Camry PDF

U.S.
Department of Energy
FreedomCAR and Vehicle Technologies, EE-2G
1000 Independence Avenue, S.W.
Washington, D.C. 20585-0121
FY 2008
EVALUATION OF THE 2007 TOYOTA CAMRY

HYBRID SYNERGY DRIVE SYSTEM
Prepared by:
Oak Ridge National Laboratory
Mitch Olszewski, Program Manager
Submitted to:
Energy Efficiency and Renewable Energy
FreedomCAR and Vehicle Technologies
Vehicle Systems Team
Susan A. Rogers, Technology Development Manager
April 2008
ORNL/TM-2007/190
Engineering Science and Technology Division
EVALUATION OF THE 2007 TOYOTA

CAMRY HYBRID SYNERGY
DRIVE SYSTEM
T. A. Burress
C. L. Coomer
S. L. Campbell
L. E. Seiber
L. D. Marlino
R. H. Staunton
J. P. Cunningham
Revised: April 2008

Publication Date: January 2008
Prepared by the
OAK RIDGE NATIONAL LABORATORY
Oak Ridge, Tennessee 37831
managed by
UT-BATTELLE, LLC
for the
U.S. DEPARTMENT OF ENERGY
Under contract DE-AC05-00OR22725
DOCUMENT AVAILABILITY
Reports produced after January 1, 1996, are generally available free via the U.S. Department of
Energy (DOE) Information Bridge:
Web site: http://www.osti.gov/bridge
Reports produced before January 1, 1996, may be purchased by members of the public from the
following source:
National Technical Information Service

5285 Port Royal Road
Springfield, VA 22161
Telephone: 703-605-6000 (1-800-553-6847)
TDD: 703-487-4639
Fax: 703-605-6900
E-mail: info@ntis.fedworld.gov
Web site: http://www.ntis.gov/support/ordernowabout.htm
Reports are available to DOE employees, DOE contractors, Energy Technology Data Exchange
(ETDE) representatives, and International Nuclear Information System (INIS) representatives
from the following source:
Office of Scientific and Technical Information

P.O. Box 62
Oak Ridge, TN 37831
Telephone: 865-576-8401
Fax: 865-576-5728
E-mail: reports@osti.gov
Web site: http://www.osti.gov/contact.html
This report was prepared as an account of work sponsored by an agency of

the United States Government. Neither the United States government nor
any agency thereof, nor any of their employees, makes any warranty,
express or implied, or assumes any legal liability or responsibility for the
accuracy, completeness, or usefulness of any information, apparatus,
product, or process disclosed, or represents that its use would not infringe
privately owned rights. Reference herein to any specific commercial product,
process, or service by trade name, trademark, manufacturer, or otherwise,
does not necessarily constitute or imply its endorsement, recommendation,
or favoring by the United States Government or any agency thereof. The
views and opinions of authors expressed herein do not necessarily state or
reflect those of the United States Government or any agency thereof.
TABLE OF CONTENTS
Page
LIST OF FIGURES .....................................................................................................................................iv
LIST OF TABLES......................................................................................................................................vii
ACRONYMS AND ABBREVIATIONS ..................................................................................................viii
1. INTRODUCTION .................................................................................................................................... 1
2. HYBRID ELECTRIC DRIVE SYSTEM DESIGN AND PACKAGING................................................ 3

2.1 PCU.................................................................................................................................................. 4
2.1.1 Power Electronics Devices ................................................................................................ 13
2.1.2 Capacitor Test Results ....................................................................................................... 19
2.1.2.1 Static capacitor test results of the 2098 μF dc-link smoothing capacitor
module ................................................................................................................. 20
2.1.2.2 Dynamic capacitor test results from large dc-link capacitor module................... 25
2.1.2.3 Static capacitor test results of single sub-cell capacitor of smoothing
capacitor module.................................................................................................. 28
2.1.2.4 Inspection into positive reactance measurement of large dc-link
capacitor module.................................................................................................. 33
2.1.2.5 Static capacitor test results of battery level filter capacitor evaluations .............. 35
2.1.2.6 Static capacitor test results of small dc-link capacitor evaluations...................... 36
2.2 TRANSAXLE................................................................................................................................. 37
2.2.1 Thermal Management and Lubrication System.................................................................. 39
2.2.2 Transaxle Gear Train .......................................................................................................... 41
2.2.3 Motor Mass and Volumetric Assessments.......................................................................... 42
2.2.4 PM Characteristic Assessments.......................................................................................... 50
3. BENCHMARKING TESTS OF CAMRY PMSM AND INVERTER................................................... 55

3.1 SUBSYSTEM PREPARATION FOR TESTING ......................................................................... 55
3.1.1 Hardware Preparation ......................................................................................................... 55
3.1.2 Instrumentation and DAQ Development ............................................................................ 57
3.1.3 Controller and Interface Development................................................................................ 60
3.2 BACK-EMF TEST RESULTS ...................................................................................................... 61
3.3 ROTATIONAL LOSS ASSESSMENTS ...................................................................................... 62
3.4 LOCKED ROTOR TEST .............................................................................................................. 64
3.5 PERFORMANCE AND EFFICIENCY EVALUATIONS ........................................................... 65
3.6 CONTINOUS LOAD DURATION TESTS .................................................................................. 72
4. SUMMARY AND CONCLUSIONS ..................................................................................................... 79
REFERENCES ........................................................................................................................................... 81
APPENDIX A: 2004 PRIUS POWER CONVERTER UNIT PACKAGING ASSESSMENTS ............... 82

2004 Prius Motor Inverter Mass and Volume Assessment........................................................... 82
2004 Prius Bi-Directional Boost Converter Mass and Volume Assessment ................................ 83
ii
TABLE OF CONTENTS (cont'd)
Page
APPENDIX B: PRIUS AND CAMRY CAPACITOR TEST RESULTS .................................................. 85
APPENDIX C: 2004 PRIUS TRANSAXLE PACKAGING ASSESSMENTS......................................... 90

2004 Prius Motor Mass and Volume Assessment ........................................................................ 90
DISTRIBUTION......................................................................................................................................... 92
iii
LIST OF FIGURES
Figure Page
1.1 Flow diagram of baseline benchmarking approach........................................................................ 2

2.1 Subsystems of the Camry inverter and converter assembly........................................................... 4
2.2 Circuit diagram of PCU.................................................................................................................. 5
2.3 Boost converter compartment with lid and IPM driver board removed......................................... 6
2.4 Inverter compartment with dc-link capacitor and casing removed. ............................................... 6
2.5 General assembly and packaging of the Camry inverter and converter. ........................................ 7
2.6 Coolant flow paths in mating converter and inverter cold plates. .................................................. 8
2.7 Converter assembly with subsections shown consistent with volume measurements. ................ 10
2.8 Non-optimized packaging used in the 2004 Prius converter........................................................ 10
2.9 Comparison of all capacitors located within Camry PCU............................................................ 11
2.10 Cutaway of dc-link capacitor removed from shell and x-ray of module...................................... 11
2.11 Battery level capacitor and x-ray of module ................................................................................ 12
2.12 Small dc-link capacitor: overhead view and x-ray of module – side view................................... 12
2.13 Side housing removed from inverter section of PCU................................................................... 13
2.14 Comparison of Camry inverter IPM and Prius inverter IPM ....................................................... 14
2.15 Camry IPM with dc-bus bars removed......................................................................................... 14
2.16 Dimensions (in mm) of Camry inverter IPM and Prius inverter IPM.......................................... 15
2.17 Close-up view of Camry inverter IGBT and Prius inverter IGBT. .............................................. 16
2.18 Cross-sectional view of Camry IGBT and Prius IGBT................................................................ 16
2.19 Close-up view of Camry inverter diode and Prius inverter diode ................................................ 17
2.20 Comparison of Camry converter IPM and Prius converter IPM. ................................................. 17
2.21 Close-up comparison of Camry converter IPM and Prius converter IPM. .................................. 18
2.22 Camry capacitor module equivalent capacitance vs. frequency................................................... 20
2.23 Prius capacitor module equivalent capacitance vs. frequency. .................................................... 21
2.24 Camry capacitor module equivalent capacitance variation with ambient temperature. ............... 21
2.25 Camry capacitor module ESR vs. frequency................................................................................ 22
2.26 Prius capacitor module ESR vs. frequency. ................................................................................. 22
2.27 Camry capacitor module ESR vs. temperature. ........................................................................... 23
2.28 Prius capacitor module ESR vs. temperature. .............................................................................. 23
2.29 Camry capacitor module DF frequency response. ....................................................................... 24
2.30 Prius capacitor module DF frequency response. .......................................................................... 24
2.31 Camry capacitor module DF temperature response. .................................................................... 25
2.32 Prius capacitor module DF temperature response. ....................................................................... 25
2.33 Ripple current test schematic. ...................................................................................................... 26
2.34 TCs on the Camry capacitor module............................................................................................ 26
2.35 Camry capacitor module steady-state temperature response vs. ripple current. .......................... 27
2.36 Prius capacitor module steady-state temperature response vs. ripple current at 4 kHz................ 27
2.37 Camry single 86 μF capacitor, capacitance vs. frequency. .......................................................... 28
2.38 Camry single 86 μF capacitor, ESR vs. frequency....................................................................... 29
2.39 Camry single 86 μF capacitor, DF vs. frequency......................................................................... 29
2.40 Camry single 86 μF capacitor, capacitance, ESR, and DF vs. temperature at 1 kHz................... 30
2.41 Camry single 86 μF capacitor, capacitance, ESR, and DF vs. temperature at 5 kHz................... 30
2.42 Camry single 86 μF capacitor, capacitance, ESR, and DF vs. temperature at 15 kHz................. 31
2.43 Camry single 86 μF capacitor, capacitance, ESR, and DF vs. temperature at 35 kHz................. 31
2.44 Camry single 86 μF capacitor, capacitance, ESR, and DF vs. frequency at 20°C. ...................... 32
iv
LIST OF FIGURES (cont'd)
Figure Page
2.45 Camry single 86 μF capacitor, capacitance, ESR, and DF vs. frequency at 140°C. .................... 32
2.46 Waveforms of the Camry and Prius capacitor module at 5 kHz. ................................................. 33
2.47 Camry and Prius capacitor module waveforms at 16 kHz. .......................................................... 33
2.48 Waveforms of the Camry and Prius capacitor module at 20 kHz. ............................................... 34
2.49 Camry capacitor module voltage and current response to frequency........................................... 35
2.50 Prius capacitor module voltage and current response to frequency. ............................................ 35
2.51 Camry battery level 378 μF capacitor, capacitance, ESR, and DF vs. frequency. ....................... 36
2.52 Prius battery level 282 μF capacitor, capacitance, ESR, and DF vs. frequency........................... 36
2.53 Camry dc-link 0.9 μF capacitor, capacitance, ESR, and DF vs. frequency.................................. 37
2.54 Prius dc-link 0.1 μF capacitor, capacitance, ESR, and DF vs. frequency. ................................... 37
2.55 Hybrid Camry transaxle isometric view and rear view. ............................................................... 38
2.56 Camry generator front view with ICE spline removed and motor rear view ............................... 38
2.57 Camry transaxle gear section open sandwich view...................................................................... 39
2.58 Trochoid oil pump view from outside rear and from inside motor section.................................. 40
2.59 Oil slinging paths formed by gear train........................................................................................ 40
2.60 2004 Prius gear train and 2007 Camry gear train......................................................................... 41
2.61 Camry speed reduction planetary and power split planetary........................................................ 42
2.62 Speed, torque, and power relationship. ........................................................................................ 43
2.63 Camry motor assembly after separation from generator assembly. ............................................. 44
2.64 Coolant channel at bottom of motor casing.................................................................................. 44
2.65 Camry motor stator....................................................................................................................... 45
2.66 Camry and Prius motor stator laminations. .................................................................................. 46
2.67 Camry rotor assembled and partially disassembled ..................................................................... 46
2.68 Camry and Prius motor rotor lamination/PM design. .................................................................. 48
2.69 Camry and Prius motor stator lamination/slot design. ................................................................. 48
2.70 Camry generator volume assessment. .......................................................................................... 49
2.71 Camry and Prius generator stator comparison. ............................................................................ 49
2.72 Camry and Prius generator rotor comparison............................................................................... 49
2.73 Walker Scientific hysteresis graph. .............................................................................................. 51
2.74 Hysteresis graph for Camry motor magnet at 110ºC.................................................................... 51
2.75 Hysteresis graph for Camry motor magnet at 198ºC.................................................................... 52
2.76 Intrinsic hysteresis graphs for Camry motor magnet at various temperatures. ............................ 52
2.77 Intrinsic hysteresis graphs for Prius motor magnet at various temperatures................................ 53
2.78 Remanent flux density for Camry and Prius motor magnet. ........................................................ 54
2.79 Coercivity of Camry and Prius motor magnet.............................................................................. 54
3.1 First test setup approach with welded Camry planetary gear....................................................... 56
3.2 Second test setup approach with modified motor rotor shaft. ...................................................... 56
3.3 Upgraded dynamometer test cell.................................................................................................. 57
3.4 TCs in the inverter power electronics heat spreaders................................................................... 57
3.5 Thermistor locations on PMSM and transaxle. ............................................................................ 58
3.6 Viewport and window locations implemented on transaxle......................................................... 59
3.7 Opal-RT controller and interface system. .................................................................................... 61
3.8 2007 Camry motor back-emf test results. .................................................................................... 61
3.9 2004 Prius motor back-emf test results. ....................................................................................... 62
3.10 2007 Camry gear train rotational torque. ..................................................................................... 62
3.11 2007 Camry gear train rotational losses. ...................................................................................... 63
v
LIST OF FIGURES (cont'd)
Figure Page
3.12 2007 Camry motor rotational losses............................................................................................. 63

3.13 2007 Camry gear losses vs. oil temperature at various ring gear speeds. .................................... 64
3.14 2007 Camry locked rotor torque vs. position for various dc currents. ......................................... 64
3.15 2007 Camry and Prius peak locked rotor torque vs. dc current. .................................................. 65
3.16 2007 Camry and Prius peak locked rotor torque vs. dc current. .................................................. 65
3.17 2007 Camry motor efficiency contours........................................................................................ 66
3.18 2007 Camry motor efficiency contours scaled by 2.47................................................................ 67
3.19 2004 Prius motor efficiency contours. ......................................................................................... 67
3.20 2007 Camry inverter efficiency contours scaled by 2.47. ............................................................ 68
3.21 2004 Prius inverter efficiency contours........................................................................................ 69
3.22 2007 Camry motor/inverter efficiency contours. ......................................................................... 69
3.23 2004 Prius motor/inverter efficiency contours............................................................................. 70
3.24 2007 Camry inverter efficiency vs. switching frequency for various dc-link voltages................ 70
3.25 2007 Camry boost converter efficiency vs. input current for various dc-link voltages. .............. 71
3.26 Prius boost converter efficiency vs. input current for various dc-link voltages. .......................... 71
3.27 Camry 50 kW continuous duration test at 5000 rpm with 50ºC coolant. ..................................... 72
3.30 Camry 33.5 kW continuous duration test at 3000 rpm with 65ºC coolant. .................................. 74
3.31 Camry 25 kW continuous duration vs. speed with 65ºC coolant. ................................................ 75
3.32 Camry 33.5 kW continuous duration vs. speed with 35ºC coolant. ............................................. 75
3.35 Camry continuous duration at 3000 rpm with various power levels and coolant temperatures... 77
A.1 Volume of entire Prius PCU assembly......................................................................................... 82
A.2 Empty inverter/converter housing showing cold plate surfaces................................................... 83
A.3 General circuit schematic of the Prius inverter. ........................................................................... 83
B.1 Prius single 141 μF capacitor from 1,130 μF module, capacitance vs. frequency. ...................... 85
B.2 Prius single 141 μF capacitor from 1,130 μF module, ESR vs. frequency. ................................. 85
B.3 Prius single 141 μF capacitor from 1,130 μF module, DF vs. frequency..................................... 86
B.4 Camry single 86 μF capacitor, capacitance, ESR, and DF vs. temperature at 10 kHz................. 86
B.7 Camry single 86 μF capacitor, capacitance, ESR, and DF vs. frequency at -40ºC. ..................... 88
B.8 Prius single 141 μF capacitor, capacitance, ESR, and DF vs. temperature at 1 kHz. .................. 88
B.9 Prius single 141 μF capacitor, capacitance, ESR, and DF vs. temperature at 5 kHz. .................. 89
B.10 Prius single 141 μF capacitor, capacitance, ESR, and DF vs. temperature at 15 kHz. ................ 89
C.1 2004 Prius motor rotor and stator................................................................................................. 90
C.2 Casing of the 2004 Prius PMSM with dimensions and volume calculations............................... 91
vi
LIST OF TABLES
Table Page
2.1 Comparison of hybrid Camry and the Prius design features and published specifications............ 3
2.2 Comparison of hybrid Camry and the Prius PCU component specifications................................. 5
2.3 Mass and volume measurements for the hybrid Camry inverter and converter. ............................ 9
2.4 Specific power and power density estimates for inverter/converter............................................... 9
2.5 Summary of power electronic device packaging.......................................................................... 19
2.6 Camry transaxle gear information................................................................................................ 42
2.7 Specific power and power density estimates for the PMSM........................................................ 44
2.8 2007 Camry/2004 Prius motor design comparison highlights. .................................................... 47
2.9 2007 Camry/2004 Prius generator design comparison highlights................................................ 50
4.1 2007 Camry/2004 Prius design comparison highlights................................................................ 80
vii
ACRONYMS AND ABBREVIATIONS
A/C air conditioning ID inner diameter

ac alternating current IGBT insulated-gate bipolar transistor
ANL Argonne National Laboratory IPM intelligent power module
AWG America wire gauge LV low voltage
CAN controller area network MG motor-generator
CT current transformer NdFeB neodymium iron boron
DAQ data acquisition system Nm Newton meter
dc direct current NTRC National Transportation Research Center
DF dissipation factor OD outer diameter
DOE U.S. Department of Energy ORNL Oak Ridge National Laboratory
ECU electronic control unit PC personal computer
ECVT electronically controlled continuously PCU power control unit
variable transmission PEEMRC Power Electronics and Electric
emf electromotive force (measured in volts) Machinery Research Center
EMI electromagnetic interference PI proportional-integral
ESR equivalent series resistance PM permanent magnet
EV electric vehicle PMSM permanent magnet synchronous motor
FCVT FreedomCAR and Vehicle Technologies rms root mean square
HEV hybrid electric vehicle rpm revolutions per minute
HV high voltage Si silicon
IC integrated circuit TC thermocouple
ICE internal combustion engine THS Toyota Hybrid System
viii
1. INTRODUCTION
The U.S. Department of Energy (DOE) and American automotive manufacturers General Motors, Ford,
and DaimlerChrysler began a five-year, cost-shared partnership in 1993. Currently, hybrid electric vehicle
(HEV) research and development is conducted by DOE through its FreedomCAR and Vehicle
Technologies (FCVT) program. The mission of the FCVT program is to develop more energy efficient
and environmentally friendly highway transportation technologies. Program activities include research,
development, demonstration, testing, technology validation, and technology transfer. These activities are
aimed at developing technologies that can be domestically produced in a clean and cost-competitive
manner.
Under the FCVT program, support is provided through a three-phase approach [1] which is intended to:
• Identify overall propulsion and vehicle-related needs by analyzing programmatic goals and
reviewing industry’s recommendations and requirements, then develop the appropriate technical
targets for systems, subsystems, and component research and development activities;
• Develop and validate individual subsystems and components, including electric motors, emission
control devices, battery systems, power electronics, accessories, and devices to reduce parasitic
losses; and
• Determine how well the components and subassemblies work together in a vehicle environment or
as a complete propulsion system and whether the efficiency and performance targets at the vehicle
level have been achieved.
The research performed in this area will help remove technical and cost barriers to enable technology for
use in such advanced vehicles as hybrid electric, plug-in hybrid electric, electric, and fuel-cell-powered
vehicles.
The Oak Ridge National Laboratory’s (ORNL) Power Electronics and Electric Machinery Research
Center (PEEMRC) performed benchmark assessments and subsystem testing of the 2004 Toyota Prius in
2004–2005. This work has been fully reported in two reports [2,3]. In 2006, the PEEMRC performed
similar work on the hybrid version of the 2005 Honda Accord [4]. This report presents the results of
benchmark testing of the 2007 Toyota Camry electric drive subsystems.
Subsystem-level testing was conducted by the PEEMRC at ORNL, which is a broad-based research
center for power electronics and electric machinery (motor) development. Located in the national user
facility known as the National Transportation Research Center (NTRC), the PEEMRC has more than
9000 square feet of space for developing and building the next generation prototypes of power electronics
and electric machine technologies.
The 2007 Toyota Camry hybrid is equipped with the Hybrid Synergy Drive system which includes an
internal combustion engine (ICE), an electric motor, and a generator. These components are integrated
with a power splitting planetary gear which provides various power flow configurations for different
modes of operation. The electric motor assists the ICE in providing mechanical drive power for the
vehicle and also acts as a generator to recharge the battery during regenerative braking. According to
published specifications, Toyota claims the motor can deliver a peak power output of 105 kW (143 hp) at
4500 revolutions per minute (rpm). Toyota has not published a rated torque (or power) versus speed graph
for the Camry electric motor and no specifications have been published for the generator. The hybrid
version of the Camry features a 2.4-liter 4-cylinder, 16-valve engine, which has a specified peak power
output of 110 kW (147 hp) at 6000 rpm [5].
1
The Camry electric drive system has several features in common with the Toyota Prius system as they
both use the Toyota Hybrid System (THS) II. Both make use of the electric motor to provide all of the
traction power during acceleration from a stop. This continues until the engine is needed for any of
several reasons (more torque, low battery state of charge, high battery temperature, etc.) Both vehicles
have a dedicated generator, which is used to route power from the ICE to the electric motor. Both use a
clutch-less, electronically controlled continuously variable transmission (ECVT) which uses a planetary
gear to manage power flow from the ICE and motor to the differential gear unit. The planetary gear
configuration of the ECVT provides the opportunity to operate the ICE throughout a more efficient
torque-speed range, depending on the torques and speeds of the generator and electric motor.
Unlike the Toyota Camry and Prius, which are both considered to be full hybrids, the Honda Accord
hybrid with its smaller motor is considered to be a “mild hybrid” since the motor never propels the
vehicle without engine power. In operation, the 2005 Accord hybrid exhibits enhanced performance
compared to the conventionally powered Accord, but little or no gain in fuel economy.
In order to fully benchmark the HEV technology used in the 2007 Camry, a design characterization study
was conducted to evaluate the vehicle’s hybrid electric drive subsystem. A flow diagram which portrays
the baseline benchmarking approach is shown in Fig. 1.1. The characterization study included (1) a design
review, (2) a packaging and fabrication assessment, (3) bench-top electrical tests, (4) back-electromotive
force (emf) and locked rotor tests, (5) performance and efficiency tests on the motor and inverter,
(6) efficiency tests on the bi-directional boost converter, and (7) continuous duration tests of the drive
motor and inverter. The design and packaging review includes comparisons to the 2004 Toyota Prius,
which was evaluated during previous years. Results from the performance and efficiency tests were used
to map electrical and thermal data for motor/inverter operation over the full range of speeds and shaft
loads that the motor is capable of producing. The results of the design and packaging assessments are
presented in Section 2 and the performance and efficiency tests results are presented in Section 3.
Fig. 1.1. Flow diagram of baseline benchmarking approach.
2
2. HYBRID ELECTRIC DRIVE SYSTEM DESIGN AND PACKAGING
The hybrid Camry includes significant advancements in HEV technology since the electric drive system
features higher peak power and more extensive continuous operation capabilities, yet the size and weight
of the system has decreased when compared to that of the Prius. This is primarily due to the care taken in
packaging the inverter/converter system and the design of a higher-speed, higher voltage permanent
magnet synchronous motor (PMSM). The Camry HEV subsystem design and packaging will be assessed
and summarized in this section.
Table 2.1 presents an overview of several general HEV design features and published specifications and
indicates similarities and differences between the Camry and Prius vehicles. The primary drive PMSM is
referred to as MG2 and the generator is referred to as MG1. Although both PMSMs can function as a
motor or a generator, MG2 is commonly referred to as “motor” and MG1 is commonly referred to as
“generator.” Both MG1 and MG2 are powered by separate three-phase inverters which share the same
direct current (dc)-link. The battery voltage is boosted by a bi-directional boost converter which feeds the
dc-link. This configuration facilitates the use of multiple power flow modes wherein the ICE, MG1,
and/or MG2 may supply energy to the battery pack, the battery pack may supply MG1/MG2 alone, or the
ICE and battery pack can simultaneously power the vehicle.
Table 2.1. Comparison of hybrid Camry and the Prius design features and published specifications
Design Feature Hybrid Camry 2004 Prius

Motor peak power rating 105 kW @ 4500 rpm (disputed to be 70kW) 50 kW @ 1200–1540 rpm
270 Newton meter (Nm)
Motor peak torque rating 400 Nm
(667 after speed reduction gear)
Top rotational speed 14,000 rpm (5,670 rpm after speed reduction) 6,000 rpm
Yes (although the motor also serves as a
Separate generator used? Same as Camry
generator during regenerative braking)
Generator specifications Not published 33 kW
Source of power to MG2 Battery and/or ICE via generator Same as Camry
Interior permanent magnets (PMs) with “V”
Interior PMs with “V”
configuration at each pole with a
PMSM rotor design configuration at each pole
reinforcement web in the laminations (middle
(no reinforcement)
of the V) to enable high-speed operation
Motor winding configuration Parallel Series
Number of rotor poles 8 Same as Camry
Boost converter output voltage 250–650 Vdc output 200–500 Vdc output
Boost converter power rating 30 kW 20 kW
Oil circulation with water/glycol heat
PMSM cooling Same as Camry
exchanger
Inverter/converter cooling Water/glycol loop Same as Camry
Planetary gears used for speed reduction and A single planetary gear
Hybrid transaxle
power split used for power split
Fan-cooled high-voltage (HV)
244.8 V, 6.5 Ah, 30 kW 201.6 V, 6.5 Ah, 20 kW
Ni-MH battery
Design and packaging assessments are presented in Subsections 2.1 and 2.2 beginning with the power
control unit (PCU) and concluding with the transaxle. Studies on the PCU include subjects such as
general design and packaging, power electronic devices, and multifaceted capacitor assessments.
Assessment topics associated with the transaxle are general design and packaging, PMSM design, and
magnet hysteresis measurements. These assessments provide useful feedback regarding technological
developments and other characteristics needed to determine packaging complexity and fabrication costs.
3
2.1 PCU
The general locations of the Camry inverters and the bi-directional boost converter assembly are shown in
Fig. 2.1. The figure makes evident the fact that the motor and generator inverters, dc-link capacitor, and
half of the cold plate comprise about 65% of the total volume. The buck/boost converter and half the cold
plate comprise the remaining 35%. The size and proportions of the PCU are similar to that of a
conventional 12 Vdc car battery.
Fig. 2.1. Subsystems of the Camry inverter and converter assembly.
Because the Ni-MH battery is rated at 30 kW, it is assumed that the boost converter will not operate at
power levels exceeding 30 kW. Similarly, it is assumed that the motor inverter has a peak power rating
that matches the measured power rating of the motor.
The overall circuit diagram of the PCU is shown in Fig. 2.2. A 250 Vdc battery supplies power to the
PCU, which is connected to the low voltage (LV) side of the boost converter. A 500 V, 378μF capacitor is
connected across the input with a 212 μH inductor between the battery and the boost converter power
electronics module. A small 53.8 kΩ resistor, 750 V, 0.9 μF filter capacitor, and 750 V, 2098μF
smoothing capacitor is located on the HV side of the boost converter. This HV bus serves as the dc link
for both the motor and generator inverter. Table 2.2 provides specification comparisons between the
components found in the Prius and Camry PCU. The boosted voltage ranges from 250–650 Vdc
depending on driving conditions such as desired acceleration and required regenerative braking and is
controlled accordingly by commands from the MG electronic control unit (ECU) and the THS ECU.
Detailed capacitor test results are provided in Section 2.1.3.
4
Fig. 2.2. Circuit diagram of PCU.
Table 2.2. Comparison of hybrid Camry and the Prius PCU component specifications
Design Feature Hybrid Camry 2004 Prius

Boost converter power rating 30 kW 20 kW
Battery voltage 244.8 V 201.6 V
Filter capacitor (LV side) 500 Vdc, 378 μF 600 Vdc, 282 μF
Inductor 212 μH at 1 kHz 373 μH at 1 kHz
Small resistor (HV side) 53.8 kΩ 64.3 kΩ
Small capacitor (HV side) 750 Vdc, 0.9 μF 750 Vdc, 0.1 μF
Smoothing capacitor 750 Vdc, 2098 μF 600 Vdc, 1,130 μF
Figures 2.3 and 2.4 show the converter and inverter compartments, respectively, prior to disassembly.
Figure 2.5 shows the general assembly and packaging of the major subcomponents of the Camry inverter
and converter. Not shown for the inverter is the dc-link capacitor and its casing which serves as a cover to
the inverter assembly. Because the inverter/converter assembly was designed to be integrated into an
existing vehicle (the non-hybrid Camry), there was a need to package the unit in a minimal volume,
which is evident in the figures provided in this section. The Camry MG ECU is located within the PCU
just above the inverter driver board. Communication between the Camry MG ECU and other ECUs is
accomplished though controller area network (CAN) digital data transmission. The Prius MG ECU is
located outside of the PCU with analog control signals being transmitted over a conductor length of
several feet. Together, the MG ECU and THS ECU monitor vehicle conditions and control the PCU
components to orchestrate proper system operation.
In the inverter compartment, the controller board includes power regulation electronics, two identical MG
microprocessors, a boost-converter microprocessor, two Tamagawa AU2802 integrated circuits (IC) for
MG speed/position detection, hardware for CAN communication, MG current measurement, safety
interlock devices, and temperature feedback. The driver board includes regulated isolation power
supplies, isolated driver electronics, and hardware to prevent faults and overlapping as well as voltage,
current, and temperature sensing circuitry for each insulated-gate bipolar transistor (IGBT). More
information regarding the power electronics modules for the inverters and converter is provided in
Section 2.1.2.
5
Fig. 2.3. Boost converter compartment with lid and intelligent power module (IPM) driver board removed.
Fig. 2.4. Inverter compartment (upside down) with dc-link capacitor and casing removed.
6
Fig. 2.5. General assembly and packaging of the Camry inverter and converter.
7
The converter compartment contains the IPM, IPM driver, power supply board, LV filter capacitor, large
inductor, bus bars, and an electromagnetic interference (EMI) shield for the circuit board. Additionally, a
small filter capacitor and resistor are located in the upper portion of the PCU and are connected to the HV
dc link. Connectors are also included for the inductor thermistor and safety interlock device feedback
signals.
The Prius has an inverter which supplies power to the air conditioning (A/C) compressor, and it is located
in the bottom section of the PCU, whereas the Camry A/C inverter is not housed within the PCU. A dc-dc
converter which converts the high battery voltage of about 200V to an accessory voltage level of about
14V is also located in the bottom section of the Prius PCU, whereas this component is located in the
battery module of the Camry.
Figure 2.6 shows the cooling plate coolant paths for the converter (top view) and inverter (bottom view)
and an enlargement of the converter coolant path configuration. The inverter photo was flipped to make
comparisons easier. The holes for the inlet and outlet coolant fittings are shown in the upper and lower
left-hand corners of the cold plates. The fins increase the surface area and heat transfer, and the ripples are
fabricated in the fin casting to create turbulence, which also increases heat transfer. Asymmetric features
in the channel walls and/or no-fin areas promote mixing. In general, the design features are likely the
result of detailed heat transfer/coolant flow modeling.
Fig. 2.6. Coolant flow paths in mating converter and inverter cold plates.
Table 2.3 provides physical measurements of the hybrid Camry inverter and boost converter. The
inverter/converter casing was unbolted and separated in order to obtain precise mass measurements that
are reflected in the total mass shown for each subsystem. The table also gives estimates for the mass and
volume of the motor inverter by itself (i.e., excluding the generator inverter).
For certain components, the dimensions and volumes are approximate since the geometries are sometimes
irregular. The entire aluminum casing unbolts and separates midway through the thickness of the cold
plate and the volumes are divided up between the inverter and converter accordingly.
The mass of the large, potted inductor used in the buck/boost converter, although an integral part of the
converter casing, is estimated to be 3.0 kg. This estimate was derived by subtracting the estimated mass of
the converter aluminum (2.0 kg) from the total measured mass of the two (5.02 kg).
8
Table 2.3. Mass and volume measurements for the hybrid Camry inverter and converter
Item Mass (kg) Volume (L) Dimensions (cm)

17.86
(17.42 without Summation of inverter and
Inverter/converter as received from OEM. 11.7
mounting converter sections (below)
brackets)
Inverters (MG1 and MG2)
DC-link capacitor mold in an aluminum case. 3.57 2.6 18.5 × 27 × 5.3
Side housing assembly containing the three-
phase bus connectors and current transformers 1.2 0.98 ~4.5 × 23 × 9.5
(CTs).
Inverter electronics and main casing to mid-
5 4.3 18.5 × 27 × 10.5
point of cooling plate.
Connector and housing at end of casing. (incl. above) 0.31 10.2 × 3.8 × ~8
Complete inverters 9.8 8.19 Combine the above
Inverter for motor only
DC-link capacitor mold in an aluminum case. 3.57 2.6 18.5 × 27 × 5.3
Side housing assembly containing the three-
0.69 0.56 ~4.5 × 12 × 9.5
phase bus connectors and CTs.
Inverter electronics and main casing to mid-
~3.3 ~2.8 11.1 × 27 × 10.5
point of cooling plate.
Complete motor inverter ~7.5 ~6 Combine the above
Converter
Boost converter inductor. ~3.5 1.14 8.8 × 17.5 × 7.4
Converter filter capacitor. 0.57 0.84 6.3 × 18.1 × 7.4
Electronics compartment and casing. ~ 0.9 + 1.7 1.51 11.3 × 18.1 × 7.4
Complete converter ~6.6 3.5 17.5 × 26.9 × 7.4
Based on the results shown in Table 2.3, Table 2.4 provides the peak power density and peak specific
power of both the motor inverter and the bi-directional boost converter. Comparisons to corresponding
estimates for the Prius are also shown. The Camry specifications in Table 2.4 are very impressive as the
power density of the motor inverter and bi-directional boost converter has more than doubled when
compared to that of the Prius. The Camry motor inverter specific power is also significantly larger, yet the
boost converter specific power is only slightly higher as the mass increase closely followed the rated
power increase.
Table 2.4. Specific power and power density estimates for inverter/converter
Parameter Camry Prius

Motor inverter peak specific power
70/~7.5 = ~9.3 50/8.8 = 5.7
(without converter), kW/kg.
Motor inverter peak power density
70/~6 = ~11.7 50/8.7 = 5.7
(without converter), kW/L.
Bi-directional boost converter specific
30/~6.6 = ~4.5 20/4.8 = 4.2
power, kW/kg.
Bi-directional boost converter power
30/3.5 = 8.6 20/5.1 = 3.91
density, kW/L.
1
This low converter power density is largely the result of the non-optimal packaging of the
converter filter capacitor in the Prius inverter/converter housing.
9
It is interesting to note that the converter mass increased by 38% and the volume decreased by 31% in
going from the 2004 Prius to the 2007 Camry PCU design. A mass increase is likely unavoidable given
that the power increased from ~20 kW to ~30 kW. Regarding the decrease of volume, the Camry
converter has a small amount of associated null space as is evident in Fig. 2.7 while the Prius converter
design (Fig. 2.8) integrates a tall inductor next to a short IPM resulting in wasted space above the IPM
and additional wasted space exists around both components. Not shown in Fig. 2.8 is the volume of the
converter portion of the monolithic capacitor module and surrounding space. Although the potential exists
for easily reducing the volume of the Prius converter, the volume calculation is based on the actual
dimensions of the casing. However, the component specifications are provided separately to allow for
various approaches of calculating specific power and power density. See Appendix A for details on the
measurement of the Prius PCU and motor inverter mass and volumetric assessment details.
Fig. 2.7. Converter assembly with subsections shown consistent with volume measurements.
Fig. 2.8. Non-optimized packaging used in the 2004 Prius converter.
All capacitors located within the Camry PCU are shown in Fig. 2.9 for a comparison of sizes. It is
important to note that that the Prius capacitor module contains both capacitors which are connected to the
HV and LV side of the bi-directional boost converter as well as a small filter capacitor, whereas the
Camry capacitors are housed in separate modules. Figure 2.10 shows the Camry dc-link capacitor and a
negative image of an x-ray of the capacitor. The x-ray shows that there are two rows of 12 discrete sub-
modules inside the entire capacitor module, with each capacitor having a capacitance of 87 µF. The
equivalent Prius capacitor has 8–142 µF capacitors in parallel to provide a total capacitance of 1,130 µF.
Figure 2.11 shows the 378 µF Camry filter capacitor and an x-ray of the capacitor which is across the
battery input of the PCU. As suggested by the x-ray image, the 378 µF capacitance is attained with two
189 µF capacitors in parallel. Similarly, the small 0.9 µF capacitor is made up of two 0.45 µF capacitors,
10
as shown in Fig. 2.12. Various characteristics of these capacitors were measured and are presented in
Section 2.1.3.
Fig. 2.9. Comparison of all capacitors located within Camry PCU.
Fig. 2.10. Cutaway of dc-link capacitor removed from shell (left) and x-ray of module (right).
11
Fig. 2.11. Battery level capacitor (upper) and x-ray of module (lower).
Fig. 2.12. Small dc-link capacitor: overhead view (left) and x-ray of module – side view (lower).
Figure 2.13 shows a view of the inverter/converter assembly upside down with the three-phase motor and
generator bus connector assembly unbolted and detached. It can be seen that CTs are used on two of the
three bus connectors. The red rubber seal inserts into the inverter housing in order to facilitate a sealed
union. In general, the inverter and converter housings are very well sealed and are water tight to prevent
contamination and corrosion from environmental effects.
12
Fig. 2.13. Side housing removed from inverter section of PCU (shown upside down).
2.1.1 Power Electronics Devices
The general packaging features of the Camry inverter and converter power electronic devices were
assessed and compared to the Prius designs. Figure 2.14 shows a comparison between the Camry and the
Prius IPMs which were photographed while sitting adjacent to each other. All power electronic devices
shown are immersed in a high-temperature, protective silicone gel. Although the power capability of the
IPM increases from the Prius to the Camry, the size of the IPM noticeably decreases. The general layout
of the IPM design is similar, wherein the shared dc bus, indicated by the red line, runs between the motor
and generator inverters. However, a closer inspection yields several differences between the two. Bus bars
are used to interconnect the dc link and phase outputs to the power electronics devices of the Prius IPM,
yet the Camry IPM does not use a conventional bus bar architecture to accomplish this task. Instead, a
more interweaved layout is used which enhances the area utilization of the IPM. The Camry IPM layout
allows for the space below the dc-bus bars to be used as extra surface area for power electronics devices,
as shown in Fig. 2.15. It can also be noticed that there is enough space for the Camry generator inverter to
contain twice the number of IGBTs than what is currently present. Toyota has made an effort to
standardize the size of the PCU and this space is likely utilized in HEVs with larger power trains, such as
the Toyota Highlander HEV or the Lexus GS450h. The maximum power supplied to the motor inverter
must be less than (due to losses) the sum of the battery power and the generator power. It is likely that the
full capability of the motor inverter is not fully utilized in the vehicle, as the generator inverter can supply
only 1/3 of the power capability to the motor inverter and any additional power must be supplied by the
battery which has a power rating of 30 kW, and the combined peak power is approximately 70 kW.
13
Fig. 2.14. Comparison of Camry inverter IPM (left) and Prius inverter IPM (right).
Fig. 2.15. Camry IPM with dc-bus bars removed.
Figure 2.16 provides a more detailed representation of the design discrepancies as well as a dimensional
comparison of the IGBTs and diodes of the Camry and Prius inverter power electronics. Since the
emitters and collectors of the Camry IPM are not connected to bus bars as they are in the Prius, additional
room is available for situating the diodes in an orientation which is more efficient in terms of surface area
14
utilization. In the Prius inverter image shown in Fig. 2.16, the IGBT emitter and diode anode are tied to a
common metal plate, which is connected to a bus bar that is partially visible on the left edge of the image.
The IGBT collector and diode cathode are connected to a bar which is just beyond the right edge of the
image. Conversely, the Camry dc link and three-phase outputs connect directly to the emitter-anode
common plate and the collector-cathode surface, thereby eliminating the use of a bus bar infrastructure.
Fig. 2.16. Dimensions (in mm) of Camry inverter IPM (left) and Prius inverter IPM (right).
As indicated by the labeled dimensions, the Camry inverter IGBT size has decreased from the Prius
inverter IGBT size in both length and width, whereas the size of the diode increased significantly.
Although the Camry IGBT chip size has decreased when compared to that of the Prius IGBT, the power
capability has increased slightly. This is primarily due to the use of a trench gate structure as opposed to
the planar gate structure used in the Prius. The planar gate structure includes areas in the drift region
below the emitter “n+” and “p” regions which do not pass as much current as the drift region areas below
the gate. This inefficiency is due to the shape of the inversion layer which is formed when using a planar
gate structure. An IGBT with a trench gate structure forms a more uniform inversion layer during on-state
operation, and therefore higher current densities are created in the drift region. Additionally, the voltage
rating of the Camry IGBT devices has increased, facilitating higher power capability and higher power
density of the generator, motor, and the respective inverters.
A detailed comparison of the Camry and Prius IGBTs is shown in Fig. 2.17. The Camry IGBT contains
two wire bonds for each emitter strip, whereas the Prius only has one wire bond for each emitter strip.
Both IGBTs include leads for semiconductor junction temperature measurement feedback, gate voltage
control, current measurement, and emitter voltage feedback. These signals are used to control the device
and prevent fault conditions from occurring. Cross-sectional views of the IGBTs are provided in
Fig. 2.18. Small metallic nodes are visible on the Camry IGBT, yet there is no evidence of these nodes on
the Prius IGBT. These nodes are likely associated with the trench gate structure, but are not directly
associated with the trench pitch as the metal nodes are separated by about 20 μm, which is much larger
than typical trench pitches.
15
Fig. 2.17. Close-up view of Camry inverter IGBT (left) and Prius inverter IGBT (right).
Fig. 2.18. Cross-sectional view of Camry IGBT (upper) and Prius IGBT (lower).
Diodes from the Camry and Prius inverters are shown in Fig. 2.19. The shape of the Camry inverter diode
is elongated providing well-suited orientation with the adjacent IGBT. Several traces are located around
the perimeter of the Camry and Prius diodes and IGBTs, which is possibly a preventative measure against
catastrophic arcing.
16
Fig. 2.19. Close-up view of Camry inverter diode (left) and Prius inverter diode (right).
The Camry and Prius boost converter IPMs are shown in Fig. 2.20. Although the power rating of the
boost converter has increased by 50% from the Prius, the area of the Camry boost converter IPM footprint
is only slightly larger than that of the Prius. Similar to the discrepancies between the inverters, smaller
bus architecture is used in the Camry boost converter IPM. There is a considerably larger amount of null
space within the Prius boost converter module. The Camry boost converter has four IGBTs and diodes in
parallel for the upper portion of the IPM and four IGBTs and diodes in parallel for the lower portion of
the IPM. The four upper IGBTs and four lower diodes are primarily used during regenerative braking and
recharging situations. The four lower IGBTs and four upper diodes are primarily used to boost the battery
voltage. The Prius boost converter IPM is similar in functionality, but has only two IGBTs and two diodes
in parallel for the upper and lower portions. The Prius boost converter IPM also only has three terminals
versus six on the Camry IPM, four of which are used to conduct power. The upper-left terminal does not
connect to the IPM, but serves as a support for the negative dc-link bus bar and capacitor connection.
The lower-left terminal serves as a support for the positive dc-link bus bar (which connects to the
inductor) and capacitor connection, and through a small wire it is connected to a pin which interfaces with
the IPM driver-board.
Fig. 2.20. Comparison of Camry converter IPM (left) and Prius converter IPM (right).
17
Figure 2.21 provides a close-up comparison of the power electronics devices of the Camry and Prius
boost converter IPMs. The fabrication process for the Camry boost converter IPM is more intricate due to
the increased number of devices that need to be interconnected, thereby increasing wire bond count and
driver and sensor feedback signals.
Fig. 2.21. Close-up comparison of Camry converter IPM (upper) and Prius converter IPM (lower).
Table 2.5 summarizes the device count, wire-to-silicon (Si) bond count on the emitter surface, and
amount of Si used in the Camry and Prius motor inverters and boost converters. The ratios of heat
spreader area to total power electronic device Si area for the Camry motor inverter and boost converter
are 10.1 and 4.3, respectively. The smaller ratio suggests a great success in size reduction (i.e., a small
heat sink relative to the amount of Si). For comparison, the corresponding ratios for the Prius are 21.7 and
6.8. Again, these improvements are the result of the challenges of implementing the hybrid system into an
already existing vehicle, whereas the Prius was strictly designed to be an HEV where volumetric
constraints were not as restrictive.
18
Table 2.5. Summary of power electronic device packaging
Motor Inverter Buck/Boost Converter
Parameter
IGBTs Diodes IGBTs Diodes
Camry
Number of devices 18 18 8 8
Emitter wire bond count per
27 12 18 17
device.
Area of total Si die per device,
120.3 71 177 104
mm2.
Summation of Si area, mm2. 2165 1278 1400 830
Heat spreader area/Si area. 34800/3443 = 10.1 9530/2230 = 4.3
Prius (for comparison)
Number of devices. 12 12 4 4
Emitter wire bond count per
20 5 21 28
device.
Area of total Si die per device,
131.9 40.7 228 119
mm2.
Summation of Si area, mm2. 1583 488 910 475
Heat spreader area/Si area. 45100/2071 = 21.7 9400/1385 = 6.8
2.1.2 Capacitor Test Results
As capacitor technologies mature, it is important to ascertain the capabilities and limitations of these new
technologies by subjecting them to standardized tests to evaluate their characteristics. Test results will
assist in benchmarking not only the capabilities and limitations of these technologies, but also to provide
a baseline for comparison. All the capacitors studied are of the metallized plastic film type and were
manufactured by Matsushita under the Panasonic brand name, except for the large Camry dc-link
capacitor which was manufactured by Nichicon. As noted in Section 2.1, the capacitor modules are made
up of several sub-capacitors. Tests were conducted upon the entire modules as well as the individual
capacitors which were destructively removed from the entire module.
Capacitor tests were conducted both statically and dynamically. The former is a capacitor parameter
assessment and the latter is a thermal characteristic assessment. The static test mode does not entail an
externally applied dc-bias voltage and ripple current, but uses only signals supplied by an LCR meter. The
capacitor parameters that are measured in the static mode are: (1) equivalent series resistance (ESR), (2)
dissipation factor (DF), and (3) capacitance value. The capacitor is placed in an environmental chamber
and the temperature is cycled through temperatures ranging from -40º to 140 ºC using steps of 20ºC.
These three parameters are measured at each temperature step.
The ESR is the real component of the equivalent impedance of the capacitor and corresponds directly
with total energy loss which is dissipated as heat during operation. The DF is the ratio of the ESR over the
capacitive reactance, which simplifies to ωRC and will be represented as the percentage of real power
associated with a particular reactive power component. Both of these parameters were measured during
static tests and the results were plotted over a wide frequency and temperature range.
In dynamic tests, a dc-bias voltage and ripple current is applied to the capacitor and the temperature of the
capacitor is monitored. When a capacitor is used on a dc bus, the maximum ripple current capability is an
important specification. The dynamic test determines the amount of heat generated by the capacitor as a
function of ripple current at various ambient temperatures.
19
2.1.2.1 Static capacitor test results of the 2098 μF dc-link smoothing capacitor module
The data acquisition system (DAQ) consists of a Dell Precision 380 Workstation running LABView 8.5.
In the static mode, the instruments used to log data are an Espec environmental chamber and an Agilent
4284 LCR meter. At the beginning of the static thermal cycling tests, after all temperature steps and test
frequencies have been entered, the program turns on the environmental chamber and commands a target
temperature to the chamber. After the target temperature has been reached and a pre-programmed delay
time has been met, the program sends a command to the LCR meter to output the first test frequency and
then begins to monitor the ESR from the capacitor under test. When that value is stable, the ESR is
recorded and the next parameter is measured and recorded. After the values for ESR, DF, and capacitance
are recorded at the first test frequency, the command is given to the LCR meter to output the next test
frequency and the monitoring and recording process is repeated. After all data are recorded at all test
frequencies, the next temperature step is commanded to the environmental chamber and the process is
repeated.
The Camry capacitor module was tested in the static mode from -40 to 140°C over a frequency range of
100 Hz to 30 kHz. ESR, DF, and capacitance was measured and recorded. An interesting fact is observed
in Fig. 2.22 as the capacitance approaches infinity beyond 16 kHz. The module can be represented as an
RLC circuit with very small resistance and inductance. At high frequencies, even a small parasitic
inductance component has a significant influence upon the entire equivalent reactance of the module.
There is a certain resonant frequency where the capacitive (negative reactance) effect and inductive
(positive reactance) effect are equal, and therefore the total reactance is zero and the imaginary
admittance or susceptance is infinite. Thus, the equivalent capacitance approaches infinity as the
frequency approaches resonant frequency. Just beyond the resonant frequency, the equivalent reactance is
inductive and increases from zero as frequency increases. Figure 2.23 shows that the Prius capacitor
module also has a similar effect near 15 kHz. As discussed shortly hereafter in Section 2.1.2.4, additional
tests were conducted to investigate this matter further.
8000
-40°C
7000 -20°C
0°C
6000
40°C
Capacitance (μF)
5000 60°C
80°C
4000 100°C
120°C
3000
140°C
2000
1000
0
0 5000 10000 15000 20000
Frequency (Hz)
Fig. 2.22. Camry capacitor module equivalent capacitance vs. frequency.
20
8000
-40°C
7000
-20°C
6000 0°C
Capacitance (μF)
40°C
5000
60°C
4000 80°C
3000 100°C
120°C
2000 140°C
1000
0
0 5000 10000 15000 20000
Frequency (Hz)
Fig. 2.23. Prius capacitor module equivalent capacitance vs. frequency.
As shown in Fig. 2.24, the response of the entire Camry module capacitance was very minimal over the
full temperature range. The largest change in capacitance over the temperature range is observed at
15 kHz, which is also the frequency at which the highest capacitance was measured. For low frequencies,
capacitance tended to increase with increasing temperature until peaking at a particular temperature and
then the capacitance decreased slightly as temperature increased above this temperature. This peak
capacitance temperature increased with increasing frequency. The Prius capacitor module showed similar
effect and slight variation of capacitance with temperature.
5000
4500
4000 600 Hz
3000 Hz
Capacitance (μF)
3500
5000 Hz
7000 Hz
3000
9000 Hz
2500 11000 Hz
13000 Hz
2000 15000 Hz
1500
1000
-50 0 50 100 150
Temperature (°C)
Fig. 2.24. Camry capacitor module equivalent capacitance variation with ambient temperature.
21
Figures 2.25 and 2.26 show the ESR response to frequency for the Camry and the Prius module,
respectively. Although the curve characteristics indicate that the influence of frequency upon the ESR of
both modules is reasonably different, the ESR measurement is similar, particularly at low frequencies.
Measurements indicate that ESR values generally increased with increasing temperature for both modules
at all frequencies.
3.5
-40°C
3 -20°C
0°C
2.5 40°C
60°C
ESR (mΩ)
2
80°C
100°C
1.5
120°C
1 140°C
0.5
0
0 2000 4000 6000 8000 10000 12000 14000 16000 18000 20000
Frequency (Hz)
Fig. 2.25. Camry capacitor module ESR vs. frequency.
3.5
2.5
-40°C
-20°C
2
ESR (mΩ)
0°C
40°C
1.5 60°C
80°C
1 100°C
120°C
140°C
0.5
0
0 2000 4000 6000 8000 10000 12000 14000 16000 18000 20000
Frequency (Hz)
Fig. 2.26. Prius capacitor module ESR vs. frequency.
To obtain a clearer perspective of the impact of temperature upon ESR, Figs. 2.27 and 2.28 shows ESR
plotted versus temperature for each frequency for the Camry and the Prius capacitor module, respectively.
22
1.8
1.6 600 Hz
3000 Hz
1.4
5000 Hz
1.2 7000 Hz
9000 Hz
ESR (mΩ)
1 11000 Hz
13000 Hz
0.8
15000 Hz
17000 Hz
0.6
19000 Hz
0.4
0.2
0
-60 -40 -20 0 20 40 60 80 100 120 140 160
Temperature (°C)
Fig. 2.27. Camry capacitor module ESR vs. temperature.
3.5
3 600 Hz
3000 Hz
2.5 5000 Hz
ESR (mΩ)
7000 Hz
2 9000 Hz
11000 Hz
1.5 13000 Hz
15000 Hz
1 17000 Hz
0.5
0
-50 0 50 100 150
Temperature (°C)
Fig. 2.28. Prius capacitor module ESR vs. temperature.
Again, the ESR of both modules generally increases with increasing temperature. Strangely, the ESR of
the Prius module has a substantial lobe at 100°C for all frequencies as well as a high ESR measurement at
-40°C and 15 kHz.
The variations of DF with frequency for the Camry and Prius capacitor module are shown in Figs. 2.29
and 2.30, respectively. These graphs are similar to the capacitance versus frequency graphs in Figs. 2.22
23
and 2.23. Since DF is the product of ESR, capacitance, and frequency, both graphs include curves which
approach infinity near the resonant frequency. The behavior of the Camry capacitor module DF as a
function of frequency is somewhat peculiar as the frequency increases beyond 6 kHz and then resumes
typical behavior beyond about 12 kHz.
100
-40°C
90
-20°C
80
0°C
Dissipation Factor (%)
70 40°C
60 60°C
80°C
50
100°C
40 120°C
30 140°C
20
10
0
0 2000 4000 6000 8000 10000 12000 14000 16000 18000 20000
Frequency (Hz)
Fig. 2.29. Camry capacitor module DF frequency response.
100
-40°C
90
-20°C
80 0°C
40°C
70
60°C
60 80°C
50 100°C
120°C
40
140°C
30
20
10
0
0 2000 4000 6000 8000 10000 12000 14000 16000 18000 20000
Frequency (Hz)
Fig. 2.30. Prius capacitor module DF frequency response.
24
The impact of temperature on the DF of the Camry and Prius capacitor modules for each test frequency is
observed in Figs. 2.31 and 2.32, respectively. Similar to the impact of temperature upon capacitance, the
DF generally increases with increasing temperature. Additionally, temperature has a greater impact on DF
at higher frequencies. Since the ESR naturally increases with increasing temperature, the amount of real
power dissipated at a particular amount of reactive power increases as temperature increases.
100
90
600 Hz
80
3000 Hz
70 5000 Hz
60 7000 Hz
9000 Hz
50
11000 Hz
40
13000 Hz
30 15000 Hz
20
10
0
-50 0 50 100 150
Temperature (°C)
Fig. 2.31. Camry capacitor module DF temperature response.
100
90
80 600 Hz
3000 Hz
70
5000 Hz
60
7000 Hz
50 9000 Hz
40 11000 Hz
30 13000 Hz
20
10
0
-50 0 50 100 150
Temperature (°C)
Fig. 2.32. Prius capacitor module DF temperature response.
2.1.2.2 Dynamic capacitor test results from large dc-link capacitor module
The Camry capacitor module was tested with a dc bias of 100 V and a maximum ripple current of 250 A.
Figure 2.33 shows the circuit used to apply a ripple current with a dc bias to the capacitor module under
25
test. A capacitor was placed across the output of the dc power supply to decrease the potential of
damaging the dc power supply with the high ripple currents generated during testing.
Fig. 2.33. Ripple current test schematic.
Thermocouples (TCs) were applied to the module in the center of both sides as shown in Fig. 2.34. Ripple
current was applied in 50 A steps starting at 100 A and stopping at 250 A [root mean square (rms)]. The
temperature was allowed to stabilize for 30 minutes before data was recorded. The two TC temperatures
were averaged and plotted to determine the module’s temperature response to high ripple current values.
These data are shown in Fig. 2.35, which shows that the average temperature rise from 0–250 A was 11°C
for the Camry capacitor module. The data for the 7.5 kHz test stopped at 190 A due to the limitations of
the alternating current (ac) amplifiers. Nonetheless, the test results indicate that the Camry is capable of
withstanding high ripple currents without reaching excessive temperatures. Results from dynamic tests
upon the Prius module at 4 kHz are provided in Fig. 2.36, and indicate that the steady state temperature of
the Prius module with a 200 A ripple current is about 52°C, whereas the steady state temperature of the
Camry module with a 200 A ripple current at 5 kHz is only 28°C.
Fig. 2.34. TCs on the Camry capacitor module.
26
35
33
Temp Average @ 1 kHz
Temp Average @ 2.5 kHz
31 Temp Average @ 5 kHz
Temp Average @ 7.5 kHz
29
Temperature (Deg C)
27
25
23
21
19
17
15
0 50 100 150 200 250 300
Ripple Current (Amps)

Fig. 2.35. Camry capacitor module steady-state temperature response vs. ripple current.
60
50
Temperature Rise (C)
40
30
20
10
0
0 50 100 150 200 250
Current (A rms)
Fig. 2.36. Prius capacitor module steady-state temperature response vs. ripple current at 4 kHz.
27
2.1.2.3 Static capacitor test results of single sub-cell capacitor of smoothing capacitor
module
The Prius and Camry capacitor modules are composed of 8 and 24 sub-cells, respectively. A sub-cell was
removed from the Camry module and tested under the same test conditions that the entire module was
subjected to. Since the module architecture greatly influences the overall function of the module and the
difference of the number of sub-cells is significant, the most informative comparison is between the entire
modules and not the sub-cells. Therefore, a comparison is not made between the Camry and Prius sub-
cells; yet the Prius sub-cell capacitor test results are provided in Appendix B. Additional Camry sub-cell
capacitor test results are also provided in Appendix B.
The large Camry capacitor module contains 24 86 μF sub-cell in parallel to form a total capacitance of
2098 μF; whereas the large Prius capacitor module contains (8) 141 μF sub-cell to form a total
capacitance of 1,130 μF. The capacitance, ESR, and DF were measured over temperatures ranging from
-40–140ºC and frequencies from 1–35 kHz and the results are shown in Figs. 2.37, 2.38, and 2.39,
respectively. The capacitance increases only slightly until subjected to frequencies above 15 kHz.
Figure 2.38 indicates that the ESR increases quickly with frequency at low frequencies and peaks at
5 kHz and slowly decreases with increasing frequency. Note again that the most common switching
frequency of the motor and generator inverters is 5 kHz. When compared to the test results from the Prius
capacitor sub-cell provided in Appendix B, the capacitance of the Camry capacitor sub-cell increases
much more slowly with frequency. The ESR of the Prius capacitor sub-cell is slightly lower than that of
the Camry capacitor sub-cell. The DF is much lower for the Camry capacitor sub-cell. Again, these
comparisons are not straightforward as the architectural discrepancies significantly impact the overall
characteristics. Nonetheless, a broad comparison of the characteristics is informative.
180.00
170.00
-40° Celsius
-20° Celsius
160.00
0° Celsius
20° Celsius
150.00 40° Celsius
60° Celsius
Capacitance(uF)
80° Celsius
140.00 100° Celsius
120° Celsius
130.00 140° Celsius
120.00
110.00
100.00
90.00
80.00
0 5000 10000 15000 20000 25000 30000 35000 40000
Frequency (Hz)
Fig. 2.37. Camry single 86 μF capacitor, capacitance vs. frequency.
28
10.00
9.00
8.00
7.00
-40° Celsius
6.00 -20° Celsius
0° Celsius
ESR(mΩ)
20° Celsius
5.00 40° Celsius
60° Celsius
80° Celsius
4.00
100° Celsius
120° Celsius
3.00 140° Celsius
2.00
1.00
0.00
0 5000 10000 15000 20000 25000 30000 35000 40000
Frequency (Hz)
Fig. 2.38. Camry single 86 μF capacitor, ESR vs. frequency.
30.00
25.00
20.00
-40° Celsius
-20° Celsius
0° Celsius
20° Celsius
15.00 40° Celsius
60° Celsius
80° Celsius
100° Celsius
120° Celsius
10.00
140° Celsius
5.00
0.00
0 5000 10000 15000 20000 25000 30000 35000 40000
Frequency (Hz)
Fig. 2.39. Camry single 86 μF capacitor, DF vs. frequency.
Figures 2.40, 2.41, 2.42, and 2.43 provide the combined capacitance, ESR, and DF responses versus
temperature for frequencies of 1 kHz, 5 kHz, 15 kHz, and 35 kHz, respectively. In comparing these
figures, it is apparent that the high temperatures have a significant effect on the capacitance, ESR, and
DF, particularly at high frequencies. Even at relatively low frequencies, the DF varies significantly with
29
temperature and frequency. Figures showing the results obtained while testing the Camry capacitor sub-
cell at additional frequencies are provided in Appendix B. Furthermore, similar test results from the Prius
capacitor sub-cell are provided in Appendix B.
88.50 9.0
Capacitance(uF)
88.00
ESR(mΩ) 8.0
Dissipation (%)
87.50
7.0
87.00
Dissipation Factor (%), ESR(mΩ)

6.0
86.50
Capacitance (uF)
86.00 5.0
85.50 4.0
85.00
3.0
84.50
2.0
84.00
1.0
83.50
83.00 0.0
-40 -20 0 20 40 60 80 100 120 140 160
Temp (Celsius)
Fig. 2.40. Camry single 86 μF capacitor, capacitance, ESR, and DF vs. temperature at 1 kHz.
90.00 10.0
Capacitance(uF)
9.0
ESR(mΩ)
89.00
Dissipation (%)
8.0
88.00
7.0
6.0
Capacitance (uF)
87.00
5.0
86.00
4.0
3.0
85.00
2.0
84.00
1.0
83.00 0.0
-40 -20 0 20 40 60 80 100 120 140 160
Temp (Celsius)
30
97.00 10.0
9.0
96.00
Capacitance(uF)
ESR(mΩ)
8.0
Dissipation (%)
95.00

7.0
94.00
6.0
Capacitance (uF)
93.00 5.0
4.0
92.00
3.0
91.00
2.0
90.00
1.0
89.00 0.0
-40 -20 0 20 40 60 80 100 120 140 160
Temp (Celsius)
172.00 35.0
Capacitance(uF)
ESR(mΩ)
170.00 Dissipation (%)
30.0
168.00
25.0

166.00
Capacitance (uF)
20.0
164.00
162.00
15.0
160.00
10.0
158.00
5.0
156.00
154.00 0.0
-40 -20 0 20 40 60 80 100 120 140 160
Temp (Celsius)
Figures 2.44 and 2.45 provide the combined capacitance, ESR, and DF responses versus frequency for
temperatures of 20ºC and 140ºC, respectively. Although a quick comparison of the figures does not
clearly reveal significant influences of temperature variation, notice the difference of scale between the
figures. While the scale for capacitance remains the same, the scale for DF and ESR changes from 0–18
to 0–35 in Figs. 2.44 and 2.45, respectively. Thus, as indicated in the comparison of the previous figures,
ESR and DF are significantly influenced by temperature variation. Since the unit is located within close
proximity of the high operation temperatures of the ICE in addition to heat generated by components
31
within the PCU (including the capacitor itself), the capacitor module is often subject to high temperatures.
Similar figures for the Camry and Prius capacitor sub-cells are provided in Appendix B.
180.00 18.0
160.00 16.0
Capacitance(uF)
140.00 14.0
ESR(mΩ)
Dissipation (%)

120.00 12.0
Capacitance (uF)
100.00 10.0
80.00 8.0
60.00 6.0
40.00 4.0
20.00 2.0
0.00 0.0
0 5000 10000 15000 20000 25000 30000 35000 40000
Frequency (Hz)
Fig. 2.44. Camry single 86 μF capacitor, capacitance, ESR, and DF vs. frequency at 20°C.
180.00 35.0
160.00
30.0
140.00 Capacitance(uF)
ESR(mΩ)
25.0

Dissipation (%)
120.00
Capacitance (uF)
20.0
100.00
80.00
15.0
60.00
10.0
40.00
5.0
20.00
0.00 0.0
0 5000 10000 15000 20000 25000 30000 35000 40000
Frequency (Hz)
Fig. 2.45. Camry single 86 μF capacitor, capacitance, ESR, and DF vs. frequency at 140°C.
32
2.1.2.4 Inspection into positive reactance measurement of large dc-link capacitor module
Further inspections into the influence of frequency upon characteristics of the Camry and Prius capacitor
modules were conducted using the dynamic test setup described above. A Yokogawa PZ4000 power
analyzer was used to monitor the current and voltage waveforms in order to determine the respective rms
values and phase differences between the two. No dc bias was applied during these tests, and all currents
and voltages discussed hereon are sinusoidal in nature and the corresponding signal levels are given in
terms of rms values.
The primary intent of these tests was to evaluate the voltage and current phase angle for various ripple
current frequencies, as well as obtain general information regarding the voltage response of the capacitor
module to a particular ripple current amplitude for various ripple current frequencies. Figures 2.46, 2.47,
and 2.48 show the phase angles between the current and voltage of the Camry and Prius capacitor
modules at various frequencies. These tests were conducted with about 25 A of current applied to the
modules at frequencies from 5–20 kHz and all signals were measured with a Yokogawa PZ4000
wattmeter. These figures are screenshots of the Yokogawa PZ4000 measurements and they include Uac1,
the rms of the ripple voltage (with no dc component), and Iac1, the rms of the ripple current, as well as
the frequency, which is shown in the upper left corner of the display. The phase angle between the current
and voltage was also measured, but the value is not shown in these figures. A leading phase angle is
obtained when the current leads the voltage, which is typical for a capacitive reactance. A lagging phase
angle is associated with an inductive reactance, in which the current lags the voltage.
Fig. 2.46. Waveforms of the Camry (left) and Prius (right) capacitor module at 5 kHz.
Fig. 2.47. Camry (left) and Prius (right) capacitor module waveforms at 16 kHz.
33
Fig. 2.48. Waveforms of the Camry (left) and Prius (right) capacitor module at 20 kHz.
Figure 2.46 shows the phase angles between the current and voltage at 5 kHz for the Camry and Prius
capacitor module with the current waveform being the orange trace and having the highest amplitude. For
both modules, the current is leading the voltage by an angle which is close to 90° indicating a typical
capacitive response at 5 kHz. The voltage ripple of the Prius module at 25 A is higher than that of the
Camry module as the capacitance of the Camry is higher than that of the Prius at this frequency.
Figure 2.47 shows screenshots obtained during a similar test with a frequency of 16 kHz for the Camry
and Prius capacitor modules. The voltage waveforms contain significant noise levels as the amplitude is
very low for these frequencies. The phase angle between the voltage and current of the Camry module at
16 kHz is nearly zero, and the current is lagging the voltage of the Prius module at 16 kHz indicating that
the reactance is positive and the module behaves like an inductor at this frequency.
In Fig. 2.48, the frequency of the ripple current applied to both modules is 20 kHz. The phase difference
between the current and voltage is large enough to easily concur that the current is lagging the voltage
significantly.
A second series of tests were conducted with the Camry and Prius capacitor modules. The overall test
procedures were similar to those just discussed, yet the amplitude of the signal generator (shown in
Fig. 2.33) waveform was kept constant. Conditions were similar but not identical between the tests on the
Camry and Prius capacitor modules and a comparison of the test results provides only general information
regarding the frequencies at which the resonance and phase angle changes occur.
Figures 2.49 and 2.50 compare the responses of the Camry and Prius capacitor modules to various
frequencies. In Fig. 2.49, as the applied frequency on the Camry module increases from 1 kHz, the current
increases and peaks at a frequency of about 2 kHz. A similar characteristic is noticed for the Prius module
and the current peaks between 2 and 3 kHz. As indicated by both figures, the current leads the voltage for
low frequencies indicating a dominant capacitive effect until a frequency of about 16 kHz for the Camry
and 15 kHz for the Prius. Leading phase measurements are indicated by the yellow traces, which end just
prior to these resonance frequencies where the capacitive and inductive effects are equal and the
corresponding reactance is zero. As the frequency increases beyond the resonant frequency, the module
behaves as an inductive load as the parasitic inductance has a greater effect on the reactance than the
capacitance does, even though the parasitic inductance is very low. These characteristics re-affirm that the
large dc-link capacitor is used for smoothing of low machine level ripple frequencies associated with the
operation the motor and generator, as opposed to being used primarily for attenuation of switching
transients.
34
100 50
90 45
Voltage (Vrms), Current (Arms)

80 40
70 35
Phase Angle (°)
60 30
Phase (leading)
50 25
Phase (lagging)
40 Vac 20
30 Iac 15
20 10
10 5
0 0
0 5 10 15 20 25
Frequency (kHz)
Fig. 2.49. Camry capacitor module voltage and current response to frequency.
100 50
90 45
Voltage (Vrms), Current (Arms)

80 40
70 35
60 30
Phase (°)
50 Phase (leading) 25
40 Phase (lagging) 20
Vac
30 15
Iac
20 10
10 5
0 0
0 5 10 15 20 25
Frequency (kHz)
Fig. 2.50. Prius capacitor module voltage and current response to frequency.
2.1.2.5 Static capacitor test results of battery level filter capacitor evaluations
Figures 2.51 and 2.52 compare the capacitor test results of the 500 V, 378 μF capacitor of the Camry
and the 600V, 282 μF capacitor of the Prius, respectively. Each capacitor resides between the battery
and the boost converter and is located within the PCU. Significant differences are noticed between
the characteristics of the two capacitors, particular when comparing the influence of frequency upon
capacitance and DF. The equivalent capacitance of the Prius begins to approach infinity near about
22 kHz, whereas the equivalent capacitance of the Camry capacitor only increased slightly as the
maximum test frequency of 30 kHz was reached. Since DF is a product of ESR, capacitance, and
frequency, the DF curves have characteristics similar to those of the capacitance curves.
35
800 16
700 Capacitance 14
ESR
600 12
Resistance (mΩ), DF (%)

DF
Capacitance (μF)
500 10
400 8
300 6
200 4
100 2
0 0
0 5000 10000 15000 20000 25000 30000 35000
Frequency (Hz)
Fig. 2.51. Camry battery level 378 μF capacitor, capacitance, ESR, and DF vs. frequency.
5000 100
4500 90
Capacitance
4000 ESR 80
Resistance (mOhms), DF (%)

3500 DF 70
Capacitance (μF)
3000 60
2500 50
2000 40
1500 30
1000 20
500 10
0 0
0 5000 10000 15000 20000 25000 30000 35000
Frequency (Hz)
Fig. 2.52. Prius battery level 282 μF capacitor, capacitance, ESR, and DF vs. frequency.
2.1.2.6 Static capacitor test results of small dc-link capacitor evaluations
Figures 2.53 and 2.54 compare the capacitor test results of the dc-link 750 V, 0.9 μF capacitor of the
Camry and the 750V, 0.1 μF capacitor of the Prius, respectively. Although the equivalent capacitance of
the small Camry capacitor appears to decrease significantly, a closer look at the capacitance axis limits
indicates that frequency has a small effect on the equivalent capacitance of the small Camry and Prius
capacitors. Unlike the ESR of the other capacitors, the ESR of both small capacitors is much higher for
36
low frequencies and decreases quickly with increasing frequency. All data points were taken at room
temperature. The DF is relatively low even at high frequencies. These characteristics indicate the high
frequency filtering capabilities of the small capacitors.
905 25
905
905 Capacitance(nF) 20
905 ESR(mΩ)
ESR (mΩ), DF (%)

Capacitance (nF)
904 Dissipation (%) 15

904
904 10
904
904 5
903
903 0
0 5000 10000 15000 20000 25000 30000 35000
Frequency (Hz)
Fig. 2.53. Camry dc-link 0.9 μF capacitor, capacitance, ESR, and DF vs. frequency.
105 45
104 Capacitance(nF) 40
103 ESR(mΩ) 35
ESR (mΩ), DF (%)

Capacitance (nF)
102 Dissipation (%)

30
101
25
100
20
99
15
98
97 10
96 5
95 0
0 5000 10000 15000 20000 25000 30000 35000
Frequency (Hz)
Fig. 2.54. Prius dc-link 0.1 μF capacitor, capacitance, ESR, and DF vs. frequency.
2.2 TRANSAXLE
The 2007 hybrid Camry transaxle assembly is comprised of motor, generator, and gear box
subassemblies. The mass of the entire unit is 108 kg, which is shown in Fig. 2.55. An ICE mounts to the
transaxle and a mechanical interface is established with spline shaft, which is slightly visible in the figure.
37
The electric motor and generator outputs are not directly accessible yet through a combination of various
gear configurations, the total motive force output of the hybrid system is supplied to the drive shafts by
the differential gear outputs which are indicated in the figure. The generator is located adjacent to the ICE
input (referred to as front end) and the motor is located in the rear portion of the transaxle, as shown in
Fig. 2.56. The gear section of the transaxle is located between the generator and motor sections, as shown
in Fig. 2.57.
Fig. 2.55. Hybrid Camry transaxle isometric view (left) and rear view (right).
Fig. 2.56. Camry generator front view with ICE spline removed (left) and motor rear view (right).
38
Fig. 2.57. Camry transaxle gear section open sandwich view.
2.2.1 Thermal Management and Lubrication System
The thermal management and lubrication system of the transaxle utilizes a mechanically driven trochoid
oil pump as well as oil slinging movement by the drive gears to circulate oil throughout the transaxle. As
shown in Fig. 2.58, the trochoid shaft is driven by the planetary carrier, which is connected directly to and
rotates at the same speed as the ICE. The black plate was removed from the rear of the trochoid system to
provide a view of the interior of the trochoid pump. An oil pickup is located near the bottom of the motor
section which serves as an input to the trochoid pump. The central component of the trochoid pump
resembles a gear and is machined such that the oil collects in the gaps of the pump teeth and forces the oil
through the trochoid shaft shortly thereafter. The motor rotor shaft is hollow and the trochoid pump shaft
extends through the hollow shaft of the motor rotor. A portion of the oil is dispensed to the motor bearing
which is located next to the toothed central component of the trochoid pump. Small holes are located
along the side of the trochoid pump shaft to supply oil to another motor bearing on the opposite end of the
motor rotor. The tip of the trochoid pump shaft supplies oil to the power split and speed reduction gear
through the same interface in which the shaft is driven. The 2004 Prius contains a similar trochoid pump
mechanism.
Note that oil is not circulated by the trochoid pump if the ICE is not active. Therefore, if the vehicle is
operating in electric mode only, then the system fully depends on the oil slinging action of the gear train
for oil circulation. A general diagram of the oil flow paths is provided in Fig. 2.59. Oil that has settled in
the bottom portion of the gear section is picked up by the differential gear teeth and is then slung to the
driver gear, to reservoirs in the upper portion of the gear section, and to the power split planetary gear
unit. The oil reservoirs are flooded if the differential gear is rotating above 900 rpm. A large reservoir is
adjacent to a heat exchanger in order to transfer heat to the cooling system. Weep holes and oil ports
distribute oil to the motor and generator stator as well as several bearings throughout the system. Note
that the circulation due to gear slinging is dependent upon vehicle speed.
39
Fig. 2.58. Trochoid oil pump view from outside rear (left) and from inside motor section (right).
Fig. 2.59. Oil slinging paths formed by gear train.
A heat exchanger is located on the one side of the motor periphery and the generator also has a small heat
exchanger which is part of a series coolant loop with the motor heat exchanger as well as the PCU heat
exchanger. According to specifications in the Camry repair manual, the coolant loop consists of a
standard 50% ethlyene glycol and 50% water mixture which flows at 10 L per minute and has an average
operation temperature of 65ºC. Vehicle tests at Argonne National Laboratory (ANL) indicate the average
coolant temperature may be considerably lower depending on operation conditions. The coolant capacity
is listed as 2.9 L. A separate radiator is dedicated to the THS coolant loop which is isolated from the ICE
coolant loop that operates at much higher temperatures.
40
2.2.2 Transaxle Gear Train
The Camry transaxle has three primary gear systems which are designed to carry out power split, speed
reduction, and drive functions. A general comparison of the 2004 Prius gear train and the 2007 Camry
gear train is provided in Fig. 2.60. The number of teeth on each gear is noted in black text on a white
background. There are several noticeable differences between the two designs. The Prius transaxle design
has a chain coupling between the power split planetary ring and the counter drive gear, whereas the
planetary ring and counter drive gears mesh directly within the Camry transaxle. Additionally, the Camry
transaxle has only three separate axes about which gear systems rotate, compared to four axes in the Prius
transaxle. The number of teeth on the planetary gear components does not differ between the two designs.
The total gear ratio from the Camry power split planetary ring to the differential output is
(80/23) × (55/54) = 3.542, where the equivalent gear ratio for the Prius is (75/26) × (44/30) × (35/36) =
4.113.
Fig. 2.60. 2004 Prius gear train (left) and 2007 Camry gear train (right).
Another significant difference between the designs is that the Camry has a speed reduction gear, which is
comprised of a planetary gear unit. The speed reduction planetary gear assembly is shown along with the
power split planetary gear unit in Fig. 2.61. The planetary gear units share the ring gear in which the
electric motor, MG2, is connected to the ring gear through the speed reduction gears. The number of teeth
is shown for each gear and the planetary carrier is fixed to the chassis. Thus, the gear ratio between the
electric motor shaft is (57/18) × (18/23) = 2.478, where the MG2 spins 2.478 times faster than the ring
gear and therefore the torque output of MG2 is effectively multiplied by 2.478 prior to reaching the ring
gear. Gear teeth on the outer periphery of the ring gear, collectively referred to as the counter drive gear,
mesh with the counter driven gear which supplies motive force to the drive shafts though the final and
differential gear units. Consequently, the speed of MG2 is directly proportional to the speed of the
vehicle. The generator rotor shaft is hollow and connects to the sun gear of power split planetary and the
input shaft from the ICE is fed through the hollow generator shaft and connects to the planetary carrier. A
locking gear mechanism is adjacent to the counter drive gear which is engaged when the vehicle gear
selector is placed in park. A summary of the number of gear teeth for each gear is provided in Table 2.6.
Since the Prius drive gear train is moderately different from the Camry drive train, a comparison is not
easily made between each component of the drive train, but is more meaningful if the overall gear ratio is
compared, as provided previously.
41
Fig. 2.61. Camry speed reduction planetary and power split planetary.
Table 2.6. Camry transaxle gear information
Power Split The No. of Ring Gear Teeth 78

Planetary The No. of Pinion Gear Teeth 23
Gear Unit The No. of Sun Gear Teeth 30
Motor Speed The No. of Ring Gear Teeth 57

Reduction The No. of Pinion Gear Teeth 18
Planetary Gear Unit The No. of Sun Gear Teeth 23
Counter The No. of Drive Gear Teeth 54

Gears The No. of Driven Gear Teeth 55
Final The No. of Drive Gear Teeth 23

Gears The No. of Driven Gear Teeth 80
Total Deceleration Ratio 3.542
2.2.3 Motor Mass and Volumetric Assessments
When comparing the Camry and Prius electric drive characteristics, the mass and volume of the primary
electric drive motor, MG2, was decreased although the peak and continuous power capabilities of the
motor increased. This was primarily accomplished by increasing the rated speed of the motor from about
42
6,000–14,000 rpm. To clarify the benefit of increasing the speed rating, the torque and power rating of the
Prius motor is given in Fig. 2.62. These are shown to present the general speed, torque, and power
relationship of any motor as power is equal to torque times speed. Therefore, a much lower torque is
required to produce a certain power level at a higher speed than for a lower speed. For example, a torque
of 382 Nm is required to produce 40 kW at 1000 rpm, whereas only 38.2 Nm is required to produce
40 kW at 10,000 rpm. Additionally, torque is proportional to current and, roughly speaking, the amount of
current required to produce a consistent power level decreases with increasing speed. While this is not
entirely an accurate statement as there are many factors to consider when making such a comparison, the
concept generally holds true.
450
50
400
40 350
Power (kW)
Torque (Nm)
300
30 250
200
20
150
100
10
50
0 0
0 1000 2000 3000 4000 5000 6000 7000 0 1000 2000 3000 4000 5000 6000 7000
Speed (RPM) Speed (RPM)
Fig. 2.62. Speed, torque, and power relationship.
There are a few negative aspects of moving to higher motor speed ratings however. Typically, the low
speed torque rating of the motor is compromised when the speed rating is increased. Therefore, the lowest
speed at which rated power can be developed, namely base speed, increases. Again, these relationships
depend upon various motor design parameters such as rotor diameter, lamination stack length, magnet
strength, and winding configuration, yet these relationships are typically pertinent. As rotor speed ratings
increase, additional mechanical design aspects must be considered. Improved support must be provided
by the rotor laminations to retain the PMs which are subjected to forces which are directly proportion to
the square of the rotor speed. Additionally, bearings must be designed to withstand the high speeds of the
rotor shaft. Depending on the characteristics of the PMs, rotor, and stator, core losses and induced back-
emf voltages typically increase with increasing rotor speed. The high speed reduction gear used in the
Camry design is required to convert the high speed of the motor to properly match the remaining
transaxle design. Although the torque rating of the Camry is only 270 Nm, the maximum torque is
converted to about 670 Nm through the speed reduction gear.
Figures 2.63 and 2.64 provide various views of the Camry transaxle and the volume associated with the
primary drive motor. Certain key dimensions are shown, but several others are needed to determine
volume since the motor terminal box, and especially the heat exchanger shown in Fig. 2.64 have complex
geometries. Since the motor casing also houses some of the gear box, the motor casing had to be well
defined to arrive at the mass and volume, excluding areas unrelated to the motor. For example, in
Figs. 2.63(a) and 2.63(b), the surface chosen for the end of the motor housing is indicated.
43
Fig. 2.63. Camry motor assembly after separation from generator assembly.
Fig. 2.64. Coolant channel at bottom of motor casing.
The results of the packaging assessment were used to generate estimates of peak specific power and peak
power density as summarized and compared to the Prius specs in Table 2.7. Because of advantages gained
by increasing the maximum speed from 6,000–14,000 rpm, Toyota was able to optimize the Camry
design such that motor volume and mass were both lower than for the Prius in spite of a higher published
peak power rating. It is appropriate to include the mass and volume of the Camry speed reduction
planetary gear in this comparison with the Prius and doing so mitigates the perceived improvement upon
the Prius design.
Table 2.7. Specific power and power density estimates for the PMSM
Parameter Camry Prius

Motor peak specific power (without
70/~41.7 = ~1.68 50/45.0 = 1.11
converter), kW/kg
Motor peak power density, kW/L 70/~14.8 = ~4.73 50/15.4 = 3.25
44
Figure 2.65 shows the stator assembly that was removed from the Camry transaxle assembly. The entire
Camry stator assembly mass is 18.0 kg, whereas the Prius stator assembly mass is 25.9 kg, with 5.7 kg
and 6.8 kg of copper within each stator, respectively. The three-phase power input (“output” during
regenerative braking) leads are clearly evident. A thermistor is embedded in the stator windings and the
corresponding leads are routed adjacently to the motor leads. There are 18 wires bundled in each motor
lead which splits in half to form a parallel winding configuration. The Camry stator windings have
9 strands of copper with 14 turns per pole and the Prius stator windings have 13 stands of copper with
9 turns per pole in parallel and series configurations, respectively. Notice that the product of the number
of strands and the number of turns has a comparable result between the Camry and the Prius.
Fig. 2.65. Camry motor stator.
A comparison of the Camry and Prius laminations is provided in Fig. 2.66. In Fig. 2.66(a), a Camry motor
stator lamination is placed on top of a Prius motor stator lamination, showing that the outer diameter
(OD) of the Camry lamination is slightly smaller than that of the Prius. Through both Figs. 2.66(a) and
2.66(b) it is shown that the Camry motor stator slots are somewhat shorter, yet they are wider than the
slots on the Prius motor stator laminations. The Camry generator stator laminations are identical to the
Camry motor stator laminations. The motor lamination stack thickness has reduced from 3.29” on the
Prius to 2.39” on the Camry. The Camry generator lamination stack thickness is 1.41”. The inner
diameters (IDs) of the Camry motor, Camry generator, and Prius motor stator laminations are equal. The
stack lengths and outer diameters of the rotors have corresponding similarities.
45
a
Camry stator winding slots are
shorter, yet wider than Prius
b
Fig. 2.66. Camry and Prius motor stator laminations.
Figure 2.67 shows details of the Camry rotor assembly and the “V” orientation of the neodymium iron
boron (NdFeB) interior PM. Looking at the pair of magnets in the foreground, it is possible to see the
narrow support bar that passes between them. This support bar, missing from the Prius motor laminations,
helps to enable the motor to operate at much higher rotational speeds.
Fig. 2.67. Camry rotor assembled (left) and partially disassembled (right).
46
In making various comparisons between the Camry and Prius motor components, the highlights shown in
Table 2.8 were generated. The table summarizes the general motor design, rotor and stator
dimensions/mass, and the wiring and slot details of the stator. Because the rotor and stator lamination
geometries are difficult to fully describe in a table, Figs. 2.68 and 2.69 provide this information. The
length of the magnets decreased along with the decrease in the stator and rotor lamination stack length.
However, the thickness and width of the Camry and Prius motor PMs are very similar. Details regarding
the magnet characteristics are provided in Section 2.2.4. Various motor characteristics were
experimentally measured and the results are presented in Section 3.
Table 2.8. 2007 Camry/2004 Prius motor design comparison highlights
Parameter Camry Prius Comments

Lamination Dimensions
Stator OD, mm 264 269 OD
Stator ID, mm 161.93 161.93 ID
Stator stack length, cm 6.07 8.4
Rotor OD, mm 160.47 160.47 OD
Rotor lamination ID, mm 105 111 ID
Rotor stack length, cm 6.2 8.36
Air gap, mm 0.73025 0.73025
Lamination thickness, mm 0.31 0.33
Mass of Assemblies
Rotor mass, kg 9.03 10.2 Including rotor shaft.
Stator mass, kg 18.0 25.9
Stator core mass, kg 12.38 19.05 Laminations only.
Stator Wiring
Number of stator slots 48 48
Stator turns per coil 14 9
Parallel circuits per phase 2 legs 0
Turns in series per phase 4 per leg 8
Number of wires in parallel 9 per leg 13
Wire size, American wire gauge
20 19
(AWG)
0.023 0.069 Average of phase-to-phase
Phase resistance at 21°C, ohm divided by two.
Total mass of stator copper, kg 5.6 6.8
Slot depth, mm 30.9 33.5
Slot opening, mm 1.88 1.93
Casing
Camry has thicker walls, larger
Motor casing mass, kg 9.5 8.9 heat exchanger, speed reduction
gear casing and bearings
Motor casing diameter, cm 30.2 29.9
Motor casing axial length, cm 17.0 20.5
Magnets (NdFeB)
Magnet dimensions, mm 60.6×19.1×6.6 83.1×18.9×6.5 One magnet (1/2 of “V” or pole).
Magnet volume, cm3 7.63 10.2 One magnet (1/2 of “V” or pole).
Magnet mass, grams 58 77 One magnet (1/2 of “V” or pole).
Total mass of magnets, kg 0.928 1.232 Entire magnet mass in rotor.
47
Fig. 2.68. Camry and Prius motor rotor lamination/PM design.
Fig. 2.69. Camry and Prius motor stator lamination/slot design.
Volume assessments of the Camry generator housing were conducted with the general approach and
calculations shown in Fig. 2.70. The total volume of the Camry generator housing is about 11 L and the
total mass is about 28 kg. The power rating of the Camry generator is approximately 40 kW compared to
the 33 kW rating of the Prius generator. Comparisons of the Camry and Prius generator stators and rotors
are provided in Figs. 2.71 and 2.72, respectively. It is evident that the generator has increased in both
diameter and stack length by about 14% and 17.5%, respectively. The size of the Camry generator is more
suited for continuous operation. A thermistor is embedded within the stator windings to provide
temperature feedback and the MG ECU will discontinue operation of the generator upon reaching an
over-temperature condition. A summary and comparison of the Camry and Prius generator specifications
is provided in Table 2.9.
48
Fig. 2.70. Camry generator volume assessment.
Fig. 2.71. Camry (left) and Prius (right) generator stator comparison.
Fig. 2.72. Camry (left) and Prius (right) generator rotor comparison.
49
Table 2.9. 2007 Camry/2004 Prius generator design comparison highlights

Lamination Dimensions
Stator OD, mm 264.16 236.2
Stator ID, mm 162.6 142.6
Stator stack length, cm 3.58 3.05
Rotor OD, mm 160.46 140.72
Rotor lamination ID, mm 95.63 85.09
Lamination thickness, mm 0.31 0.33
Mass of Assemblies
Rotor mass, kg 5.188 4.01 Including rotor shaft
Stator mass, kg 12.09 9.16
Stator Wiring
Number of stator slots 48 48
Number of wires in parallel 18 12
Wire size, AWG 20 20
2.2.4 PM Characteristic Assessments
Hysteresis tests were conducted on the Camry motor PMs using the Walker Scientific AMH-40
hysteresisgraph (shown in Fig. 2.73), which is capable of producing 30,000 Oersteds of magnetic field
strength. A hysteresis test provides information regarding the remanent flux density and coercivity of a
magnet and these tests are often conducted over a wide range of temperatures. The remanent flux density,
Br, is the remaining flux density created by the magnet with no external field applied. Or simply stated the
remanent flux density represents the strength of the magnet. The remanent flux density decreases to some
extent with increasing temperature. The coercivity of a magnet is defined to be the intensity of the
magnetic field required to reduce the magnetization of the magnet, and the coercivity of a magnet
decreases with increasing temperature. Coercivity is an especially important characteristic of PMs that are
used in HEV applications, which often involve high temperature and high field weakening operation
conditions.
During hysteresis tests, the magnet is subjected to positive and negative magnetic fields while the impact
of the externally applied magnetic field on the magnet is observed. A graph from hysteresis tests
conducted on a magnet from the Camry motor rotor at 110ºC is shown in Fig. 2.74. Two curves are
shown on the graph with the burgundy trace being the normal, or total magnetic flux density, and the
turquoise trace representing the intrinsic magnetic flux density. The intrinsic flux density is the flux
density solely created by the magnet, which is obtained by subtracting the applied flux density from the
total flux density. The remanent flux density is the value at which both curves intersect with the y-axis,
which is at y = 11.86 kG in this case. The value of coercivity is found at the point where the intrinsic
curve crosses the x-axis, which is x = 14.46 kOe in this case. A similar graph is provided in Fig. 2.75,
with the hysteresis evaluation conducted at an ambient temperature of 198ºC. Notice that the curves are
much closer to the origin when compared to Fig. 2.74. As the temperature increased from 110–198ºC, the
remanent flux density and coercivity decreased from 11.86–10.29 kG and 14.46–5.757 kOe, respectively.
The drastic decrease of the coercivity indicates that the magnet is much more susceptible to
demagnetization at high temperatures.
50
Fig. 2.73. Walker Scientific hysteresis graph.
Fig. 2.74. Hysteresis graph for Camry motor magnet at 110ºC.
51
Fig. 2.75. Hysteresis graph for Camry motor magnet at 198ºC.
Intrinsic curves from hysteresis tests conducted with a Camry motor magnet at various temperatures have
been combined onto one plot in Fig. 2.76. These curves further exemplify the diminishing nature of the
remanent flux density and coercivity with increasing temperature. A similar graph containing the intrinsic
curves obtained from the Prius motor magnet hysteresis tests is provided in Fig. 2.77. Although it is
difficult to make a detailed comparison using these figures, a general comparison indicates that the
coercivity of the Camry magnet is significantly lower than that of the Prius.
Fig. 2.76. Intrinsic hysteresis graphs for Camry motor magnet at various temperatures.
52
Fig. 2.77. Intrinsic hysteresis graphs for Prius motor magnet at various temperatures.
To provide a more straight forward comparison of the remanent flux density characteristics of the Camry
and Prius motor magnets, the values were plotted versus temperature in Fig. 2.78. The strength of the
Camry motor magnet is marginally higher than the strength of the Prius motor magnet, except for
temperatures beyond about 215ºC. A similar plot is shown in Fig. 2.79 for the coercivity versus
temperature relationship of the Camry and Prius motor magnets. The coercivity of the Camry motor
magnet is about 70% of the coercivity of the Prius magnet, except for temperatures above about 200ºC
where the coercivity values are a bit closer.
Although the Camry motor magnet is slightly stronger than the Prius magnet, the coercivity is much
lower and, therefore, is much more susceptible to demagnetization when subjected to high temperatures
and/or high strength magnetic fields which oppose the direction of the magnetic field of the magnet. It is
common to apply opposing magnetic fields to the magnets when operating at high speeds, wherein high
back-emf voltages are subdued to enable high speed operation within the voltage constraints. It is very
likely that the cost of the Camry motor magnet is lower than the cost of the Prius magnet and although it
has a lower coercivity, demagnetization can be avoided using preventative control techniques.
53
Fig. 2.78. Remanent flux density for Camry and Prius motor magnet.
Fig. 2.79. Coercivity of Camry and Prius motor magnet.
54
3. BENCHMARKING TESTS OF CAMRY PMSM AND INVERTER
This section describes the hybrid Camry subassembly preparation and the back-emf, rotational loss,
locked-rotor, performance/efficiency, and continuous load duration tests. With the exception of the back-
emf, rotational loss, and locked-rotor tests which did not require inverter operation, each was performed
using both the PMSM and inverter subsystems.
3.1 SUBSYSTEM PREPARATION FOR TESTING
As design and packaging assessments of the transaxle and PCU were being conducted, preparation for
testing of the subsystems was commenced, including hardware design and fabrication, controller/interface
development, and instrumentation for the DAQ. Two separate approaches were taken to adapt the Camry
PMSM to separate dynamometer setups as each provided a benefit in terms of implementation time
versus unmitigated testing capability. Although basic framework was available from previous
benchmarking efforts, a large portion of the efforts devoted to controller/interface and DAQ development
required innovation.
3.1.1 Hardware Preparation
The first test setup approach provided a quick resolution and means to initiate testing, yet it did not
provide testability throughout the entire operation range of the PMSM. Therefore, while the initial tests
were being conducted with the first test setup, a second approach was instituted. Additionally, the
dynamometer test cell was upgraded to suit the high speed and power demands of the Camry PMSM.
Similar to the approach taken with the Prius transaxle hardware setup, the first approach involved welding
the power split planetary gear as shown in Fig. 3.1. This causes the ICE input shaft, which is easily
accessible, to rotate at the same speed as the ring gear of the power split planetary gear unit. For the Prius
design, the ring gear rotates at the same speed of the primary drive motor, MG2. However, the MG2 of
the Camry is connected to the ring gear through the speed reduction planetary gear unit. Thus, although
the first approach provided a quick means to access the motive force of the motor, uncertainties were
associated with the influence of speed reduction gear losses upon power and efficiency measurements. To
obtain accurate motor efficiency measurements, it is crucial to accurately measure the mechanical power
produced directly by the motor. Nonetheless, the setup provided a quick means of test initiation and
valuable insights, such as controller development aspects. Additionally, the setup eliminated the
immediate need for equipment with high speed testing capabilities, which was not installed until a later
time during the project duration.
The second approach involved design and fabrication of a shaft and faceplate (shown in Fig. 3.2) which
adapts to the Camry motor rotor, transaxle mounting pattern, and the high speed torque cell face mount
and spline shaft. Design considerations were carefully surmised to ensure that no part of the system
operation was hindered or enhanced. Particularly, intense focus was placed on maintaining the operational
characteristics of the thermal management and lubrication system mostly by ensuring that the oil flow
pattern throughout the transaxle remained unchanged. This approach avoids uncertainties associated with
the losses of the high speed reduction gear, thereby providing accurate measurements of motor power
output.
55
Fig. 3.1. First test setup approach with welded Camry planetary gear.
Fig. 3.2. Second test setup approach with modified motor rotor shaft.
To accommodate the high speed and power testing requirements of the Camry PMSM, the test cell was
upgraded with a 400 hp dynamomter and speed reduction gearbox, as shown in Fig. 3.3. The speed
reduction gearbox was required to match the speed range of the dynamometer to a speed of up to
18,000 rpm on the high speed input shaft. A high speed torque transducer (not shown) attaches to the
high speed input shaft of the gearbox via a face mount support. The spline shaft output of the modified
Camry rotor shaft and the fabricated faceplate mates with the input of the high speed torque transducer.
56
Fig. 3.3. Upgraded dynamometer test cell.
3.1.2 Instrumentation and DAQ Development
The PCU, PMSM, and transaxle were instrumented with thermistors and TCs in order to monitor and
record thermal conditions and characteristics of the subsystem during various operation conditions. In
preparation for performance, efficiency, and continuous duration tests in the ORNL laboratory, the motor
inverter was instrumented with three TCs placed in the edges of the three power electronics heat spreaders
as shown in Fig. 3.4. Two other TCs were placed in the (1) aluminum wall against the potted inductor of
the buck/boost converter, and (2) edge of the heat spreader for the buck/boost converter power electronics
module.
Fig. 3.4. TCs in the inverter power electronics heat spreaders.
57
Thermistors were installed in various locations on the stator windings and transaxle housing, as indicated
in Fig. 3.5. Each location has a unique letter identifier and additional location information. For example,
thermistor “D: ML 1” is located on the motor lead side of the stator at the 1 o’clock position indicated by
the red circle. All thermistor locations on the front side of the stator are indicated by red circles and
thermistor locations on the rear of the stator are marked with cyan circles. Case thermistor locations are
indicated by light green circles and oil thermistor locations are indicated by orange circles. Note that
thermistor “E: ML 2” is the original factory installed thermistor. The resistance of an identical thermistor
was evaluated over a wide range of temperatures in order to generate a lookup table to be used with the
DAQ software.
Fig. 3.5. Thermistor locations on PMSM and transaxle.
The Camry transaxle was modified to include windows and temporary viewports in the locations shown
in Fig. 3.6 to facilitate observations of the oil system characteristics. These observations provided useful
information regarding oil circulation and oil levels in various locations throughout the transaxle. It was
especially necessary to assess these characteristics as the design, fabrication, and implementation of the
second test setup approach was carried out and that these characteristics remained unchanged during
58
operation. It was determined that the oil reservoirs as well as the upper oil channels are nearly full when
the ring gear is rotating at speeds near 1,000 rpm.
Oil paths to
stator
Window
Viewport
Viewport
Oil
reservoir
Heat
exchanger
Fig. 3.6. Viewport and window locations implemented on transaxle.
A program written in the Visual Basic programming language was developed to collect data from various
measurement instruments and merge the data into a convenient spreadsheet format. Torque, speed, and
mechanical power were obtained using a Himmelstein torque and speed transducer. Temperature
measurement signals from thermistors and TCs were collected with a Keithly 2700 acquisition unit. A
Yokogawa PZ4000 power analyzer was used to carry out electrical measurements such as ac and dc rms
voltage, rms current, real power, and many other properties of the waveforms.
The DAQ system was important not only for data logging, but it also served as a real time feedback
system in which system conditions were monitored to ensure that operational limitations were not
violated and to ensure that the system was being operated in an optimal manner. Temperature and current
constraints were the primary limitations of the entire drive system. Efficiencies were calculated using the
mechanical and electrical data and were used to verify that the controller operation was optimized.
59
3.1.3 Controller and Interface Development
The hybrid subsystems of the 2007 Toyota Camry were evaluated for performance and efficiency
characteristics such as peak and continuous power ratings and motor/inverter efficiency. Continuous
operational capabilities are greatly affected by the heat transfer characteristics of the motor, which is
cooled through oil circulation and an ethylene-glycol coolant loop. The hybrid subsystems were operated
over a broad torque-speed range in order to collect and analyze thermal and electrical data at a multitude
of operation points. For each operation point during efficiency evaluations, the system was held in steady
state for at least 20 seconds as a minimum of 5 data samples were taken.
ORNL’s dynamometer test cell and Opal-RT-based speed and current feedback controller were adapted to
provide the torque needed at each reference speed. Thus, as the applied torque from the dynamometer was
varied manually, the controller regulated the torque producing current appropriately. The current
controller consists of two standard proportional-integral (PI) controllers for the direct and quadrature
currents, id and iq, respectively. These d-q components are obtained by applying the d-q transformation to
the three-phase currents which have a fixed reference. The transformation converts the three-phase
currents into two-phase vectors, which have a reference that rotates with the rotor. Therefore, precise
rotor position feedback is used during this transformation.
The steady state torque equation for the salient PM machine is expressed by:
τ L = n p ( Ld − Lq ) id iq + n p Kiq , (1)
where
np is the number of pole pairs,

Ld is the d-axis inductance,
Lq is the q-axis inductance, and
K is the back-emf and torque-current factor.
The total torque given by Eq. (1) consists of two torque terms which are reluctance torque and PM torque,
respectively. PM torque is produced only by the current component along the q-axis. If current is applied
which results with a negative component along the d-axis, positive reluctance torque is developed since
the difference, Ld -Lq, in the first term is negative and all remaining variables and constants are positive in
the motoring region. In theory, there is an infinite amount of d-q current combinations that will satisfy a
particular operation condition. There is an optimal d-q current combination in which the motor efficiency
is maximized for each particular torque. It is difficult to determine the optimal current trajectories for the
entire torque-speed range, as complex factors such as effects of saturation and harmonics must be
considered. Therefore, the DAQ was used to monitor the system efficiency to ensure that the controller is
operating optimally.
A high level diagram of the Opal-RT controller and interface system is shown in Fig. 3.7. A console
personal computer (PC) is used to develop and modify control schemes in a block diagram format, which
is converted and compiled to traditional programming languages (mostly C) and sent to the target nodes
through an ethernet network. The target nodes execute the controller code in real time and the console PC
is used to control the mode of the controller operation as well as to update various parameters if desired.
While the development process is more straightforward than microprocessor programming, the PC based
architecture of the target nodes introduces a significant degradation in terms of control sampling
frequency. Therefore, the system resources must be utilized in an efficient manner.
60
Fig. 3.7. Opal-RT controller and interface system.
3.2 BACK-EMF TEST RESULTS
The non-energized Camry PMSM was spun at various speeds over its full design range to obtain back-
emf data. The shaft speed and back-emf data (line-to-neutral rms averaged for the three phases) is plotted
in Fig. 3.8. This test was repeated after the performance/efficiency mapping test to verify that no
magnetic strength was lost in the rotor during performance tests or extended periods of high-temperature
testing. As expected, there were no significant differences in the data obtained from the second back-emf
test.
450
400
Voltage, V_L-n (V RMS)
350
300
250
200
150
100
50
0
0 5000 10000 15000
Motor Speed (rpm)
Fig. 3.8. 2007 Camry motor back-emf test results.
61
A plot containing the line-to-neutral back-emf voltage measurements observed during benchmarking
evaluations of the Prius is shown in Fig. 3.9. A comparison of Figs. 3.8 and 3.9 indicates that the volts-
per-hertz ratio of the Prius back-emf is higher. However, the maximum back-emf voltage of the Camry,
observed at a shaft speed of 14,000 rpm is much higher than the maximum back-emf voltage of the Prius
at 6,000 rpm. Note that the ratio of these maximum voltages is approximately equal to the ratio of the
maximum dc-link voltages, which is 650/500.
Voltage, V_L-n (V RMS) 350

300
250
200
150
100
50
0
0 1000 2000 3000 4000 5000 6000
Speed (RPM)
Fig. 3.9. 2004 Prius motor back-emf test results.
3.3 ROTATIONAL LOSS ASSESSMENTS
The rotational losses of the entire gear train and the Camry motor alone were evaluated by rotating the
input shafts with an external drive motor. Gear train loss assessments began with the entire gear system
installed, which includes the drive gears and planetary gears. The generator and motor rotor were
removed in order to assess only the gear losses and motor core loss assessments were conducted at a later
time. Torque and power measurements were measured and plotted versus ring gear speed as shown in
Figs. 3.10 and 3.11, respectively. The average oil temperature was slightly higher than room temperature
during these tests. Note the maximum speed shown in this plot is 5,500 rpm and since the speed reduction
gear ratio is 2.47, the corresponding motor speed is 13,585 rpm and the vehicle speed associated with this
extreme condition is approximately 113 miles per hour.
9
8
7
Torque (Nm)
6
5
4
3
2
1
0
0 1000 2000 3000 4000 5000 6000
Drive Gear Speed (rpm)
Fig. 3.10. 2007 Camry gear train rotational torque.
62
5000
4500
4000
Power Loss (W)

3500
3000
2500
2000
1500
1000
500
0
0 1000 2000 3000 4000 5000 6000
Drive Gear Speed (rpm)
Fig. 3.11. 2007 Camry gear train rotational losses.
Results from the rotational loss tests conducted with the shaft connected directly to the motor rotor are
shown in Fig. 3.12. These losses are mostly comprised of core (or iron) losses, which are associated with
eddy currents and hysteresis effects. In combining the gear losses shown in Fig. 3.11 and the core losses
shown in Fig. 3.12, it is inferred that a significant amount of losses are associated with the rotational
motion of the drive system. Nonetheless, these losses are exceptionally low when compared to that of a
conventional automatic transmission, which includes multiple planetary gears and clutching mechanisms.
1200
1000
Spinning Loss (Watts)
800
600
400
200
0
0 2000 4000 6000 8000 10000 12000 14000
Speed (rpm)
Fig. 3.12. 2007 Camry motor rotational losses.
Further gear loss assessments were conducted to obtain information which more specifically describes the
losses associated with each component, as well as the influence of oil temperature upon the gear losses.
The first test was conducted with all gears installed and the motor rotor removed and the oil temperature
initially low. The Bay-Voltex external coolant temperature regulator was used to increase the oil
temperature in order to observe the effects of increasing oil temperature as the rotor speed was held
constant. These tests were conducted at ring gear speeds of 2000, 3500, and 5000 rpm. A gear system was
removed after each test was conducted, and therefore the associated losses of the removed gear system
can be assessed. The results from these tests are shown in Fig. 3.13, with the 2000, 3500, and 5000 rpm
traces shown in shades of blue, green, and red, respectively. After tests were conducted with no motor
rotor, the drive gears were removed, then the planetary gears were removed. Losses of the entire gear
train tend to decrease with increasing oil temperature due to deceasing oil viscosity and losses are
minimal near the common operation temperatures of the transaxle, namely 50–80ºC. Extended tests at
63
3500 rpm reveal that the gear losses increase beyond a temperature of 90ºC. In comparing the loss
difference of each of the tests, the drive gears have the highest amount of losses associated with spinning
motor and the planetary gears contribute only a small amount.
4000.00
2000 RPM - Minus motor rotor
3500.00 2000 RPM - Minus drive gears
2000 RPM - Minus plan gears
3000.00
2500.00 3500 RPM - Minus drive gears
Power (W)
3500 RPM - Minus plan gears

2000.00
1500.00
1000.00
500.00
0.00
0.00 20.00 40.00 60.00 80.00 100.00 120.00
Oil Temperature
Fig. 3.13. 2007 Camry gear losses vs. oil temperature at various ring gear speeds.
3.4 LOCKED ROTOR TEST
A series of locked rotor tests were performed to determine the torque-producing behavior of the PMSM
motor. During locked rotor tests, the rotor is rotated throughout an electrical cycle and the shaft is rigidly
locked at each position as dc current is applied. Torque values were obtained from an in-line shaft torque
sensor when the motor windings were energized. The resulting data was used to produce torque versus
electrical position plots, which are shown in Fig. 3.14 for various current levels. These tests were not
conducted at extremely high dc currents to avoid high temperatures which could potentially damage the
stator windings or rotor magnets. The peak torque for each current level was chosen and is indicated by
the blue trace in Fig. 3.15. Peak locked rotor torques for the Prius are also included in the figure, and are
much larger than that of the Camry. However, the equivalent torque of the Camry motor on the low speed
side of the speed reduction gear is much higher, as indicated by the yellow trace in Fig. 3.15.
Fig. 3.14. 2007 Camry locked rotor torque vs. position for various dc currents.
64
800
Camry
700 Prius
Camry w/ Gear
600
500
Peak Torque
400
300
200
100
0
0 100 200 300 400 500
Current
Fig. 3.15. 2007 Camry and Prius peak locked rotor torque vs. dc current.
Interior PM motors have two torque components and there is an optimal current angle at which the two
torque components produce the maximum torque per current. Figure 3.16 shows the PM torque,
reluctance torque, and the total sum of the two torque components for a constant exemplar current
magnitude. Although the maximum PM torque is produced at a current angle of zero electrical degrees
and the maximum reluctance torque is produced at 45 electrical degrees, the maximum total torque is
produced at about 35 electrical degrees. This optimal current angle varies with many conditions such as
current magnitude and speed.
Fig. 3.16. 2007 Camry and Prius peak locked rotor torque vs. dc current.
3.5 PERFORMANCE AND EFFICIENCY EVALUATIONS
This section provides information about the performance and efficiency evaluations of the 2007 Camry
hybrid subsystem components. The motor and inverter characteristics were evaluated simultaneously with
power provided directly to the dc link as opposed to the power being supplied through the boost
converter. Therefore, the 33 kW power rating of the inverter did not limit the capabilities of the motor and
65
inverter during the performance and efficiency assessment process. The dc-link voltage was maintained at
650 Vdc and a switching frequency of 10 kHz was used. Motor efficiency was measured throughout most
of the entire torque-speed range in which the motor is capable of operating. For each operation point, the
motor was controlled in steady state for at least 20 seconds and at least 5 data points were averaged to
generate the efficiency map in Fig. 3.17.
Fig. 3.17. 2007 Camry motor efficiency contours.
Efficiencies of the Camry motor are above 90% for a large portion of the torque-speed operation region.
The high speed or high torque regions tend to incorporate lower operation efficiencies. To provide a
straightforward comparison with the Prius motor efficiency map, the torque and speeds were scaled by
2.47 appropriately, as shown in Fig. 3.18. A comparison of the scaled Camry efficiency map and the Prius
efficiency map in Fig. 3.19 reveals that the Camry has higher efficiencies for much of the map. Similar
scales, axis limitations, percentage ranges, and color schemes are used to further ease the comparison.
Note that a portion of the Camry high torque region is not visible. The most significant improvement over
the Prius is observed at low speeds, where the Prius efficiencies drop quickly as torque increases and
speed approaches zero. This region is frequented often since the electric motor is utilized to develop the
torque required to accelerate the vehicle at low speeds or from standstill.
66
Fig. 3.18. 2007 Camry motor efficiency contours scaled by 2.47.
Fig. 3.19. 2004 Prius motor efficiency contours.
Thermal limitations hindered the extent of the Camry motor evaluations at low speeds and high torques,
where steady state testing was limited to about 210 Nm in order to prevent over-temperature conditions. It
has been verified through dynamic, short duration tests that the motor is capable of producing the
published rated torque of 270 Nm, yet for only a very short time duration of a few seconds. According to
the Camry service manual, the operation temperature of the motor is typically below 90ºC. With this
limitation, the extent of maximum torque of 270 Nm is much shorter as temperature limitations were set
at about 150ºC during the tests. The maximum power capability is limited by voltage constraints to about
67
70 kW at 4500 rpm. This matches the expected power rating based on the size of the generator inverter
and boost converter. Estimated power rating techniques based on properties such as the mass, volume,
speed, and thermal characteristics also result in similar power ratings for this motor. It is assumed that the
published 105 kW power rating of the transaxle is the sum of the motor and generator power ratings as no
rating is published for the generator.
Similar to the process used with the Camry motor efficiency contours, the Camry inverter efficiency
contours were generated and scaled by the high speed gear ratio of 2.47 in order to provide a
straightforward comparison with the Prius. A comparison of the inverter efficiency maps, shown in
Figs. 3.20 and 3.21, indicates that the efficiency of the Camry inverter is somewhat lower than the
efficiency of the Prius inverter for speeds below 2000 rpm. Similar scales, axis limitations, percentage
ranges, and color schemes are used in these efficiency maps to further ease the comparison. The low
speed, high torque capabilities of the Camry inverter appear to be greater than that of the Prius inverter as
the speed reduction gear facilitates the use of a much lower current to generate the same amount of
equivalent torque on the low speed side of the speed reduction gear. For low torques above 2000 rpm, the
Camry inverter efficiency contours indicate that the efficiency decreases slightly as the torque approaches
zero, whereas the Prius inverter efficiency remains consistent beyond 2000 pm. Higher losses of the
Camry inverter is possibly due to the 30% increase of the dc-link voltage and the 50% increase in the
number of power electronic devices when compared to that of the Prius.
Fig. 3.20. 2007 Camry inverter efficiency contours scaled by 2.47.
68
Fig. 3.21. 2004 Prius inverter efficiency contours.
The lower inverter efficiencies of the Camry cause the combined motor/inverter efficiencies to be
somewhat closer than the motor efficiencies alone, as indicated in Figs. 3.22 and 3.23. Although there are
striking similarities between the two efficiency contour maps, benefits of the high speed motor in the
Camry are evident for the low speed, high torque regions.
Fig. 3.22. 2007 Camry motor/inverter efficiency contours.
69
Fig. 3.23. 2004 Prius motor/inverter efficiency contours.
To further explore the various influences on inverter efficiency, tests were conducted with various dc-link
voltages and inverter switching frequencies. In Fig. 3.24, a constant operation condition at 1000 rpm and
80 Nm was maintained throughout the tests. As expected, the inverter efficiency decreases with
increasing dc-link voltage and increasing switching frequency. The Camry controller can operate at
switching frequencies of 1.25 kHz, 2.5 kHz, 5 kHz, and 10 kHz, yet it operates at 5 kHz for much of the
time when controlled by the MG ECU in vehicle. The impact of these parameters is lower for some
operation conditions and high dc-link voltages are unavoidable at high speed, high torque conditions
wherein back-emf voltages demand a high supply voltage.
98
96
94
Efficiency (%)
92
90
275 Vdc
88
350 Vdc
86
500 Vdc
84
650 Vdc
82
80
0 2 4 6 8 10 12
Inverter Switching Frequency (kHz)
Fig. 3.24. 2007 Camry inverter efficiency vs. switching frequency for various dc-link voltages.
Boost converter efficiency assessments were conducted with various dc-link voltage and current output
levels. The evaluations were conducted as the boost converter supplied the motor inverter, both of which
70
were operated at 5 kHz. The motor speed was maintained at 1000 rpm as the torque was increased to
increase the boost converter current. According to the data from full vehicle tests conducted at ANL, the
input voltage to the boost converter is commonly near 275 Vdc. Camry and Prius boost converter
efficiencies are plotted versus boost converter input current in Figs. 3.25 and 3.26. Curve characteristics
are undoubtedly different between the two, and the Prius boost converter has a slightly higher efficiency
for much of the operation range. Low dc-current limitations were set for initial tests to prevent potential
failures from preventing completion of general tests. Nonetheless, a failure occurred with an output
voltage of 500 Vdc and input current of about 70 A and anticipated tests at 650 Vdc as well as a second
series of tests at 10 kHz were not conducted. The failure was associated with an extremely high ripple
current and the Toyota controller likely uses a switching frequency of 10 kHz in these conditions.
102
100
98
Efficiency (%)
96 275 Vdc
350 Vdc
94 500 Vdc
92
90
88
0 10 20 30 40 50 60 70
Boost Converter Input Current (A DC)
Fig. 3.25. 2007 Camry boost converter efficiency vs. input current for various dc-link voltages.
99.0%
233Vdc Out
98.5% 342Vdc Out
500Vdc Out
98.0%
Efficiency
97.5%
97.0%
96.5%
0.00 20.00 40.00 60.00 80.00 100.00 120.00
Boost Converter Input Current (Amps)
Fig. 3.26. Prius boost converter efficiency vs. input current for various dc-link voltages.
71
3.6 CONTINOUS LOAD DURATION TESTS
The Camry motor was operated at continuous power levels for different speeds and coolant temperatures
to observe the continuous capability of the motor under various conditions. Continuous tests were
conducted at 25, 33.5, and 50 kW at 3000, 5000, and 7000 rpm with coolant temperatures of 20, 25, 50,
and 65ºC. For each case, motor temperatures were allowed to stabilize prior to initiating the continuous
tests and the Bay-Voltex coolant temperature regulation unit was used to regulate the coolant at a desired
temperature.
TCs and thermistors were installed in various locations throughout the transaxle and PCU with locations
indicated in Fig. 3.5. Coolant temperatures are indicated by the traces labeled “To Camry” and “From
Cam.” A thermistor labeled “D: ML 1” was installed within close proximity of the factory thermistor of
the Camry, which is labeled “E: ML 2”. Additionally stator, internal/external case, and inverter
temperatures were monitored and recorded.
Figure 3.27 shows the continuous test results conducted at a power level of 50 kW, a motor speed of
5000 rpm, and a coolant temperature of 50ºC. The traces are labeled and identified by color in the key
provided on the left. Mechanical power, “Pmech,” is represented by a blue trace. The symbol at each
datapoint is not distinguishable in the figure since the time scale is so large and the datapoints are close
together. A table indicates the time duration for the associated stator temperature limitation in the upper
left corner of the figure. For example, with a coolant temperature of 50ºC, the motor can operate at 50 kW
for about 8:05 (minutes:seconds) prior to reaching a stator temperature of 130ºC. If the stator temperature
limit is raised to 140ºC, the motor can operate for about 10:18 under these conditions. Thus, the extent of
continuous operation capabilities depends greatly upon stator temperature limitations.
Fig. 3.27. Camry 50 kW continuous duration test at 5000 rpm with 50ºC coolant.
Stator temperature limitations are set to prevent potential damage within the stator windings and to
prevent demagnetization of the PMs. Long term effects which affect the life of the PMSM must also be
considered when choosing a stator temperature limitation. According to full vehicle tests at ANL, the
Camry stator temperature is generally kept well below 90ºC. The temperatures of the original Camry
72
thermistor and the installed thermistor remained close during the continuous duration tests. Results show
that the original Camry thermistor, “E: ML 2,” is placed in one of the hottest locations of the motor.
Figures 3.28 and 3.29 show the plots from tests conducted with a coolant temperature of 65ºC at power
levels of 50 kW and 25 kW and 7000 rpm and 5000 rpm, respectively. Note that in the latter test the
coolant temperature was reduced after thermal steady state was reached, as indicated in Fig. 3.29.
As expected, a comparison of Figs. 3.27 and 3.28 shows that the time extent of operation at 50 kW
decreases as the test speed changes from 5000–7000 rpm and coolant temperature increases from
50–65ºC. All tests conducted at 5000 rpm produced the most efficient operation and thus the longest
duration of operation at any power level when compared to test results obtained while operating at other
73
speeds and a similar coolant temperature. The highest temperatures stabilized near 130ºC in Fig. 3.29 at
5000 rpm and 25 kW. Note that it took over an hour to reach this temperature. Another continuous
duration graph from tests conducted at 3000 rpm and 33.5 kW with a coolant temperature of 65ºC is
shown in Fig. 3.30.
Fig. 3.30. Camry 33.5 kW continuous duration test at 3000 rpm with 65ºC coolant.
Graphs similar to Fig. 3.30 were generated for each test condition and while they are informative, it is
difficult to fully study the impact of various power levels, motor speeds, coolant temperatures, and stator
temperature limitations upon the extent of time at which the motor can operate under these conditions.
Therefore, the duration times shown in the upper left corner of the previous figures were used to produce
graphs which provide more meaningful comparisons. In Fig. 3.31, the time durations associated with
operation at 25 kW with a 65ºC coolant temperature is graphed versus speed. Each trace represents the
time durations which correlate to a particular stator temperature limit. Through inspection of this graph, it
is more evident that the duration of operation is greater at 5000 rpm than for any other speed. Note again
that in this case, the stator temperature stabilized near 130ºC at 5000 rpm, which is represented by the
upper most trace. Thus, the trace for a temperature of 150ºC would theoretically have a data point that
approaches infinity. Also note that tests under these conditions were discontinued after 30 minutes of
operation at 7000 rpm and therefore stator temperatures did not reach higher than 120ºC.
Similar graphs are shown in Fig. 3.32 for a power level of 33.5 kW and a coolant temperature of 35ºC. At
5000 rpm, the motor operated for over 30 minutes before reaching a stator temperature of 125ºC. For low
stator temperature limits, the duration capability does not change significantly as the motor speed
increases from 5000–7000 rpm.
74
Fig. 3.31. Camry 25 kW continuous duration vs. speed with 65ºC coolant.
Fig. 3.32. Camry 33.5 kW continuous duration vs. speed with 35ºC coolant.
Similar graphs are shown in Figs. 3.33 and 3. 34 for continuous tests conducted at 50 kW with coolant
temperatures of 35ºC and 65ºC, respectively. The trace color and stator temperature limitation
correlations are consistent for all of the graphs to provide ease in comparisons. The impact of coolant
temperature is quite visible as the time for stator temperatures to reach 135ºC at 5000 rpm is about
13 minutes with a 35ºC coolant temperature versus about 8 minutes with a 65ºC coolant temperature.
75
The previous graphs provided a clear indication of the influence of speed and coolant temperature upon
duration capabilities. To provide a more clear assessment of the impact of stator temperature limitations
and coolant temperature upon duration capability, duration time is plotted versus the stator temperature
limit in Fig. 3.35 for operations at 3000 rpm. Each trace represents a particular power level and coolant
temperature combination. In studying the figure, it is evident that coolant temperature does not affect
operation duration greatly at 33.5 kW, yet coolant temperature greatly affects operation duration at a
power level of 25 kW.
76
Fig. 3.35. Camry continuous duration at 3000 rpm with various power levels and coolant temperatures.
Characteristics similar to what are observed in Fig. 3.35 are observed in Fig. 3.36 where duration time is
plotted versus stator temperature limit at 5000 rpm for various power and coolant temperature
combinations. As the power level increases, the impact of coolant temperature upon duration decreases.
For tests wherein thermal steady state was nearly reached, the curves approach a vertical asymptote which
is located at the steady state stator temperature.
77
A graph of duration versus stator temperature limit at 7000 rpm with various power and coolant
temperature combinations is shown in Fig. 3.37. Again, with increasing power, coolant temperature has a
decreasing effect upon the duration of motor operation under these conditions. Note that all power-
coolant temperature combinations and color correlations do not remain the same among Figs. 3.35, 3.36,
and 3.37. A comparison of these figures indicates that coolant temperature has a greater impact upon the
duration of operation at higher speeds. For example, the duration of operation is influenced more by
coolant temperature at 7000 rpm and 25 kW than it is for 25 kW power levels at 3000 rpm and 5000 rpm.
Note that in some of the tests in which lower coolant temperatures were used, the duration of operation
was longer than that of a lower power level with a higher coolant level. This occurred because system
temperatures were allowed to stabilize at each coolant temperature. For example, the duration of the
33.5 kW and 35ºC coolant temperature test condition is longer than that with the 25 kW and 65ºC coolant
temperature test conditions if the stator temperature limitation is set at a low temperature. However, if the
stator temperature limitation is set at a high level, the motor can operate longer at the 25 kW and 65ºC
coolant temperature test condition. Thus, with 33.5 kW operation and a 35ºC coolant temperature, the
entire system begins at a cooler temperature, yet the entire system temperature increases more quickly and
eventually becomes hotter than if the system were operating at 25 kW operation with 65ºC, even though
the system begins at a higher temperature (65ºC).
It is clear that defining the continuous operation capability of a motor depends on many variables, which
even includes the definition of continuous operation. For example, continuous operation could be defined
to be operation at a particular power level for an infinite amount of time, or a particular time restraint
could be included in the definition. For some HEV applications, it is crucial that the motor is capable of
operating continuously within particular power levels. For example, the Camry hybrid system requires
torque from the motor and generator for the engine to supply power to the drive wheels. Many aspects
must be considered for these conditions, and if properly designed, the volume and mass of the electric
drive system can be optimized to match the demands of the system.
78
4. SUMMARY AND CONCLUSIONS
The 2007 Camry HEV subsystem assemblies were physically evaluated and comprehensively tested in
the laboratory to fully assess their performance, efficiency, design, and packaging characteristics. The
laboratory evaluations included back-emf, rotational loss, locked rotor, efficiency mapping, and extended-
duration load tests. Overall, the Camry motor outperformed the Prius motor in terms of efficiency and
equivalent torque and power, while the mass and volume of the system decreased significantly. Increased
operation speed of the motor is the primary factor which provides the enhanced characteristics of the
hybrid system. Additionally, the improved packaging and trench gate structure used in the PCU IGBTs
led to a higher power density of the PCU components. The Camry inverter efficiencies are slightly lower
than those of the Prius, yet the combined motor and inverter efficiencies of the Camry are still higher,
especially for the low speed, high torque operation region. The lower motor efficiencies of the Prius have
a much higher impact on the combined motor and inverter efficiencies than the impact of the slightly
lower efficiencies of the Camry inverter.
Evaluations illustrate the benefits of moving to a high speed motor through improved packaging, higher
efficiencies, increased performance, and improved continuous operation test results. Motor efficiencies
are above 90% for a great portion of the operation range. Low speed efficiencies of the Camry are much
higher than those of the Prius, which is particularly noticeable when the speed reduction gear ratio of 2.47
is applied to the efficiency map, and therefore the resulting torque is increased significantly. The peak
power of the primary Camry motor is about 70 kW at 5000 rpm, which is much lower that the published
rating of 105 kW. There are no specifications published for the generator and the published power rating
may be for both the motor and generator. Simulations of this motor and a comparison of specific power
characteristics with other high speed motors also suggest that the power rating is near 70 kW.
Continuous duration varies significantly with speed, specified stator temperature limit, and coolant
temperature. A power level of 33.5 kW was maintained at 5000 rpm for about 30 minutes with 65°C
coolant, at which a stator temperature of 140°C was reached. There is no standard for establishing
continuous or peak power rating specifications for motors designed for HEV applications such as the
hybrid Camry and Prius motors. For example, the Camry PMSM is able to sustain a power level of 50 kW
at 5000 rpm for about 13 minutes with a stator temperature limit of 135ºC and coolant temperature of
35ºC; however, the duration is only about 8 minutes if a coolant temperature of 65ºC is applied. In
addition to the unexpectedly low motor power rating, this highlights a very important reason for
performing benchmarking tests on HEV subsystems – current technology must be verified objectively
under consistent operating parameters before the results are used by the FCVT program and researchers.
The influence of HEV specifications on technical goals and program planning would be drastically
different if unclear published specifications of HEV systems were used as a baseline.
Among the vast amount of information obtained during benchmarking efforts, the significant findings are
summarized in Table 4.1. Some of the information obtained from the Prius and Camry benchmarking
studies has been and will be used by researchers to improve/verify analytical models. As a next step in the
benchmarking efforts at ORNL, focus will be placed on an HEV system with improved power
capabilities. These efforts are expected to be performed in FY2008 and will focus on the Lexus LS 600h
hybrid subsystems, which are expected to include significant design improvements. Technological
advances which facilitate higher power capability, power density, specific power, efficiency, and cost
effectiveness of HEV components are essential as HEVs become increasingly dependant upon the
electrical portion of the drive system and as the electric vehicle (EV) and HEV market continues to
expand in the future.
79
Table 4.1. 2007 Camry/2004 Prius design comparison highlights

Transaxle
Published Camry power rating
Motor power rating. 70 kW 50 kW is likely a combination of motor
and generator ratings.
Motor lamination stack length. 6.07 cm (2.39″) 8.40 cm (3.3″) Reduced by 2.33 cm (0.91″).
Motor mass. 41.7 kg 45.0 kg 7.3% reduction.
Motor volume. 14.8 L 15.4 L 3.9% reduction.
Motor specific power. 1.68 kW/kg 1.11 kW/kg Improved by a factor of 1.5.
Motor power density. 4.73 kW/L 3.25 kW/L Improved by a factor of 1.5.
Camry torque value from speed
Equivalent torque rating. 667 Nm 400 Nm
reduction gear (short duration).
Motor speed rating. 14,000 rpm 6,000 rpm
Motor winding configuration. Parallel Series
Power split.
PCU
Up to 20% higher during
Nominal battery voltage. 244.8 V 201.6 V
operation.
Maximum dc-link voltage. 650 Vdc 500Vdc
Boost converter power rating. 30 kW 20 kW
500 Vdc, 600 Vdc,
Filter capacitor (battery level).
378 μF 282 μF
Smoothing capacitor 750 Vdc, 600 Vdc,
(boosted level). 2098 μF 1,130 μF
Entire PCU mass. 17.86 kg 21.2 kg 15.8% reduction.
Prius PCU includes 12V dc-dc
Entire PCU volume. 11.7 L 17.8 L and compressor inverter, Camry
PCU does not.
Motor inverter mass. 7.5 kg 8.8 kg 14.7% reduction.
Motor inverter volume. ~6 L 8.7 L 31% reduction.
Motor inverter specific power. 9.3 kW/kg 5.7 kW/kg Improved by a factor of 1.6.
Motor inverter power density. 11.7 kW/L 5.7 kW/L Improved by a factor of 2.1.
Boost converter mass. 6.6 kg 4.8 kg 37.5% increase.
Boost converter volume. 3.5 L 5.1 L 31.3% reduction.
Boost converter specific
4.5 kW/ kg 4.2 kW/kg Improved by 7%.
power.
Boost converter power density. 8.6 kW/L 3.9 kW/L Improved by a factor of 2.21.
80
REFERENCES
1. Source: http://www.eere.energy.gov/vehiclesandfuels/technologies/systems/index.shtml.
2. J. S. Hsu, S. C. Nelson, P. A. Jallouk, C. W. Ayers, R. H. Wiles, S. L. Campbell, C. L. Coomer,

K. T. Lowe, and T. A. Burress, Report on Toyota Prius Motor Thermal Management, ORNL/TM-
2005/33, UT-Battelle, LLC, Oak Ridge National Laboratory, Oak Ridge, Tennessee, February 2005.
3. R. H. Staunton, C. W. Ayers, J. Chiasson, T. A. Burress, and L. D. Marlino, Evaluation of 2004

Toyota Prius Hybrid Electric Drive System, ORNL/TM-2006-423, UT-Battelle, LLC, Oak Ridge
National Laboratory, Oak Ridge, Tennessee, May 16, 2006.
4. R. H. Staunton, T. A. Burress, and L. D. Marlino, Evaluation of 2005 Honda Accord Hybrid Electric
Drive System, ORNL/TM-2006-535, UT-Battelle, LLC, Oak Ridge National Laboratory, Oak Ridge,
Tennessee, September 2006.
5. Camry Hybrid Vehicle New Car Features 2007, Toyota Motor Corporation, March 2006.
81
APPENDIX A: 2004 PRIUS POWER CONVERTER UNIT
PACKAGING ASSESSMENTS
2004 Prius Motor Inverter Mass and Volume Assessment
Total Volume of PCU casing is 17.8 L as depicted in Fig. A.1.
Fig. A.1. Volume of entire Prius PCU assembly.
Total volume of “inverter only” is 8.7 L and is derived as follows:
• Mid section volume is reduced by 55% after the exclusion of the volume associated with the
converter components, the generator portion of the inverter board (a 1/2 reduction), the generator
portion of the power module (a 1/3 reduction) and the corresponding portion of the cold plate (a 60%
reduction). See Fig. A.2 as an aid in visualization.
82
Fig. A.2. Empty inverter/converter housing showing cold plate surfaces.
• It is also reduced by the deletion of the bottom compartment (unrelated circuitry).

• The lid and capacitor volume, although not optimized, is scaled by a factor of 1130 μF/(282 μF +
1130 μF) = 4/5 to account for the separation of the boost converter and inverter capacitors.
Total mass of PCU casing as depicted in Fig. A.1 is 21.2 kg.

Total mass of “inverter only” is 8.8 kg and is derived in a similar manner that volume was derived (80%
of capacitor mass, 45% of mid-section mass, and 0% of bottom section mass). The reduction of mid-
section mass is actually more severe than the “45%” suggests since the high-mass converter transformer
and converter power module were first fully excluded.
2004 Prius Bi-Directional Boost Converter Mass and Volume Assessment
The general circuit diagram for the converter is shown in Fig. A.3, showing that the converter topology is
that of a conventional bidirectional boost converter. The figure also supplies mass and volume figures of
the inductor and IPM.
Fig. A.3. General circuit schematic of the Prius inverter.
83
During periods of maximum power demand from the driver, the converter provides up to ~20 kW to the
inverter/motor system and the vehicle generator supplies up to ~30 kW. This power split allows a
reduction in size of the high-cost HV battery and the converter, while the extent of fully electric operation
is compromised.
Excluding the cold plate, the volume and mass totals for the ~20 kW boost converter IPM and inductor
are ~1.12 L and 3.23 kg, respectively.
Volume calculation approach A: The volume of the cold plate under the 2 converter components
(including adjacent mounting holes) is 19 mm × 100 mm × 250 mm = 0.48 L. However, this approach
ignores the space enclosed by the converter portion of the casing.
Volume calculation approach B: It is more pertinent to consider the entire volume that includes the
cold plate, items mounted on it, and surrounding space enclosed in the casing. This volume is 335 mm ×
100 mm × 140 mm = 4.7 L.
It is estimated that 15% of the casing mass is dedicated to the bidirectional converter. Since the casing
shell is 6.45 kg, the mass reduction is ~1 kg.
Capcitor: The entire capacitor volume is 1.8 L. The volume associated with the boost converter is
approximately 1.8 × 282 μF/(282 μF + 1130 μF) = 1.8/5 = 0.36 L. The capacitor mass is approximately
2.79 kg /5 = 0.56 kg.
Conclusions: Accounting for the inductor, IPM, casing, and capacitor the converter system volume is
5.1 L and the mass is 4.8 kg.
84
APPENDIX B: PRIUS AND CAMRY CAPACITOR TEST RESULTS
-40 Celsius 20 Celsius 120 Celsius

800.00
700.00
600.00
Capacitance (uF)
500.00
400.00
300.00
200.00
100.00
0.00
0 2000 4000 6000 8000 10000 12000 14000 16000
Frequency (Hz)
Fig. B.1. Prius single 141 μF capacitor from 1,130 μF module, capacitance vs. frequency.

6.00
5.00
ESR (MilliOhms)
4.00
3.00
2.00
1.00
0.00
0 2000 4000 6000 8000 10000 12000 14000 16000
Frequency (Hz)
Fig. B.2. Prius single 141 μF capacitor from 1,130 μF module, ESR vs. frequency.
85
30.00
25.00
20.00
15.00
10.00
5.00
0.00
0 2000 4000 6000 8000 10000 12000 14000 16000
Frequency (Hz)
Fig. B.3. Prius single 141 μF capacitor from 1,130 μF module, DF vs. frequency.
92.00 10.0
Capacitance(uF)
ESR(mΩ) 9.0
8.0
90.00

7.0
6.0
Capacitance (uF)
89.00
5.0
88.00
4.0
3.0
87.00
2.0
86.00
1.0
85.00 0.0
-40 -20 0 20 40 60 80 100 120 140 160
Temp (Celsius)
Fig. B.4. Camry single 86 μF capacitor, capacitance, ESR, and DF vs. temperature at 10 kHz.
86
116.00 16.0
Capacitance(uF)
ESR(mΩ)
115.00 Dissipation (%) 14.0
114.00
12.0

113.00
10.0
Capacitance (uF)
112.00
8.0
111.00
6.0
110.00
4.0
109.00
108.00 2.0
107.00 0.0
-40 -20 0 20 40 60 80 100 120 140 160
Temp (Celsius)
136.00 25.0
Capacitance(uF)
134.00 ESR(mΩ)
Dissipation (%) 20.0

132.00
15.0
Capacitance (uF)
130.00
10.0
128.00
5.0
126.00
124.00 0.0
-40 -20 0 20 40 60 80 100 120 140 160
Temp (Celsius)
87
180.00 12.0
160.00
Capacitance(uF)
ESR(mΩ) 10.0

120.00 8.0
Capacitance (uF)
100.00
6.0
80.00
60.00 4.0
40.00
2.0
20.00
0.00 0.0
0 5000 10000 15000 20000 25000 30000 35000 40000
Frequency (Hz)
Fig. B.7. Camry single 86 μF capacitor, capacitance, ESR, and DF vs. frequency at -40ºC.
145.00 5.00
Capacitance (uF) ESR (MilliOhms) Dissipation Factor (%)

144.00 4.50
143.00 4.00
ESR (MilliOhms), Dissipation Factor (%).

142.00 3.50
141.00 3.00
Capacitance (uF)
140.00 2.50
139.00 2.00
138.00 1.50
137.00 1.00
136.00 0.50
135.00 0.00
-50 -30 -10 10 30 50 70 90 110 130 150
Temp (Celsius)
Fig. B.8. Prius single 141 μF capacitor, capacitance, ESR, and DF vs. temperature at 1 kHz.
88
160.00 5.00
159.00 4.50
158.00 4.00

157.00 3.50
Capacitance (uF)
156.00 3.00
155.00 2.50
154.00 2.00
153.00 1.50
152.00 1.00
151.00 0.50
150.00 0.00
-50 -30 -10 10 30 50 70 90 110 130 150
Temp (Celsius)

750.00 30.00
725.00 25.00

700.00 20.00
Capacitance (uF)
675.00 15.00
650.00 10.00
625.00 5.00
600.00 0.00
-50 -30 -10 10 30 50 70 90 110 130 150
Temp (Celsius)
89
APPENDIX C: 2004 PRIUS TRANSAXLE PACKAGING ASSESSMENTS
2004 Prius Motor Mass and Volume Assessment
The configuration of the motor rotor and stator are shown in Fig. C.1.
Fig. C.1. 2004 Prius motor rotor and stator.
Figure C.1 shows three views of the PMSM housing with selected dimensions called out. As indicated,
about 3/4th of the casing houses the PMSM rotor and stator and the remainder encloses part of the
gearbox. The portion of the casing that houses the PMSM is roughly cylindrical and that is the geometry
used for the purposes of estimating the PMSM casing volume. Based on the average of 3 measurements
of casing diameter, the cylinder diameter can be assumed to be ~29.9 cm. The depth of the PMSM
housing extending from the end plate to the surface identified in the upper left photo in Fig. C.2 is
20.5 cm. This results in a volume of 14,400 cm3 excluding the three-phase terminal block housing and the
cooling passages that protrude from the surface of the casing. Including all 3 volumes results in a total of
15,400 cm3 (15.4 L). This volume and the 50 kW peak power specification result in a peak power density
of 3.25 kW/L.
The mass of the PMSM was determined in order to estimate the specific power. The masses of the
components of the Prius PMSM are:
Stator: 25.9 kg
Rotor: 10.2 kg
Case: 6.36 kg (machined – see below)
Case cover: 2.49 kg
Total mass of motor: 44.95 kg
The casing for the PMSM has (1) a portion that encloses one end of the gear box, and (2) compartments
that enclose the gear-shift-lever-to-shift-plunger linkages. These areas are unrelated to the PMSM and
were therefore machined off of the structure to obtain a new 6.36 kg mass (the original casing mass was
13.9 kg). The resulting specific power for the PMSM is 1.11 kW/kg.
90
Fig. C.2. Casing of the 2004 Prius PMSM with dimensions and volume calculations.
91
DISTRIBUTION
Internal
1. D. J. Adams 8. J. S. Hsu
2. T. A. Burress 9. L. D. Marlino
3. S. L. Campbell 10. M. Olszewski
4. C. L. Coomer 11. L. E. Seiber
5. J. P. Cunningham 12. R. H. Staunton
6. E. C. Fox 13. Laboratory Records
7. K. P. Gambrell
External
14. F. D. Barlow III, University of Idaho, Department of Electrical & Computer Engineering,
Buchanan Engineering Laboratory, Rm. 211, 607 Urquhart Avenue, Moscow, ID 83844-1023,
fbarlow@uidaho.edu.
15. R. Al-Attar, DCX, raa9@dcx.com.
16. S. J. Boyd, U.S. Department of Energy, EE-2G/Forrestal Building, 1000 Independence Avenue,
S.W., Washington, D.C. 20585.
17. T. Q. Duong, U.S. Department of Energy, EE-2G/Forrestal Building, 1000 Independence
Avenue, S.W., Washington, D.C. 20585.
18. R. R. Fessler, BIZTEK Consulting, Inc., 820 Roslyn Place, Evanston, Illinois 60201-1724.
19. G. Hagey, Sentech, Inc., 501 Randolph St., Williamsburg, Virginia 23185.
20. E. Jih, Ford Motor Company, Scientific Research Laboratory, 2101 Village Road, MD-1170, Rm.
2331, Dearborn, Michigan 48121.
21. K. J. Kelly, National Renewable Energy Laboratory, 1617 Cole Boulevard, Golden, Colorado
80401.
22. A. Lee, Daimler Chrysler, CIMS 484-08-06, 800 Chrysler Drive, Auburn Hills, Michigan 48326-
2757.
23. F. Leonardi, Ford Motor Company, 15050 Commerce Drive, North, Dearborn, Michigan 48120-
1261.
24. F. Liang, Ford Motor Company, Scientific Research Laboratory, 2101 Village Road, MD1170,
Rm. 2331/SRL, Dearborn, Michigan 48121.
25. M. W. Lloyd, Energetics, Inc., 7164 Columbia Gateway Drive, Columbia, Maryland 21046.
26. J. Maquire, General Motors Advanced Technology Center, 3050 Lomita Boulevard, Torrance,
California 90505.
27. M. Mehall, Ford Motor Company, Scientific Research Laboratory, 2101 Village Road, MD-2247,
Rm. 3317, Dearborn, Michigan 48124-2053.
28. J. A. Montemarano, Naval Surface Warfare Center, Carderock Division; Code 642, NSWD, 9500
MacArthur Boulevard; West Bethesday, Maryland 20817.
29. N. Olds, United States Council for Automotive Research (USCAR), nolds@uscar.org
30. S. A. Rogers, U.S. Department of Energy, EE-2G/Forrestal Building, 1000 Independence
Avenue, S.W., Washington, D.C. 20585.
31. G. S. Smith, General Motors Advanced Technology Center, 3050 Lomita Boulevard, Torrance,
California 90505.
32. E. J. Wall, U.S. Department of Energy, EE-2G/Forrestal Building, 1000 Independence Avenue,
33. B. Welchko, General Motors Advanced Technology Center, 3050 Lomita Boulevard, Torrance,
California 90505.
34. P. G. Yoshida, U.S. Department of Energy, EE-2G/Forrestal Building, 1000 Independence Avenue,
92

Camry PDF

Uploaded by

Document Informationclick to expand document information

Copyright:

Available Formats

Camry PDF

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Camry PDF

Uploaded by

Copyright:

Available Formats

U.S.

EVALUATION OF THE 2007 TOYOTA CAMRY

Oak Ridge National Laboratory

Mitch Olszewski, Program Manager

Susan A. Rogers, Technology Development Manager

Engineering Science and Technology Division

EVALUATION OF THE 2007 TOYOTA

Revised: April 2008

Web site: http://www.osti.gov/bridge

National Technical Information Service

Office of Scientific and Technical Information

This report was prepared as an account of work sponsored by an agency of

LIST OF FIGURES .....................................................................................................................................iv

ACRONYMS AND ABBREVIATIONS ..................................................................................................viii

2. HYBRID ELECTRIC DRIVE SYSTEM DESIGN AND PACKAGING................................................ 3

3. BENCHMARKING TESTS OF CAMRY PMSM AND INVERTER................................................... 55

4. SUMMARY AND CONCLUSIONS ..................................................................................................... 79

APPENDIX A: 2004 PRIUS POWER CONVERTER UNIT PACKAGING ASSESSMENTS ............... 82

APPENDIX B: PRIUS AND CAMRY CAPACITOR TEST RESULTS .................................................. 85

APPENDIX C: 2004 PRIUS TRANSAXLE PACKAGING ASSESSMENTS......................................... 90

1.1 Flow diagram of baseline benchmarking approach........................................................................ 2

3.12 2007 Camry motor rotational losses............................................................................................. 63

A/C air conditioning ID inner diameter

Fig. 1.1. Flow diagram of baseline benchmarking approach.

Design Feature Hybrid Camry 2004 Prius

Fig. 2.1. Subsystems of the Camry inverter and converter assembly.

Design Feature Hybrid Camry 2004 Prius

Item Mass (kg) Volume (L) Dimensions (cm)

Parameter Camry Prius

Fig. 2.8. Non-optimized packaging used in the 2004 Prius converter.

Fig. 2.9. Comparison of all capacitors located within Camry PCU.

2.1.1 Power Electronics Devices

Fig. 2.15. Camry IPM with dc-bus bars removed.

2.1.2 Capacitor Test Results

Fig. 2.22. Camry capacitor module equivalent capacitance vs. frequency.

Fig. 2.23. Prius capacitor module equivalent capacitance vs. frequency.

Fig. 2.25. Camry capacitor module ESR vs. frequency.

Fig. 2.26. Prius capacitor module ESR vs. frequency.

Fig. 2.27. Camry capacitor module ESR vs. temperature.

Fig. 2.29. Camry capacitor module DF frequency response.

Fig. 2.31. Camry capacitor module DF temperature response.

Fig. 2.32. Prius capacitor module DF temperature response.

Fig. 2.33. Ripple current test schematic.

Fig. 2.34. TCs on the Camry capacitor module.

Ripple Current (Amps)

Fig. 2.37. Camry single 86 μF capacitor, capacitance vs. frequency.

Fig. 2.39. Camry single 86 μF capacitor, DF vs. frequency.

Dissipation Factor (%), ESR(mΩ)

Dissipation Factor (%), ESR(mΩ)

Dissipation Factor (%), ESR(mΩ)

Dissipation Factor (%), ESR(mΩ)

Dissipation Factor (%), ESR(mΩ)

Voltage (Vrms), Current (Arms)

Voltage (Vrms), Current (Arms)

Resistance (mΩ), DF (%)

Resistance (mOhms), DF (%)

2.1.2.6 Static capacitor test results of small dc-link capacitor evaluations

ESR (mΩ), DF (%)