Thermal management is an issue with all electrical machines, including electric vehicle drives and wind turbine generators. Excessively high temperatures lead to loss of performance, degradation and deformation of components, and ultimately loss of the system.
Cooling of Rotating Electrical Machines: Fundamentals, modelling, testing and design provides a foundation of heat transfer and ventilation for the design of machines. It offers a range of practical approaches to the thermal design, as well as design data and case studies. Chapters cover fundamentals of heat transfer, fluid flow, thermal modelling of electrical machines, computational methods for modelling ventilation and heat transfer such as finite element methods and computational fluid dynamics, thermal test methods, and application of design methods.
Intended for engineers and researchers working in either academia or machine design companies, this book provides sound insights into the phenomena of heat transfer and fluid flow, giving readers an understanding of how to approach the thermal design of any machine.
Inspec keywords: heat radiation; stators; computational fluid dynamics; cooling; convection; pipe flow; thermal conductivity; finite element analysis; electric machines
Other keywords: thermal conductivity; convection; finite element analysis; pipe flow; electric machines; cooling; heat radiation; computational fluid dynamics; heat transfer; stators
Subjects: Convection and heat transfer; Conference proceedings; Thermal conduction in nonmetallic liquids; General fluid dynamics theory, simulation and other computational methods; Refrigeration and cooling (energy utilisation); Flows in ducts, channels, and conduits
Thermal analysis is becoming a more important component of the electric motor and generator design process due to the increasing demands of power output and efficiency for reduced weight and cost. Moreover, the applications for electric machines are increasing dramatically with CO 2 and urban pollution reduction involving developments of cleaner eMobility and renewable energies.
Heat transfer deals with the rate of heat flow as a result of temperature differences. There are three principal mechanisms: conduction, convection, and radiation, and all are relevant to electrical machines. Conduction occurs principally in solids and the rate of heat flow is determined by the temperature gradient and the thermal conductivity of the material. In electrical machines, it is of concern in dissipating heat in solid regions, such as the electrical conductors, magnetic iron, insulating materials, and frame.
Convection is the dominant form of heat transfer in liquids and gases and is associated with the transfer of heat by movement of the fluid. Air or liquid cooling systems are used on most electrical machines and the nature of the fluid flow, in terms of the type of fluid and flow pattern over the surfaces being cooled, determines the rate of heat transfer by convection.
Radiation heat transfer is in the form of electromagnetic radiation and in electrical machines occurs principally between solid surfaces separated by air gaps. Solids and liquids used in electrical machines can be considered to be opaque to thermal radiation and air can be considered to be transparent and not interact with radiation heat transfer.
This chapter gives a basic introduction to heat transfer and focuses on aspects that are particularly relevant to electrical machines. Many textbooks give more detailed coverage of the subject that may be referred to and examples are given in Refs [1-5].
This chapter reviews the fundamental principles of fluid flow and explains how this affects the convection cooling of electrical machines. For all fluid flow problems, the governing equations can be solved numerically based on the conservation of mass, momentum, and energy. Since the fluid flow equations are non-linear, they are being solved iteratively commonly using computational fluid dynamics (CFD) software. The main drawback of using the CFD method is its expensive computational cost and also the user needs to have good skills in CFD before useful solutions can be obtained, see Section 5.2.
Nevertheless, the relationship between the flow rate, fluid pressure, and the resistance to fluid flow can be simply explained by the flow equations given in this chapter. The flow rate passing through a cooling duct is limited by two main factors. One is the pressure gain from an impeller (fan/pump). Euler's turbomachinery formula is a useful equation that provides a fundamental understanding of the key parameters that affect the pressure gain from an impeller while the fan affinity laws are a useful tool for estimating the relationships between pressure, flow rate, and impeller power with impeller size and speed. The other factor that affects the flow rate is the flow resistance due to duct wall friction and flow separation effects. Besides the pressure losses in stationary ducts, some cooling ducts such as the annular air gap and rotor ducts suffer additional rotational pressure losses due to the effects of Coriolis force and centrifugal force. This means that for a constant fan pressure, the amount of flow passing through the air gap and rotor ducts is less than the stationary condition. Many correlations that describe the variation of pressure loss coefficient with the rotational speed based on the experimental investigation have been provided in this chapter. These correlations are useful tools that allow electrical machine designers to calculate the convective cooling performance.
Thermal modelling and analysis is an important topic for the electrical machine design process due to the demands for high machine power output with reduced weight, reduced cost and increased efficiency. Also, there is a strong interaction between electromagnetic and thermal design. For instance, the losses are dependent on machine temperatures and vice versa; an increase in magnet temperature will lead to a decrease in flux and thus reduce the output torque; the electrical resistance of copper windings increases with temperature and hence elevated winding temperatures give much higher copper losses and hence reduce the machine efficiency. Therefore, it is essential for electrical machine engineers to consider both electromagnetic and thermal design before an optimum design can be obtained.
The use of advanced computational methods: finite-element analysis (FEA) and computational fluid dynamics (CFD) will be considered in this chapter. FEA is useful to model conduction heat transfer, while CFD can additionally model fluid flow. Both methods rely on modelling the actual geometry of the full machine or machine component by discretizing the individual parts into smaller mesh elements, volumes or cells over which the heat transfer or flow variables are assumed constant. Triangular and quadrilateral mesh element shapes are typically used for 2D geometries, and tetrahedral, quadrilateral pyramid, triangular prism, and hexahedral shaped cells for 3D geometries.
The FEA method is often applied to individual components in the electrical machine where the heat transfer mode is by conduction, such as the slot and winding. Useful models can often be formed in 2D. 3D FEA models for a particular part can also be created but are less commonly used due to the additional time to set up and solve.
CFD is mainly of use to model the fluid flow and convection from surfaces in the machine and the models are nearly always 3D in nature. Models can be created for individual components of the machine or for the full machine, although this can be very complex to put together and solve. The model may also include conduction heat transfer.
In this chapter, Section 6.1 looks at the most common devices and methods used to measure temperature, a heat flux, and air flow. The advantages of the different methods are highlighted. Section 6.2 shows a test method that can be used to measure winding anisotropic thermal conductivity. As an example, typical values of the thermal conductivity through the wire-to-wire insulation system to the slot wall and along the conductors to the end-winding are given for different sizes of copper and aluminum conductors with varnish and epoxy resin impregnation material. Also for compressed windings. Section 6.3 gives some basic details of test methods that can be used to measure the different components of loss in the machine. Segregation of losses is a large topic in itself so this is only covered in basic detail here and reference is made to different international standards commonly used for the measurement of losses in electrical machines. The measurement of windage losses in more detail is described as this relates to air flow. In Section 6.4, guidance is given on how best to calibrate a thermal model using test data. This can be very useful to increase model accuracy and provide a useful insight of how a machine compares with other similar machines in terms of manufacturing goodness and quality of the design. Finally, in Section 6.5, details are given of a relatively new test method in which a short thermal transient test is used to estimate the winding-to-stator thermal resistance and winding thermal capacitance.
In this chapter, examples are given of practical electrical machine applications with different types of cooling such as Totally Enclosed Non-Ventilated (TENV), Totally Enclosed Fan-Cooled (TEFC), through ventilation, internal circulation flow with heat exchanger, housing water jackets, flooded stator cooling, and oil spray cooling. An indication is given of the reason that the cooling method was chosen for the particular application and some details of the advantages of one cooling type over another are explained. Apart from active cooling, examples are also given of electrical machine applications using more thermally conductive insulation materials. With enhanced passive cooling, the improvements in machine performance and torque/power density are explained.
Finally, an example is given of a high-performance electric motor designed to allow multiple cooling types to be used. The motor was for electric motorsport and the cooling can be tailored to the particular race circuit to give optimum overall performance.