This study presents preliminary results of the modeling of all different testing components of a masonry wall when these components are subjected to unconfined compression: mortar cube, mortar cylinder, and concrete block compression (meso-scale structures), by the use of 2D and 3D Discrete Element Method (DEM). The purpose of this investigation is to calibrate the mortar cylinder and the concrete block using their corresponding numerical DEM model and to perform a parametric analysis for each of their sets of micro-parameters. The micro-parameters' influence on the model behavior is analyzed, and a broad comparison between the materials micro-properties is presented. The specimens were calibrated with experimental data obtained from previous experimental tests realized at Universidad Autonoma de Yucatan (Mexico). The initial calibration was completed based on meso-parameters populated from the experimental data. Besides the materials' micro-parameters, other control variables are analyzed such as shape (cube and cylinder) and the modeling dimension approach (2D and 3D) for the case of the mortar. Models showed to be more brittle than the experimental data. A parametric analysis was carried out to understand the independent influence of each micro-parameter on the macro-behavior of each specimen. PFC 2D/3D by Itasca was the software used to perform the simulations of all compression tests. Results showed difficulties when predicting the stress - strain curve, being able to predict the peak stresses. Also, comparison between geometry and dimension approach showed inconsistency when comparing 2D and 3D in the mortar case. Results will be used in a future study for the probabilistic multi-scale calibration of the masonry wall systems, from the independent wall components to the full macro-scale system response submitted to lateral cyclic loads.
