IEEE TRANSACTIONS ON INDUSTRY APPLICATIONS, VOL. 47, NO. 5, SEPTEMBER/OCTOBER 2011 2023 Thermal Modeling of a Segmented Stator Winding Design Rafal Wrobel, Phil H. Mellor, and Derrick Holliday Abstract—This paper presents a thermal analysis of a seg- mented stator winding design. As the thermal performance is one of the main factors limiting a machine’s output capability, a thermal test on a complete prototype machine is an essential part of the design process. However, for the segmented stator winding design, a test-informed thermal analysis on a single stator tooth can be performed prior to the manufacture of the full machine. This approach allows for a rapid and inexpensive assessment of the thermal performance of the complete machine and early identification of design modifications needed. The research has been applied to the design of a highly efficient and compact permanent-magnet traction motor. A thermal model for a single tooth was developed and supported by tests to identify key heat transfer coefficients. A number of winding assemblies were com- pared, and the most promising was selected for the final motor prototype. The results from the approach are compared with thermal test results from the complete machine. Index Terms—Permanent-magnet (PM) motor, segmented stator, thermal modeling. I. I NTRODUCTION T HERMAL PERFORMANCE is one of the main factors limiting an electrical machine’s output capability. To as- sess the thermal envelope for a new motor design, a number of tests on a prototype are usually required. As the design process tends to involve a series of iterations, this may require more than one prototype to be manufactured and tested. Such an approach can be expensive and time consuming. There are a number of analytical and numerical tech- niques that can be used to determine the temperature distrib- ution within electrical machines. Preferred techniques include finite-element (FE)- and lumped-parameter-based approaches [1]–[4], [15]–[26]. However, these thermal models tend to be inaccurate if not informed by test data obtained from a previous prototype. This relationship between a model and the prototype machine is very important since there are many manufacture- dependent factors that can only be accurately obtained by experiment. The segmented stator winding design has well-known bene- fits of high copper fill factor, compact end winding, and simple Manuscript received February 21, 2011; accepted May 9, 2011. Date of publication July 14, 2011; date of current version Septmeber 21, 2011. Paper 2011-EMC-059, presented at the 2010 IEEE Energy Conversion Congress and Exposition, Atlanta, GA, Sep. 12–16, and approved for publication in the IEEE TRANSACTIONS ON I NDUSTRY APPLICATIONS by the Electric Machines Committee of the IEEE Industry Applications Society. The authors are with the Department of Electrical and Electronic Engi- neering, University of Bristol, Bristol BS8 1UB, U.K. (e-mail: r.wrobel@ bristol.ac.uk; p.h.mellor@bristol.ac.uk; derrick.holliday@eee.strath.ac.uk). Color versions of one or more of the figures in this paper are available online at http://ieeexplore.ieee.org. Digital Object Identifier 10.1109/TIA.2011.2161741 manufacture [6], [7]. Potentially, there is an additional benefit of undertaking thermal analyses on a single stator segment that could significantly accelerate and simplify the thermal design process. This simplified approach of a single-tooth thermal model informed through test is investigated in this paper. The single-tooth thermal model employs a hybrid modeling technique that is based on the use of anisotropic lumped regions within a 3-D thermal FE analysis (FEA) solver [14], [15]. The method significantly simplifies the model definition and consequently reduces the solution time. The anisotropic thermal properties of the lumped regions need to be derived experimen- tally [14], [15]. However, the measured material data improve the accuracy of the model as compared to more conventional modeling techniques. Input loss data are obtained using iron loss estimates from a computationally efficient voltage model [9] combined with test results to obtain ac copper loss correlations. The iron loss model accounts for both rated flux and field-weakening operations and is based on two discrete-time step 2-D magnetostatic FEA for open- and short-circuit operations of the machine. The parameters obtained from these analyses are used alongside the standard d–q equivalent circuit model to generate a map for the iron loss across the entire machine working envelope. The calculated results from the calibrated single-tooth ther- mal model are compared against test data from a complete motor assembly showing good agreement. Furthermore, the thermal model has been used to evaluate the potential benefits coming from the use of a winding encapsulant with improved thermal properties. II. PROTOTYPE MOTOR CONSTRUCTION The research has been applied to the design of a highly efficient (97% peak efficiency) and compact (2 kW/kg continu- ous rated) water-jacket-cooled permanent-magnet (PM) motor designed for a large-vehicle application. The motor comprises 8 poles and 12 slots with a double-layer concentrated winding (Fig. 1). The segmented stator core is cooled via an outer water jacket. The laminated stator segments are made of 0.35-mm silicon iron. Due to the high maximum speed of operation, a Litz wire was used for the winding to minimize ac loss effects. To improve the thermal path between the winding and the casing, the stator segments are vacuum impregnated with a high thermal conductivity resin (Fig. 2). III. THERMAL MODEL Thermal FEA is a well-established technique for modeling conduction across a 3-D structure. However, it has not been 0093-9994/$26.00 © 2011 IEEE