Design for punching shear in concrete: critical review of Canadian Standard CSA-A23.3-94: Discussion

The authors review the provisions in CSA-A23.3-94 for punching shear in concrete slabncolumn connections. They identify two items that are missing from this standard: (i )a value of αs = 2 for corner columns when determining the fac- tored shear stress resistance for the concrete and (ii) an upper limit for the stirrup spacing of s ≤ 0.5d when stirrups are used for shear reinforcement. The writers agree that these two omit- ted items should have been included. The authors also indicate they believe that the provisions for slab shear reinforcement when such reinforcement is re- quired have been changed from the previous edition of the Standard in a manner that can lead to unsafe design and refer to three tests that they claim substantiate this belief. These two items are discussed separately. The authors assert that CAN3-A23.3-M84 limits the shear stress due to the combined effect of Vf and (γvMf)x and y to 0.2λφc√ fcon a critical section outside the shear-reinforced zone. In the opinion of the writers, this is highly debatable. Clearly, this is the interpretation provided in Chapter 10 of Concrete Design Handbook (CPCA 1985) where, for an edge column, the shear moment interaction is calculated on a sec- tion outside the shear reinforcement. It is of interest to note that a similar example (18.4) can be found in Notes on ACI 318-89 (PCA 1990). This is relevant, since both CAN3- A23.3-M84 and ACI 318-89 use almost identical language regarding combined shear and moment transfer at slabncol- umn connections. In the PCA document, however, only the average shear stress due to Vf is considered at the critical section outside the shear-reinforced zone. In the opinion of the writers, neither CAN3-A23.3-M84 nor CAN3-A23.3-M94 re- quires the combined effects of shear and moment to be con- sidered at any section other than that located d/2 from a support, concentrated load, or change in slab thickness. Ideally, this discussion should depend not on semantics but on a consensus understanding of the behaviour of slabncolumn connections. The writers do not believe that there is as yet such a consensus. At best, the procedure in the Standard provides a reasonable metric of the severity of loading on a connection. It has been calibrated for a critical section d/2 from the column face. To extrapolate the procedure to sections located at greater distances from the column places inordinate faith in the fun- damental soundness of the procedure and ignores profound changes in the behaviour of the slab as distance from the col- umn support increases. In analyzing the test results, the authors do not appear to have considered all of the provisions that may govern the strength of a connection. Clause 13 of A23.3-94 gives three possible limits on the strength of a shear-reinforced interior slabncolumn connection. The first of these is a straight-line shearnmoment interaction calculated for the critical section located d/2 from the column face. The end points of the inter- action are given by the shear capacity of the connection with no unbalanced moment and the maximum unbalanced moment capacity with no net vertical shear. The second limit is on the magnitude of the average vertical shear outside the shear-re- inforced zone, called Vcode by the authors. Finally, Clause 13.11.2 requires that flexural reinforcement sufficient to resist γfM be placed within a band c + 3h. This provision places a flexural upper limit, Mflex, on the magnitude of the unbalanced