Numerical study of the behaviour of hollow unreinforced concrete masonry subject to eccentric axial load and examination of North American design standard provisions

Inaccuracies and inconsistencies in international design standard predictions of eccentrically loaded slender concrete masonry wall (SCMW) behaviour have long been noted. Scarce experimental data has limited the development of accurate provisions. As SCMW experimental work is complex, a numerical model presents value as a rational guide for future experiments and to compare against current SCMW design provisions. In this study, a three-dimensional detailed micro-model is developed for HSCMWs. The model is validated against recent HSCMW experimental strength and stiffness data. Different uniaxial compression stress-strain curves and initial imperfections are considered. Assumed compressive stress-strain behaviour is found to especially influence numerical HSCMW flexural stiffness; initial imperfections are found to reduce wall axial capacity up to 15 % among considered cases. A preliminary comparison of CSA S304–24 and TMS 402/602–22 HSCMW provisions against each other and the numerical model is then presented. Differences in maximum axial compressive stress ( f c ) between both standards are found to particularly drive differences in predicted strength. By comparing both standards against the numerical results, it is suggested both standards permit linear elastic analysis of cross-sectional stresses without penalty, and that the use of a higher f c in CSA S304–24 be considered. • A 3D detailed micro-model is developed for hollow concrete masonry walls. • Chosen uniaxial stress-strain curve particularly affects numerical flexural stiffness. • Effect of load eccentricity, end fixity, effective height, imperfection presented. • Normalized standard strength values differ from FE by up to 0.71 (CSA), 0.46 (TMS). • Changes to peak axial stress, linear elastic analysis clauses advised for CSA S304.