No abstract

This paper addresses the heat transfer coefficient (HTC) associated with the hollow-shaft rotor cooling of a traction motor. In such a hollow-shaft cooling system, the coolant is made to flow through the shaft hole in order to cool the rotor. The HTC is estimated using computational fluid dynamics (CFD), where the effect of rotational velocity as well as the coolant flow rate have been accounted for. Experiments are then used to validate the accuracy of CFD models with the assistance of an analytical lumped-parameter thermal network (LPTN) approach. On the basis of CFD simulations and the experimental prototype testing, it is concluded that the rotational speed can significantly increase the convective heat transfer of the shaft hole about 3.8 times at 4500 rpm above the level of the stationary condition. As a result of such analysis, a new dimensionless correlation of the Nusselt number with the Reynolds number is derived.

This paper addresses the heat transfer coefficient associated with a shaft-cooling of traction motors. In such shaftcooling systems, the 50-50 ethylene glycol-water is made to flow through the shaft hole in order to cool the machine. The heat transfer coefficient is estimated using a CFD (computational fluid dynamics) method, where the effect of rotational velocity as well as liquid flow rate have been accounted for. The results from two different turbulence models were compared. As a result of the simulations, it is concluded that the rotational speed can significantly increase the convective heat transfer in the shaft hole above the stationary condition. Finally, the benefits of implementing a shaft-cooling to an existing ferrite magnet traction motor, in terms of continuous torque capability, is described.

No abstract

Fans are commonly used to refrigerate electrical machines when air forced cooling is required. They are rather simple solutions that can considerably improve the thermal performance of electrical machines, so the fan analysis could become an important point during the thermal design of electrical machines. This paper presents an analysis by computational fluid dynamics (CFD) simulations with the goal of characterizing radial fans, used in electrical machines. Different boundary conditions and parameters that affect to CFD results are evaluated in order to get an efficient way to simulate radial fans. The main objective is to obtain the flowpressure characteristic curve of fans and airflow velocities by CFD simulations. Finally CFD simulation results of 2 fans are experimentally validated. One of them has been analysed in a fan testing wind tunnel. And the second one has been analysed in an auto-ventilated machine.Index Terms-computational fluid dynamics (CFD), cooling systems design, fan characterization, thermal optimization, electrical machines, lumped-parameters thermal models, airflow pressure, fan testing wind tunnel, auto-ventilated machine. he is currently an Associate Professor. His current research interests include electrical machine design, modeling, and control. He has participated in various research projects in the fields of wind energy systems, lift drives, and railway traction. he is currently an Associate Professor. His current research interests include electrical machine design, modeling, and control. He has participated in various research projects in the fields of wind energy systems, lift drives, and railway traction. Gaizka Ugalde received the B.Eng. and Ph.D. degrees in electrical engineering from the University of Mondragón, Mondragón, Spain, in 2006 and 2009, respectively. Since 2009, he has been with the Department of Electronics, Faculty of Engineering, University of Mondragón, where he is currently an Associate Professor. His current research interests include permanent-magnet-machine design, modeling, and control. He has participated in various research projects in the fields of lift drives and railway traction.