Discussion of “Lateral Stiffness of Shear Walls with Openings” by A. Neuenhofer

Abstract

The author presents a parametric study devoted to investigatingthe accuracy of a simplified hand method for calculating the lat-eral stiffness and deflection of shear walls with openings whencompared with results obtained from finite-element analyses. Theauthor concludes that the simplified hand method recommendedin several design guidelines for practicing structural engineersconsistently underestimates the impact of the opening on the re-duction of the stiffness, thereby producing a lateral stiffness largerthan that obtained from detailed finite-element analysis.The discusser fully agrees with the author, since he had alsoevaluated this simplified hand method before using finite ele-ments and found it inaccurate, inconsistent, and unreliable i.e.,Tena-Colunga 1992b; Tena-Colunga and Abrams 1992 . The dis-cussed method is presented and illustrated in detail as Method I inSchneider and Dickey 1987 . The discusser evaluated the accu-racy of the estimates of the lateral deflection of the simplifiedhand method presented by Schneider and Dickey Method I forthe four walls depicted in Fig. 1, which are part of example 10–2of Schneider and Dickey 1987 . Subject walls were meshed withfour-noded plane-stress isoparametric finite elements using theprogram POLO-FINITE Lopez et al. 1987 . Top average deflec-tions obtained from the finite-element analysis and the simplifiedhand method Method I for the four walls with openings depictedin Fig. 1 are presented in Table 1, where the error estimate pro-posed by the author in Eq. 8 is also presented. As can be ob-served, the results obtained with the simplified hand method un-derestimate the lateral deflection overestimate the lateralstiffness of the perforated cantilever walls under study whencompared with results obtained with finite-element analyses. Be-sides, from analysis of the results given in Table 1, one finds thatthere is no consistency in the method. According to the back-ground of the simplified hand method, which is well explained bythe author i.e., Fig. 5 , one may assume that the best approxima-tion of the simplified method will be obtained for Wall B Fig. 1 ;however, the higher error is precisely obtained for that wall Table1 . On the other hand, one may assume a priori that the worstcorrelation should be obtained for Wall A Fig. 1 , but for thiswall the simplified hand method had the best approximation, witha reasonable error of 9.9% Table 1 .Recently, the discusser directed a study Vergara-Sanchez2005 that assessed the accuracy of the simplified hand method among other methods with finite-element analyses by usingSAP-2000 for 16 cantilever walls with a single opening for threedifferent wall aspect ratios