Electric Machine Simulation Technology Steve Hartridge Director, Electric & Hybrid Vehicles

Electric Machine Simulation Technology Steve Hartridge Director, Electric & Hybrid Vehicles

Agenda Intro/Session description Todays demands/motivations EMAG and Thermal modeling Combined workflow Examples

Motivation for Analysis Over the last decade it is noticeable that there is a growing need for electric machines with High torque or High power density along with a High efficiency demand or/and Reduction in size, weight, cost Leading to higher temperature gradients with a higher demand on the materials in general, but esp. on the insulation materials shorter lifetime expectation due to a higher risk of thermal damages (esp. in the insulation materials). A higher risk of demagnetization of the magnets Source graphics: NREL

Motivation for Analysis Component lifetime estimates [1]: 22% of failures due to thermal damages in insulation 17% further thermal damage in other components Lifetime depends on temperature history; Temperature depends on losses and cooling Insulation lifetime L can be modeled by the Arrhenius chemical equation [2]: L A Montsinger s rule taken from transformer oil and solid insulation materials shows that the lifetime L decreases by 50% with increase of temperature T by 10 K [3]: L T 10K 0.5 L T So insulation breakdown is likely to be the problem associated with high temperatures. This problem may be tackled by either improving the insulation material and allowing the temperatures to rise or improving the cooling performance of the windings and limiting the maximum temperature. Source: [1] Bruetsch, R., Tari, M. Froehlich, K. Weiers, T. and Vogesang, R., 2008. Insulation Failure Mechanisms of Power Generators IEEE, Electrical Insulation Magazine, 24(4) [2] Dakin, T.W., 1948, Electrical Insulation Deterioration Treated as a Chemical Rate Phenomena, AIEE Trans., Part 1, 67 [3] Binder, A., TU Darmstadt, EW, 2008, Script Large Generators & High Power Drives

Motivation for Analysis To accomplish today s demand the new machine designs have to eliminate the safety factors of the over-sizing designs of the past to finally ensure the requested high power densities. The need to have an optimized thermal design besides an optimized electromagnetic design.

Electric Machine Simulation Technology Electromagnetic Simulation Electrical/mechanical performance of design Design studies of different types of machine IMD vs. BDC Torque and efficiency requirements are met Build efficiency map for machine Detailed geometric design of components 2D/3D Optimize magnet position/shape/material Include a simple/conduction only thermal model Coupled Problem Thermal Simulation Understand the efficiency of the cooling system Optimize a flow paths for a given cooling system Predict maximum component temperatures at given different operating points Consider Conduction/convection/radiation system Include temperature dependent properties Machine Designer/Electrical Engineer Thermal analyst/mechanical engineer

Losses in Electrical Machines The heat generated inside the motor originates from two main sources: Electrical losses include the copper losses - also I 2 R losses - in the windings (heating effect due to copper resistance), core losses and (magnetic hysteresis (~ B k f) and eddy currents (~ B 2 f 2 ) in iron cores) eddy current losses in other parts of the machine being electric conductive, e.g permanent magnets, end shields, housing parts, Mechanical losses, such as frictional losses generated by the bearings as well as windage losses

Thermal Modeling in Electrical Machines Brushless generator Conjugate Heat Transfer Analysis of Integrated Brushless Generators for More Electric Engines Marco Tosetti, Paolo Maggiore, Andrea Cavagnino, Senior Member IEEE, and Silvio Vaschetto, Member IEEE Politecnico di Torino, Corso Duca degli Abruzzi 24, 10129 Torino Italy

Thermal Modeling in Electrical Machines Brushless generator Winding Temperature Stator Core Temperature Conjugate Heat Transfer Analysis of Integrated Brushless Generators for More Electric Engines Marco Tosetti, Paolo Maggiore, Andrea Cavagnino, Senior Member IEEE, and Silvio Vaschetto, Member IEEE Politecnico di Torino, Corso Duca degli Abruzzi 24, 10129 Torino Italy

Achieving Coupled Models Electromagnetic Simulation Thermal Simulation Manual Transfer of losses Rotor, Stator, Windings Homogeneous application Coupled Problem Template based design codes Simple circuit models Mapping of distributed losses Segmented by parts Maintain distribution of losses Typically from Finite element codes Codes often use a temperature Finite Volume flow/thermal codes Homogeneous losses on bodies Rotor Stator Windings Heterogeneous losses Map between grids

Losses in Electric Machines Brushless DC motor, 10KW max power Homogeneous application of losses per component Copper losses = 43 W Iron losses stator = 345 W Magnet losses = 0.74 W Heterogeneous application of losses See image Comparison of Solution Heterogeneous Homogeneous

Losses in Electric Machines Comparison of maximum temperature Heterogeneous Homogeneous Heterogeneous mapped losses lead to higher maximum temperatures

Data Transfer to STAR-CCM+ - Losses The B-field variation allows iron loss estimation GoFER of 72 rotor positions /elec. revolution Modified Steinmetz method in SPEED applies also to non sinusoidal currents Front part of the tooth sees stronger field variations which is reflected in the higher iron loss density The iron loss density can be visualized SPEED Select the Plot Tab 13

Data Transfer to STAR-CCM+ - Geometry SPEED geometry for: stator, slot windings, rotor, rotor bars. CAD geometry for: end-windings, endrings, all non-active components (fan, housing, etc )

SPEED > STAR-CCM+ Industrial Example Induction machine, overblown with fan on the shaft SPEED Model > loss distribution STAR-CCM+ > Temperature profiles

Simulation Steady State Temperatures 2 2 1 1 Rotor Bar Avg=148.4 C, End Ring 1 Avg=144.7 C, End Ring 2 Avg=147.6 C Shaft Min Temp=55.8 C, Shaft Max Temp=148.3 C SPEED model with rotor temp @ 148 C requires 52.5 % of copper conductivity for consistent losses and performance at this load point. 16

SPEED > STAR-CCM+ Industrial Example Comparison with Measurements Client measurements on aux and main winding at 2 circumferential locations, both for the fan (cold side) and exhaust (hot side) of the end winding. Compare with mean and standard deviation of temperature in outer 5mm of endwinding End Winding 2 (cold side) Measurement Simulation % Error Mean 91.5 C 93.1 C 1.74 % STD 1.88 C 2.14 C End Winding 1 (hot side) Measurement Simulation % Error Mean 111.4 C 111.9 C 0.45 % STD 3.03 C 1.30 C

SPEED > STAR-CCM+ Industrial Example Heat Flow: Rotor and Stator

SPEED > STAR-CCM+ Workflow Import SPEED geometry and surrounding CAD for non-active components in to STAR-CCM+ Compute electromagnetic losses in SPEED for specific load point and import into STAR-CCM+ Define appropriate physics and boundary conditions in STAR- CCM+ Solve conjugate heat tranfer problem for specific load point in STAR- CCM+ Specify new operating point and recompute temperatures Low speed, high torque High speed, low torque

SPEED > STAR-CCM+ Industrial Example What if study: Vented Stator iteration New CAD geometry imported Remessed and case rerun End Winding 2 (cold side) Orig Design Vented Stator % Mean 93.1 C 76.6 C 17.7 % STD 2.14 C 1.63 C End Winding 1 (hot side) Orig Design Vented Stator % Mean 111.9 C 85.9 C 23.2 % STD 1.30 C 0.97 C

Temperature Dependent Resistivity of Copper Winding Copper winding modeled with temperature dependent resistivity, results in higher local heating where the coil is hotter. Vented stator shows reduction in coil temp and total heat load from 197 W to 180 W of copper losses.

SPEED > STAR-CCM+ Workflow Import SPEED geometry and surrounding CAD for non-active components in to STAR-CCM+ Compute electromagnetic losses in SPEED for specific load point and import into STAR-CCM+ Define appropriate physics and boundary conditions in STAR- CCM+ Solve conjugate heat tranfer problem for specific load point in STAR- CCM+ Specify new load point and recompute temperatures Change Geometry and recompute

JMAG > STAR-CCM+ Example Copper loss density distribution JMAG Low speed: 600 rpm Loss density Iron loss density distribution JMAG Magnet loss density distribution JMAG Low speed Medium speed High speed

JMAG > STAR-CCM+ Example Mapped imported heat loss distribution STAR-CCM+ Low speed: 600 rpm High speed: 8,000 rpm Temperature distribution STAR-CCM+

Combined Workflow Links with other FE supplier: JMAG (JSOL, Japan) and FLUX (Cedrat, France) SPEED provides initial design Data export for further electromagnetic and thermal analysis FE calculation For detailed EMAG and loss calculation and export of loss data STAR-CCM+ cooling analysis Conjugate heat transfer using liquid and/or gaseous coolants Import of thermal loading from EMAG tool 2D or 3D Loss distribution data is mapped onto STAR-CCM+ grid PC-FEA

Thermal Modeling (7) Links with Motor-CAD (Motor-Design, UK) 1. Creation of the Motor-CAD model based on geometry parameters and winding scheme or import from SPEED 2. FE-analysis and fitting of the analytical model 3. Run thermal calculations in Motor-CAD to check the model 4. Preparation of the geometry in STAR-CCM+ by running a Java script FE-grid SPEED 7. Solving and post processing in STAR-CCM+ FV-grid STAR-CCM+ Data transfer 6. Mapping process for rotor and stator heat losses is carried out separately and automatically with transfer of the values from neighbor grid node in SPEED to STAR- CCM+ 5. Transfer of the heat loss distribution from the FE-analysis to STAR-CCM+ via the sbd-file

STAR-CCM+ EMAG solver Applications often allow 2D reduction Available in STAR-CCM+ 8.06 Validated with PC-FEA

Achieving Coupled Models Electromagnetic Simulation Thermal Simulation Coupled Problem Mapping of distributed losses Heterogeneous losses Map between grids Iterations Iterations EMAG Solution Thermal Solution EMAG Solution Thermal Solution EMAG Solution Iterations Iterations Solution Progress

Electric Machine Simulation Technology Steve Hartridge Director, Electric & Hybrid Vehicles

Besides CD-adapco internal material this presentation is based on the following publications: Bauarten von elektrischen Antrieben und deren Kühlung, Verluste, Vor- und Nachteile, Univ.-Prof. Dr. phil. Dr. techn. habil. Harald Neudorfer, Traktionssysteme Austria GmbH, Kolloquium Elektrische Antriebe in der Landtechnik, Wieselburg, 26. Juni 2013 Austria Keith R Pullen, Professor of Energy Systems, Brunthan Yoheswaren, PhD Researcher Energy and Transport Research Centre School of Engineering and Mathematical Sciences, Cooling of Electrical Machines, EMTM 13, 12 September 2013 Nottingham University UK Conjugate Heat Transfer Analysis of Integrated Brushless Generators for More Electric Engines Marco Tosetti, Paolo Maggiore, Andrea Cavagnino, Senior Member IEEE, and Silvio Vaschetto, Member IEEE, Politecnico di Torino, Corso Duca degli Abruzzi 24, 10129 Torino Italy Electric Motor Thermal Management, U.S. Department of Energy, Kevin Pennion, May 11, 2011 US End Winding Cooling in Electrical Machines, Christopher Micallef, BEng (Hons), PhD Thesis submitted to the University of Nottingham, September 2006 UK Script Large Generators & High Power Drives, Prof. habil. Dr.Ing. A. Binder, A., TU Darmstadt, Inst. f. Elektrische Energiewandlung, 2008 Germany