Dowels for Contraction Joints in Concrete Walls

[Hokie93]

I am planning to use smooth round dowel bars for shear (out-of-plane) transfer purposes at vertical contraction joints in cast-in-place concrete walls. Basically, I am making the horizontal reinforcing discontinuous at the joints (which include reveals/chamfers on both wall faces to provide a weakened plane) and adding the smooth dowels at the center of the wall thickness. In designing the dowels I referenced ACI 224.3R-95 (Joints in Concrete Construction). This document was reapproved in 2013. In Chapter 8 (Walls), the document states the following: "PCA (1975) recommends using dowel bars that provide 0.015 times the cross-sectional area of the wall and extending 30 bar diameters each side of the joint as shown in Fig. 8.4". The PCA (1975) reference is "Basic Concrete Construction Practices". I do not have this reference. My question concerns the 0.015 factor. Taking a 12" thick wall and providing dowel bars at 24" on center results in a required dowel bar area equal to 4.32 square inches. Say what?? Clearly, this cannot be correct. Has anyone run into this before? I am guessing the correct factor is 0.0015 (i.e., a "0" was dropped from the published document). Or is the requirement to provide large bars (#9) at a very close spacing (6") for a 12" thick wall? I find that very hard to believe.

 

[JedClampett]

The .015 looks like a typo. I might believe .0015.
But I'm not a believer in any smooth dowels at contraction (or expansion) joints. If correctly designed, the walls shouldn't move different from each other. And if they do move, those dowels aren't going to help. They seem like a good idea, but I just can't justify them.

 

[TXStructural]

I guess it depends on what mechanism you are trying to use. I would assume that the concern leading to unbonding of dowels is preventing excessive restraint. Are these contraction joints or expansion joints?

Dowels that aren't bonded act in shear, but very quickly act in bending as the concrete around them crushes or the joint opens. In this case, the dowel needs to be stiff enough to resist bending. That said, I am not sure why one would need 30 bar diameters embedment for an unbonded dowel.

Bonded dowels require full development on both sides in order to engage shear friction (resistance using aggregate interlock across the joint), which creates a much better resistance to displacement than dowels alone.

As to As of dowels, and spacing, I don't think I would go to 24" spacing, since the resulting load per dowel could be substantial, depending upon the purpose of the dowels.

 

[Hokie93]

The dowels are intended to limit any tendency for two adjacent wall segments to have differential out-of-plane movement. I am slightly concerned about this since I am discontinuing 100% of the horizontal reinforcement at the contraction joint. The horizontal dowel bars would thus, hopefully, limit the out-of-plane movement while allowing in-plane movement due to shrinkage (the dowel bars are smooth and coated to prevent bond with the concrete).

 

[StructuralEd]

Even though we specify for both sides of walls be be evenly backfilled, we know that is never going to happen so the dowels do help keep a straight wall plane in that instance.