Shear strength across cold joint without shear friction?

[Lion06]

I'm having a serious struggle with a waterproofing contractor and would like to get some other opinions.

I'll preface the statement with my understanding of a basic aspect of what is critical (in my mind) to the discussion. My understanding of the shear capacity across a cold joint across a concrete pour is based on shear friction only. ACI does not state anywhere that you can simply use the capacity of steel bars in shear crossing a joint. I recognize this is conservative. For example, if you have bars crossing a cold joint where you have a frictionless surface then shear friction would say you have no shear capacity. That's obviously not correct, but I would feel like my hands are somewhat tied.

Now to the actual problem - waterproofing and mastic between drilled pier and grade beam pours (there are concrete shear walls above the grade beams). This would be the biggest concern at drilled piers taking lateral loading. I feel that in order to get the shear force from the grade beam into the drilled piers I MUST use shear friction. If there is something (waterproofing and mastic) with no reliable coefficient of friction how can a shear capacity be calculated?

I'm being told by the waterproofing sub that they've waterproofed this way on every drilled pier job they've done for 30 years. Am I missing something?

 

[dcarr82775]

I don't agree. Shear friction gives you a value for a non-roughened surface, 0.6. It doesn't say frictionless surface, but it isn't assuming any friction for that condition anyway. It is all the steel in that condition.

I have never allowed anything between the top of the drilled pier and the concrete bearing on that pier. I have never had anyone ask either. If you have any sort of 'gap' between (2) surfaces I have looked at the shear transfer as the bending capacity of the rebar across that gap (fixed each end), or the shear strength of the bar itself. The bending typically controls when I have done it.

 

[Lion06]

dcarr -

That's my concern. ACI doesn't allow you to use the steel in shear. It requires you to use shear friction. When you look at the steel bars in shear across a joint without using shear friction, do you look at the bars with Appendix D in mind considering breakout of the bars near edges?

I'm not comfortable using a coefficient of friction for concrete cast against concrete if there is a later of mastic and waterproofing between the two pours.

The point I was trying to make with the frictionless surface example was that if you have a theoretically frictionless surface (say you cast a teflon plate on top of the drilled pier and another on the bottom of the grade beam - not that you would do this, but just for talking purposes) then the shear friction equation would say that you have no capacity.

 

[KootK]

I believe that the commentary to the Canadian code describes the conditions associated with the various coefficients of friction. For the "smooth surface" option, they state that it's basically dowel action. So:

1) I agree with Dcarr that shear friction provisions do make some account of dowel action.
2) I'm not sure that I agree that, because of dowel action, the shear friction model is grossly conservative.

I was doing some research on precast connections recently and came across a method for calculating shear transfer based explicitly on dowel action. If I can find that, I'll post a reference. I've never seen it anywhere else before.

Can the drilled piers project up into the grade beams in such a way that you can create a keyed connection of some sort? I'd be okay with the mastic in between if there were shear keys in the pile cap such that there was some serious amplitude in the resulting surface. I suspect that your drilled piers are already in place and precluding that however.

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

 

[Lion06]

I agree it is grossly conservative. That's the point I was making with the frictionless surface example. What I'm getting at is that ACI doesn't give you any other options. It would not be prudent to just disregard ACI requirement because I don't like it. We all make judgments in specific circumstances for unique conditions, but I don't think it wise to completely disregard a code requirement.

With that said, the commentary in ACI does state that for concrete cast against concrete without intentional roughening that the primary transfer is dowel action. I suppose this can be interpreted that you could never have a coefficient of friction less than 0.6 (for purposes of shear friction calculations) regardless of what is between the two pours.

 

[KootK]

Lion,

If you were referring to my comments, you may have misread me. I don't know that shear friction values are grossly conservative.

I'm also not sure that agree with your minimum COF proposal. Here's what I'm thinking:

1) If you're truly down to just dowel action, the capacity ought not depend on clamping force. COF would no longer be relevant.
2) With true dowel action, I start to become concerned with where the dowels are located. I'd be inclined to discount any "dowel" near to and pushing towards a free edge.

I've always wondered if that commentary blurb that we've both mentioned was a bit of a misstatement. I feel as though it ought to say something like "As the interface becomes smoother, dowel action plays a more important role". The format of the equation, and the fact that clamping force is still part of it, make me think that a dependable friction surface is still required.

KootK.

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

 

[dcarr82775]

As you note, ACI says the resistance is primarily through dowel action for the 0.6 case, not friction. So IMO it does give you clear guidance. In this condition friction isn't the issue.

Under your condition you don't have concrete against concrete so you could easily have less than 0.6, and what that # is up to you to figure.