is for straight bars, not hooks. I believe spacing, clear cover and confinement addressed in 25.24.2.3 are more about getting enough concrete to encase the bar on all sides and address 1st failure mechanism illustrated by R13 above.

You wont see a 3" OC vertical rebar in a typical spread footing, but with anchor reinforcement, I got 4 bars closely spaced around each anchor rod. Not sure about moment load and effect of compression side, but if it were a pure uplift, breakout would very likely to occur if each hook barely meets l_dh.

The calculated l_DH accounts for rebar spacing and cover limitations, after that, it is there to ensure the rebars will develop code specified tensile strength. If you need the 4 bars to resist the load, you shall make sure the l_DH is calculated based on the bars in group, not the individual bar. Also, when utilize rebars to carry load from anchor, you need to pay attention to the distance limits between rebars and the anchor.

Quote (r13)

you shall make sure the l_DH is calculated based on the bars in group, not the individual bar

I am not familiar with any provision of ACI which addresses groups of standard hooks. In fact, quite opposite, per R25.6.1.5A: "The development of bundled bars by a standard hook of the bundle is not covered by the provisions of R25.4.3"

And per R25.6.1.7: "The increased length of lap required for bars in bundles is based on the reduction in the exposed perimeter of the bars"

Lets consider the following example:
single #4 90 degree hook in 4ksi concrete, no confinement and Cb of >=1.25in requires Ldh=6.64" and provides 11.78 kips of tensile strength
anchor embedded 6.64" into 4ksi concrete give us 19.48kips of breakout strength.

Now, lets consider 2 of the same 90 degree hooks in 8" OC: Cb is still >=1.25, hence the same Ldh and 23.56kips capacity
2 anchor 8" OC have breakout capacity of 27.3 kips, so still not a controlling factor.

And now, lets consider 4 of the same 90 degree hooks in 8"x8 pattern: Cb is still >=1.25, hence the same Ldh and 47.12kips capacity
While 4 anchor in 8"x8" pattern limit breakout capacity to 38.27 kips. Breakout!

So KootK appears to be correct:

Quote (KootK)

1) Generously spaced bars: development is usually enough/.

2) Gaggles of bars: development is often not enough

Bar group ≠ bundle bars. The minimum base slab thickness is governed by the longer l_DH calculated for the #8 (in bar group), and the #6 (individual). Hope this ends the discussion.

I am not seeing any ACI provisions talking about increasing Ldh for a "group" of bars. Only anchorage section considers closely spaced groups. However what we are not considering is a compression force from the moment arm, and its effects on the breakout. Due to that force breakout might not occur, but I am still not comfortable specifying 4 closely spaced bars with minimal Ldh and near their maximum capacity.

I am even more concerned with load distribution and soil bearing. While overall moment per ft of length is relatively low, I got a single concentrated moment applied in a middle.

From a practical standpoint....wouldn't other things become more critical before development? For starters, how does the wall footing handle this much shear? Say you've got 4 #5's closely spaced.....that sends nearly 75 kips of force into the wall footing for shear. That's going to get you close to a 3' wall footing. (Which is outrageous in my neck of the woods unless we are talking industrial work or frost depth.)

I've generally observed the 3 bar diameter spacing in the code for hooks (without using the allowed reduction). I don't know if that meets Kootk's "Generously spaced" criteria.....but usually problems show up elsewhere before that becomes one.

Quote (WARose)

I've generally observed the 3 bar diameter spacing in the code for hooks (without using the allowed reduction). I don't know if that meets Kootk's "Generously spaced" criteria.....but usually problems show up elsewhere before that becomes one.

Nope. Obviously, the problem isn't going to show up if you're not checking for it.

So what is the Kootk recommended spacing? (Before problems show up from development alone.)

Quote (WARose)

how does the wall footing handle this much shear? Say you've got 4 #5's closely spaced.....that sends nearly 75 kips of force into the wall footing

Yeah, its going to be a problem if it sends it over 8"x 8" area. That's why the first question is how the load is distributed along the wall, as 75 kips along 10ft is only 7.5 kips/ft. I am having a hard time imagining that only area directly under the anchors will see the load. What do you guys think? Or any tricks to make sure stem wall distributes point load over the entire length of the footing?

Quote:

I am having a hard time imagining that only area directly under the anchors will see the load. What do you guys think? Or any tricks to make sure stem wall distributes point load over the entire length of the footing?

I'm not sure what your stem wall height is (or it's width) but eyeballing the pic....with such a short height (relative to the 20' length) combined with such a thin footing....I think the odds of the shear & moment equally distributing (in magnitude) over the full length are pretty low.

Would something like this work? Anchors extend to the middle of the stem wall, stirrups are fully developed both above and below anchor heads, so they pickup full load and transfer it to the stem wall which is now designed as a beam, and then several fully developed 90 degree hooks transfer the load down to the footing. So the original breakout cone is not only extended, but its reinforced by 5 well spaced fully developed hooks. Any flaw in my load path?

P.S. The stem wall is 24" tall and 14" wide. Illustrations are not to scale.

Not a bad idea....the distribution would still be a question (i.e. how much shear winds up in elements attached; FEA could resolve).

You could always use shear reinforcing in the footing.

Quote (WARose)

So what is the Kootk recommended spacing? (Before problems show up from development alone.)

I don't have one I'm afraid. For me, it's more about trying to distinguish between situations that are characteristic of anchorage versus those that are characteristic of reinforced concrete design. The reinforced concrete stuff basically means anchorage being provided by a compression strut coming into the development region at a reasonably shallow angle, consistent with how the development length testing is carried out.

Quote (RabitPete)

However what we are not considering is a compression force from the moment arm

There are anchorage design provisions that give account to that benefit. If you want to explore that and have trouble finding the provisions, let me know. Basically, it turns the governing mode from tension breakout to something more like pryout, as you'd expect.

Quote (RabitPete)

I am even more concerned with load distribution and soil bearing. While overall moment per ft of length is relatively low, I got a single concentrated moment applied in a middle.

That doesn't bother me at all. Your latest sketch is basically what I was getting at when I made the statement below. And, with a defined reinforcing path, I'd be perfectly happy to open that up to 2:1 or more. Beyond the anchorage, I see this is a plastic system capable of considerable redistribution.

Quote (KootK)

1) I feel that your logic is sound with that and that it could be made to work with the right attention to detail. Even without the beam detailing, it's pretty common to assume a 1:1 load spread from your base plate down to the footing for bending.

This is for the design of stem wall reinforcement to resist the applied loads. Not for the anchorage.

1) Assume the concentrate load is longitudinally spread over an effective length, which is a 45° spread on each side of the load.
2) Get soil pressure under the assumed strip.
3) Turn the stem wall up-side-down, now you have a loaded T-beam to design for.

Quote (KootK)

There are anchorage design provisions that give account to that benefit. If you want to explore that and have trouble finding the provisions, let me know. Basically, it turns the governing mode from tension breakout to something more like pryout, as you'd expect.

Please share, I know I have read that their is current research being done on this that the ACI plans to incorporate in future revisions, but I hadn't seen anything that is available now.

Sure, it comes from this document which, as I understand it, is kind of the source document for AppD: Link

In summary:

1) Work done by Zhao in 1993. I couldn't get my hands on the original stuff.
2) As expected, for the right combinations of lever arm versus embedment depth, there are marked improvements.
3) There's a proposed evaluation method but it requires determining a parameter (z) based on an elastic analysis. Not sure what to do with that in practice.

Thanks for sharing - more reading material for me to get through. Maybe I'd have more time for this sort of thing if I could only stop myself doing things like redrawing lines through bricks.

Quote:

Sure, it comes from this document which, as I understand it, is kind of the source document for AppD...

Not to nitpick but I think Appendix D evolved out of ACI 349 (Appendix B) first (before Eligehausen's book came out).

I have that reference....excellent source.

For a shallow footing 12" or so deep, do you guys see any issues with detailing a single layer of reinforcement in a middle (assuming moment capacity checks out)?

Quote (rabitpete)

For a shallow footing 12" or so deep, do you guys see any issues with detailing a single layer of reinforcement in a middle (assuming moment capacity checks out)?

I think that it can be done but there are challenges to be addressed:

1) Your one way and two way footing shear capacities will be halved.

2) Within the joint, you've compressed the geometry substantially which will amplify the shears there.

3) You'll be inviting a lot of top side footing flexural cracks right where you'll have your mission critical tension breakout frustum.

4) The side of the footing that would be resisting uplift will have a very challenged shear joint where it ties into the vertical tension bars which will be, effectively, hanger bars at that location.

For these reason, I'd much prefer two layers of rebar myself.

Makes sense, just trying to find the most cost effective solution. Typically, upsizing rebar or decreasing spacing on a single layer is a lot cheaper than additional labor of putting 2nd layer. And it is not just a stand alone footing, it extends further and becomes a floor slab, kind of a flexible mat foundation, so the area is somewhat large. I was also thinking of adding a 2nd layer just under the column, a strip about 4 ft wide (assuming a conservative 1:1 distribution from 24" stem wall) and all the way across. Red rebar on the plan is a top layer, the remaining area has bottom steel only. Something does not feel right about it making me think it might crack near transitions between dual and single layers

1) It would be a great help if you could post a section through this drawn to scale.

2) Based on your last sketch, it seems as though the portion of the footing to the left of the column would be of little relevance other than to help with development and anchorage of the bars on the right.

3) I'm sure that regions where the reinforcement scheme changes are slightly more prone to cracking but this is something that is done all the time without incident so I'd not worry about it.

4) I understand the economic advantage of a single layer of reinforcing. That said, it only works if it works.

Here you go (minus anchor reinforcement). Yes, 12" portion of the footing to the left of the stem wall is for anchorage and development.
:

Thanks. In my mind, this arrangement means that:

1) You will be counting on the opening behavior of the joint to resist the overturning associated with outward shear.

2) Centrally placed reinforcing isn't likely to pan out for the opening moment as shown below.

How about turning the L other way around like you suggested in earlier post? Would not we also need to worry about development length in compression for the vertical bar on the compression side? That alone would dictate a deeper footing or may be a 2" thicker section under the wall only

Quote (RabitPete)

How about turning the L other way around like you suggested in earlier post?

I don't see it:

1) While there is a preferred hook direction from a mechanics perspective, the strut and tie model is agnostic to hook directions. It's just part and parcel of generic hook development.

2) As you can see, there's a strut coming in from both directions requiring restraint. A single hook wouldn't be able to restrain both struts unless you're alternating hooks etc.

3) My earlier proposal wasn't a hook but, rather, a lapped rebar bend around a corner. You could do that here but that would require moving the hook legs up to the level of the centrally placed rebar. And that would obviously weaken the tension anchorage.

Quote (RabitPete)

Would not we also need to worry about development length in compression for the vertical bar on the compression side?

Only if you actually need to utilize the bars in compression. I very much doubt that's the case.

Something like this:

You're only doing this at a local area so maybe just go with the Cadillac and be done with it. Potential detailing problems with yours:

1) Hooks facing the wrong way.

2) Unrestrained tension kinks in your footing bottom bars.

For less critical applications, one can certainly back away from my detail towards yours. That's a matter of judgment though: yours.

Yeah, I did not like the offset bend in the bottom bar either, smelled like trouble. Actually it is not just one localized area, there are 12 of those along the wall, and then there is also a corner. I only showed 1 middle column. I like the idea of U shaped reinforcement to make the hooks face inwards without having separate bars all next to each other.

Another option might be to extend a thickened 14" section all the way to the right. I have to do some math, but with 14" thickness and 2 layers, it might only need to be 36 or even 24" wide, the idea is to get it long and wide enough to keep soil bearing pressures under control. I have not even accounted for the rest of the slab which will surely share some of that load too.

Here is what I meant, may be make a 3ft wide x 14" deep grade beam centered under each column to help with distributing moment over a larger area of the slab? 2 U-shaped ties on each side of each anchor (as per KootK details), standard L hooks 12/24"OC along the remaining section of a stem wall. Something about that rebar piercing the grade beam feels neither right no elegant. Every time I think the good solution is here, it seems to be slipping away...

This might be a better option:

Could someone please explain a little or point to a good source of information on the compression strut-tie system. What shear forces do we need to deal with? Would not I have to add shear reinforcement between compression/tension sides and the same thing in the other direction, so moment in both directions transfers from anchors to the beam?

I am still struggling to come up with a good and simple solution. As the moment can be reversed and also applied in the orthogonal direction, I always end up with some hooks pointing not in the optimal direction.
And as far as U hooks go, should not the U be at least as wide as 2x standard hook lengths for it to be fully developed in tension on both sides?

Any potential issues with the attached detail? Do I need to worry about shear due to moment from the column?
Anchor Reinforcement Details

• https://files.engineering.com/getfile.aspx?folder=af2d059b-c747-42e2-a75c-2

Sorry, we dropped the ball on this one. If you haven't yet resolved this, report back and we'll try to get it sorted over the next few days. Fundamentally, I feel that your long, thin strip foundation proposal is a little weird and perhaps ill advised. That said, maybe I don't fully understand your intent with that.

This is more of a thin mat foundation, not just a strip under the wall. So the idea was to utilize that entire slab to keep it from overturning or exceeding soil bearing capacity. I posted some of the 3D details in another thread, but cant say I am happy with those. What I have so far is a T-shaped thickened section (14" deep, 36" wide) under each column. I would rather see that moment fully transferred into the stem wall and then distributed over a larger area of the slab, instead of taking it straight down into the slab and relying on a small core.

KootK, here is what I have so far:
Reinforcement Detail
Looks like an overkill. Any better ideas on how to transfer bending moment from the columns to the 12" slab 2ft below?

• https://files.engineering.com/getfile.aspx?folder=fa057835-f92f-4917-894f-0

KootK:
Your prolific posting has been great gain to many, including myself - particularly for reinforced concrete and challenging anchorage problems.
Thank you and keep it up.
My one complaint: your posted sketches look like you're blindfolded, writing upside-down, using a turd stick. We can take a collection here on eng-tips and get you a better drawing program.

Quote (ATSE)

My one complaint: your posted sketches look like you're blindfolded, writing upside-down, using a turd stick. We can take a collection here on eng-tips and get you a better drawing program.

1) Duly noted.

2) With regard to a drawing program, what would you recommend? I use Bluebeam for this stuff currently. I also have AutoCAD and Revit but don't find either of those option to be speedy enough. Like most, I'm threading this pro-bono stuff in amongst the assignments that pay the bills.

3) I've been operating under the assumption that the main thing about forums sketches is that they be made to exist in the first place, regardless of their quality. So, as long as I'm able to get my point across, my second prerogative is expediency. There was a time where I was known for the quality of my hand sketches here on Eng-Tips. Some miss that and I miss it too. For speed though, Bluebeam really speeds things up for me which allows me to do more at a lower cost. Being able to take OP sketches, screen them down, draw over them, and upload them here in a minute or two is a game changer for me.

4) If you'd kindly start that collection, I'd be totally game for a Tekla license.