Prediction of the thermal behavior of electric motors in the early design stage is crucial in any design process. The most popular prediction methods are analytical, and based on the lumped parameter model approach. These methods require experimental data in order to obtain accurate results, but this data is often not available. This paper deals with the problem of the lack of experimental data for an Open Self-Ventilated (OSV) Induction motor and reviews some of the most controversial parameters in thermal modeling, such as the bearings model and the axial conductivity of the lamination stack. Due to the nature of the OSV machine, through ventilation is also investigated, and a hydraulic model with improvements focused on rotational effects observation is presented. Moreover, the heat transfer in end spaces and ducts is studied, using dimensionless analysis correlations, along with focusing on new hydraulic phenomena, such as the development of the flow and the roughness effect. An implementation of a thermal circuit for an OSV machine that has good agreement with reference results is used to compare heat transfer coefficients used regularly for Totally Enclosed Fan Cooled (TEFC) enclosures. Finally, a sensitivity analysis is carried out on some parameters to determine their importance.

The prediction of the thermal behaviour of electric motors in the early stages of their design is a critical factor for reducing time and cost in the design process. In complex machine topologies such as open self-ventilated machines, there are several phenomenas to take into account in order to predict the correct thermal behaviour of the machine. In this study, a thermal model coupled with a hydraulic model is presented. These models provide information of the thermal behaviour of the machine. First, the complete thermal circuit is described, with some emphasis in the specially modelled parts. Then, the heat transfer coefficients for each surface inside the machine are presented, by the use of dimensionless correlations that avoids the need of previous knowledge. Moreover, the hydraulic model of the machine is studied, and also the coupling methodology between the two models is described for both steady state and transient calculations. Finally, the results from the model are validated using the data from two experimental runs, the first one with constant torque and speed, and the other with variable power, in a standardised service cycle, with a difference in the rotor bars and the stator winding below +/- 10 degrees C.

The design process for permanent magnet linear motors (PMLM) is similar to that used for their rotary counterparts. A number of electro-technical equations are used to get the best design for a specific application. Nevertheless, there are two issues that differ from the design of a standard rotary machine that must be considered. The first one is the intrinsic asymmetry of the magnetic circuit, which makes the flux distribution different from pole to pole. The second one is the particular geometry and constraints of the thermal circuit, which is crucial for the design of miniaturize linear motors. This paper presents an analytical method for computing the flux distribution in a surface-mounted PMLM. The method, based on the equivalent magnetic circuit method, considers the saturation of the material, the leakage at the end of the motor and the asymmetrical distribution of the magnetic flux. Additionally, an analytical lumped parameter thermal model for linear drives that gives results for both transient and steady-state temperatures is described. Validation of the models is performed by using Finite Element and Computer Fluid Dynamics software. Finally, both magnetic and thermal circuits have been used to design a miniaturized PMLM to drive a sliding door to experimentally verify its goodness