The thermal and multiphase flow physics are among the leading determinants of the performance, durability and life of gearbox systems and are crucial to devising optimization guidelines to minimize the power loss associated with their operation. Despite the significance of the thermo-fluids of the gearbox system, limited capability has been developed to predict and characterize these phenomena in a wholistic fashion. This is, in part, due to the significant complexity and multi-scale nature of the physical phenomena involved. This paper demonstrates the application of a recently developed modeling methodology (Yazdani and Soteriou, 2014) for predicting the time-dependent thermo-fluid state of the gearbox system on a set of three interlocking gears that serve as a validation vehicle. Along with the modeling approach which is applicable to any multiscale problem of this nature, a novel gridding methodology is presented for simulating interlocking gears with substantially improved accuracy. The long-term (time-independent) temperature distribution within the gearbox is obtained through the proposed simulation method and compared to the experimental measurements. In addition, the model is shown to be able to reproduce the temporally cyclic temperature oscillations once the system reaches the stationary-state condition. (C) 2014 Elsevier Ltd. All rights reserved.