Shear Friction - Concrete to Rock

[JAE]

We have a situation where we'd like to install a concrete wall/plug in a rock tunnel to temporarily resist water pressure on one side of the plug.

The concept proposed is to install rebar dowels into the rock with adhesive and then place the concrete.

The plug is >30 ft. in diameter and 10 ft thick and will take a head of water >200 ft.

We are questioning the following:
1. Can the ACI shear friction provisions be applied to the shear connection between the rock and concrete?
2. Would shrinkage of the concrete for a 30 ft. dia plug be an issue on the sides and top? The plan is to place concrete in three lifts (10 ft each +/-)
3. The AASHTO bridge code also has shear friction provisions (where they apply it to slab-girder connections) but they add a second component based on cohesion. ACI includes the cohesion added value in their commentary in ACI 318 but require a very high ratio of reinforcement (ρfy >200) to allow this addition. Would the added cohesion value be too unconservative to include in the capacity calculation?
4. If the ACI or AASHTO provisions for shear friction do not apply to rock-concrete interfaces - what does?
5. The rock is reportedly a hard granite and has 4 to 6 inches of variation in profile - much greater than the 1/4" roughened surface suggested in the codes.
6. Would expansive concrete mixes be helpful?

 

[JAE]

bump - anyone?

 

[RFreund]

An attempt to get something started:

1. Seems like it should be pretty similar to me and with 4"-6" 'roughness' that should help.
2. Are you referring to 'radial' shrinkage? With the 4"-6" variation I'm still not sure this would be an issue but once again i'm not too familiar with that thick of a mass of concrete.
3. Not familiar with the background of this provision to provide comment.
4.Run calcs as if the rebar (or use threaded rods/bolts) were bolts in shear and use the lesser of the allowable shear for rock and concrete?
5. That's good news...I think.
6. Hmm, I would be nervous about the rock failing not necessarily the concrete or even the connection. I'm not sure if expansive concrete would add to the this concern or help alleviate it...

EIT

http://www.HowToEngineer.com

 

[cvg]

will you ever have backpressure?

In my opinion, shrinkage of the concrete is an issue, it will reduce the friction / shear by an unknown amount
the 4-6 inch rock will shear off, do you really want to rely on that?
seems that with enough rock bolts, you should not need to rely on friction except for additional FS.

 

[hokie66]

I tend to agree with cvg. In any case, I consider shear friction to be a bit of a black art, and would not like to rely on it for something like this. To make shear friction work, a lot of drilling is required, so you may as well use a recognized technique in tunnel construction. As the pressure is only on one side, I would look at using inclined rock anchors.

 

[aggman]

JAE,
We have used the ACI shear friction provisions for concrete to rock interface several times. Nothing along the magnitude of what your talking about, but definitely designing for loads of several hundred kips. We have not had any failures of our stuff that we designed but we have fixed several other similar designs that didn't survive the test of time. We are conservative with our approach and assume "concrete placed against hardened concrete not intentionally roughened", and honestly overkill the design. This came from AASHTO giving foundation sliding friction between mass concrete and clean sound rock as 0.70. During our experiences with this we have found the rock unpredictable at best. We stay concerned with the bond between the rock and the dowels mostly. We have seen very good bond in one location, move over 4' and rock fractures around the hole when drilled and you get virtually nothing. (This is with using epoxy grouted rebar with shallow holes) Water seems to seep over time between the two faces and lower the friction between the surfaces in many of our applications. 100% competent rock and perfect conditions and it shouldn't be a problem, but we have seen failures with foundations pinned to rock in our industry many times. (We design a lot of crusher foundations for mines, etc.) I have learned to prefer rock dowels that penetrate deeper into the rock (4' - 7' passive dowels), grouted, and then designed for shear through the bars and bearing. If we truly think we will see tension along the interface then we use post tensioned only because passive dowels have to lift to engage the grout and therefore shear friction is greatly reduced. We have always maintained that the interface had to stay tight for the thing to work like we were designing so it had to be weighted down, or pulled together with post tensioned rock anchors. You may be fine with the profile you mentioned but rock is far from concrete when it comes this type of design so be conservative. We have not found very much data on the subject.

 

[haynewp]

I would look into the feasibility of getting some tests done of the anchors into rock, I like the idea of shrinkage compensating mix.

 

[TXStructural]

The purpose of shear friction it to maintain engagement of the concrete and aggregates across a crack or boundary. In order to engage shear friction, the reinforcement must be developed on both sides of the boundary, and as you note, the surface roughness must be adequate. With the amplitude of the surface roughness you mention (4-6"), you should have no trouble engaging the boundary (like a series of keys), but the rock itself may be fairly smooth. I would suggest that the dowels be the largest feasible size, and they should be sufficiently anchored into both the rock and concrete, since the failure mechanism will likely be withdrawal.

You will be placing mass concrete in a confined (insulated) space, so you should have that evaluated. Thermal expansion/contraction and shrinkage will be a consideration for any mass concrete design. The mix will probably use very large aggregate and low cement content, which will minimize shrinkage anyway.

How do you propose to remove the plug at the end of the "temporary" period? That might demand certain design considerations, such as avoiding long, heavy rebar/dowels.