Shear Friction Reinforcement 1

[spats]

According to ACI 318, shear friction reinforcement needs to be anchored to develop yield strength on both sides of the shear plane. I assume this means development length per Chapter 12. Question is, can you multiply development length by As required/As provided?

 

[hokie66]

It may be because there are old guys on the committee, who like me, still don't believe in shear friction as a reliable means of transferring force across a joint. Shear friction depends both on clamping and dowel action, and the contributions of each are not well understood or quantified.

 

[Lion06]

One other thought to consider. If you ever get mill certs on your reinforcement, a 60 ksi bar almost always has an actual yield stress of 66 ksi - 70 ksi+. 60 ksi is the MINIMUM required yield stress of the bar. For this same reason, I see nothing to preclude using a lower yield stress in the calculations for purposes of calculating a shear friction capacity.

 

[KootK]

@ Lion: that's an excellent point that I had not considered. However, development length testing was done with real rebar, presumably having realistic over strength variation. Might that phenomena already be baked into that particular cake?

@ Hokie: Trust me, I have plenty of my own issue with shear friction. However, I earned my stripes in a post-shear friction world. I don't know anything else. Without using shear friction, how would you resist sliding in, say, a horizontal shear wall cold joint? Regular friction? Dowel action? Shear keys? Part of the problem for me is that I've never seen an established, North American, capacity check procedure for the last two mechanisms. I have seen a dowel strength method in some European precast literature.

The greatest trick that bond stress ever pulled was convincing the world it didn't exist.

 

[hokie66]

Yes, dowels or shear keys. Primarily shear keys, like slabs supported on the cores of high rise buildings. I don't think anyone uses just shear friction for that.

 

[Ingenuity]

slabs supported on the cores of high rise buildings

FYI - for those that don't peak Australian, "cores" are primarily an Aussie-term for elevator and stair shafts. Not to be confused with the voids in hollow-core slabs.

 

[KootK]

So how do you check the capacity of the shear keys hokie? Is there guidance in your codes? Rationally, I feel as though it would be (considering your highrise slab example now):

1) bearing on the underside of the key.
2) Diagonal tension in the slab based on a depth from the underside of the key to the top of the slab.
3) Vertical shearing through the width of the key. I don't know how you'd accomplish this as it's some form of shear that is not diagonal tension but is also not to be addressed by shear friction.
4) Modulus of rupture bending in the key since the load and the bearing area are not concentric.

I would very much like to see a recognized design procedure for this. Does your code provide a method for figuring dowel capacity as well?

The greatest trick that bond stress ever pulled was convincing the world it didn't exist.

 

[hokie66]

Maybe now it does...I don't know. I have been retired for a few years, and am sure I have missed some recent additions to the code. Possibly someone younger will chime in. But it is essentially just a bearing check, like on unreinforced masonry. There are nominal tie bars used through the joint, bendout bars or coupled bars, Rebox or similar.

http://www.danley.com.au/Product_Connections_Rebox.htm

 

[Lion06]

@ kootk - I can't say if that's baked into the development length equations, but I would assume it isn't. I haven't seen testing data, but I would imagine they based fy on the load imposed at yield of the bar. Even if it does account for some increase in actual fy over design fy, there is no limit placed on this ratio, therefore there is nothing, in my mind, to preclude using a design fy of 40 ksi, 42, ksi, or 51 ksi, with an actual fy of 63 ksi.

 

[TXStructural]

What it boils down to, as with every other Code provision, is that an engineer can violate it IF they can provide evidence to the building official that their solution provide equivalent or safer design. The fact that this provision stays through each cycle makes me think there is good reason. I will not rehash my previous posts explaining my understanding of the provision and why it is required, but I still think the provision is valid and should be observed.

I doubt there are very many cases where there would be significant cost savings for cutting this corner without creating risk. I will do a little research and if there is room for changes, I will talk to 318 members about reviewing the provision next cycle.

 

[KootK]

I, for one, am not attempting to cut corners. It's not as though I'm concerned about situations where I've got all the space in the world and I want to shave a few inches off of my shear friction dowels. I'm concerned about situations -- and there are many of them -- where you simply can't get the job done if you don't use partial development in your shear friction bars. Here are two examples:

1) Field bent dowels in basement wall to main floor slab conditions. Seem my post and sketch on June 2nd.

2) The vertical joints between the panels of shear walls that we design as composite sections. My code (Canada) specifically states that interconnected wall assemblies be checked to ensure that they've got the mettle to truly behave compositely. Makse sense. And that means shear friction where "flange" panels meet "web panels". By the book, all I've got going for me there is shear friction. And if I stick with developing the shear friction bars for fy, then I'll usually need at least a 250 wall for 10M horizontal bars and at least a 300 wall for 15M horizontal bars. Obviously, the exact value depends on several parameters. The point is that satisfying this provision in this instance is a tough row to hoe and I don't really have any other options to turn to.

Developing for fy has been a thorn in the side of designers ever since I started designing a decade and a half ago. If we're going to continue to live with the inconvenience, it would be nice to have a little commentary elucidating the necessity of it. Surely someone had a rationale for the fy development business when they introduced it in the code? Can't that same party now just pull back the curtain and share that with the design community? And as we all know, the "test out" option for convincing building code officials of things is a non-starter for all but the largest projects designed at the largest firms. I won't be performing shear friction tests in my garage any time soon.

@TX: please don't interpret my frustration with this code provision as frustration with you. We've shaken hands before -- I know that you're a good guy. And I know that you're just relaying what you know about this particular issue, which is alot. I hope that you will table this issue before the ACI committee folks and either get the clause changed or have some explanation added to the commentary. If you can pull it off, there's a grain fed steak dinner in it for you the next time that CRSI sends you up north.

The greatest trick that bond stress ever pulled was convincing the world it didn't exist.

 

[TehMightyEngineer]

I'll second KootK's frustration though I usually get around it by headed anchor rods or some other mechanical means of developing fy. Either that or I ignore shear friction and design them as dowels per ACI appendix D.

Maine EIT, Civil/Structural.

 

[KootK]

It never occurred to me to use APP.D for rebar dowels. Thanks for that TME. What I generally do is hijack the provision that allows you to assume that 15M or smaller beam stirrups can be considered developed at the bend so long as they're hooked around a bar going the other way. Of course, this isn't explicitly allowed in the code. I'll be kicking myself pretty hard the day that I find out that my cheat is invalid. No... I'll be kicking the commentary writers for forcing me to shame myself in such a fashion.

The greatest trick that bond stress ever pulled was convincing the world it didn't exist.

 

[TehMightyEngineer]

Without diving into the code I can't recall anything that makes your "cheat" invalid. My understanding is stirrup sized bars can be considered developed if given hooks around anchoring bars.

Maine EIT, Civil/Structural.

 

[TXStructural]

Jut to add to the discussion about As,reqd/As,prov, ACI 318-11 12.5.3(d) uses the phrase:
(context is reduction factors for Ldh)

"(d) Where anchorage or development for fy is not
specifically required, reinforcement in excess of that
required by analysis.... (As,reqd/As,prov)"

The commentary restates this:
"The factor for excess reinforcement in 12.5.3(d) applies
only where anchorage or development for full fy is not
specifically required."

Which clearly do not allow reduction for hook length so the very least would be 8 db.

As far as using development around a bar, this is not explicitly permitted. 12.13.2.1 applies only to web reinforcement. It does seem reasonable, but we need to fix 318 if it is to be permitted.

 

[TehMightyEngineer]

Good info, thanks.

Maine EIT, Civil/Structural.