Punching of Reinforced and Post-Tensioned Concrete Slab-Column Connections

Data collected from shaketable tests of two, approximately 1/3-scale, two-story flat plate frames using shear reinforcement, as well as data from previous tests, were evaluated to assess the interstory drift ratios when punching failures occur for reinforced concrete and post-tensioned slab-column connections with and without shear reinforcement. The drift ratios at punching failures for the two shaketable specimens were approximately equal to values reported for quasi-static tests of isolated specimens without shear reinforcement; but substantially less than reported for quasi-static tests of connections with shear reinforcement. The review of test results indicates that the bilinear relation for lateral drift versus gravity shear ratio to assess the need for shear reinforcement at slab-column connections approved for ACI 318-05, Section 21.11, is generally conservative for typical connections of all connection types. The data also were used to assess parameters required for a simple shear-strength degrading model for slab-column connections.

Keywords: columns; reinforcement; seismic; slabs.

INTRODUCTION

Slab-column frames are commonly used to resist gravity and lateral loads in regions of low-to-moderate seismicity and well-established design requirements exist to avoid punching failures at the slab-column connections.1 To avoid punching shear failures at slab-column connections, the shear stress on the slab critical section due to direct shear (V^sub u^/b^sub o^d) and eccentric shear (γ^sub v^M^sub u,unb^c/J^sub c^) cannot exceed the nominal shear stress capacity of the critical section (v^sub n^ = v^sub c^ + v^sub s^), where γ^sub v^ is the portion of the unbalanced moment M^sub u,unb^ transferred by eccentric shear (for example, typically 40% for square, interior columns), v^sub c^ and v^sub s^ are the nominal shear stress capacity provided by the concrete and the shear reinforcement, respectively. If the calculated shear stress exceeds the nominal shear stress capacity, a punching failure is anticipated, and the design must be modified until the stress is acceptable (for example, thicker slab, larger column).

For slab-column frames subjected to lateral displacements due to earthquakes, punching failures are possible even if the shear stress on the slab critical section does not exceed the nominal shear stress (that is, no stress-induced failure). In this case, it is hypothesized that the shear stress capacity of the critical section degrades (for example, Pan and Moehle2 and Hawkins and Mitchell3), and punching failure occurs when the shear capacity degrades to the point where it equals the demand (Fig. 1). In this paper, this is referred to as a driftinduced failure to differentiate it from a stress-induced failure discussed in the preceding paragraph.

Experimental studies of reinforced concrete flat plate slabcolumn connections (for example, Pan and Moehle,2,4 Moehle,5 and Robertson and Durrani6) have shown that the magnitude of the gravity shear stress on the slab critical section adjacent to the column significantly influences the drift level at which a punching failure occurs. Although slabcolumn frames are commonly used for nonparticipating systems1 in zones of high seismicity, no guidance is provided in ACI 318-02,7 Section 21.11, for design of these systems.

A code change was approved for ACI 318-05,1 Section 21.11.5, to clarify the intent of the code with respect to checking the potential for punching failures of slab-column connections of nonparticipating frames. The new code provision assesses the need for shear reinforcement at slabcolumn connections based on the interstory lateral drift ratio and the gravity shear stress on the slab critical section. Alternatively, calculations can be made to show that the connection is capable of sustaining the drift associated with the design displacement without punching. The latter approach requires either a detailed analysis of the nonparticipating slab-column frame subjected to imposed lateral displacements or a limit analysis where maximum connection demands are determined. The use of a limit analysis appears attractive given the complexities of doing the detailed analysis, especially for cases where a fuse is used to limit the connection demands. This latter approach does not address the potential for shear strength degradation noted in Fig. 1, whereas use of the first approach, which is based on test results, incorporates potential strength degradation.

The relationship between gravity shear stress ratio, lateral drift ratio, and punching failure in ACI 318-051 as well as a best-fit line for test data derived from tests of isolated, reinforced concrete, slab-column connections without shear reinforcement (Table 1), are depicted in Fig. 2(a). Test results indicate that the lateral drift ratio at punching decreases as the gravity shear ratio increases, and results are conservative for the database of existing tests (only 4 of 76 tests fall below the ACI relation), which represent details used in typical construction of slab-column connections in areas of high and low-to-moderate seismic risk. For unusual geometries and quantities of reinforcement, a more detailed assessment of the potential for punching failure should be conducted. It is noted that relatively little data exist for gravity shear ratios greater than 0.6; however, for such large gravity shear ratios, the new ACI provision requires shear reinforcement unless the interstory drift is less than 0.005.