Shear Strength of Dowel from Continuous Footing into Wall

[dcarr82775]

I am looking at a 12" concrete basement wall. Its connection to a mat foundation consists of a single row of dowels from the mat into the wall that align with the outer mat of rebar in the wall. In my way of thinking these dowels provide almost no shear resistance at the bottom of wall to soil and groundwater loads. This is because the dowels are located at the loaded face of the wall. What would others use for shear strength of the connection under this condition? No shearkey is present

 

[KootK]

I detail it the way that you've proposed (sometimes with a shear key at the inside face if I'm feeling cautious). However, provided that the shear surface is roughened, I am comfortable using shear friction for the shear capacity. This reflects the assumption that:

1) There will be flexure at the joint that will stress the dowels in tension on the exterior side.
2) The tension in #1 will be balanced by a compressive force on the interior side.
3) The compression in #2 will provide the necessary clamping force for non-dowel action shear friction to develop.

I share your concern, however. I always have to fight the urge to place some dowels on the interior face as well. When I have dowel action stuck in my head, it feels as though the exterior dowels will just pop out of the wall.

Another thing that causes me grief is that the shear friction mechanism that I described above DEPENDS on a moment being developed at the joint. If the load on the wall is well distributed, as one would expect, then developing the necessary moment ought not be a problem. If, hypothetically, the resultant lateral load were to be applied one foot above the base of the wall, however, you'd be in trouble.

I worry about a similar thing for ramp slabs cast against CIP walls. I feel good about shear friction when I imagine cars driving down the centre of the ramp. I feel less good about cars driving beside the walls.

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

 

[KootK]

Also, I usually have a slab on grade for the wall to press against which feels good.

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

 

[FootNMouth]

What is your demand capacity ratio on the dowels? Also, the location of the dowels on the tension face is ideal for shear-friction as well. For example, if you were to dowel a simply supported beam into a support you would want to concentrate the dowels for shear friction at the top. This way as the top of the beam contracts the dowels will hold it tight so that the gap does not become excessive. If I were to design this wall from scratch I would provide enough reinforcement on the outside face to resist the moment in addition to the reinforcement required for shear friction.

 

[dcarr82775]

In this instance there is no roughening, so thinking shear friction you are in the world of dowel action. The steel provided is much less than minimum steel for bending. So the bar yields due to bending and nothing is left for shear friction. Even assuming there was steel sufficient for bending+shear, you still have dowel action and concrete breakout of the wall which is tiny for 1.5" of cover. Again, a roughened surface is a different animal.

 

[KootK]

I'm surprised to hear that your surface will be smooth Dcarr. Usually, when a footing is poured with dowels projecting from it, the area around the dowels is quite rough -- easily enough to qualify for 1/4" intentional roughening.

It's also important to note that, in this situation, you can use 100% of your flexural reinforcement as shear friction steel and vice versa. This is because, when the bars are stressed in tension due to flexure, you get a corresponding compression force which provides the clamping required for shear friction. You get to double dip so to speak.

It's not As = A_flex + A_sf.

Rather, it's As = max (A_flex, A_sf).

Since you likely designed your wall as pinned at the bottom anyhow, it really just comes down to A_sf.

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

 

[dcarr82775]

It isn't my wall/design and it is already built. When I want a roughened surface I call for one or at the least show it rough, the drawings show it smooth as a babies bottom. Given the dowels are 24"o.c. on the outside face and the rest of the mat is broom finished I have no confidence it is rough.

I don't agree on the double dip with a smooth surface. With a rough surface maybe, but not with a smooth surface.

 

[KootK]

Now that I know a bit more about your situation, I agree with your concern regarding the roughness. Still, localized roughness around the dowels and even the broom finish texture will give you something. Like dudley said, come up with a demand / capacity ratio based on 1/4" amplitude and As for your dowels. If that's close to the mark, be concerned; if you've got tons of reserve, consider letting it go as a judgement call.

Like you, I don't agree with the double dip -- or even the logic of a clamping force -- for a smooth surface. However, ACI does. ACI even explicitly recommends putting your dowels in the tension zone. While this might not alleviate your conscience altogether, it should be a comfort to know that you'd be fine in court.

If you're desperate, you could:

1) Use your dead load as a permanent clamping force in lieu of rebar. Even just the weight of your basement walls will add up to something.
2) You could use a seperate cohesion value of 0.25 MPa like the Canadian code allows. Obviously, using provisions from another jurisdiction's code requires careful consideration.

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

 

[dcarr82775]

I told the they have to move the surcharge farther from the building. At this point it is a mental exercise.

 

[precast78]

Does not matter where the steel is. Shear friction is purely friction between concrete and concrete. It has nothing to do with the dowel. The dowel is just there to provide tension/normal force for the friction force.

 

[dcarr82775]

With all due respect precast, the commentary of 318 says differently when the surface is smooth.

 

[FootNMouth]

I agree with dcarr. The commentary states they have provided artificially high values of the coefficient of friction so that the calculated shear strength will be in reasonable agreement with tested results. In other words dowel action does provide shear strength but they are accounting for it indirectly by increasing the friction coefficient.

 

[KootK]

@Dudley,

I read up on this recently. Apparently shear friction has two components: cohesion and friction. The code mu values are inflated to incorporate the cohesion component, not because of dowel action. If you have MacGregor's book, he gets into this in some detail.

KootK

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

 

[precast78]

So if you use the .6 value for smooth concrete, the shear value is not good enough? How do you analyze the wall? I hope you didnt analyze it as a cantilevered wall. I would at least design it as a wall supported by three sides especially if the walls aren't long. Your shear value at the bottom of the wall will be a lot lower.

 

[hokie66]

precast78,
I am not a proponent of shear friction, but your statement that the location of reinforcement crossing the joint is obviously incorrect for a situation where significant moment is present. Without reinforcement on the tension side, there is nothing to resist the compressive force required to create friction. There was a recent discussion in this forum of the same subject.