In this study, a series of spray coating experiments were carried out in a pilot scale Forberg-style paddle mixer to validate a previously published multi-dimensional population balance (PB) model (Li et al., 2011). Experiments were carried out using unimodal distributions of large (mean mass 0.45 g) and small (mean mass 0.27 g) particles, as well as bimodal distributions with a 1:1 mass ratio of the large and small particles. Experiments were performed at Froude numbers from 0.6 to 2 with the final coating mass to core particle mass ratio being 0.05. A fully two-dimensional distribution of the coated particles, with respect to solid core mass and coating mass, was obtained for each batch operation. For unimodal core particle distributions, the coating mass coefficient of variation was proportional to t(-0.5) where t is the coating time, consistent with both the PB model and other published coating models. For the bimodal experimental series, the coating growth rate varied with particle mass m according to the relationship G=m(r) where r=0.37. This value for the growth expression exponent r is smaller than that predicted for random coating (r=2/3) or observed experimentally in fluidized coating systems (r=1). This preferential spraying of small particles in the mechanical mixer indicates that the size dependence of coating mass growth rate is a strong function of the equipment geometry and particle flow field. When a growth exponent r=0.37 was used, the coating distribution prediction from a discrete element method-population balance simulation compares well with the experimental results. Published by Elsevier Ltd.