Rebar Development Wall to Diaphragm Connection 1

[AThor]

Okay, I have been puzzling over this all day, and after pouring through ACI 318, Hilti, and other Eng-Tips threads, I finally decided to just ask. I am replacing an existing concrete roof slab in a simple building with 6" concrete walls and roof. The attached detail shows the planned replacement - chip away the existing slab around the existing embedded hooked dowel that connects the roof to the wall, and pour the new roof slab, using the existing dowel for diaphragm shear transfer to the shear wall.

The issue is that, with the roof being only 6" thick, I do not have enough length to develop the bar as a standard hook past the cold joint. Therefore, I can't use shear friction for the shear strength across that joint. I have enough shear strength in the rebar as only shear dowels, but I don't have the Chapter 17 edge breakout strength. From what I gather by reading past threads (see links below), I can't use a prorated shear friction strength based on reduced hook development length. Per ACI 22.9.4.5, I can use permanent net compression to reduce the required shear friction reinforcement, and there is net compression here from the dead load of the slab. If I combine the net compression times a conservative friction factor, plus the steel shear strength of the dowels, that's enough capacity. But, I'm leaning towards thinking that is combining two sections of code, neither of which I'm fully complying to, in order to get a combined total strength.

My detail is identical to the existing detail before replacing the slab, since the original design had a construction joint at the top of the wall where this new joint is. Was that just a bad detail, or am I missing some other code-supported method to demonstrate shear capacity across this joint? Do I have to detail this differently to get sufficient strength? I'd rather not have to drill in a bunch of new post installed dowels in such a thin wall. And even so, they would need to extend into some kind of parapet to get the development length.

Relevant other threads:

https://www.eng-tips.com/viewthread.cfm?qid=389772

https://www.eng-tips.com/viewthread.cfm?qid=365678

https://www.eng-tips.com/viewthread.cfm?qid=365104

 

[KootK]

A common technique to use in this situation is to put a longitudinal #5 in the knuckle of the bent bars and then make recourse to the beam stirrup provisions that allow one to consider stirrups effectively anchored as they come off of longitudinal beam bars in the flexural compression zone. This gets you back into shear friction territory ostensibly. This method probably isn't 100% kosher but there aren't many good competing options and this one will generally be a saleable story to tell anyone reviewing your design.

My real feeling on the mater is that, if a damn concrete wall connected to a concrete slab with a whole bunch of little, developed, bent bars can't transfer some measly diaphragm shear... I'm out. I'll get into real estate or dentistry instead. The original detail was a good detail, and still is.

 

[EZBuilding]

Is this for a repair or is it for an alteration?

If for a repair the existing building code would provide you the following out:

406.2.1Repairs for less than substantial structural damage.
For damage less than substantial structural damage, the damaged elements shall be permitted to be restored to their predamage condition.

 

[Celt83]

I've reached out to ACI on this and got no response, their example 3 in MNL-17(21) designs the dowel quantity using shear friction and then conveniently never goes through the bar development calculation nor justifies the detailing they adopt in the solution:

 

[AThor]

Thanks for the replies, each gives me some justification, from three different angles. I will add the longitudinal bar, that makes sense and is easy enough. Referencing the IEBC also makes sense, this is just a minor repair. And I'm glad to see a similar detail in an ACI publication.

 

[KootK]

A fourth angle could be recourse to ACI methods used for coaxing shear transfer out of thin toppings on prestressed girders. They make do with much less than full development in very aggressive situations. I've even seen some research indicating that no reinforcing at all works quite well.

In the interest of keeping it real, it's probably also worth nothing that the world is chock full of basement wall to main floor slab connections that look like this and are, ostensibly, transferring "hand of god" level backstay forces from the core walls out to the perimeter.

 

[lexpatrie]

Include an image in the post, rather than a link.

I'm not sure I'm following you, you need to develop the bar in tension for fy on both sides of the interface for shear friction, that's the jagged edge at the bottom of the slab.

The development you need is a) into the existing, where you are showing a little over 1'-1" into the existing, (you don't give a bar size), I think this development length can be adjusted down if you provide "excess" reinforcement, which is fairly likely for lower diaphragm loads. Standard hook used to be 12db for a 90 degree bend, (7.1.2 in ACI 318-95) (18.8.5 in 318-2019? looks like it's down to 6" min?) so you've probably got a fair bit more than the length past the standard hook, unless you are viewing the development length into the wall as a straight bar like it's going into a footing from a shear wall? I wouldn't count the part of the bar that's in the new concrete, because it doesn't make a lot of sense as you need it developed at the roughened surface, but you look like you have 1'-1" or more.

The trick, I suppose, is the shear friction provisions want the bar developed for fy, so the more bar you put in, the longer the development length you'll need. But if it's old enough it might even be 40,000 psi bar. And I wouldn't be too surprised to see #4 or #3 bars in the existing, unless there's some reason for there to be a larger bar, not enough information in what you've provided to gauge that.

The other development length is 3'-0" +/- into the 6" roof slab, which is fresh concrete, if you need, higher psi concrete will shorten the development length a little but 3'-0" looks pretty viable.

When you talk about edge breakout strength, I'm not convinced that applies, this isn't a post-installed (epoxy) anchor. This is rebar. Hilti has done some seminars on the subject because the two sets of provisions don't really mix well, and are at odds with each other if misapplied.

https://ask.hilti.com/article/post-installed-reinforcing-bars-vs-post-installed-anchors/e222ff

Regards,
Brian

 

[KootK]

The other development length is 3'-0" +/- into the 6" roof slab, which is fresh concrete, if you need, higher psi concrete will shorten the development length a little but 3'-0" looks pretty viable.

That is the fundamental issue at play here. Stock ACI provisions do not allow you to consider bar extensions around corners as adding to the the development of the bar. That is allowed in the concrete standards of some other jurisdictions such as Britain. But, there, some attention is paid to having an appropriate bar bend radius that avoids crushing the rebar in the knuckle. Technically, you could do the same thing in North America using a procedure known as the curved bar node STM model. That's too much hassle for a routine connection like this, however. Additionally, specifying any bend radius other than the "normal" radii in North America is a tough go as folks just aren't accustomed to having to do that.

 

[AThor]

Thanks for the reply lexpatrie. I agree, embedded images are nicer than links, I just don't know how to do that. Would someone mind giving some quick instructions on how to do that?

Agree, the bars need to be developed for fy on either side of the shear plane (the cold interface between wall and slab in this case). These bars are #4, and are in fact Grade 40. The development into the existing wall is fine, it is into the new slab that is the issue. I was trying to develop it as a standard hook. The 3' "l_ext", the extension after the bend, is sufficient, it is the "l_dh", length before the bend, that I do not have, which has a floor of 6" per ACI 318-19 25.4.3.1.

Since I do not meet the explicit code for shear friction, I thought to see if I could prove strength via anchoring provisions. These provisions apply to cast-in as well as post-installed anchors, and rebar is permitted to be used as the anchor. The thought is that, if using shear friction, the planes do not slip because of the rebar clamping force, so there can be no anchor-type failure modes, like breakout or pryout. However, if I can't use shear friction, the planes can slip, so the anchoring into the concrete needs to resist these failure modes. With such a small edge distance, those do not work either.

 

[lexpatrie]

When you are creating the post, there's an image icon with a camera, click on that and you can upload an image directly into the post.

I don't think you need to develop the new bar going downward into the existing. It's developed on the horizontal in the newly cast slab, and it's developed in the existing concrete that's remaining.

If you can't stomach it, what about a L and some headed studs into the bottom of the new concrete and some nice special inspection epoxy anchors into the existing?