About Web Crushing of Structural Walls

Figure A. Thin-Webbed Members & Systems

The availability of rational assessment models for both shear-failure mechanisms permits the incorporation of a new genre of acceptable ductile failure mechanisms particularly suitable for structures in moderate seismic zones. This new genre consists of ductile shear failures, which manifest themselves as web-crushing failures or yielding of the transverse reinforcement at relatively high levels of displacement ductility. The accuracy of models already created offers the possibility of designing structural members according to performance level and introducing ductile shear failures as viable alternative to ductile flexural failures. Structural walls designed for this type of behavior can increase the system stiffness, minimize flexural damage to the edge elements, and provide for large stable shear capacities.

Shear strength of structural walls is governed by their transverse reinforcement, the wall thickness, and the concrete strength. The transverse reinforcement can be designed effectively to resist diagonal tension while the thickness and concrete strength must be chosen to resist diagonal compression (web-crushing) demands. Existing models for the web-crushing capacity of structural walls are based on the notion of average shear stresses distributed evenly across an effective section. Results from a recent research project conducted by the co-PI showed that web crushing failures occurring under inelastic displacement demands were concentrated inside the plastic hinge region at the interface of the structural wall with the compression boundary element. These types of inelastic failures can be termed as “flexure-shear” web-crushing failures or “ductile shear” failures.

Contrary to the assumption of average stresses on an effective section, web crushing in structural walls with highly-confined boundary elements under inelastic deformation demands occurs primarily where struts converge near the base of the compression boundary element within the plastic hinge region. The development of a web crushing failure within the region subjected to plastic deformations indicates that it is helpful to distinguish between elastic, or standard shear, and inelastic, or flexure-shear, web crushing capacity. The first type of failure, typically associated (but not limited) to the elastic range, occurs at very high levels of shear stress. Flexure-shear web crushing occurs in the inelastic rage and depends heavily on deformation demands.

Figure B shows a flexure-shear web-crushing failure, revealed by the convergence of struts into the compression toe within the inelastic region, seen in the in-plane test of a structural wall with confined boundary elements. The test is part of the studies conducted by the co-PI on the shear behavior of structural walls confined with boundary elements. Test observations show that as the test plastic hinge region grows up the wall height at moderate levels of displacement ductility (mD = 2), cracks near the base become too large for struts to carry compression directly to the footing. Consequently, the struts realign within the plastic hinge region to transfer shear into the compression boundary element. Narrowing of the compression struts is then linked to the tendency for web-crushing to occur in this zone, which finally dictates the failure mode for the wall. Figure C shows the hysteretic behavior for this test unit. Sufficient test data describing flexure-shear web-crushing just after yield between displacement ductility levels of mD = 1 and mD = 4 does not exist to demonstrate how, if at all, standard web and flexure-shear web-crushing behaviors interact at these levels.

Figure B. Web-Crushing Failure of Structural wall with Confined Boundary Elements

Figure C. Behavior of Structural Wall with Confined Boundary Elements with Web-Crushing Predictions

Development of the aforementioned models explicitly arrives at expressions for standard shear web-crushing as a function of the concrete compressive strength f'c, contrary to the codified limit state for shear stress as a function of sqrt(f’c) . This is reasonable since web-crushing failures are compressive failures. Such a relationship implies that high-strength concretes could increase the web-crushing capacity of structural walls with boundary elements as effectively as making these walls several times thicker. Verifying this experimentally has yet to be determined and it is thus the objective of this research.