PAR Page 1 of 22

Project no. FP6-019848

Project acronym: HOPE

Project title: High Density Power Electronics for FC- and ICE-Hybrid Electric Vehicle Powertrains

Thematic Priority: 4- Aeronautics and Space 6.1.ii Sustainable Energy Systems 6.2 Sustainable Surface Transport

Publishable final activity report Period covered: from 1.1.2006 to 31.12.2008 Date of preparation: 14.2.2008 Start date of project: 1.1.2006 Duration: 36 month Project coordinator name: Dr. Kai Kriegel Project coordinator organisation name: Siemens AG Final

PAR Page 2 of 22

Content: 0. Glossary 1. Project execution 1.1 Summary description of project objectives 1.2 Contractors involved 1.3 Work performed and end results 1.3.1 Common specifications and standardization evaluation 1.3.2 Mission profiles and reliability assessment

i) Mission profile definition ii) Translation of drive cycles to mission profiles iii) Accelerated power cycling tests, accelerated mission profile tests iv) Summary mission profiles and accelerated testing

1.3.3 Key technologies for high power density electronics i) Agreement on requirements ii) Characterization and selection of active, passive, and sensor

devices and components iii) Substrates, and interconnect technologies for high temperature Si-

PEBBs, Si- IML and control boards iv) Evaluation of cooling concepts v) Reliability testing of components and interconnects vi) High density, ultra low communication cell with SiC devices and

integrated passives vii) Summary in key technologies

1.3.4 Power electronics building blocks and inverters Summary power electronic building blocks PEBB and inverter

1.3.5 High temperature control board Summary High Temperature Control Board

1.3.6 Perspective of inverter integration into powertrain 1.4 Assessment of the three main goals of HOPE – performance, low cost, and reliability 2. Dissemination and its use 2.1 HOPE Web site 2.2 Conferences 2.3 Publishable results 2.4 Patents applied

34466677889999

10

111113

13141616171718

1919192021

PAR Page 3 of 22

Glossary Accelerated test For testing the end-of-life of electronics, accelerated test procedures are necessary.

The most common is to test at elevated temperatures and extrapolate to operating temperatures

ARTEMIS FP5 project “Assessment and Reliability of Transport Emission Models and Inventory Systems”

DCB Direct copper bonding – without solder - on ceramic substrate which acts as insulator

DC-link Interconnects (bus bars) which carry the direct current Driver circuit In electronics, a driver is an electrical circuit to control another circuit or

other component, such as a high-power transistor. Heat pipe A heat pipe is a heat transfer mechanism that can transport large quantities

of heat with a very small difference in temperature between the hotter and colder interfaces

IGBT Insulated Gate Bipolar Transistor IML Inserted molded leadframe Inverter An inverter is an electrical device that converts direct current (DC) to

alternating current (AC) Junction temperature

Junction temperature is the highest temperature of the actual semiconductor in an electronic device

LTJ Low temperature joining by using silver particles, moderate temperatures (~230°C), and high pressure to join e.g. a Si chip with a DCB. This interconnect technology is also named silver sintering

Mission profile All the stresses which are applied to the powertrain during operation of a vehicle Mechatronics Mechatronics (or Mechanical and Electronics Engineering) is the

combination of mechanical engineering, electronic engineering and computer engineering.

MOSFET The metal–oxide–semiconductor field-effect transistor (MOSFET) is a device used to amplify or switch electronic signals

PCB Printed circuit board consisting of a glass fibre meshwork, epoxy material and copper traces and vias

PEBB Power electronics building block: This is a functional unit with a certain power rating which can easily put together to achieve higher power ratings

Power cycling Reliability test method where a semiconductor chip is turned on and off many times to produce alternate thermal losses. This procedure is mainly used for testing the wire bond reliability

Powertrain Drive unit in a vehicle consisting of an electric machine and a converter which provides the sinusoidal electric current by pulse width modulation

Reliability The ability of an item to perform a required function under given conditions for a given time interval (IEC 50(191): 1990)

Ribbon bond Is recently used to interconnect chips electrically by a ribbon instead of a wire Robustness Validation

Process for validation of electronics which is based on the physics-of-failure and end-of-life approach

SAE Society of Automotive Engineers, USA SiC Silicon carbide single crystals are wide band gap semiconductors which can

withstand higher voltages and temperatures in contrast to silicon Thermosyphon A liquid is evaporated at the hot part of an electronics. It condenses at the cold

part of the system and is brought back as a liquid to the hot part by gravity

PAR Page 4 of 22

1. Project execution 1.1 Summary description of project objectives

According to the “Joint Call on component development and system integration of hydrogen and fuel cells for transportation and other applications” the main aim is to develop generic technology and modular systems, built up from components that can be manufactured in essentially similar configurations, but with different qualities, to meet the specific performance, lifetime and cost requirements of the different applications (e.g. FC stacks, Membrane Electrode Assemblies, batteries, power electronics)”. Since HOPE was started three years ago, the global warming issue was more and more spotlighted. Recently the UN intergovernmental panel on climate change IPCC published the IPCC FOURTH ASSESSMENT REPORT, CLIMATE CHANGE 2007 (http://www.ipcc.ch/graphics/graphics.htm). This report has drawn a lot of attention by the commission and the media as well. As a consequence on December 19, 2007, the European Union as well as the US have made very important actions to declare activities against the global warming: • On 19 December 2007, the European Commission adopted a proposal for legislation to reduce the

average CO2 emissions of new passenger cars which account for about 12% of the European Union's carbon emissions. The proposed legislation is the cornerstone of the EU's strategy to improve the fuel economy of cars and ensure that average emissions from the new passenger car fleet in the Community do not exceed 120 g CO2/km through an integrated approach.

• In the U.S. the DOE, companies from the oil industry, and carmakers started the project “FreedomCAR” to achieve energy independence. The key to this is the development of hybrid electric and fuel cell vehicles that are economically justifiable for the average consumer.

The project HOPE is addressing power electronics. It is based on previous EU research projects like the recently finished FP5 HIMRATE (high-temperature power modules), FP5 PROCURE (high-temperature passive components), and MEDEA+ HOTCAR (high-temperature control electronics) and other EU and national research projects.

Fig 1 The acceptance of HEVs and FCEVs largely depends on the costs of the power train.

PAR Page 5 of 22

The aspects of vehicle electrification are to be seen in Fig 1. CO2 reduction means, however, additional costs. The general objectives of HOPE are:

• cost reduction • meet reliability requirements • reduction of volume • reduction of weight

This is a necessity to bring the FC- and ICE-hybrid vehicles to success. In Fig 2 the concept of two alternative power electronic building blocks PEBB is shown:

i) The insert molded leadframe aims for a low-cost technology omitting the costly insulating ceramic substrate

ii) The SiC-PEBB aims for high-temperature operation to gain an additional 100 K temperature margin for cooling the SiC semiconductor device. This can be used to apply a passive cooling instead of an active liquid cooling.

The idea of using standardised building blocks is to develop a construction kit for user defined power ratings.

280260240220200180160140120100

8060

-40

Tem

pera

ture

[°C

]

3kW Inverter withSiC PEBB,

Integrated Passives,Driver,ECU

Additional 100K temperature margin

Max. vehiclecoolanttemperature

IMLPower Module

VESL SCT

280260240220200180160140120100

8060

-40

Tem

pera

ture

[°C

]

3kW Inverter withSiC PEBB,

Integrated Passives,Driver,ECU

Additional 100K temperature margin

Max. vehiclecoolanttemperature

IMLPower Module

VESL SCT

Fig 2 Two alternative concepts for power electronic building blocks PEBB by Valeo (VESL) and Siemens.

PAR Page 6 of 22

1.2 Contractors involved The partners in the HOPE consortium remained the same over the whole project duration of the project, although the legal status of some of the partners has changed. Participant no. Participant name Participant short name Country

1 Siemens CT SCT D

2 Daimler DC D

3 ETH Zurich ETHZ CH

4 Fraunhofer IISB FHG-IISB D

5 INRETS INRETS F

6 Magna Steyr MSF A

7 Renault Renault F

8 Robert Bosch BOSCH D

9 Continental Automotive Conti D

10 Valeo Valeo F

11 Volkswagen VW D

12 University of Technology Belfort-Montbéliard UTBM F

13 Warsaw University of Technology WUT Pl

The HOPE project was partitioned into eight workpackages:

WP Workpackage name:

1 Common specifications and standardisation evaluation 2 Mission profiles and reliability assessment 3 Key technologies for high power density electronics 4 Power electronic building blocks and inverters 5 High temperature control board 6 Perspectives of inverter integration into powertrain 7 Coordination & management incl. review & assessment 8 Dissemination and exploitation

1.3 Work performed and end results

In the following the description and assessment of the corresponds to the six R&D workpackages.

1.3.1 Common specifications and standardization evaluation

defines specifications common to OEM’s for FC- and ICE-hybrid vehicle drive systems; identification of common key parameters (power, voltage, size) that allows consequent standardisation. Developing a scalability matrix for power electronic building blocks PEBBs.

PAR Page 7 of 22

The OEM’s have prepared very early a “common specification from OEM’s”. It contains vehicle architectures, the electro-mechanical architecture, a battery hypothesis, options for the DC bus voltage, a needs matrix, and finally the mid term/ long term view.

Fig 3 Agreement on common specifications for HEV and FCV worked out by the OEMs.

Summary Common specifications and standardization evaluation A need Matrix with common specifications for HEV and FCV had been worked out by the OEMs. The suitability of the power modules developed within HOPE (PEBBs of WP4) for a modular solution satisfying the OEMs` needs of Fig 3 has been assessed at the End of the Project. The IML IGBT modules are well suited for such a modular solution under the assumption that the still running reliability tests will show that the modules are satisfying the reliability requirements. With only two modules we can cover the needs all the considered hybrid and FC vehicles, On the other hand, the SiC modules are not convenient for application in the drive inverters and are expected to be applied in the DC/DC converters. The extremely low switching losses of the SiC semiconductors allow for advantageous high switching frequencies with higher power densities and higher efficiencies.

1.3.2 Mission profiles and reliability assessment

A big study was undertaken to define “load patterns for reliability assessments”: It starts with the current, voltage and power factor versus time for different drive train architectures. Temperature profiles are described next which are based on measurements on existing cars. Applicable test procedures for power electronics systems deal with the very difficult topic “accelerated reliability tests”. It contains firstly the reliability test conditions extracted from the mission profile. Secondly, the test definition is described concerning major test parameters for reliability and expected working conditions, high temperature methodology, failure criteria, test protocol.

FCVParallel Serie - parallel (split) Serie

Battery voltage 125V to 380V 200V to 400V 200V to 400VFuel cell voltage 150V to 400VDrive train MT/AT/RMT iVT reduction gearHigh Voltage DC/DC not needed Step up (10 to 40kW) suitable Step up/down (20 to 40kW) needed

Number of motors 1 motor 2 motors 1 motor Power rating 10 - 30 kW 20 - 50 kW 35 - 80 kW Cooling Water Water and/or direct oil WaterTorque/speed Limited by MT/AT/RMT Limited by iVT Whole vehicule torque/speed rangeCooling Water Water and/or direct oil Watertechnology

Housing integration into other parts

Expensive because Drive train redesign is needed

(already in mass production)Affordable in drive train Affordable in front/rear axle

or wheels

Reduction ratio Discret Continous Fixed

Numbers of inverters 1 inverter 2 in one inverter 1 inverterPulse frequencySwitch technologyDC voltage 125V to 380V 200V to 400V 200V to 400VPhase current 50A to 200A 50A to 400A 100A to 600ADies combination checked at end of project checked at end of project checked at end of projectCapacitor technology foil or electrolytics foil or electrolytics foil or electrolytics

Housing integration into other parts

Expensive because Drive train redesign is needed

(already in mass production)Highly suitable Highly suitable

HEV

Syst

em

10-20 kHz for motor inverter / 50-100 kHz for DC/DC converter

IM or PMM

Pow

er E

lect

roni

cs X ? 600 V breakdown

Mot

or

PAR Page 8 of 22

i) Mission profile definition: The OEMs in the HOPE project – Daimler, Renault and Volkswagen – have agreed on the Artemis mission profile (EU FP5 project). Artemis stands for “Assessment and Reliability of Transport Emission Models and Inventory Systems”. The “Workpackage Road traffic characteristics” deals with emissions from road vehicles which very much on the way they are operated: on the traffic composition, vehicle speeds, road configuration, etc. ARTEMIS provides statistics on road traffic operation in EU and Eastern European countries and examines the effects of local traffic. A drive cycle was defined as a mission profile which is representative for the life of a car. It is the basis for designing reliability tests. The life cycle consists of: 4 jam-, 3 urban- 2 road-, and 1highway-cycle. A “vehicle life time” of 360.000 km corresponds then to 4.412 combined drive cycles. ii) Translation of Drive Cycles to Mission Profiles:

If the vehicle drive cycle is chosen the translation from the mission profile of the vehicle down to the power electronics system, the subsystem, and the components, materials and interconnects has to be made (see Fig 4). This ensures that the real stresses which are causing damage during operation of the vehicle are used for designing reliability tests and to carry out lifetime predictions.Please note that the knowledge of reliability data from supplier companies is essential for the decision if the technology can withstand the required mission profile over the given lifetime.

Fig 4 Translation of the vehicles mission profile to the mission profile at the components and materials level (this means all kinds of stresses which are applied during vehicle operation).

PAR Page 9 of 22

iii) Accelerated Power Cycling Test, Accelerated Mission Profile Test: According to the new Robustness Validation process, end-of-life testing is a major part to figure out the failure mechanisms. For doing this in a reasonable time accelerated tests have to be used. The standard procedure is to use elevated temperatures for testing just to increase the test time. For high-temperature electronics, however, it cannot be recommended to increase test temperatures because materials and components may be damaged. A good way seems to be the accelerated mission profile test. The basic idea is to take the mission profile - e.g. the temperature stresses - at the critical part of the power electronics system and filter out all those parts which are not harmful. Fig 5 discloses results of a power cycling test. The temperature changes are produced by turning-on and turning-off the power semiconductor switch. The ambient cooling temperature is 95°C (lower temperature) and the junction temperature is 175°C caused by current flow and corresponding heat losses. A fter 50.000 power cycles an increase of the VDSsat and Rth values is observed. The state-of-the- art for a temperature swing of 90 K between 60°C and 150°C is just 30.000 cycles. A silver-LTJ die attach yields to an even higher reliability because of the lower junction temperature.

Fig 5 Accelerated reliability test of a SiC-PEBB: Power cycling test between 95°C (lower temperature – left scale) and 180°C (right scale). The failure criterias are almost reached for the thermal resistance Rth and the saturation voltage VDSsat. The results shows that the wirebond reliability is more than two times higher than the state of the art (INRETS). Summary Mission profiles and reliability assessment • The power cycling reliability tests on test modules and PEBBs result in a higher reliability by at

least a factor of two. This is most likely due to a better thermal management of the packaging. • The mission profile tests are a promising way to test the real operating condition of vehicles in an

accelerated way.

1.3.3 Key technologies for high power density electronics

i) Agreement on requirements: The HOPE consortium agreed on the partitioning of the rated power forming power electronic building blocks (PEBB) and footprints. Two examples are shown for the SiC- and the IML- technologies.

PAR Page 10 of 22

ii) Characterization and selection of active, passive, and sensor devices and components: The active Si-devices like IGBTs, MOSFETs, and diodes were investigated regarding their high-temperature behaviour already in detail in the FP5 project “HIMRATE”. It turned out that 600V IGBTs and 80V MOSFETs are capable to operate safely at 200°C. The packaging, however is the real bottleneck because of the die attach and the bond wires. SiC JFET switches are mature power devices operating in a wide temperature range of -40°C to >300°C having small on-state and switching losses in comparison to Si devices. They should be used in such applications where a high voltage capability and a high temperature capability are required. For automotive applications the best use are DC/DC converters where the high operating frequency of SiC JFETs will allow very compact and less costly systems. This may compensate the cost of SiC JFETs which are today 10 times more costly as IGBTs. The newly designed high temperature current sensor is shown in Fig 6. It is based on a Hall sensor with a differential core for a closed loop. The differential core allows a lower current for sensing which keeps all elements below 150°C. Additionally the size of the new component is more compact .

Key Technologies for High Power Density Electronics

HOPE Final Assessment, Dec 17, 2008page 23

Fraunhofer-IISB

Fig 6 One of the highlights of HOPE is the compact high temperature current sensor which is a key component in an FC- and/or HEV- inverter (WUT).

iii) Substrates, housings, and interconnect technologies for high-temperature for SiC-PEBBs, Si-IML, and control boards:

A comprehensive test matrix including several substrates as well as joining technologies was the basis for selecting suitable technologies. In Fig 7 some of the main results are shown. Two substrates still have an outstanding quality and reliability after 1700 temperature cycles: Al2O3 DCB ceramic with dimples along the copper edges and Si3N4 ceramic with active metal brazed copper metallisations. The standard alumina- and AlN- substrates are failing already after 300 cycles completely.

PAR Page 11 of 22

Fig 7 Test matrix for qualifying ceramic substrates with different solders (Bosch, FhGIISB). A new packaging technology has been simulated and tested intensively – the IML package (insert molded leadframe). Fig 8 shows the test package where several technologies, like wire bonding and ribbon bonding are realised. The benefit of such combinational test structures is obvious: The test results can be compared quite easily.

Fig 8 Test package for the IML technology (insert molded leadframe) containing wire bonds as well as ribbon bonds. The IML package does not have an insulation which has to be provided by a insulating layer underneath the package (Valeo).

iv) Evaluation of cooling concepts, thermal and thermo-mechanical analysis:

Two extreme situations for cooling were investigated: Natural air convection as a fully passive cooling and double-side liquid cooling with water-glycol. For exchanging heat to the ambient air it is necessary to have a large temperature difference as well a big area. If the thermal path is considered from the pn-junction as the hottest point to the ambient air, the thermal resistance should be as low as possible. The heat-pipe has a very high thermal conductivity – at least 30 times higher than copper. The idea was to integrate the SiC JFET directly with a heat-

0%

10%

20%

30%

40%

50%

60%

70%

80%

90%

100%

0 500 1000 1500 2000 2500No. of Cycles

Failu

re R

ate

Al2O3 DCB (Curamik)

Al2O3 dimpled DCB (Curamik)

AlN DCB (Curamik)

AlN Cu-AMB (Anceram)

AlN DAB (Denka)

Si3N4 Cu-AMB (Kyocera)

Si3N4 Cu-AMB (Kyocera)

PAR Page 12 of 22

pipe. To make an operation possible in the temperature range from -40°C to 175°C a combination of methanol and water heat-pipes was developed and tested. Figs 9 shows the test setup containing the SiC-PEBB as heat source, the integrated heatpipe as outstanding heat conductor and the array of fins as cooling device. It has to be noted that only natural convection is applied and that the ambient temperature is ~ 130°C. This yields to a rather big area for the fins. The cooling performance of this setup is: 120 W to an ambient of 25°C and 90 W to an ambient of 125°C.

Fig 9 Experimental setup for characterizing the heatpipe. The heatsource consists of two SiC JFETs in parallel which generate heat in a very small volume. The heatspreading is done by an integrated water-methanol heatpipe allowing an operating temperature range from -40°C to 175°C (SCT). v) Reliability testing of components and interconnects:

Virtual performance assessmentCircuit simulation power losses

Thermal simulation Rth, TjEMC simulation emission level

Comparison with data sheet, standards

Virtual reliability assessment

Physics of failure degrading mechanismsPhysical models lifetime as a function of stress

Robustness margins (Derating)

Robustness validation of the PE system

Non anticipated failures

Mission profile tests; max. ratings tests

SOA?

Exis

ting

Kno

wle

dge

RequirementsFunctional performance,Mission profile, Reliability

First design

Bui

ldin

g-in

relia

bilit

yPe

rfor

man

ce

optim

isat

ion

Fig 10 Process flow how to secure reliability. The new process ”Robustness Validation” has been implemented in the middle the HOPE project (SCT).

PAR Page 13 of 22

At the beginning of the HOPE project the joint initiative between the German ZVEI, the SAE (society of American automotive engineers and the Japanese JSAE to work out and standardize a new validation process was not known to the consortium. When it became known an improved strategy to secure reliability was worked out which is based on the Robustness Validation Handbook. This was published in April 2007 by ZVEI. The new process is based on the physics of failure and includes end-of-life testing of components and systems (see Fig. 10). An important part plays life time prediction when new components and/or materials are introduced. Results from case studies applying new materials with well matched coefficients of thermal expansion and high thermal conductivity show that the reliability can be significantly improved. It is necessary, however, to rely on well suited physical models for the damage. Using a model for thick Al wire bond lift-off, the correlation between wire diameter and number of power cycles can be simulated. The result is shown in Fig 11 which concludes that many thin wires are more reliable than one thick wire.

Diameter of thick Al‐wire

Num

ber o

f cycles to failure 

Fig 11 Lifetime prediction for a wire bond lift-off failure. The cause of the failure mechanism is due to the mismatch of the coefficients of thermal expansion CTE which differs almost by a factor of 10 for the Si chip and the Al wire (SCT).

vi) High density, ultra low commutation cell with SiC devices and integrated passives

An in-depth research was done on a high density, ultra low parasitic commutation cell with SiC devices and integrated passives which should show the impact on switching frequency and losses. For application oriented measurements a half bridge DCB layout was realised using a standard single layer

Fig 12 The low parasitic SiC commutation cell with integrated passives is an excellent example for system integration. The gain in performance due to short interconnects is remarkable.

PAR Page 14 of 22

DCB technology, high speed 600 V IGBTs and 600 V SiC diodes. The module is designed for a peak phase current of 30 A at a switching frequency of 100 kHz (Fig 12).

vii) Summary key technologies

In summary progress was made in many of the key technologies like:

• A small closed loop current sensor with low losses and 150°C maximum operation temperature (WUT)

• New technologies for die attach like nano- silver sintering for small chips (Bosch, FHG-IISB)

• The insert molded leadframe package IML designed for mass production is one key product of the development roadmap (Valeo)

• A data sheet for the SiC PEBB is available but there is no industrial standard yet (SCT)

• A new passive cooling using a heatpipe capable allows to operate in a temperature range between -40°C to +175°C (SCT)

• A thermosyphon cooler allows cooling of hot spots at a high-temperature printed circuit board (Conti)

• The improved communication cell for very fast switching contains Si-IGBTs, SiC-diodes, low inductive busbars and integrated passives on a DCB substrate (FhG-IISB)

It can be expected that some of the results will be used in future products while for others the costs are still too high. This may be changed, however, if dedicated systems are evaluated.

1.3.4 Power electronic building blocks and inverters

Within HOPE two PEBBs have been developed: A PEBB inverter (silicon carbide semiconductor JFET devices instead of Si devices) and an IML-PEBB (injection molded leadframe).

i) SiC power inverter made out of SiC- PEBBs:

Fig 13 explains the realization of the high temperature PEBB which is based on SiC JFET switches. The characterization of the JFETs indicates that an operation at 275°C does not yield any problems. The limiting elements are the interconnects at the top and the bottom of the JFETs. For the bottom die attach three different types were used and tested:

• High temperature Pb solder • AuSn solder • Ag LTJ (low temperature joining – or recently named silver sintering)

Another key component is the driver circuit for SiC JFETS which has to provide a negative voltage of about 30V. Because the distance between the driver output and the SiC switches should be as short as possible, the driver has to operate at high ambient temperatures of about 150 – 175°C. Fig 14 shows design and implementation of the driver circuit. The operating junction temperature is limited by the packaging and not by the SiC device itself.

PAR Page 15 of 22

Fig 13 Many new processes and process steps are used for the SiC- PEBB. They were selected according to the building-in principle.

Fig 14 Driver circuit for SiC-PEBBs capable of operating at 200°C (ETH Zurich). In Fig 15 the major features of the IML technology are to be seen: Cu/Al ribbons for the emitter bonding of the IGBTs instead of thick Al wire monds (small insert). Due to the use of Cu there is an

PAR Page 16 of 22

increase of the electrical as well as the thermal conductivity. The use of ribbons instead of wires improves the electrical as well as the thermal conductivity of the system.

Fig 15 Measures how to improve the performance of the mechatronic IML module by replacing Al thick wire bonds by Cu/Al ribbons (Valeo).

Summary Power electronic building blocks and inverters

• The integration of the SiC- inverter with a two-liquid heatpipe necessitates many new processes and process steps

• The electrical performance is improved by a factor of five while the thermal resistance is lowered by at least one factor of magnitude

• The consequent application of the building-in principle as a part of the new robustness validation process secures reliability even in high temperature operation

• The circuit driver for the SiC JFET switches is capable to operate at 200°C ambient • The IML packaging technology turns out to have a good cost saving potential. It will be

transferred to a product family in the near future

1.3.5 High temperature control board

The technology for a 140°C operation has been developed and tested very carefully for the SiC-PEBBe electronic control unit (ECU). The work done is based on the results of the Medea+ project "HOTCAR". Fig 16 shows the capability of devices, interconnects and substrates to withstand high temperature operation. The maximum operation temperature is considered to be 140°C. Of course always the minimum temperature has to be kept in mind which is typically -40°C. It can be seen from Fig 16 that there are solutions available for most of the parts, e.g. the use of thermal vias and copper inlets for substrates. There is, however, one group of devices like EEPROMs and some logic devices which have a lower maximum junction temperature than 140°C. Several methods of local cooling were investigated - one of it is the use of Peltier elements. Because of the low efficiency quite a lot of heat is generated additionally which yields to an increase of the ambient temperature.

PAR Page 17 of 22

Fig 16 Components and materials which are capable to operate at an ambient temperature of 140°C. For EEPROMs and logic devices a special cooling has to be applied (Continental Automotive). A better solution is the use of a thermosyphon. Like in a heat pipe the heat is removed from the hot part by evaporating a liquid. The vapor carries the heat very easily to the condenser which has to be at a lower temperature than the maximum junction temperature. The liquid droplets return to the hot part by gravity. The thermosyphon technology is well known refrigerators. Summary High temperature Control Board A high temperature control board with a maximum temperature capability of 140°C has been specified. The basic technologies were investigated and tested and the components interconnects and substrates validated by specific reliability tests. The results are very promising and show that an electronic control unit can be operated reliably and safely. The operation of some of the micro-electronic devices - which have a lower maximum junction temperature - can be guaranteed by using a local thermosyphon cooler.

1.3.6 Perspective of inverter integration into the powertrain Extensive studies were made on integrating the new technologies into power train systems for ICE-hybrid and FC-vehicles. Fig 17 gives an overview about different integration steps for power electronics systems. Concepts for integration in an axial arrangement performed as a circular double inverter between two electric machines have been carried out. These concepts are well suited for ICE _Full Hybrid application with two e-motors.

PAR Page 18 of 22

Fig 17 Joint Roadmap and several solutions for integrating power electronics into the power train (Daimler, MagnaSteyr, Fraunhofer IIS) For FCV another integration concept (two segments radial integration as shown in Fig 18) have been introduced. The inverter is divided into two flat sections for the dc capacitor and the power electronics DCBs with microcontroller. The introduction of the packaging technologies (WP3) and high temperature controller (WP5) to these DCBs allow higher temperatures and better reliability. This activity is strongly correlated to the project HYSIS (FP6) to make sure that all results of HOPE can be used as an input for HYSIS .The advantages of this attached solution over the fully integrated solution are:

• Flexible packaging • Good accessibility and serviceability • Shared cooling with E-machine • Electric synergy

Summary Perspective of inverter integration into the power train - The integration of the drive inverter into the e-motor allows a noticeable reduction of volume and system costs -Two innovative integration concepts have been introduced: 1. A disc inverter for axial integration in the power train which is well suited for ICE Full hybrid vehicles (Patent application) 2.Two segment radial integration concept which allows a flexible packaging as well as good accessibility and serviceability compared to the known integration concepts. This concept represents a good solution for power trains of FCV and EV.

PAR Page 19 of 22

Fig 18 Radial integration of the power electronics. This concept is used by the FP6 integrated project HYSIS (Daimler, Fraunhofer IISB, MagnaSteyr).

1.4 Assesment of the three main goals of HOPE – performance, low cost, reliability In summary it can be said that HOPE has achieved good progress: Performance: There are many examples where the performance has been increased, like:

• Thinner die attach to lower the thermal resistance • Measures to reduce the CTE mismatch • Specific packaging to reduce on-state and switching losses in SiC inverters • Low Rth heatpipe coupling to the SiC inverter • High temperature driver circuit • Local heating by thermosyphon • Integrated SiC commutation cell with very high power density

Lower cost: Three examples demonstrate that lower costs can be achieved:

• IML technology has the potential of 30% cost reduction by replacing the ceramic DCB by a plastic insulator

• The high temperature control board has a potential of 50% costs saving in contrast to the ceramic substrate

• Integration itself has a potential of 20 to 30% because chip area and other materials can be saved.

Reliability: Improvements were made in the following areas: • Partially implementation of the new “Robustness Validation “ process • Methodology like the mission profile test • Translation of drive cycle to stress profile at the weakest parts of a system • Robust high temperature test equipment

PAR Page 20 of 22

2. Dissemination and use of knowledge

2.1 HOPE Web site (www.fp6-hope.eu) 2.2 Conferences

2nd APE 2007, 26-27 September 2007 in Paris: The programme of the Automotive Power Electronics Conference APE was strongly dependent on contributions from the HOPE consortium. Also the organization of the conference was supported by HOPE partners. HOPE – ECPE Joint Workshop, 28 September 2007 in Paris: The aim of the workshop was to bring together experts from the HOPE project and scientists working for ECPE projects to have discussions on roadmaps and reliability issues. CIPS 2008 – Conference on the integration of power electronics systems, March 11-13, 2008, Nuremberg: The CIPS 2008 highlighted all the issues which are important for system integration. Many of the HOPE partners were active by presenting scientific and technical results as well as by serving in the organizing and programme HOPE - ECPE Symposium on „Automotive Power Electronics“ 125 experts attended the Symposium on “Automotive Power Electronics” on October 7 and 8, 2008 at the Daimler Conference Centre, Sindelfingen.

PAR Page 21 of 22

The core of the 20 presentations came from the HOPE consortium. Six presentations were dealing with results from other European projects as well as from a French and a German national funded project. The content and results of the FP6 IP HYSIS project were explained as well. ECPE - HOPE Symposium Automotive Power Electronics - High Density Power Electronics for Hybrid Traction

7 - 8 October 2008 Stuttgart-Sindelfingen, Germany

Tuesday, 7 October 2008 10:00 Start of Registration 10:20 Opening, Welcome Address and Introduction T. Harder (ECPE), J. Mittnacht (Daimler) 10:45 Powertrain Electrification at Volkswagen R. Plikat, Volkswagen (D) 11:15 Kinetic Energy Recovery System for F1: Hints on the Electrical Solution G. Catona, Centro Ricerche Fiat (I) 11:45 General Requirements on Quality and Reliability from the View of an OEM W.Wondrak, Daimler (D) 12:15 Lunch 13:15 Intelligent Testing based on Mission Profiles G. Coquery, INRETS (F) 13:45 Reliability of High Temperature Electronics in PCB Technology M. Rittner, Robert Bosch (D) 14:15 High Temperature Control Board A. Rekofsky, Continental (D) 14:45 Coffee Break 15:15 Reliability of Today´s Power Module Technologies A. Roth, Fraunhofer IISB (D) 15:45 Robustness Validation E. Wolfgang, ECPE (D) 16:15 ECPE Automotive Power Electronics Roadmap M. Maerz, Fraunhofer IISB (D) 16:45 Panel Discussion Impact of Automotive Power Electronics on Industrial Applications – Synergies and Competition 17:45 End of 1st Days Programme 19:30 Dinner

Wednesday, 8 October 2008 8:30 IML Mechatronic Packaging Technology, an Innovative Solution for Automotive Power Electronics J.M. Morelle, Valeo (F) 9:00 New Technologies for Liquid-Cooled Power Modules A. Schletz, Fraunhofer IISB (D) 9:30 High Temperature Current Measurement in Automotive Power Electronics F. Grecki, W. Koczara, Warsaw University (PL) 10:00 Coffee Break 10:30 Power Electronics Building Blocks with SiC Devices and Advanced Cooling K. Kriegel, Siemens CT (D) 11:00 High Temperature Gate Drive for SiC-JFETs S. Waffler, ETH Zurich (CH) 11:30 Perspectives of Inverter Integration in Vehicle Powertrains A. Schmidhofer, Magna Steyr (A) 12:00 General Discussion 12:15 Lunch Information on Current Research Projects in Europe: 13:15 HYSYS: Fuel Cell Vehicle System Components (EC - FP6) J. Wind, Daimler (D) 13:45 HI-CEPS: Highly Integrated Combustion Electric Propulsion System (EC - FP6) Centro Ricerche Fiat (I) 14:15 HyHEELS: Hybrid High Energy Electrical Storage (EC - FP6) R. Knorr, Continental (D) 14:45 Coffee Break 15:15 MOVEO – A French Initiative on Power Mechatronics G.-M. Martin, Valeo (F) 15:45 InGA: Power Electronics Integration in the Gearbox/Drivetrain (German BMBF) K. Kriegel, Siemens CT (D) 16:15 End of the Symposium The HOPE

2.3 Publishable results Publications from partners:

• K. Kriegel, Siemens CT, Germany 'HOPE – High Density Power Electronics for FC- and ICE- Hybrid Electric Vehicle Powertrains', European Hydrogen and Fuel Cell Review Days 2007, Brussels, 10 - 11 October 2007

PAR Page 22 of 22

• F. Grecki, Warsaw University of Technology, Poland, et al. 'Current Sensor Dedicated for High Temperature Automotive Power Electronics' Automotive Power Electronics (APE), Paris, 26 & 27 September 2007

• E. Wolfgang, Siemens CT, Germany, et al. 'Securing Reliability – a Key Issue for Power Electronics in Automotive' Automotive Power Electronics (APE), Paris, 26 & 27 September 2007

• M. Maerz, Fraunhofer IISB, Germany 'Mechatronics: Technology Choices for Automotive Applications' Automotive Power Electronics (APE), Paris, 26 & 27 September 2007

• JM. Morelle, Valeo, France, et al. 'Innovative connectivity for power dice in mechatronic packaging of automotive power electronics' Automotive Power Electronics (APE), Paris, 26 & 27 September 2007

• M. Rittner, Bosch, Germany 'Lead-Free Interconnection Technologies in Power Electronics Modules – An Approach from the EU-Funded Project HOPE' Automotive Power Electronics (APE), Paris, 26 & 27 September 2007

2.4 Patents applied

Partner Subject

Bosch, Siemens Low temperature joining/ silver sintering

Magna; FhG-IISB; Daimler Thin electric motor

Siemens Test procedure for SiC devices

WUT Magnetic circuit for compensation of the flux produced by measured current

WUT Magnetic circuit for current measurment system

WUT Magnetic circuit for compensation of the flux produced by measured current in the wide frequency range