Modelling the interaction between particles and boundaries is of great interest and importance in many Discrete Element Method (DEM) applications where boundary failure conditions dictate the flow behaviour and velocity gradients of a bulk material. Bulk wall friction angles in DEM models are dependent on the mechanical properties of the particles and boundary material as well as the constraints of the contact models used, especially the tangential and rolling torque components. This paper examines the parameters determined from the calibration inclination test by conducting direct wall shear tests to validate the accuracy of the calibration parameters and the sensitivity of particle-to-boundary parameters of the material model using the Hertz-Mindlin (H-M) and H-M with linear cohesion contact models. The influence particle geometry has on wall friction is also examined where results from the DEM simulations are compared against the corresponding results from a large-scale shear tester. The 3-sphere cluster particle developed in this study is found to provide the most representative bulk shear behaviour where the wall yield loci predicted by DEM agree well with all experimental results.