Electronics in Motion and Conversion May 2010 · Lighting High-Efficiency Converters with PFC for LED Street Lighting at 48 V and 130 W; By Claudio Spini, STMicroelectronics, Davide

ZKZ 64717

05-10ISSN: 1863-5598

Electronics in Motion and Conversion May 2010

Hot Show EventStop at the Podium at its Best

Wednesday, the 5th of May,

12:20 to 13:20 at booth 377

“Passive Components for System Efficiency”

Bodo Arlt, Editor Bodo’s Power Systems

C O N T E N T S

1www.bodospower.com May 2010 Bodo´s Power Systems®www.bodospower.com May 2010 Bodo´s Power Systems®

Viewpoint

Fresh Asparagus in Nuremberg in May . . . . . . . . . . . . . . . . . . . . . . 4

Events . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4

News . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6-12

Blue Product of the Month

Win a Microchip nanoWatt XLP 16-Bit Development Board! . . . . . 14

Green Product of the Month

Rapid Response Service on B2B eCommerce Portal . . . . . . . . . . .16

Guest Editorial

New Power Technologies Set to Improve Energy Efficiency

in Data Centers

By Peter Oaklander, Senior VP, Intersil Corporation . . . . . . . . . 18-19

The Experts View

Energy Harvesting: Breaking Through to Commercial Viability

By Donald E. Paulus, VP and GM Power Products, Linear Technology Corporation . . . . . . . . . . . . . . . . . . . . . . . . . . . . 20

Market

Electronics Industry Digest

By Aubrey Dunford, Europartners . . . . . . . . . . . . . . . . . . . . . . . . . . 22

Market

Vehicle Electrification Moving in Different Directions

By Linnea Brush, Senior Research Analyst, Darnell Group . . . 24-25

Cover Story

Easy Parallel Connection of IGBT Modules at the Next SCALE Level

By Jan Thalheim, Olivier Garcia, Sascha Pawel, Heinz Rüedi, CT-Concept Technologie AG, Switzerland . . . . . . . . . . . . . . . . . 26-28

Technology

New Generation of GaN Based Power Stage Products

By John Lambert, PM, International Rectifier Corp., El Segundo, California, US . . . . . . . . . . . . . . . . . . . . . . . . . . . . 30-31

IGBT Modules

Enhancing an IGBT Module for High Temperature & High Current

Operations; By B. Aydin, C. Corvasce, L. Feller, S. Hartmann, ABB Switzerland Ltd, Semiconductor . . . . . . . . . . . . . . . . . . . . 32-35

Power Modules

How to avoid errors when applying thermal paste

By Dieter Esau, Process Engineer and Dr. Michaela Strube, Manager Service Engineering, Semikron . . . . . . . . . . . . . . . . . 36-38

Measurement

A New Class of Rogowski Coil Split-Core Current Transducers

By Pierre Turpin, Project Manager, LEM Energy & Automation 40-44

IGBT Drivers

IPS Drivers Combine Highest Performance and Design Flexibility

By Robert Hemmer, Pavel Kviz and Marita Wendt, InPower Systems GmbH . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 46-47Power Supply

The Fundamentals of Flyback Power Supply Design

By Sameer Kelkar, Staff Applications Engineer, Power Integrations, Inc. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 48-50

Power Supply

High Efficiency, Low-Profile AC-DC Power Supply Design

By Steve Mappus, Principal Systems Engineer, Fairchild Semiconductor . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 52-54

Lighting

High-Efficiency Converters with PFC for LED Street Lighting

at 48 V and 130 W; By Claudio Spini, STMicroelectronics, Davide Giavarini and Wolfgang Dreipelcher, Epcos . . . . . . . . . 56-57

Design and Simulation

Ultra low latency HIL simulator for Power Electronics Applications

By Eric Carroll and Ivan Celanovic, Typhoon RTDS GmbH . . . 58-59

Thermal Management

Cold Plates for Water Cooling Electronic Components

By L Dubois/ JL Dubelloy, Ferraz Shawmut Thermal Management, La Mure, France . . . . . . . . . . . . . . . . . . . . . . . . . 60-61

IGBT Modules

Direct Liquid Cooling IGBT Module for Wind Power Applications

By Neil Markham, Hitachi . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .62-64

New Products

High Current low RDS(on)

Trench MOSFET SIX-PACKNew DCB based surface mount package

for automotive applications

TYPE VDSS ID25 RDS(ON)typ

V A m�GMM 3x180-004X2-SMD 40 180 1,9

GMM 3x160-0055X2-SMD 55 150 2,2

GMM 3x120-0075X2-SMD 75 110 4,0

GMM 3x100-01X1-SMD 100 90 7,5

GMM 3x60-015X1-SMD 150 60 17

Features• Low RDS(on)

• Low stray inductance

• Multi chip packaging

• Isolated DCB base plate

• Suited for SMD mounting

• Low thermal impedance

• Trench MOSFET2nd generation

Applications• Automotive

• Battery powered systems

• Industrial motor control (robotics)

• DC-DC converter

• Battery charger

Benefits• Highest reliability

• Optimized layout

• 3 identical half bridges

Customized configurations possible!

SMD: Surface Mount Device

For more information pleaseemail [email protected] call Petra Gerson: +49 6206 503249

Bodo´s Power Systems® May 2010 www.bodospower.com2

TThhee GGaalllleerryy

GvA Leistungselektronik GmbH | Boehringer Straße 10 - 12 | D-68307 Mannheim

Phone +49 (0) 621/7 89 92-0 | www.gva-leistungselektronik.de | [email protected]

ACCELERATINGYOUR PROJECTSWelcome to the House of Competence.GvA is your expert in individual problem solutions for all sectors of power electronics – state of the art know how and profound experience as an engineering service provider, manufacturer and distributor.

Consulting – Design & Development – Production – Distribution

Bodo´s Power Systems® May 2010 www.bodospower.com

Consider the many paths to better efficiency

– they are all important!

Digital control is no longer new, but it is now

reaching a broader range of acceptance at

all levels of design. Darnell’s Digital Power

Forum provided great insight on future

usage, acceptance of digital control and digi-

tal communication between power supplies.

Overall, the focus was on education and

simplicity. A major user can have design

work done by a digital control manufacturer

but this reaches only a segment of the mar-

ket. Extending the benefits of these efficient

designs to a wider audience will require edu-

cating the market and streamlining process-

es.

One approach is through a user interface for

software that can define the application

parameters. Examples are already on the

web, complete with reference designs and a

bill of material for your power converter.

Independent workshops like the Biricha Digi-

tal Workshop are helping to spread digital

control expertise. CUI, for example, has a

simplified digital approach that will open the

mind of many potential users. The article in

my April issue on pages 40 and 41 provides

a perfect introduction.

In the past, digital control was used to bal-

ance currents between power supplies in

telecom applications. It is now moving into

the regulation loop. This implies fast digital

control circuits for the response speed that

such systems require.

And what about the usage of GaN switches

in future designs? These will definitely run

at higher frequencies and temperatures to

fully utilize the capability of GaN or SiC. We

are facing a totally new mindset thanks to

these state-of-the-art materials.

Passive components in designs will have to

keep up with semiconductor improvements.

Without optimized passives, design solutions

will not reach their full potential. So note the

following date in your calendar: Wednesday,

May 5th, 12:20 to 13:20 at Booth 12 / 377,

for the Forum discussion at the PCIM

Europe highlighting:

“Passive Components for System Effi-

ciency”

Passives need to accompany semiconduc-

tors in moving to new materials and higher

temperature limits. System solutions must

build on these developments.

You will meet experts in passive capacitor

and inductor components willing to share

their competence in a personal discussion

with you.

PCIM is the leading edge power electronics

show and conference, both in Europe and in

China and my May issue will reflect this. I’ll

be featuring all of my supporters, with their

logos, on a floor plan in the Chinese Cabinet

on my cover. This extra attention is meant to

reflect world-wide acceptance of my publica-

tion for power design engineers. Bodo’s

Power Systems is now working with nearly

two dozen shows and conferences around

the globe with a focus on power and its

applications.

My Green Power Tip for May:

Eating locally-grown white asparagus from

the Bavarian countryside helps reduce pollu-

tion, as its transportation to the PCIM in

Nuremberg is minimal.

Enjoy the fantastic Schrobenhausener

asparagus, in season, while you’re there!

See you at the PCIM conference in Nurem-

berg

Best regards

Fresh Asparagus inNuremberg in May

V I E W P O I N T

4

A MediaKatzbek 17a

D-24235 Laboe, Germany

Phone: +49 4343 42 17 90

Fax: +49 4343 42 17 89

[email protected]

www.bodospower.com

Publishing EditorBodo Arlt, [email protected]

Creative Direction & ProductionRepro Studio Peschke

[email protected]

Free Subscription to qualified readers

Bodo´s Power Systems

is available for the following

subscription charges:

Annual charge (12 issues) is 150 €

world wide

Single issue is 18 €

[email protected]

circulation

printrun

25000

Printing by:

Central-Druck Trost GmbH & Co

Heusenstamm, Germany

A Media and Bodos Power Systems

assume and hereby disclaim any

liability to any person for any loss or

damage by errors or omissions in the

material contained herein regardless of

whether such errors result from

negligence accident or any other cause

whatsoever.

EventsPCIM Europe Nuremberg Ger.

May 4-6 www.mesago.de

Sensor+Test, Nuremberg,

May 18-20 http://www.sensor-test.de

SIC Workshop Kista Sweden, May 18-19

www.acreo.se/SICseminar-10-05-18--19

SMT Hybrid Nuremberg Ger.

June 8-10, www.mesago.de

Digital Power Workshops Dortmund Ger.

June 2, www.biricha.com

Intersolar Europe 2010 Munich Ger

June 9 to 11 www.intersolar.de

SEMICON West San Francisco USA

July 14-16 www.semiconwest.org

Digital Power Workshops Stockholm

Sweden Aug. 24 www.biricha.com

Solar Energy Valencia Spain

Sep. 6-10 http://www.photovoltaic-conference.com

Husum Wind Energy Ger.

Sep. 21-25 www.husumwindenergy.com

Innotrans Berlin Ger.

Sep. 21-24 www.innotrans.com

Digital Power Workshops Munich Ger.

Oct. 5 www.biricha.com

Several current ranges from 6 to 50 ARMS

PCB mounted Up to 30% smaller size (height)Up to 8.2 mm Clearance / Creepagedistances +CTI 600 for high insulation

+5 V Single Supply Low offset and gain driftHigh Accuracy @ +85 C Access to Voltage Reference Analog Voltage output

www.lem.com At the heart of power electronics.

Future precision. Future performance.Now available.

The transducers of tomorrow. LEM creates them today. Unbeatable in size, they are also adaptable and adjustable. Not to mention extremely precise. After all, they have been created to achieve great performance not only today but as far into the future as you can imagine.

CAS-CASR-CKSR

PCIM

Europe 2010

Hall 12-402

6 Bodo´s Power Systems® May 2010 www.bodospower.com

N E W S

• Industrial/Ph.D. Course in Power Electronics for Renewable Ener-

gy Systems – in theory and practice,

3 – 6 May 2010, Aalborg, Denmark

programme

• Industrial/PhD Course in Photovoltaic Power Systems - in theory

and practice

10 – 13 May 2010, Aalborg, Denmark

programme

• Workshop Silicon Carbide Power Electronic Applications,

18 – 19 May 2010, Kista, Sweden

Programme Workshop Silicon Carbide

• IET Seminar Power Electronics 2010: Improving the efficiency of

the power grid,

9 June 2010, Birmingham, U.K.

Programme

• 1st International Congress - Automotive Electronics: driving the

future of powertrain and electrification,

10 – 11 June 2010, Torino/Venaria Reale, Italy

Programme Automotive Electronics Congress

PCIM Booth 12 / 569

www.ecpe.org

Upcoming Events supported by ECPE

Rogers Corporation will highlight its RO-

LINX®busbars and advanced thermal solu-

tions at the upcoming PCIM Europe 2010

Exhibition. Representatives from Rogers

BVBA Power Distribution Systems Division

will be on hand to detail the company's

broad portfolio of connection techniques for

their RO-LINX Laminated Busbars. Laminat-

ed busbars are an essential part in power

modules as they serve as an interconnection

part that links components, cables and mod-

ules together. Having the right connection

technique is important to guarantee an opti-

mal electrical performance of the laminated

busbar and the complete power module.

Rogers will also showcase its Thermal Man-

agement Solutions Division HEATWAVE™

metal matrix composite (MMC) material for

overcoming difficult thermal management

challenges through effective and efficient

dissipation of excess heat.

Members of Rogers Advanced Circuit Mate-

rials Division will be on hand at Booth # 12 /

439 to offer advice and guidance on the use

of their high performance PCB materials,

including Rogers’ halogen-free, RoHS-com-

pliant Theta circuit materials that offer supe-

rior thermal and electrical performance. With

a stable dielectric constant of 3.90 and low

dissipation factor of 0.008 at 1 GHz, Theta

materials are ideal for high-speed telecom-

munications and computing applications, and

are an environmentally-friendly solution for

the high reliability, high speed digital market.

Tomorrow’s high reliability requirements are

addressed today with a very low coefficient

of thermal expansion (CTEz) of about 50

ppm/ºC (approximately 30% lower than stan-

dard FR-4).

PCIM Booth 12 / 439

www.rogerscorp.com

Busbars and Advanced Thermal Solutions

CamSemi has announced the opening of a

new application design centre and business

development office in Shenzhen that will be

instrumental in growing the company’s cus-

tomer base and sales of its highly ‘cost effi-

cient’ off-line power management ICs within

China.

The new offices in Shenzhen Academy of

Aerospace Technology – within one of the

city’s most important electronics districts -

will allow the company’s engineering team to

work even more closely with an increasing

number of customers and distributors on

major designs. The new facility, three times

larger than the company’s initial Shenzhen

office, includes a state of the art power sup-

ply design laboratory for development proj-

ects plus offices and meeting rooms to host

training seminars and other events.

CamSemi’s new China office will be man-

aged by Kim Tey, regional director for Asia,

who also oversees the company’s design

centre in Taipei, Taiwan. The new facilities

were opened formally by David Baillie, CEO

at a launch party for local media, distributors

and commercial partners.

www.camsemi.com

China Support by Application Design Centre

Ferraz Shawmut participates to PCIM that will be held at Exhibition

Centre Nuremberg, Germany from 4th to 6th of May 2010. PCIM

Europe Conference is one of the most leading conferences address-

ing the fields of Power Electronics, Intelligent Motion, Power Quality,

Energy Management and topics of common interest. From latest

developments of power semiconductors, products for thermal man-

agement, new materials, sensors as well as servo-technology and

the wide area of power quality and energy-management, PCIM

Europe exhibition offers a comprehensive, focused and compact

presentation of products all under one roof.

During this event, Ferraz Shawmut will showcase his solutions, share

his expertise and pursue the dialogue opened some decades ago

with the Power Electronics Community, especially with the design

engineers.

PCIM Booth 12 / 510

www.ferrarzshamut.com

Welcome to the World of Thermal Management

Cree and the Cree logo are registered trademarks of Cree, Inc. Z-Rec is a trademark of Cree.Inc.

PART NUMBER If (A) Vf (V) IR (μA) Qc (nC) TJ (˚C) TYPICAL TYPICAL TYPICAL MAX

CPW3-1700S010B 10 1.8 @ 25˚ C 10 @ 25˚ C 80 175 3.2 @ 175˚ C 20 @ 175˚ C

CPW3-1700S025B 25 1.8 @ 25˚ C 25 @ 25˚ C 210 175 3.2 @ 175˚ C 50 @ 175˚ C


N E W S

International Rectifier has announced that its

HiRel Business Unit (BU) has expanded its

portfolio of leaded products to provide a con-

tinued long term viable source for customers

in the military, aerospace and biomedical

markets.

IR’s HiRel Business Unit has acquired a

large portion of the leaded product offerings

previously available from the company’s

commercial divisions. As part of its ongoing

commitment to offer leaded products, IR has

entered into manufacturing agreements with

key assembly subcontractors to 2013, which

combined with its own assembly capabilities

allows the company to offer customers of its

leaded products long term supply agree-

ments.

PCIM Booth 12 / 202

www.irf.com

Long Term Solution for HiRel Leaded Products

The German photovoltaic specialist IBC

SOLAR will support the People's Republic of

China by training and certifying photovoltaic

installers. The photovoltaics specialist

received a delegation from the Chinese Min-

istry of Construction this week for the signing

of the cooperation contract that also includes

the certification of photovoltaic systems. The

visit marks a successful impetus for the

expansion of renewable energy sources in

China over the coming years. The Chinese

government plans to establish one of the

world's largest photovoltaic markets by 2015.

Yao Bing explained: "We have decided to

work with IBC SOLAR due to their very good

reputation and their long-term experience."

The cooperation with the Ministry of Con-

struction of the People's Republic of China

offers great opportunities for both sides: IBC

SOLAR will share its experience in the train-

ing of installers in building-integrated photo-

voltaic (BIPV) systems as well as the devel-

opment of standardized certification guide-

lines for PV systems. The Chinese Ministry

of Construction plans to introduce jointly

developed training standards for Chinese

installers.

www.ibc-solar.com

Advise in the Expansion of Renewable Energy Sources

Last quarter saw two more successful Digital Power Workshops

delivered by Biricha Digital Power in conjunction with Texas Instru-

ments. The US workshop was hosted by Arrow Electronics in Solon,

Ohio and the UK workshop ran at Texas Instruments' UK Headquar-

ters in Reading. Both workshops attracted a sell-out audience of

both hardware and software engineers. Biricha and Texas Instru-

ments were also very honoured to host Dean Venable at the Ohio

workshop.

Participants were taught the skills necessary to quickly start using the

TI F28x family to design stable digital power supplies. By the end of

the course the participants had designed and coded four closed loop

digital converters and had received the necessary templates, libraries

and skills to implement multiple converters with minimal coding. In

response to customer feedback, the original four day workshop has

now been revised to run over a period of three days. The three day

long laboratory based workshop is written with analog power supply

designers in mind and aims to significantly reduce the development

time of digital power supplies.

Further US and European workshops are planned for Dallas (May

2010), Dortmund (June 2010) and Stockholm (August 2010). For

more information regarding these workshops, full syllabi and other

forthcoming events, please visit:

www.biricha.com

Successful Digital Power Workshops in UK and US

Completing a merger transaction between

the former NEC Electronics Corporation and

Renesas Technology Corp., the newly estab-

lished Renesas Electronics Corporation

(TSE: 6723) announced it has commenced

business operations.

Upon approvals from the Renesas Electron-

ics Board of Directors, Junshi Yamaguchi

became the Chairman and Yasushi Akao

became the President of the Board of Direc-

tors of Renesas Electronics. Renesas Elec-

tronics also announced that the company

issued shares of its common stock to NEC

Corporation (NEC; TSE: 6701), Hitachi, Ltd.

(Hitachi; TSE: 6501), and Mitsubishi Electric

Corporation (Mitsubishi Electric; TSE: 6503)

in exchange for an aggregate of approxi-

mately 134.6 billion yen.

www.renesas.com

Renesas Electronics Corporation Commences Operations

Ericsson Power Modules has been awarded the 2009 World Vertical Market Pene-

tration Leadership Award in the Board Mount DC-DC Converter market from global

research organisation Frost & Sullivan.

The Vertical Market Penetration Leadership Award is prestigious recognition of Eric-

sson Power Modules' accomplishments in the board mounted DC-DC converter

market. It is an unbiased, third party recognition. The award is presented each year

to the company that has demonstrated excellence in capturing the fastest measured

rate of change of market share of a specific vertical market. This award recognizes

how fast a company increases its vertical penetration of a market, in terms of rev-

enues or units as specified.

www.ericsson.com

Vertical Market Penetration Leadership Award

2SP0115T Gate DriverUnleash the full power of your converter design using the new 2SP0115T Plug-and-Play driver. With its direct paralleling capability, the scalability of your design into highest power ratings is unlimited. Rugged SCALE-2 technology enables the complete

the size of 17mm dual modules. Combined with the CONCEPT advanced active clam-ping function, the electrical performance of the IGBT can be fully exploited while keeping the SOA of the IGBT. Needless to say that the high integration level provides the best possible reliability by a minimzed number of components.

FeaturesPlug-and-Play solution1W output power15A gate current<100ns delay time± 4ns jitterAdvanced active clampingDirect- and halfbridge modeDirect paralleling capability2-level and multilevel topologiesDIC-20 electrical interfaceSafe isolation to EN50178 UL compliant50.- USD @ 1000 pieces

www.IGBT-Driver.com

SAMPLES AVAILABLE!

CT-Concept Technologie AG, Renferstrasse 15, CH-2504 Biel, Switzerland, Phone +41-32-344 47 47

UnleashSheer Power!

PCIM Booth 12 / 102

N E W S


Micropelt, an innovator in thermal energy

harvesting technology, has collaborated with

Royal Dutch Shell to prove the concept of

converting waste heat into a sustainable,

maintenance-free power supply for wireless

sensor devices. Shell sees wireless sensor

networks as a strong contributor to increas-

ing productivity and lowering maintenance

costs through better status information. The

concept of energy harvesting eliminates the

need for primary batteries in many wireless

sensors. Deployment in restricted access or

even explosive areas, and those previously

considered too costly, become commercially

viable if both wiring and battery maintenance

can be eliminated.

www.micropelt.com

Thermoharvester Uses Waste Heat to Power Wireless Sensors

Thanks to rapid growth in the high-end serv-

er, notebook, mobile handset and wired

communication segments, the Gallium

Nitride (GaN) power management semicon-

ductor market is expected to reach $183.6

million in revenue in 2013, up from virtually

nil in 2010, according to iSuppli Corp.

GaN is an emerging process technology for

power management chips that recently

moved beyond the university-based testing

phase and into the commercialization stage.

The technology represents an attractive mar-

ket opportunity for suppliers by providing

their customers with capabilities that may be

out of the reach of present semiconductor

process materials.

Component suppliers have begun offering

GaN parts. International Rectifier Corp., for

instance, released its first GaN technology-

based Point-of-Load (POL) solutions in Feb-

ruary, while Efficient Power Conversions

Corp. (EPCC) is placing all its bets on GaN

technology, releasing 10 power MOSFET

devices this month.

The attached figure presents iSuppli’s GaN

power management revenue forecast for the

period of 2008 through 2013.

The adoption of GaN devices will be driven

by the improved efficiency and small form

factors enabled by the material. Such bene-

fits are in particularly high demand for

portable electronic products, including

mobile PCs and smart phones. They also

provide advantages for power-hungry elec-

tronic equipment, such as enterprise servers

and wired communications infrastructure

gear.

Discover more about the emerging power

management technologies with Vukicevic’s

new report entitled: World of Unlimited Pos-

sibilities — GaN Devices to Capture Market

Share.

www.isuppli.com

GaN Power Management Chip Market Set for Boom

Microsemi announced that it has entered

into a definitive agreement to acquire White

Electronic Designs Corporation

(Nasdaq:WEDC) through a cash tender offer

at $7.00 per share for a net transaction

value of approximately $100 million, net of

White Electronic's projected cash balance at

closing.

White Electronic is a leader in design,

assembly, and test integration. They have

extensive offerings and experience in Multi-

Chip-On-Board solutions that are integrated

into Defense and Aerospace applications.

Their technology integrates surface mount

technologies, microelectronics, and Anti

Tamper technologies into one solution. Their

market focus is where size, weight, and per-

formance create a market advantage. A sig-

nificant area of market expansion where they

have developed unique technology is in the

Anti Tamper market.

PCIM Booth 12 /422

www.microsemi.com

Microsemi to Acquire White Electronic

0,020,040,060,080,0

100,0120,0140,0160,0180,0200,0

2008 2009 2010 2011 2012 2013

Mill

ions

ofU

.S.D

olla

rs

iSuppli Figure: Global Gallium Nitride (GaN) Power ManagementSemiconductor Revenue Forecast (in Millions of U.S. Dollars)

At the 4th Photovoltaic Fab Managers

Forum, held on March 8 in Berlin, SEMI PV

Group Europe announced the formation of a

new European group for crystalline solar

cells. This founding group of eight crystalline

solar cell manufacturers (Q Cells, Deutsche

Cell, Bosch Solar Energy, Schott Solar,

Sovello, Sunways, SolarWatt/Systaic Cells

and Solland) is working together in a pre-

competitive environment to address the

technology challenges facing the photovolta-

ic industry.

The CTM Group has established a crys-

talline solar cell technology roadmap up to

the year 2020, which was announced at the

PV Fab Managers Forum on March 8, 2010.

The roadmap describes the development of

crystalline solar cell technology with focus on

materials, manufacturing processes, and

product development.

The priority of the CTM Group will be the

definition of the development processes for

raw materials, cell technology and cell man-

ufacturing. The goal is to strengthen Euro-

pean competitiveness in the global market-

place by improving efficiency and quality

while also reducing cost. The group also

aims to optimize the interfaces within the

entire manufacturing supply chain to help

achieve this goal.

www.pvgroup.org

European Group for Crystalline Solar Cells

N E W S

www.bodospower.com

ABB has agreed to acquire the semiconduc-

tor business of Polovodièe a.s. in the Czech

Republic. The additional production capacity

for high-power semiconductors will help ABB

to cope with the expected rising demand

fueled by growth in renewable energy and

efforts to improve energy efficiency.

Polovodièe a.s. has been making power

semiconductors since the mid-1950s and

had revenues in the low double-digit millions

of US-Dollar, mostly from its power semicon-

ductor activities. Together with the semicon-

ductor assets about 200 employees with

strong technical capabilities will join ABB.

“Power semiconductors are central to the

development of smarter electricity networks

in which greater use of renewable energy is

combined with better control of power flows

for more reliability and efficiency,” said Peter

Leupp, head of ABB’s Power Systems divi-

sion. “The business of Polovodièe will

strengthen our market position as one of the

leading producers of power semiconductors.”

ABB is already investing $150 million over

three years to expand its semiconductor

plant in Lenzburg, Switzerland. The acquisi-

tion will improve economies of scale at each

facility and will increase flexibility of produc-

tion and delivery.

PCIM Booth 12 / 408

www.abb.com

ABB Expands its Power Semiconductor Business

Bicron Electronics is a designer and manu-

facturer of ultra high reliability, high frequen-

cy transformers to supply power to IGBT cir-

cuit systems required to operate at voltages

typically greater than 1200V.

Bicron provides dependable isolation for

operating voltages up to 20KV. Common

applications include rail and marine drive

controls, wind and solar power controls as

well as controls for very large motors used in

a broad range of industries.

CMS is a sales organization represented in

almost all EU and CIS countries in Europe.

CMS specializes in the joint development

and manufacture of high performance con-

tacts, gasket sealing systems, RF transform-

ers and related parts and components. It

maintains close personal and electronic con-

tact with Bicron Electronics’ USA office for

high levels of customer support.

www.cmscontact.com

Bicron Electronics

Representative for

Europe

Telephone: +49 (27 71) 9 [email protected]

www.isabellenhuette.de

Innovation from tradition

Low-ohmic precision resistors VMx

Features: 3 watt power loss (size 2512)max. 25 A constant currentTcr < 20 ppm/KRthi < 20 K/W

This is unrivalled quality

The winning pitch

in Nuremberg, GermanyMay, 4 - 6 2010

Hall 12, Stand 629


230 power electronics experts attended the CIPS 2010, the 6th Inter-

national Conference on Integrated Power Electronics Systems. The

conference topic was on power electronics systems, high and medi-

um power modules and reliability. The program included 71 technical

papers: Two keynotes, 11 invited, 42 oral and 16 posters. The author-

ship was well balanced: 26 from industry, 28 from academia, 7 from

both, and 13 from research institutes.

In the first keynote Prof. Johann Kolar/ ETH Zurich outlined a mathe-

matical approach for a generic power electronics systems roadmap.

The procedure relies on a multi-objective optimization of converter

systems. As a result efficiency and power density are presented in a

map in which the pareto fronts are disclosing the state-of-the-art for

individual topologies. The power electronics design engineer can

easily read from the plan what can be achieved with today’s (and

tomorrows) technologies. The third major parameter in the plan

would be cost.

Prof. Dushan Boroyevich/ CPES, Virginia Tech., USA, summarized

the results on system integration achieved by the CPES consortium

in the last 4 years. Active device integration was done for SiC- JFETs

and –MOSFETs into high-temperature power modules using wire

bond as well as planar interconnect technologies. Passive power

components were integrated like EMI filters and energy storage

capacitors. Finally converter integration was shown and discussed.

85 references may give insight in details of integration.

Dr. Michel Mermet-Guyennet/ Alstom Transport, France, discussed

railway traction reliability. He pointed out that present methods for

life-time estimation of traction IGBT modules lead to high level of

uncertainty related to observed failures in the field. A better link to the

failure criteria of the power cycling test (VCEsat, Rthjc, leakage cur-

rent) is necessary in the future.

Prof. Chris Bailey/ University of Greenwich, UK, explained the current

status of prognostic techniques and application to power electronics.

Three techniques are used: data driven, model driven and a combi-

nation of both – fusion approach.

Jens Goehre/ Fraunhofer IZM, Berlin, and Samuel Hartmann/ ABB

Semiconductor, Switzerland, presented degradation data on Al wire

bonds and chip solder respectively. For presenting their results they

were rewarded with the “ECPE Young Engineer Award”. The best

Poster Award was given to Christoph Marxgut/ ETH Zurich for

“Design of a Multi-Cell, DCM PFC-Rectifier for a 1 mm Thick, 200 W

Off-Line Power Supply”.

Another highlight has been presented by Dr. Dirk Siepe, Infineon

Technologies. In his paper “The future of Wire Bonding is? Wire

Bonding!” he was able to demonstrate an increase of wire bond relia-

bility by a factor of 20 using copper wires as well as a copper chip

metallization. This high reliability is very much appreciated but one

has to keep in mind that other failures might be dominant in copper

bonded systems.

An own session with 5 papers dealt with silver sintering, a technology

which is superior to the solders used today, mainly because of the

high melting point of 960°C. The sintering technology is based on

applying high pressure at moderate temperatures to the chips and

substrates to be interconnected. Good progress was shown by Wolf-

gang Schmitt/ Heraeus, Hanau, Germany, regarding silver pastes

which allow a sintering process with lower pressure.

The last session was dedicated to the future of power integration

considering two very different aspects: requirements for more electri-

fication in the societies, and device developments including new

semiconductor materials.

Dr. Gerhard Miller (Infineon Technologies) demonstrated the close

interaction of device developments (high power density, high temper-

ature, fast switching) and improved hybrid integration technologies.

Prof. Ichiro Omura/ Kyushu Institute of Technology, Japan, discussed

the “Future Role of Power Electronics”. Because of the strong inten-

tion of Japan to become a highly electrified society, roadmaps were

presented at device, module and systems level.

The second keynote at the end of the conference entitled “Is it the

End of the Road for Silicon in Power Conversion?” was presented by

Dr. Alex Lidow, CEO of Efficient Power Conversion; El Segundo,

USA. He demonstrated clearly the integration capability of GaN

devices, like power switch and driver. His solution to get rid off para-

sitics in the package is surprising: “Use no package at all”. He

showed also some promising reliability results. As in all new tech-

nologies, reliability has to be proven. The presented results are prom-

ising.

The CIPS 2010 was organized by VDE ETG/ VDE Conferences and

the European ECPE Network. Technical co-sponsorship was provid-

ed by IEEE PELS and ZVEI.

The Proceedings are published by VDE- Verlag, ETG Fachbericht

121, ISBN 978-3-8007-3212-8. They will be available for downloading

from IEEE Xplore digital library soon.

www.ecpe.org


N E W S


CIPS 2010 - a Real Success!By Prof. Eckhard Wolfgang and Prof. Dieter Silber, ECPE e.V.

Flexible IGBT-based power electronics platform

SEMiXBOX™

10 kW – 100 kW

For AC/DC drives, solar power, UPSand power conversion applications

1 or 3-phase inverter, rectifier andchopper up to 180 A

Hall-effect current sensors

Australia +61 3-85 61 56 00 Belgium +32 23 00 07 93 Brasil +55 11-41 86 95 00 Cesko +420 37 80 51 400 China +852 34 26 33 66 Danmark +45 58 10 35 56 Deutschland +49 911-65 59-0 España +34 9 36 33 58 90 France +33 1-30 86 80 00 India +91 222 76 28 600 Italia +39 06-9 11 42 41 Japan +81 68 95 13 96 Korea +82 32-3 46 28 30 Mexico +52 55-53 00 11 51 Nederland +31 55-5 29 52 95 Österreich +43 1-58 63 65 80 Polska +48 22-6 15 79 84 Russia +7 38 33 55 58 69 Schweiz +41 44-9 14 13 33 Slovensko +421 3 37 97 03 05 Suid-Afrika +27 12-3 45 60 60 Suomi +358 9-7 74 38 80 Sverige +46 8-59 4768 50 Türkiye +90 21 6-688 32 88 United Kingdom +44 19 92-58 46 77 USA +1 603-8 83 81 02 [email protected] www.semikron.com

Bodo´s Power Systems® May 2010 www.bodospower.com14

B L U E P R O D U C T O F T H E M O N T H

Win a Microchip nanoWatt XLP16-Bit Development Board!

Bodo’s Power Systems is offering its readers the chance to win a

Microchip XLP 16-bit Development Board. The XLP 16-bit board per-

mits users to explore and evaluate extreme low-power features, and

learn low-power software and hardware techniques. It provides a low-

cost, highly configurable development system for Microchip’s extreme

low power 16-bit PIC24F microcontrollers which feature sleep cur-

rents down to 20nA and various headers are available to measure

both microcontroller and component power consumption.

The board supports development on PIC24F16KA102,

PIC24FJ64GA102 and PIC24F64GB002 families of MCUs. The

board can be powered by over five sources including batteries and

energy harvesting modules and supports a variety of common com-

ponents that can be selectively enabled. It was also expandable

through its modular interface, providing for the addition of advanced

interfaces and connectivity methods. The board supports a wide volt-

age range; from 1.8 V to 5.5V.

The XLP 16-Bit Development Board functions as a demonstration

platform on initial power-up. The included demonstration software

takes a temperature measurement, datalogs information to the serial

data EEPROM and displays information to a host PC via a USB con-

nection. Additional software is provided to demonstrate low-power

techniques and IC interface routines.

Today’s portable products need to operate longer with less power

and more functionality. Microchip’s nanoWatt XLP microcontrollers

contain features that are ideally suited for applications such as

remote sensors powered by energy harvesting or sealed-battery

applications, which can run for more than 20 years from a single bat-

tery. nanoWatt XLP technology gives designers the flexibility to cus-

tomize their applications for the lowest power consumption through

multiple internal wake-up sources, such as Real-Time Clock and Cal-

endar alarm, Brown-Out Resets, interrupts and Watch-dog Timers, all

while maintaining the I/O states.

For your chance to win a Microchip XLP 16-Bit Development Board

with a preprogrammed PIC24F16KA102 microcontroller installed, visit

Microchip and enter your details in the online entry form.

PCIM Booth 12/363

www.microchip-comp.com/BP-XLP

Learn more about our power electronics solutions by visiting us at Booth #12-439

USA +1-480-917-6137 EUROPE +32-9-235-3611 ASIA + 65-6747-3521

Powerful Products For Powerful Electronics.

PDS

Power Distribution Systems specializes in the design and

manufacture of custom designed busbars. RO-LINX®

busbars serve as power distribution highways. Rogers

laminated busbars provide a customized liaison between

IGBT modules, capacitors and the power source.

Low Inductance

Design for Controlling Partial Discharge

Compact Design

UL Rating

•

•

•

•

TMS

Rogers’ HEATWAVE™ high performance AlSiC materials

combine excellent thermal conductivity and controlled

thermal expansion with low density and high stiffness to

match the performance characteristics of modern power

semiconductor device packaging solutions and systems.

Improves Longterm Reliability

Reliable Thermal Performance

Ease of Integration into Applications

High Stiffness & High Mechanical Stress

•

•

•

•

Power Distribution Systems

Thermal Management Solutions

For more information on our Thermal Management Solutions (TMS),

visit us at www.rogerscorp.com/tms. You can find more information

on our Power Distribution Systems (PDS) at www.rogerscorp.com/pds.

Empowering your energy.

Every day more than 250 power electronics customers from across

the globe use the extensive information service or consult directly

with experts on the B2B eCommerce portal provided by Sindopower,

a holding firm in the Semikron Group. One of the merits of this serv-

ice is that more than 95% of the technical questions asked are

answered immediately.

Sindopower GmbH now strengthens its wings on the international

customer service front: the company’s extensive technical consulta-

tion service comprising a telephone hotline service, power electronics

TechChat and forum, and a knowledge base is now going multi-lin-

gual. Sindopower’s comprehensive information service has already

proven to be a huge success, offering customers different access

paths to the services in the power electronic portal. What is more,

customers are guaranteed speedy solutions to their problems. And

the customer can decide himself whether he prefers to use verbal or

written contact channels. The resulting service quality has won over

customers of all sizes and order volume.

Positively surprising, was how quickly customers of all sizes from all

over the world came to appreciate this 24-hour service in the area of

power electronics. The slogan – ‘Power Electronics in the Web’ – is

not intended to simply reflect the presence of power electronics on

the internet, but will set new standards in customer service and give

customers the feeling that they have a competent partner to answer

their questions. Orders from all five continents of the globe in the first

six months after the portal went live speak for themselves. Sindopow-

er is the only portal in the power electronics sector that links eCom-

merce with a technical advice and consultation service.

To continue to back this trend, the company’s sales and marketing

department has been equipped with a new and dedicated US-Dollar

portal. Payment options have also been extended to include credit

card payment and bank transfers to US bank accounts. In the near

future, the portal will also be available in Russian and Portuguese;

the introduction of Portuguese additional to the already existing

Spanish into the portal language bank is hoped to boost overall cus-

tomer service levels in the Latin American power electronics market.

Traditionally customers were looked after either by sales representa-

tives in the field providing a personalized service mainly to the larger

customer segment and the smaller customers are looked after by

selected distributors. Of utmost importance for all customers is the

reaction time and the time it takes until an answer is received. Sales

representatives usually have to face the dilemma of not being avail-

able for consultation whilst consulting a customer on site. Additional-

ly, the long distances that must be travelled in some countries further

reduce the time available for consulting. This new kind of customer

service, however, also causes a paradigm shift in business communi-

cation. Customers who need written quotations had formerly to cre-

ate a request for a quotation and then wait for an answer. On the

Sindopower website the customer can create his quotation in PDF

format within two minutes. He can then print out his own quotation in

business letter format. Also the best answer to a general power elec-

tronics question does not necessarily have to come from the supplier

himself. In the Forum all experts have the possibility to exchange

information independently from their business relationships.

To find out more about Sindopower, check out their stand at the

PCIM in Hall 12, Stand 411, and meet the people behind the tele-

phone hotline service, the TechChat power electronics chat room and

forum in person.

PCIM Booth 12/411

www.sindopower.com

G R E E N P R O D U C T O F T H E M O N T H


Rapid Response Service on B2B eCommerce Portal

95% of all the technical questions are answered immediately

Solutions for windpower systemsEnergy-efficient components for high system reliability

THE INFINEON PRODUCT PORTFOLIO provides components for the highest energy efficiency in windmill power converter and pitch control solutions.

OUR POWER MODULES with newest 1200V/1700V trench fieldstop IGBT4 and Emitter Controlled diode chip technology offer best in class power density solutions in conjunction with extended lifetime. The modules feature low on state losses, opti-mized soft switching behavior and a wide operation temperature range up to 150°C maximum junction operation temperature. The newly introduced stack assembly ModSTACK™ HD leads to more than 50% higher power density at same footprint.

The following benefits are provided to our customers:� Extended module utilization by 150°C maximum junction operation temperature� Highest power density� Supreme power cycling and thermal cycling capability

[ www.infineon.com/highpower ]

DC- Link Circuit

AC to DC Rectifier

DC to AC IGBT Inverter

ACSource

==

PCIM Booth 12 / 404


Reducing the energy required in data cen-

ters is a top priority. Data center electricity

consumption now approximates 2.5 percent

of total energy use in the United States,

according to a 2009 report from Lawrence

Berkeley National Laboratory. It will continue

to climb rapidly as the mobile Internet, cloud

computing, and other technological trends

mature. As of 2009, data center energy con-

sumption was rising at about 12 percent per

year, according to a 2009 VMWare Corpora-

tion report. Total power costs in the US

alone are now close to $3.4 billion annually.

As a result, strategies to reduce power con-

sumption, manage capacity, and promote

environmental responsibility are critical

objectives.

These strategies are vital as the number of

servers in data centers grows by approxi-

mately 10 percent annually, according to a

2009 McKinsey and Company report. The

new generations of servers are complex and

potentially power-hungry. For example, the

number of DC-DC regulators in a typical

server now is huge, with 5 or 6 phase regu-

lators used for the CPU Vcore. Altogether,

they deliver up to 150A peak at 1V or 150

watts per CPU. In addition, memory rails

can dissipate between 25 to 120 watts more.

For other rails, dissipation is more modest,

at a few hundred milliwatts to 5 watts each.

But the total adds up fast.

The proliferation of servers from corporate

and IP service providers to embedded appli-

cations such as wireless base station net-

work controllers or routers requires new,

highly efficient power management tech-

niques. One solution is powering-off excess

servers in the data center, which can deliver

immediate energy cost savings by conserv-

ing power. Fewer systems clearly translate

into less power and reduced operating costs.

A native workload on an entry-level server

with low utilization will consume 50W of

energy costing around $600 annually, where-

as a virtual machine workload on a server

hosting 16 virtual machines uses only a frac-

tion of that power — 5W, costing around $45

annually.

Server virtualization can help by reducing

the use of hardware as the loading decreas-

es. Another way to reduce consumption is to

increase the light load efficiency of the

power chain from AC to DC conversion to

the point of load. A typical server spends

lots of time operating at loading points with

low efficiency. A typical personal computer

likewise operates at relatively low power uti-

lization much of the time. Virtualization soft-

ware improves efficiency, maximizing utiliza-

tion in a server farm by ensuring that each

server operates at peak MIPS rates.

Efficiency also can be improved by imple-

menting digital core controllers that compa-

nies such as Intersil are developing and

enhancing. This kind of power management

technology has been developed for use in

mobile and computing applications. Consider

improvements in light load efficiency. In this

case, the CPU core regulator is able to

achieve over 90 percent efficiency from full

load which can be over 100 amps to light

load at close to 1 amp, or two orders of

magnitude. In the data center, this type of

high power load exists for core CPUs and

dense memory in servers and also for cus-

tom ASICs that handle the network data traf-

fic.

To help control the dissipation, companies

like Intersil are developing new multiphase

and point of load architectures to Improve

server DC:DC efficiency. Multi-phase regu-

lators such as Intersil’s VR12 6 phase regu-

G U E S T E D I T O R I A L

New Power Technologies Set to Improve Energy Efficiency

in Data Centers Increases in Power Demand, Costs, Require Strategies and Tactics to Reduce Consumption and Promote Environmental Responsibility

By Peter Oaklander, Senior Vice President, Intersil Corporation

1.2V OUTPUT, 300kHz, 230nH1.2V OUTPUT, 300kHz, 230nH

94

1.2V OUTPUT, 300kHz, 230nH

93

94


92

93

94


92

93

94


92

93

94


91

92

93

94

%)


90

91

92

93

94

y (%

)


90

91

92

93

94

ncy

(%)


89

90

91

92

93

94

ienc

y (%

)


88

89

90

91

92

93

94

ffici

ency

(%)


6-Phase5-Phase88

89

90

91

92

93

94

Effic

ienc

y (%

)


6-Phase5-Phase4 Ph

87

88

89

90

91

92

93

94

Effic

ienc

y (%

)


6-Phase5-Phase4-Phase3-Phase87

88

89

90

91

92

93

94

Effic

ienc

y (%

)


6-Phase5-Phase4-Phase3-Phase2 Ph86

87

88

89

90

91

92

93

94

Effic

ienc

y (%

)


6-Phase5-Phase4-Phase3-Phase2-Phase1-Phase

85

86

87

88

89

90

91

92

93

94

Effic

ienc

y (%

)



85

86

87

88

89

90

91

92

93

94

Effic

ienc

y (%

)



85

86

87

88

89

90

91

92

93

94

0 20 40 60 80 100 120 140 160 180

Effic

ienc

y (%

)



85

86

87

88

89

90

91

92

93

94

0 20 40 60 80 100 120 140 160 180

Effic

ienc

y (%

)



85

86

87

88

89

90

91

92

93

94

0 20 40 60 80 100 120 140 160 180

Effic

ienc

y (%

)



85

86

87

88

89

90

91

92

93

94

0 20 40 60 80 100 120 140 160 180

Effic

ienc

y (%

)



85

86

87

88

89

90

91

92

93

94

0 20 40 60 80 100 120 140 160 180

Effic

ienc

y (%

)



85

86

87

88

89

90

91

92

93

94

0 20 40 60 80 100 120 140 160 180

Effic

ienc

y (%

)



85

86

87

88

89

90

91

92

93

94

0 20 40 60 80 100 120 140 160 180

Effic

ienc

y (%

)



85

86

87

88

89

90

91

92

93

94

0 20 40 60 80 100 120 140 160 180

Effic

ienc

y (%

)



85

86

87

88

89

90

91

92

93

94

0 20 40 60 80 100 120 140 160 180

Effic

ienc

y (%

)



85

86

87

88

89

90

91

92

93

94

0 20 40 60 80 100 120 140 160 180

Effic

ienc

y (%

)



85

86

87

88

89

90

91

92

93

94

0 20 40 60 80 100 120 140 160 180

Effic

ienc

y (%

)



85

86

87

88

89

90

91

92

93

94

0 20 40 60 80 100 120 140 160 180

Effic

ienc

y (%

)



85

86

87

88

89

90

91

92

93

94

0 20 40 60 80 100 120 140 160 180

Effic

ienc

y (%

)



85

86

87

88

89

90

91

92

93

94

0 20 40 60 80 100 120 140 160 180

Effic

ienc

y (%

)



85

86

87

88

89

90

91

92

93

94

0 20 40 60 80 100 120 140 160 180

Effic

ienc

y (%

)



85

86

87

88

89

90

91

92

93

94

0 20 40 60 80 100 120 140 160 180

Effic

ienc

y (%

)



85

86

87

88

89

90

91

92

93

94

0 20 40 60 80 100 120 140 160 180

Effic

ienc

y (%

)



85

86

87

88

89

90

91

92

93

94

0 20 40 60 80 100 120 140 160 180

Effic

ienc

y (%

)



85

86

87

88

89

90

91

92

93

94

0 20 40 60 80 100 120 140 160 180

Effic

ienc

y (%

)



85

86

87

88

89

90

91

92

93

94

0 20 40 60 80 100 120 140 160 180

Effic

ienc

y (%

)



85

86

87

88

89

90

91

92

93

94

0 20 40 60 80 100 120 140 160 180

Effic

ienc

y (%

)



85

86

87

88

89

90

91

92

93

94

0 20 40 60 80 100 120 140 160 180

Effic

ienc

y (%

)



85

86

87

88

89

90

91

92

93

94

0 20 40 60 80 100 120 140 160 180

Effic

ienc

y (%

)



85

86

87

88

89

90

91

92

93

94

0 20 40 60 80 100 120 140 160 180

Effic

ienc

y (%

)



85

86

87

88

89

90

91

92

93

94

0 20 40 60 80 100 120 140 160 180

Effic

ienc

y (%

)



19www.bodospower.com May 2010 Bodo´s Power Systems®

lator are designed specifically to improve

efficiency during light load conditions using

Newly developed algorithms such as auto

phase dropping, diode emulation mode and

gate voltage over threshold can improve effi-

ciency by as much as 20 percent at 10 per-

cent load conditions — and even more as

utilization drops. Efficiency can maintained

over two orders of magnitude from a few

amps to nearly 100 amps.

There are other new integrated power

stages such as DrMOS that allow higher

switching frequencies with less loss, due to

lower Ron figures and less parasitic FET

capacitance. For the other rails, newer reg-

ulators borrow techniques from portable

systems, providing other means of improv-

ing efficiency, including switching from PWM

to PFM, and integrating FETs for higher

switching speed and density.

Considering the high power CPU, memory

and ASIC power rails – as well as the prolif-

eration of other rails for field programmable

gate arrays, auxiliary analog, I/O and stand-

by circuits — the total benefit of these archi-

tectures can be significant.

Another impetus to improve efficiency is to

add intelligence to the power chain. Digital

power management technology in conjunc-

tion with virtualization can help concentrate

CPU activity to a subset of data center

servers, so large numbers of idled servers

can be reduced to low power states easily.

Digital power also allows the monitoring of

input and load current, voltage and power,

with diagnostic functions like over

voltage/current and temperature. This

allows the data center system controller to

monitor the efficiency and adjust based on

real-time conditions. Digital power manage-

ment ICs such as the Zilker Labs ZL2106

provide advanced algorithms that adapt the

conversion to different load situations and

communicate back information to the host.

Digital power converters are being used in

communications infrastructure systems

where high-performance conversion and

management of power is critical.

With these kinds of products and technical

capabilities, the challenge of reducing ener-

gy consumption and optimizing power use in

data centers can be met, even while the

number of data centers and servers per cen-

ter continues to expand.

www.intersil.com

www.circuitprotection.com© 2009 Tyco Electronics Corporation. All rights reserved.

www.tycoelectronics.com

PolySwitch, PolyZen, TE (logo) and Tyco Electronics are trademarks of

the Tyco Electronics group of companies and its licensors.

SuperSpeed USB Circuit Protection Solutions

USB 3.0 delivers 10 times the data rate of USB 2.0 and can

use nearly twice the power. So protecting your circuit from

overcurrent, overvoltage and ESD damage is all the more

critical to help assure reliable performance.

You can rely on Tyco Electronics Circuit Protection for a

complete range of products and the applications expertise

you need.

• Innovative PolyZen overvoltage protection

• The latest in silicon-based and polymer ESD protection

• Industry-leading PolySwitch resettable overcurrent protection

For the latest information, go to www.circuitprotection.com/usb3

Biricha Digital Power

offering

Digital Power Supply Workshop

based on TI's F28x family.

For more information and your free

drill hole stencil please visit

www.biricha.com


The laws of thermodynamics are

not in immediate danger, but

devices that create usable power

from ambient energy are increas-

ingly capable of providing that

illusion. Transducers that create

electric energy from latent physi-

cal sources such as temperature

differentials (thermoelectric gen-

erators and thermopiles),

mechanical vibration or strain

(piezoelectric and electro-

mechanical devices) and light

(photovoltaic solar cells) are

becoming viable sources of

power for many applications,

supplementing, or in many cases

replacing, wired power and batteries. Increased safety and accessi-

bility, lower maintenance costs, improved energy efficiency and sys-

tem flexibility are just some of the benefits attainable with harvested

energy. Energy harvesting technology is enabling a wide variety of

remote, autonomous and wireless sensing and monitoring/control

systems in diverse applications such as transportation infrastructure,

automotive and avionics, remote metering, industrial process control

and building automation.

Energy transducers tend to fall into two broad classes when modeled

electrically. The first class presents a relatively low output voltage

and low impedance to the power conditioning circuit. Examples

include thermoelectric generators (TEGs) and solar cells. For these

elements, the power conditioning circuit must start up with a small

applied voltage in the tens to hundreds of millivolts range. As an

example, the LTC3108 Step-Up Converter and Power Manager oper-

ates from inputs as low as 20mV, enabling operation with TEGs in

response to temperature differentials as small as a degree or two.

The second class of transducers presents a relatively higher voltage,

high impedance source to the power conditioner. A typical piezo-

electric element can generate a load line with an open circuit voltage

of 5V to 40V and a short circuit current of 10μA to 50μA. The

LTC3588 Piezoelectric Energy Harvesting Power Supply, for exam-

ple, includes an integrated low-loss full wave bridge rectifier to opti-

mally mate with piezo elements and features a 20V maximum input

voltage to allow for operation near the transducer’s peak power

transfer point with high density energy storage using 25V-rated

capacitors.

Along with advances in energy transducer technology, power man-

agement products compatible with the low output levels achievable in

many environments are catalysts for commercial adoption. Since the

ambient energy and generated power levels may be quite small,

there are several critical attributes of an effective power management

system. First, low quiescent (active, no load) current is a key deter-

minant of energy harvesting performance. The transducer output

must overcome the operating current required by the power manage-

ment circuits before excess power can be applied to the intended

application. To address this need, the Linear Technology products

mentioned above feature ultra-low quiescent currents ranging from

6uA to less than 500nA.

Second, extracting useful amounts of power from these transducers

remains a challenge due to the low power operating regime. There-

fore high power conversion efficiency across a wide range of load

conditions is a key design goal. This topic has been successfully

addressed in the LTC3588, for example, in which the use of an

advanced hysteretic step-down switching regulator topology enables

efficiency approaching 90% for loads from 60μA to 100mA at 3.3Vout

– a three decade-wide load current range.

Finally, the ability to manage and store the harvested power is as crit-

ical as high energy conversion efficiency. This is especially true

when the input energy source is not continuously available or in

applications where the average power requirement is matched to the

harvesting transducer, but the peak application power draw exceeds

the transducer output capability. An example is a remote sensor

installation or asset tracking tag that is usually in a low power sleep

mode, but then wakes up to gather and transmit telemetry data in a

relatively high power burst. In this case the ability to manage power

between the primary output and a storage element (energy reser-

voir) is valuable. The use of supercapacitors for high density energy

storage and long charge cycle lifetime is an attractive choice. The

LTC3108 Power Manager provides this functionality autonomously

and with a minimum of external components for thermal- or solar-

based systems.

Finally, the power management device should provide standard, reg-

ulated outputs in order to match the power requirements of the sen-

sors, data converters, processors, radio transceivers and other cir-

cuitry that make up the application load. And of course carefully inte-

grated solutions optimize manufacturability, reliability and perform-

ance while simplifying the design process and shortening develop-

ment time.

The availability of power management products that interface effec-

tively with transducers and condition the harvested energy into

usable power are enabling performance breakthroughs that lead to

commercially viable energy harvesting-based products and systems

in many diverse markets. Recognition and acceptance of these new

products will drive excellent growth opportunities in 2010 and

beyond.

www.linear.com

T H E E X P E R T S V I E W

Energy Harvesting: BreakingThrough to Commercial Viability

By Donald E. Paulus, Vice President and General Manager Power Products, Linear Technology Corporation

- 66

86 -

03-

2010

PCIM Booth 12 / 510


GENERAL

Bad times for automo-

tive electronics suppli-

ers: in 2009, total vehi-

cle production in

Europe (cars, trucks,

buses) decreased by

17.3 percent com-

pared to 2008 and by

23 percent compared

to the pre-crisis level of 2007, so the ACEA.

Passenger car production dropped by 13

percent to 13.4 million units, or the lowest

level in fourteen years. But the production of

passenger cars went up 22.8 percent in the

fourth quarter compared with the low last

quarter of 2008. Germany remained by far

the largest vehicle producer (5.2 million units

in 2009) in the EU, despite a 13.8 percent

decrease. Following a 20.2 percent drop,

France fell to third rank while Spain (-14.6

percent), with 2.2 million vehicles, became

the second largest manufacturing country in

2009. The UK (-33.9 percent) ranked fourth

with more than 1 million units.

SEMICONDUCTORS

The semiconductor business now is more

profitable than it has been at any time in the

last decade, reflecting the industry’s increas-

ingly aggressive management of costs,

capacity and competitive positioning, so

iSuppli. Overall semiconductor supplier oper-

ating profitability rose to 21.4 percent in

fourth quarter of 2009, the highest level

since the fourth quarter of 2000 when it

reached 24.7 percent.

The German semiconductor market is

expected to recover this year and grow by

13 percent to € 8 billion, so the ZVEI. The

growth will be supported by the 15 percent

turnover in discrete elements, optoelectronic

semiconductors and sensors/actuators, as

well as the 12 percent growth in the IC seg-

ment.

Intersil, a supplier of analog and mixed sig-

nal semiconductors, has entered into a

definitive agreement to acquire Techwell, a

fabless semiconductor company that sells

mixed signal video solutions for the security

surveillance and automotive infotainment

markets.

In the presence of the German Federal

President, the new 200 mm semiconductor

fab at the Bosch location in Reutlingen has

gone into operation. At a total cost of € 600

M, the new facility, which will manufacture

semiconductors and micromechanical com-

ponents, is the largest single investment in

the history of the Bosch group.

MagnaChip Semiconductor filed a registra-

tion statement on Form S-1 with the U.S.

Securities and Exchange Commission relat-

ing to the proposed initial public offering of

its common stock and plans to raise up to $

250 M. Headquartered in South Korea,

MagnaChip Semiconductor is a designer

and manufacturer of analog and mixed-sig-

nal semiconductor products for high volume

consumer applications.

The two European clusters of Dresden and

Grenoble in nanoelectronics and nanotech-

nologies announce the foundation for a

structured and strengthened cooperation, in

the areas of R&D, education, industry &

institutions.

Synopsys, a world leader in software and IP

for semiconductor design, verification and

manufacturing, and the Belgian nanoelec-

tronics research centre, IMEC, have entered

into a collaboration to use Synopsys TCAD

(Technology Computer-Aided Design) finite-

element method tools for characterizing and

optimizing the reliability and electrical per-

formance of through-silicon vias (TSVs). The

collaboration will accelerate the development

of 3D stacked IC technologies.

The worldwide semiconductor materials mar-

ket contracted 19 percent in 2009, at $ 34.6

billion, as the semiconductor industry react-

ed quickly to deteriorating market conditions

in the first part of the year, so SEMI. Total

wafer fabrication materials and packaging

materials were $17.9 billion and $16.8 bil-

lion, respectively.

OPTOELECTRONICS

The revenues of global LED packaging man-

ufacturers reached $ 8.05 billion, a 5 percent

growth compared to 2008, so LEDinside.

Nichia maintained its No. 1 position in the

world in 2009, followed by Osram Optoelec-

tronics, Cree, and Samsung.

PASSIVE COMPONENTS

Germany's printed circuit board industry

made an unexpected recovery in December

2009, growing 10 percent compared to the

previous month and posting a book-to-bill

ratio of 1.36-the highest since more than

three years, so the ZVEI. Incoming orders

for December were up 120 percent com-

pared to the same period last year, although

sequentially, it was down 16 percent. Still,

Germany's overall PCB industry for 2009

was down 30 percent compared to 2008.

DISTRIBUTION

Avnet Memec, the highly specialised semi-

conductor distributor of Avnet Electronics

Marketing EMEA, has been honoured by

Actel as “European Distributor of the Year“.

Avnet Memec has the sales franchise for

Actel products in Germany, France and the

UK.

Avnet has entered into a definitive agree-

ment to acquire certain assets of value-

added distributor Servodata in the Czech

Republic.

RS Components has signed a two year glob-

al supply agreement with Cobham, one of

the world’s largest aerospace and defence

contractors. Under the agreement, RS Com-

ponents is Cobham’s distributor of choice for

electronic and electromechanical compo-

nents in the UK, France, USA and Denmark,

for development and small batch production

quantities.

This is the comprehensive power related

extract from the « Electronics Industry Digest

», the successor of The Lennox Report. For

a full subscription of the report contact:

[email protected]

or by fax 44/1494 563503.

www.europartners.eu.com

M A R K E T

ELECTRONICS INDUSTRY DIGESTBy Aubrey Dunford, Europartners

PCIM Booth 12 / 321


Activity is heating up in the vehicle electrification sector, and not all of

it is in electric vehicles (EVs). There is a broad trend toward the elec-

trification of all vehicle functions, which extends the opportunities for

power supply makers beyond automotive EVs, hybrid-electric vehi-

cles (HEVs) and plug-in hybrid electric vehicles (PHEVs). For exam-

ple, environmental concerns are often more important than “efficien-

cy.” Issues with hydraulic fluids, along with truck anti-idling regula-

tions, are driving inverter sales for turf care vehicles and heavy-duty

trucks, for example. The current worldwide market for “auxiliary”

inverters alone is approximately $500 million (€370 million).

Europe’s Euro 5 and 6 standards are expected to help bring conver-

gence to global emissions standards, ending the need for multiple

engine platforms. Heavy-duty trucks represent a growing market for

inverters in Europe, although it is still a relatively low penetration rate.

The use of inverters in heavy-duty trucks gained strong interest 5-10

years ago, as on-board devices started to proliferate. In the early

2000s, Volvo, Freightliner, Navistar and Western Star started offering

inverters as a factory option. The majority of truck original equipment

manufacturers now offer ac power infrastructure and shore power

connections. Many Volvo trucks include the shore power option, but

not all use inverters. Recently, Magna Steyr announced that it will

produce lithium-ion battery systems for Volvo Group. The battery sys-

tems will be integrated into Volvo’s city buses, heavy-duty distribution

vehicles and refuse trucks..

The market for on-board auxiliary power inverters in Europe varies

considerably from application to application. On the one hand, Euro-

pean countries are heavily subject to regulations that favor the use of

low-emission, efficient and “green technology” – which inverters are,

compared with generators. Some applications, like emergency vehi-

cles, have high market penetration rates. In other segments, like

heavy-duty trucks, inverters are viewed more as a “luxury” feature.

Due to heightened security concerns, fire rescue departments across

Europe have focused on purchasing specialized equipment needed

to respond to terrorist attacks and natural disasters. Instant power is

needed for scene lighting and tools, including ventilation fans to clear

smoke from a building or hydraulics for rescue workers. Vehicles in

Europe tend to be smaller and more compact than those in North

America. In Europe, a custom fire chassis usually isn’t used. They

tend to use Mercedes Benz, Volvo, Scania and others.

Recreational vehicles are a “mid-sized” opportunity for on-board aux-

iliary inverter unit sales in Europe. “Caravaning” is not the ubiquitous

activity that it is in North America, due to higher fuel costs and colder

weather throughout much of the region over the year. The three

largest motor caravan markets in Europe are Germany, France and

Italy. French and other European vehicles dominate this market.

An alternative powering technology is fuel cells. Inverter use has

increased over the past few years, primarily due to (1) the increased

use of on-board devices; and (2) efficiency and environmental con-

cerns. Fuel cells are another technology that can address these con-

cerns and reduce generator use. In January, 2008, SFC Smart Fuel

Cell AG announced major orders from France, England, Ireland, Italy

and Benelux countries for products designed for recreational vehi-

cles. The company said it had increased its “competitive edge as a

supplier of on-board power equipment for motor homes in the Euro-

pean markets.”

With the delivery of two fuel-cell trucks to the Linde Gases Division in

February, 2010, The Linde Group - Linde Material Handling, stated

that it had taken an “important step” on the road to more intensive

use of automotive technology. The fuel cell trucks were developed

over the past two years with long-term partner, Hydrogenics, the

Canadian fuel cell manufacturer. The benefit of the fuel-cell drive for

Linde Gas is the “zero emissions” that these trucks produce when

used. Another benefit is that there is no longer a requirement for bat-

tery replacement or a battery charging process.

Bi-directional dc-dc converters are typically included in vehicle electri-

fication architectures, as well. They provide a “bridge” between high-

voltage and low-voltage sections of the system, with typical ratings

up to 5kW. The load profile of specific functions affects whether they

are suitable for electrification. “Start-stop” loads are the most suitable,

since they enable the maximum benefit from regenerative energy

capture. These include vehicles such as buses, refuse trucks, lift

buckets and tractors.

The traditional automotive market (i.e. passenger vehicles and light

trucks) gets the most press in terms of EV technology, but other vehi-

cle classifications are expected to provide more opportunities in the

short-term. For example, due to business and regulatory incentives,

the heavy-duty truck market is expected to grow 55.2% between

2008 and 2013. Off-road utility vehicles and agriculture/construction

vehicles are also expected to have significant sales over the next few

years. These sales are not at the level of passenger cars, but the

market for inverters, dc-dc converters and battery chargers is com-

mercially viable now. That makes them an attractive, immediate mar-

ket.

There is plenty of activity around the EV, HEV and PHEV markets,

however. These vehicles are an “emerging” segment, and most of the

opportunities still rely on various partnerships, regulations, research,

and incentives for companies. Many of these initiatives are occurring

in Europe and currently support components, batteries and other

energy storage technologies.

For example, IBM recently announced an agreement with the Energy

Technologies Institute (ETI) to evaluate the potential impact of elec-

tric vehicles on the UK electricity grid. The project will also assess

the infrastructure required to achieve a mass market for electric and

plug-in hybrid electric vehicles in the UK. Tata Motors European

Technical Centre plc announced a €10 million loan under the United

Kingdom government’s Automotive Assistance Program to develop

and manufacture the Tata Indica Vista Electric Vehicle in the UK with

an investment of €25 million.

M A R K E T

Vehicle Electrification Moving in Different Directions

By Linnea Brush, Senior Research Analyst, Darnell Group

Late last year, Infineon Technologies AG launched what is described

as, “Europe’s largest research project to advance the development of

electric vehicles.” The E3Car (Energy Efficient Electrical Car) project

brings together 33 automotive companies, key suppliers and

research facilities from a total of 11 countries to collaborate on boost-

ing the efficiency of electrically-driven vehicles by more than a third.

In fact, numerous partnerships have been formed that will help sup-

port the burgeoning EV market in Europe and globally. Schneider

Electric, Legrand & Scame announced that they were forming the EV

Plug Alliance to promote the use of a high safety plug and socket

solution for electric vehicle charge infrastructure. The plug endorsed

by the Alliance will ensure compatibility between multiple suppliers’

products.

Companies are also forming partnerships to further electric vehicle

technology. Volkswagen and Varta Microbattery are reportedly collab-

orating on research to develop next-generation lithium-ion battery

systems for EVs. Magneti Marelli and STMicroelectronics have

signed a memorandum of understanding that lays the foundation for

an agreement in the sector of power electronics modules and compo-

nents for inverters to be fitted on hybrid and electric vehicles. And

Eltek Valere announced that it will be delivering battery chargers to

THINK for its THINK City electric vehicle.

These steps toward EV-related products and components indicate

that vehicle electrification, in all its forms, is growing and will be a sig-

nificant part of the transportation market over the next few years.

www.darnell.com/hybrid

www.darnell.com

NDM1-12

NDM1-25

www.novumdigital.com

V-Infinity’s new 12 A and 25 A digital DC-DC Point-of-Load (POL) modules are aimed at the emerging digital

power management and control market. The Novum product line is focused on providing a complete, easy-to

implement solution, with the goal of making the benefits of digital power accessible to a wide array of users.

Rail/marine drive controls

Wind power & solar power controls

Large motor drive controls

Bicron Electronics specializes in the design and manufacture of custom high frequency transformers for critical-use applications with frequencies up to 1 Mhz.

High IsolationSwitchmodeLoad Leveling

Gate DrivesSignal ConditioningPulse

When failure is not an option, choose Bicron.

Bicron offers the following transformer types:

[email protected]+49(0)2871 7374

BICRONElectronics

Ultra-reliableIGBT transformers

. . . isolation for operating voltages up to 20KV!

26 Bodo´s Power Systems® May 2010 www.bodospower.comBodo´s Power Systems® May 2010 www.bodospower.com

The goal of high efficiency drives modern

power converters to greatly increased com-

plexity and highly optimized solutions at dif-

ferent power levels. Reuse or re-targeting of

building blocks can increase design produc-

tivity and offers a solution to decreasing

time-to-market while in particular meeting

increasing diversity and complexity. As a

consequence, the SCALE-2 driver chipset

[1] offers the option of easily scaling the con-

verter power by direct parallel operation of

standardized subsystems composed of

IGBTs and dedicated gate drivers [2].

Plug-and-Play Drivers for High-Power and

High-Voltage IGBTs

Thanks to SCALE-2 technology, the new

1SP0635 and 1SP0335 families comprise

highly integrated, high-performance, com-

plete and extremely compact single-channel

IGBT drivers equipped with DC/DC convert-

ers, short-circuit protection, Advanced Active

Clamping, regulated turn-on gate driving

voltage, and supply-voltage monitoring.

Users need only mount them onto the corre-

sponding IGBT module. The system can

then be put into immediate operation with no

further development or matching effort.

These new drivers are perfectly matched to

130 x 140mm and 190 x 140mm IGBT mod-

ules from various manufacturers. The

1SP0635 is designed for a voltage class of

1200 to 3300V and a current range up to

3600A. See Fig. 1.

The similar 1SP0335 gate driver (Fig. 3)

focuses on 3300V modules with 10.2kV iso-

lation voltage as well as 4.5kV and 6.5kV

modules and uses the ISO5125I new dedi-

cated external DC/DC converter for up to

5W gate power.

Direct Paralleling by Means of a Master-

Slave Architecture

Furthermore, these drivers offer a master-

slave architecture allowing direct paralleling

of IGBTs by simple means, see Fig. 2. The

master is equipped with a fiber-optic inter-

face and global fault management.

The slaves are connected to the master by a

bus cable which distributes the common

command signal and the secondary-side

supply voltages for the DC-DC converter.

C O V E R S T O R Y

Easy Parallel Connection ofIGBT Modules at the Next

SCALE LevelA Unified Direct Paralleling Approach Applied to

IGBT Voltage Classes 1.2 to 6.5kV

The SCALE-2 implementation of high-voltage and high-power IGBT gate drivers offerscompetitive advantages such as a dramatically reduced component count, exceptionalcost performance and wide application and topology diversity: from single-switch to

parallel connection within multi-level converters for industry and traction applications,renewable energy and HVDC power transmission.

By Jan Thalheim, Olivier Garcia, Sascha Pawel, Heinz Rüedi, CT-Concept Technologie AG, Switzerland

Figure 1: SCALE-2 Plug-and-Play driver 1SP0635 for 1200V to 3300V IGBTs

Figure 2: Driving parallel-operating IGBTs with individual drivers

Thanks to the extremely low jitter and negli-

gible variance of propagation delay of the

SCALE-2 chipset, all IGBTs operate at virtu-

ally the same gate-driving voltage. The main

advantage over the use of a central gate

driver is the unlimited and easy scalability for

a wide range of applications and power lev-

els at optimum performance.

High Performance at Lowest Cost

A particular advantage is the optimum scal-

ing of the Advanced Active Clamping func-

tion which enables full utilization of single-

switch performance within a parallel connec-

tion of IGBTs. Application specific integrated

circuits (ASICs) are used here because they

reduce system complexity and therefore

lower manufacturing costs while increasing

reliability and system performance.

It is the strength of CONCEPT as an inde-

pendent and highly experienced gate-driver

supplier for medium and high-power applica-

tions to overcome the obstacles of monolith-

ic integration in this highly specific market by

means of broad application coverage and a

large combined quantity of drivers delivered

to a wide variety of customers.

Combined with the inherently correct scaling

of driver output stages and supply capaci-

tors, the master-slave architecture offers

superior total-cost-performance ratio of the

overall system.

Relevant cost reduction also results from a

dramatic reduction of time-to-market, devel-

opment effort and production cost compared

to custom-specific low-volume IGBT drivers.

Maximum Utilization and Extended Safe

Operation by Advanced Signal Processing

The architecture of the master and slave

gate drivers is shown in Fig. 4. The new

plug-and-play drivers introduce the following

application advantages:

• Dynamic Advanced Active Clamping DA2C

temporarily allows extremely high DC-link

voltages. This is a particular advantage for

traction, windmill and solar converters.

• Dynamic short-circuit detection to protect

the IGBTs fully from any kind of short-cir-

cuit at any level of DC-link voltages while

fully utilizing the collector current capabili-

ty of both slow conduction-loss optimized

and fast-switching IGBTs.

• Centralized monitoring of gate-emitter volt-

ages of all individual drivers by the master

to ensure correct parallel operation.

A Reliable and Long-Term Available Com-

ponent

The SCALE-2 implementation offers compet-

itive advantages such as exceptional cost

performance, long-term availability and tried-

and-tested SCALE technology. The chipset

has been developed on the basis of two

independent semiconductor processes (true

second source) while retaining full functional

and parametric compatibility.

In spite of increased functional complexity

compared to the similar 1SD536F2 SCALE

driver, the overall component count and the

estimated failure rate are reduced through

an exceptional level of integration achieved

with the SCALE-2 chipset.

www.bodospower.com May 2010www.bodospower.com May 2010

Figure 3: SCALE-2 Plug-and-Play driver1SP0335 for 3300V to 6500V IGBTs

Figure 4: Master slave architecture of SCALE-2 Plug-and-Play driver 1SP0635


Beyond that, a long service life and safe operation are achieved

thanks to increased thermal and gate current capability at high ambi-

ent temperatures and superior EMI immunity.

Outstanding signal integrity has been achieved by using PCB-inte-

grated inductances within the power supply buses and differential

15V CMOS logic signal processing using planar transformers [3] to

decouple the individual IGBT gate and emitter potentials during col-

lector current redistribution. This decoupling ensures safe signal

transfer even under extreme conditions such as severely asymmetric

operation (e.g. as a consequence of failure), extremely high dV/dt

and di/dt or dH/dt, or a command change during a switching transi-

tion.

Clearance and creepage distances comply with both IEC 60077-1

and EN 50178 for pollution degree 2 and overvoltage category 2.

The paralleling bus interface is realized by a miniaturized high-relia-

bility connector system which is used in the widest range of applica-

tions in automotive, industrial and medical sectors to achieve high

vibration and shock-load capability as well as broad temperature

capability.

To satisfy the requirements of multi-level converter topologies, the

external ISO5125I DC/DC converter is available in different versions

up to a specified operating voltage of 12kV (partial discharge extinc-

tion voltage above 9.4kV AC to IEC 61287).

Experimental verifications and field data gathered from products

shipped in large item numbers since 1999 have shown no critical

impact of the IGBT baseplate and junction temperature on gate-driver

reliability for typical industrial and traction applications.

Assuming that the degradation of optical output power is the main

failure criterion, the estimated lifetime of the fiber optic transmitter for

a surface temperature of 85°C is above 208’000 hours, which is more

than 23 years of permanent operation. (Based on time performance

according to manufacturer data).

Superior Switching Behavior

The easy adaptation of the drivers permits an optimum setup to han-

dle the special demands of a wide range of applications.

Synchronous switching of two parallel operating FZ1500R33HE3

IGBT modules from Infineon is shown in Figs. 5 to 6.

The SOA compliance has been verified for a collector current of up to

twice the nominal current, and under short-circuit conditions for a

junction temperature up to 150°C and a maximum permitted DC link

voltage VDC of 2200V.

Worst-case conditions for the active clamping effect have been veri-

fied by setting the temperature of the active clamping circuitry to

125°C, but the IGBT junction temperature to 25°C. Thanks to the

advanced driver architecture with integrated active clamping, IGBT

operation is kept within the safe operating area with a sufficient mar-

gin up to a total DC link inductance of 180nH.

For 1700V IGBTs, the maximum DC-link voltage is specified to

1200V and may be increased up to 1450V and beyond in the off-

state by DA2C to enhance the safety margin for traction, wind and

solar power applications.

For the 6.5kV version of the 1SP0335, the maximum permitted DC

link voltage for permanent switching is 4450V. In the off-state, the DC

link voltage may approach 5240V.

This exceptional performance is made possible by keeping the MOS

channel conducting during turn-off. The feedback signal is applied to

both the driver input and the IGBT gate to improve the efficiency of

the active clamping devices. This tried-and-tested architecture has

become a virtual standard ever since CONCEPT presented a plug-

and-play driver solution for a high-voltage IGBT for the first time ten

years ago [4].

A total gate drive capability of 6W is available for the 1SP0635, which

allows the parallel operation of three 3.3kV / 1500A modules con-

nected in parallel operating at more than 1500Hz. The single IGBT

gate drive capability is limited to 3W and 35A. The 1SP0335 also

allows the connection of more than one external ISO5125I DC/DC

converter to increase the available total gate power. The operating

ambient temperature range of the driver is defined as -40°C to 85°C.

Customized Solutions

Plug-and-play drivers are also offered in customized versions for

applications produced in volume quantities.

For example, an option will be available to separate the master from

the IGBT module. This will allow the system lifetime to be further

increased by reducing the ambient temperature of the fiber-optic

transmitter, which may be advantageous for ambient temperatures

above 60°C (depending on the mission profile). Customizing of

dynamic short-circuit detection is also available upon request.

Pricing and Availability

The pricing of the drivers is very competitive, thanks to the exception-

al level of integration achieved with the SCALE-2 chipset. In spite of

increased functionality compared to the similar SCALE driver

1SD536F2, the price level is 40% lower.

The drivers are now being shipped in sample quantities.

Volume production is planned to start in Q3 and Q4 2010 for the

1SP0635 and 1SP0335 respectively.

References

[1] J. Thalheim, H. Rüedi: Universal Chipset for IGBT and Power-

MOSFET Gate Drivers, PCIM Europe, 2007

[2] J. Thalheim, O. Garcia, S. Pawel: Fast Gate Drivers Simplify Par-

allel Operation of IGBTs, PCIM Europe, 2009

[3] S. Pawel, J. Thalheim: 1700 V Planar Transformers for High

Power Gate Drives, PCIM Europe, 2009

[4] H. Rüedi, P. Köhli: SCALE Driver for High Voltage IGBTs, PCIM

Europe, 1999

PCIM Booth 12/102

www.IGBT-Driver.com

Bodo´s Power Systems® May 2010 www.bodospower.comBodo´s Power Systems® May 2010 www.bodospower.com

Figure 5: Turn-on of two parallel-operating IGBTs with 1SP0635

Figure 6: Turn-off of two parallel-operating IGBTs with 1SP0635

C O V E R S T O R Y

Naturalmatch!

Features+15V/-10V gate voltage

3W output power

20A gate current

80ns delay time

Direct and half-bridge mode

Parallel operation

Integrated DC/DC converter

Electrical isolation for 1700V IGBTs

Power supply monitoring

Short-circuit protection

Fast failure feedback

Superior EMC

2SP0320 is the ultimate driver platform for PrimePACKTM IGBT

modules. As a member of the CONCEPT Plug-and-play driver

family, it satisfies the requirements for optimized electrical

performance and noise immunity. Shortest design cycles are

achieved without compromising overall system efficiency in

any way. Specifically adapted drivers are available for all

module types. A direct paralleling option allows integrated

inverter design covering all power ratings. Finally, the highly

integrated SCALE-2 chipset reduces the component count

by 80% compared to conventional solutions, thus signifi-

cantly increasing reliability and reducing cost. The drivers are

available with electrical and fiberoptic interfaces.

PrimePACKTM is a trademark of Infineon Technologies AG, Munich

2SP0320

SAMPLES AVAILABLE!

CT-Concept Technologie AG, Renferstrasse 15, CH-2504 Biel, Switzerland, Phone +41-32-344 47 47 www.IGBT-Driver.comPCIM Booth 12 / 102


To meet new performance goals beyond the reach of silicon, IR has

developed GaNpowIR™, the industry’s first family of commercial inte-

grated power stage products utilizing IR’s revolutionary Gallium

Nitride-on-silicon (GaN-on-Si) epitaxial power device technology plat-

form. GaN has a critical electric field which is substantially greater

than Si. Combined with device structures which yield excellent con-

ductivity, this provides GaN-based devices with a far superior trade-

off between operating voltage and on-resistance. GaN-based

devices are also capable of operating efficiently at much higher fre-

quencies than comparable silicon-based devices since they present a

significant drop in gate charge (Qg) and device switching

RDS(on)*Qg Figure-of-Merit (FOM) is much lower than for

silicon1.This allows the reduction in size of power conversion solu-

tions, a key enabling feature in today’s power hungry end application

markets. In addition, very efficient lateral devices are possible using

GaN-based technology which allows relatively simple scaling with

operating voltage, as well as for improved integration with other cir-

cuit functions, compared to state-of-the-art vertical Silicon based

power devices.

New GaNpowIR Product Family

Presently aimed at point-of-load (POL) and multiphase regulators for

servers, routers, switches, and general-purpose converters, the first

two members of this commercial GaN-based power stage GaNpowIR

product family includes the iP2010 and iP2011. These devices inte-

grate a dedicated PowIRtuneTM driver IC matched to a multi-switch

monolithic GaN-on-Si based power device as illustrated in Figure 1.

The proprietary PowIRtune™ driver has a superfast sensing scheme

to precisely tune deadtime for optimal performance and maintain the

optimal deadtime with variations in load current, input voltage and

temperature. The incorporated high-side (Q1) and low-side (Q2) GaN

power switches are combined in a monolithic power switch in order

eliminate the parasitic switching losses that are inherent with the cop-

per interconnects used in traditional power stage solutions that utilize

discrete high and low side switches. This along with the intrinsically

low on-resistance and low gate charge of the GaN-based technology

provide the GaNpowIR devices the ability to switch at frequencies as

high as 5MHz.

The iP2010 features an input voltage range of 7V to 13.2V and out-

put voltage range of 0.6V to 5.5V with an output current up to 30A

and can operate at up to 3MHz. Operating up to 5MHz, the pin-com-

patible iP2011 features the same input and output voltage range but

is optimized for up 20A output current. The GaNpowIR devices are

housed in a small LGA package measuring only 7.7 mm x 6.5 mm x

1.7 mm. By offering multiple current rating devices in a common

footprint, the product family provides flexibility for meeting a broad

range of customer requirements in terms of current level, perform-

ance and cost. To deliver high efficiency and more than double the

switching frequency of the state-of-the-art silicon-based power MOS-

FETs, the GaNpowIR components are mounted in a flip chip package

T E C H N O L O G Y

New Generation of GaN BasedPower Stage Products This generation heralds a era in high density

and highly efficient power Conversion

With the introduction of commercial HEXFET® power MOSFETs more than three decadesago, International Rectifier started a revolution in switch-mode power supplies (SMPS).Now, 30 plus years later, as silicon matures, the demand from industry for smaller and

more efficient DC-DC converter solutions continues.

By John Lambert, POL Marketing Manager, International Rectifier Corp., El Segundo, California, US

Figure 1: PowIRtuneTM driver IC matched to a multi-switch monolithicGaN-on-Si based power device

Figure 2: iP2010 and iP2011 deliver over 90 percent peak efficiencies


platform that eliminates the need for wirebonds. Combining GaN-

based technology advances with novel packaging enables the inte-

grated GaNpowIR devices to be optimized for operation in the

600kHz to 1.2MHz frequency range.

Taking into account the power stage, inductor and PC-board losses,

Figure 2 demonstrates how the iP2010 and iP2011 deliver over 90

percent peak efficiencies while operating at 1.2MHz with; 12V input,

1.2V output and using a 90nH inductor.

GaNpowIR Compared to Traditional Silicon-based Solutions

Although most designers associate GaN technology with the ability to

switch at higher frequencies than traditional silicon solutions, the

GaNpowIR devices can be operated at lower frequencies to achieve

highest possible efficiencies and energy savings. In fact, by operat-

ing the iP2010 at 600kHz, it can deliver much higher efficiencies than

the competing leading-edge silicon based commercial power stage

solutions operating at the same frequency.

As shown in Figure 3, the peak efficiency performance of iP2010 is

over 93 percent in the 11 to 14A output current range, which offers

more than 2 percent improvement over the nearest competing sili-

con based power stage device. This gap widens further as the out-

put current is increased to 30A. At this output current, the conversion

efficiency for the GaN based DC-DC converter is about 90 percent,

which is at least 4.5 percent higher than the best competing power

stage solution and 6.8 percent higher than the other competing solu-

tion.

Since GaNpowIR devices can operate at higher frequencies than tra-

ditional silicon-based solutions, they can provide maximum power

density when space is at a premium. By operating a power stage at

these higher switching frequencies the value and size of output

capacitors and inductors can be reduced, which in-turn reduces the

overall board space. Figure 4 shows the potential space saving of an

iP2010/11 power stage solution operating at 800kHz when compared

to a discrete solution or a DrMOS power stage solution operating

400kHz. In this example the discrete solution using a 4x4mm discrete

MOSFET driver, a 4x5mm high side MOSFET and a 5x6mm low side

MOSFET requires about 495mm2 of board space. The DrMOS power

stage integrates the driver and high and low side MOSFETs which

reduces space compared to the discrete solution, but still requires

450mm2 of board space. Since the GaNpowIR solution can operate

at double the frequency of the silicon solutions with about the same

efficiency it only requires 280mm2 of board space. This translates to

a GaNpowIR solution board space savings of more than 40 percent

compared to a typical discrete solution and more than 35 percent

compared to a typical DrMOS power stage solution. The space esti-

mate for the GaNpowIR solution includes a 3x3mm P-Ch MOSFET

and a 3x3mm Negative Voltage Generator IC. These components are

included in the space measurements because one or both may be

required for bias, sequencing and protection. However, many of

today’s system power boards already have provisions for the bias,

sequencing and protection and therefore may not be required.

Summary

The pioneering GaN-based power device technology platform is the

result of five years of research and development by IR based on the

company’s proprietary GaN-on-Si epitaxial technology. The high

throughput, 150mm GaN-on-Si, together with subsequent device fab-

rication processes which are fully compatible with IR’s cost effective

silicon manufacturing facilities, offers customers a world-class, com-

mercially viable manufacturing platform for GaN-based power

devices. By taking advantage of the inherent high frequency capabili-

ties of GaN, the iP2010 and iP2011 will allow the industry to provide

the world’s highest density solutions with little or no sacrifice in effi-

ciency when compared to traditional silicon solutions. These GaN-

powIR products can also enable world highest energy saving solu-

tions when operated in the higher end of the frequency range of

today’s silicon based solutions.

References

1. By Tim McDonald, International Rectifier, “GaN Based Power

Technology Stimulates Revolution in Conversion Electronics”, Bodo’s

Power Systems, April 2009, p.2

PCIM Booth 12/202

www.IRF.com

Figure 3: Efficiency performance of iP2010 is over 93 percent

Figure 4: Potential space saving of an iP2010/11 power stage solution

Biricha Digital Power offering





www.biricha.com


Introduction

In power electronics IGBTs are gaining more

importance since their introduction to the

market, not only from the installed number of

devices but also from the served applica-

tions.

Today’s high-power IGBT modules cover a

range of applications from industrial inverters

up to large traction motor drives, windpower

and HVDC converters.

The trend in IGBT modules continues to be

towards higher power densities. One way to

achieve this is by increasing current ratings

on the same footprint. This requires higher

power dissipations and operating tempera-

tures.

To meet these requirements, each part in the

construction of an IGBT-Module has to be

optimized. This article explains four of such

optimizations made on a HiPak module and

proven with a final product rated at 3600A

and 1700V.

HiPak Technology

The HiPak modules are high-power IGBTs in

industry-standard housings with the popular

190, 70 or 130 x 140 mm footprint as shown

in Figure 1. They cover a wide voltage range

from 1700 V to 6500 V and a current range

from 400 A up to 3600 A. In addition, three

different voltage categories for isolation volt-

ages of 4, 6.2 and 10.2 kVRMS are offered.

They are built in single IGBT, dual IGBT,

dual diode and in chopper configurations.

An IGBT module consists of IGBTs and

Diodes, built on substrates that are soldered

to a base plate. Terminals are conductor

leads, which provide the electrical connec-

tion from the electronic circuit on the sub-

strate to contacts outside the module. The

chipset and the terminals are protected with

a silicone gel moulding, an epoxy layer and

the housing.

Implemented Improvements

To achieve reliable operation under higher

currents and temperatures, the capabilities

of the chipset, terminals, soldering and the

silicon gel have been improved. This section

explains these improvements in detail.

Chipset

The development of high current modules

operating at high temperature is very

demanding with regards to both IGBT and

diode chip design. Soft and controllable

switching behaviour is essential when the

chips are utilised in high current modules

because the combination of high currents

and larger stray inductances will normally

result in higher overshoot voltages and

snappy behaviour during turn-off. To achieve

higher current on the same footprint the

1700V technology platform has been

upgraded from SPT to SPT+. Compared to

SPT, the SPT+ IGBT technology offers about

15% lower on-state losses while keeping

similar turn-off losses as shown in Figure 2.

The targeted 150°C junction temperature

requires stable and reliable operation of the

devices well beyond such limit. This imposes

a proper optimization of the termination

design and of the diode lifetime killing in

order to reduce the high temperature leak-

age. Figure 3 shows the cooler temperature

range where both IGBT and diode have

been proven to be stable without thermal

runaway under the application of a DC volt-

age of 1400V and 1700V over a time longer

than 300sec.

Finally, the chip set was qualified by stan-

dard reliability tests including HTRB (High

Temperature Reverse Bias) performed at full

voltage and Tj=150°C, HTGB (High Temper-

ature Gate Bias) and THB (Temperature

Humidity Bias 85°C/85% relative humidity).

Package

The packaging technology has to serve four

I G B T M O D U L E S

Enhancing an IGBT Module for High Temperature & High

Current OperationsA new generation of modules for demanding applications

The ever increasing requirements on high-power IGBT modules to operate at higherpower densities need a series of improvements on the total value chain. During the

development phase a team of engineers from different expertises contributes to theseimprovements. This article explains four of such improvements.

By B. Aydin, C. Corvasce, L. Feller, S. Hartmann, ABB Switzerland Ltd, Semiconductor

Figure 1 : The HiPak IGBT power modulefamily

Figure 2: Trade-off curve for SPT and SPT+for different stray inductances and gateresistors.

main functions. First, it must provide a cur-

rent path from the busbar to the chip and

then back. Secondly, it has to cool away the

heat generated in the module. Thirdly, the

package has to isolate the electrical contacts

from each other. Finally, the same package

needs to ensure its mechanical robustness.

Following improvements on the Gel, module

soldering and terminals enabled a robust

new product with 1700V blocking and 3600A

current rating.

High Temperature capable Gel

Silicone gels are used as dielectrics to pre-

vent partial discharge and to seal the system

against moisture and atmospheric contami-

nants. In addition to the trends towards oper-

ation at higher junction temperatures, envi-

ronmental needs are also targeting storage

temperatures down to -55°C, where the

power modules have to remain fully opera-

tional.

The existing insulation material is a silicone

gel (gel R) which is specified for an operat-

ing range of -40…150°C from the supplier.

The new requirements of -55°C … 175°C

and the new operational temperature of the

chips evoked a veri-

fication of the mate-

rial characteristics

of two promising

alternatives (gel S

and gel E).

The extended tem-

perature ranges of

the gels in the

datasheet together

with the dielectric

properties were

important require-

ments for the selec-

tion of the potential alternative gels. The

potential gel candidates have undergone

several tests and investigations.

In order to evaluate the thermal stability of

the selected silicone gels a thermo-gravimet-

ric analysis (TGA) and differential scanning

calorimetry (DSC) were carried out. TGA

showed that both the reference gel R and

gel S had similar weight losses at 150°C.

Gel E has the earliest onset for the weight

loss.

DSC analysis as in Figure: 4 showed that

Gel S had no phase transition, while gel E

had a phase transition peak at -44+- 1°C

and the reference gel at -40.6 +-0.6°C.

These phase transitions can be attributed to

the glass transition temperature of the gel. A

harder gel would transmit more thermo-

mechanical stress towards the bond wires

and with that to the bond-chip interface.


Figure 3: 1.7kV SPT+ IGBT and diode chip stability test data: leak-age current is measured as stable over t=300sec at fixed heat sinktemperature and Rth= 1.25W/K (at Vdc=1400V up to Tj=172°C; atVdc= 1700V up to Tj=162°C).

Figure 4: DSC measurements

Whilst you harness the power of the sun and wind, HITACHI liquid cooled IGBTs harness the power of water for

next generation high density Solar and Turbine renewable energy systems.

Hitachi liquid cooled IGBTs. 35% lower thermal resistance. AlSiC life-time reliability. Optional cooling jackets. System

solutions capable of delivering upwards of 500kW even at 5kHz operation within 30% of the footprint area of some

conventional solutions. Could you be missing out?

Isn’t it time you switched to HITACHI?

Hitachi Europe Ltd. Power Device Division Tel: +44 1628 585000 E-mail: [email protected]

Power Device Division

PCIM Booth 12/355


This has negative influence on the lifetime of

the system. In this respect Gel S has no

abrupt change in its mechanical properties.

Finally the physical characterization focused

on the hardness of the isolation material and

on the tackiness or adhesion to other materi-

al components of the system. The aim is to

have a soft insulating material with a good

sealing. The lowest penetration depth is

measured for Gel S, which is a negative indi-

cation for increased hardness. Comparing

adhesion forces between the different gels

that of gel E with the other components are

clearly the highest followed by gel S and gel R.

To conclude as summarized in Table: 1, gel

E was selected to be the best candidate for

high temperature operation modules. In

addition, the outstanding softness and the

tackiness of the gel makes gel E into the

most promising alternative to the current gel.

Module Soldering with Spacer

Higher operation temperatures critically

increase the requirements of different pack-

aging technologies in order to maintain high

reliability and long lifetime of the IGBT mod-

ule. Some of the identified lifetime limiting

failure mechanisms are terminal solder

joints, large area solder joints and wire bond

contacts. Therefore an additional step has

been introduced to the soldering process of

the substrates to the base plate. In this step

flat aluminum bonds are soldered on the

base plate on positions where substrate

edges are attached (Figure 5). These bonds

give a reproducible and mechanically suffi-

cient stable spacer to guarantee a minimal

thickness of the solder. In this way the tilting

of the substrate could be decreased.

To prove the benefit on reliability, modules

with and without spacers have undergone

temperature swings. In all modules after

cycling cracks in the substrate solder near

some substrate corners can be observed.

Correlating the crack growth rate with the

solder thickness at the corresponding loca-

tion clearly shows that the locations with the

thinnest solder have the highest growth rate.

(Figure: 6) Hence the use of spacers

improves the power cycling capability.

High Current Terminals

With higher current ratings of semiconductor

chips, the contribution of resistive losses to

the power module’s losses is getting higher.

High currents cause unwanted power dissi-

pation through the power terminals to the

connected bus bar. Furthermore they can

lead to reliability problems due to the over-

heating of the internal conductor leads.

Therefore the current path had to be investi-

gated.

Beside dominant conduction and switching

losses, resistive losses occur at several

points. On a 2400A 1700V module these

losses contribute by 14% to the overall loss-

es. With 39% the terminal contributes the

most to the resistive losses. The bond wires,

the chip metallization and the wire bonds are

minor contributors.

To lower the losses generated in the termi-

nals, the electrical resistance must be low-

ered. Since there is no affordable material

with higher conductivity than copper, the only

option left is changing the geometry of the

conductor.

The terminals as used in today’s HiPak mod-

ules are shown left and the new design on

the right in Figure 7.

As a result a considerable reduction of elec-

trical resistance by 18% is achieved by bal-

ancing the current density in the conductor

plate and making the current paths shorter.

At the same time the mechanical reliability is

maintained. With the new terminal design

continuous phase currents of up to

1800Arms are reasonably possible.


Figure 5: Top: Base plate with the name ofthe positions for the spacers. Bottom:Images of the spacers.

Figure 6: Correlation of crack growth rateand solder thickness

Figure 7:Terminals (left : standard, right :improved)

Figure 8: 1.7kV SPT+ IGBT module meas-ured under nominal conditions at Tj=150°C:a) IGBT turn-off, Eoff=1.75J b) IGBT turn-on,Eon=1.6J and c) diode reverse recovery,Erec=1.2J.

Table 1: Qualitative rating of the investigatedgels regarding bulk, thermal and mechanicalproperties

Gel R Gel S Gel E

Bulk properties 0 0 +

Low temperature

range- + 0

High temperature

range+ + +

Hardness + -- ++

Tackiness 0 0 ++

Electrical Results

To verify the performance of the new 1.7kV SPT+ HiPak module,

extensive measurements have been carried out.

Figure 8 shows the IGBT turn-off, turn-on and diode reverse recov-

ery waveforms as measured on module under nominal conditions

(900V/3600A) at Tj=150°C. The IGBT and the diode both exhibit

controlled switching characteristics as well as short current tails.

This behavior is enabled from the combination of SPT buffer design

and silicon resistivity used in SPT+ technology, which provides fast

switching with low losses and low overshoot.

The IGBT turn-off tested at a DC-link voltage of 1200V for a collec-

tor current value of 7200A at Tj=150°C is shown in Figure 9 proving

the ruggedness of the SPT+ IGBT design when paralleled in the

HiPak module.

The short circuit SOA test at Tj=150°C and for a DC-link voltage of

1350V can be seen in Figure10 No thermal runaway after short cir-

cuit test has been observed and excellent short circuit capability

has been measured at chip level for Tj=150°C and pulse times up

to 30us. Moreover the SPT buffer and anode designs employed in

the SPT+ IGBT have been optimised in order to obtain a high chip

short-circuit SOA capability even at gate voltages exceeding the

standard gate drive voltage of 15V.

PCIM Booth 12/408

www.abb.com/semiconductor

www.bodospower.com

Figure 9: 1.7kV SPT+ IGBT module turn-off measured under SOAconditions at Tj=150°C.

Figure 10: 1.7kV SPT+ IGBT module short-circuit characteristicsmeasured under SOA conditions at Tj=150°C.

MAKING MODERN LIVING POSSIBLE

DANFOSS SILICON POWER

SILICONPOWER.DANFOSS.COM

The future of cool designMove into the fast lane with customized

power modules

It cannot be stressed enough:Efficient cooling is the mostimportant feature in power modules.Danfoss Silicon Power’s cutting-edgeShowerPower® solution is designed to secure an even cooling across base plates. In addition, our modules can be customized to meet your automotive requirements in detail, offering: Low weight, compactdesign, extended life and very low life cycle costs In short, when you choose Danfoss Silicon Power as yoursupplier you choose a thoroughly tested solution with unsurpassed power density. Day in and day out. Please go to siliconpower.danfoss.com for more information.

ShowerPower®®

Meet us at

PCIM!Hall 12;

Booth 325

36 Bodo´s Power Systems® May 2010 www.bodospower.comBodo´s Power Systems® April 2010 www.bodospower.comBodo´s Power Systems® May 2010 www.bodospower.com

We are working on closing this “gap” on two levels. Firstly, by offering

a professional thermal paste application service – Pre-applied Ther-

mal Paste for Power Modules – a service that has already proven

rather successful, with over 700,000 power modules having been

printed with a thermal paste layer. In addition, Semikron is developing

its know-how and expertise in the area of thermal conductive media

application and function.

Designated use of thermal conductive media

Thermal conductive media normally consist of a plastic carrier materi-

al (e.g. silicon oil) and thermal conductive filler substances such as

zinc oxide, graphite or silver. They are available in the form of pastes,

adhesives, phase-change materials and foils. Thermal interface

materials conduct heat better than air and typically have a specific

thermal conductivity (Lambda) of 0.5 - 6 W/m·K. In other words, the

thermal conductivity of thermal interface materials is approximately

20 - 200 times better than that of air. To enable the thermal conduc-

tivity properties of thermal interface materials to be categorised,

Table 1 shows the specific thermal conductivity of materials common-

ly used in power modules. The thermal paste P12 from the company

Wacker has been taken by way of example. The thermal resistance

values R(th) shown are based on the module-specific thermal

spreading.

If the thermal conductivity of thermal paste is compared with the ther-

mal conductivity of other components in a power module (see Table

1), the thermal paste does not rate particularly well. The extent to

which thermal paste contributes to the overall thermal resistance

R(thjs) of the module amounts to around 20-65%, depending on the

module and the combination with the heat sink. The thermal paste

layer therefore has to be as thin as possible but as thick as neces-

sary (see Figure 1).

Too thin a thermal paste layer results in air pockets between the under-

side of the module and the top of the heat sink, causing a high thermal

resistance Rth(cs). Once the optimum has been reached, the thermal

resistance between the case and the heat sink increases quickly again

in line with the increase in thermal paste layer thickness. This happens

because the specific thermal conductivity of thermal conductive media

is very low compared with other materials in a power semiconductor

module. The minimum value is different for every heat-sink-mounted

module and has to be defined in appropriate tests.

Power ModulesHow to avoid errors when applying thermal paste

The power electronics sector is continually striving to boost the reliability of power mod-ules. The main focus of research work in this sector is on semiconductor chips, packagingtechnology and the DBC substrate. The weak point of heat-sink-mounted power modules,however, is the “gap” between the module and the heat sink which results from uneven-

ness on the contact surfaces and which has to be filled with a thermal conductive mediumin order to get rid of the air pockets.

By Dieter Esau, Process Engineer and Dr. Michaela Strube, Manager Service Engineering, Semikron

P O W E R M O D U L E S

Table 1: Specific thermal conductivity of materials commonly used ina power semiconductor module

Material Spec. thermal

conductivity Lambda

Thickness

[μm]

Portion R(th) of

SKiM modules

Chip 106 120 2,92%

Chip solder 57 70 3,65%

DCB (Copper) 394 300 1,94%

DCB (Al2O3) 24 380 32,91%

DCB (Copper) 394 300 1,31%

Thermal paste (P12

from WACKER)

0,81 30 57,26%

Figure 1: Dependence of thermal resistance on thermal interfacematerial layer thickness

The importance of thermal paste composition

R(th) tests have shown that the thermal conductivity of a thermal

paste in actual application does not only depend on its specific ther-

mal conductivity, but also on it its composition. The larger the filler

particles in a thermal paste are, the higher the specific thermal con-

ductivity. The particle size of the filler determines the minimum layer

thickness. In other words, the thermal paste layer applied cannot be

thinner than the largest particles in the paste. After several tempera-

ture cycles, a paste with small particles (e.g. P12: particle size

0.04μm - 4μm) allows almost for metal-to-metal contact at points

where the pressure is particularly high, resulting in a substantial

reduction in R(thcs).

Thermal paste application

Thermal paste can be applied either to the module or to the heat

sink. This is done using a roller or in printing processes. In roller

application, a rubber roller is normally used, while the most common

printing method used is silk screen printing or stencil printing.

Applying thermal paste with a rubber roller can lead to sufficient

results, provided this critical step is performed by experienced profes-

sional staff with relevant training. This process does, however, have

shortcomings such as inhomogeneity, poor reproducibility and the

risk of contamination.

In stencil and screen printing, far better results can be achieved than

with the roller process, provided automatic printing methods are

employed. Manual printing, for its part, can lead to considerable

process deviations. The development of a process with an automatic

stencil printer featuring continuous process monitoring, as is the case


ABB FranceCurrent & Voltage Sensors Departement

e-mail: [email protected]

Shared experience

creates a shared

success?

Certainly.

The ES range has become our bestseller; this is due to an optimisation of its design using our

shared experience with our customers.

These upgrades allow us to offer the most cost effective sensor in high current measurement.

As drives become more and more compact, we also have enhanced the ESM range in terms of

magnetic immunity and dynamic response.

Thanks to these improvements, we are able to offer our clearest signal increasing the

performance of your equipment. www.abb.com

38 Bodo´s Power Systems® May 2010 www.bodospower.comBodo´s Power Systems® May 2010 www.bodospower.comBodo´s Power Systems® May 2010 www.bodospower.com

at SEMIKRON, requires substantial investments, however, which in

economic terms only makes sense for large production quantities.

In addition to complying with recommended layer thickness, care

should be taken when applying the thermal paste to ensure that the

thermal paste layer is spread on the underside of the module or the

heat sink surface evenly and homogenously. An inhomogeneous

thermal paste layer (extreme case: application of one or more ther-

mal paste blots) can result in DBC ceramic substrate breakage (Fig-

ure 2). This applies to modules with and without a base plate alike. In

addition to this, thermal paste inhomogeneity can also lead to local

overheating resulting from the air pockets between the underside of

the module and the upper side of the heat sink surface.

Measuring the thickness of thermal paste layer

The thickness of a thermal paste layer can be measured directly or

indirectly. An indirect way of measuring the thickness is, for example,

to weigh the thermal paste by performing a Tara weight measurement

using suitable scales. An example of a direct contact-free measure-

ment of the thermal paste layer is a measurement using an optical

profilometer such as the μSCAN from Nano Focus. Other measure-

ment equipment that could be used to measure the thermal paste

layer directly includes, for example, thickness gauges such as wet

film combs (e.g. from Zehntner (ZND 2051) or Elcometer Instruments

or BYK Gardner (PG-3504)) or wet film wheels (e.g. from Zehntner

(ZWW 2100-2102) or BYK Gardner). The downside of these, howev-

er, is that they may cause damage to the layer in places.

Determining the optimum thickness for thermal paste layer

The optimum minimum thickness for a specific thermal paste in com-

bination with a specific heat sink surface can be determined in a

defined process which starts at a minimum thickness of around 10μm

and is increased in 10μm steps (another option would be to alternate

the steps). Here, the thermal paste is applied to the module or the

heat sink or to an aluminium plate in accordance with the specifica-

tions of the module manufacturer. When tightening the mounting

screws, the tightening torques specified by the module manufacturer

must be observed. To achieve a relaxed system state the mounted

and secured module should undergo three thermal cycles

(20°C/100°C/1h).

After thermal cycling, a module with no base plate can not be easily

removed without causing destruction, since the module is pressed

onto the heat sink/aluminium plate and the sticky thermal paste is

distributed in the space between, producing an enormous adhesive

force. To ensure non-destructive removal, the module should there-

fore be left untouched at room temperature for 12 hours after the

screw has been loosened or should undergo 1-2 thermal cycles.

Once the module has been unscrewed, the imprint pattern on the

underside of the module gives an indication of whether the thermal

paste layer provides optimum contact between module and heat sink.

Figure 3 (left) shows the underside of a power module containing

large areas with untouched thermal paste. This indicates that the

thermal paste layer is in fact too thin (approx. 30μm). By way of com-

parison, Figure 3 (right) shows the underside of a module that is cov-

ered entirely with thermal paste, with the exception of certain high-

pressure points where metal-to-metal contact is achieved. This is

indicative of optimum thermal paste application (approx. 50μm).

By optimising the thermal paste layer thickness for the individual

heat-sink-mounted module and using automated application process-

es to guarantee quality standards, the shortcomings of thermal con-

ductive media can be compensated for to a certain extent. The prob-

lem with the “gap” that emerges between the power module and the

heat sink, however, still bears the biggest potential for improvement.

Thermal paste application service

The thermal paste application service provided by Semikron simpli-

fies the module assembly onto the heat sink. Customers no longer

have to include this production step and can therefore reduce costs.

The production staff’s gloves are not at risk of contamination from the

thermal paste and the thermal paste cannot accidentally find its way

into production. The optimised module-specific thermal paste layer

thickness reduces the overall thermal resistance and the risk of DBC

breakage. The thermal paste is applied in an automated printing

process, and the module-specific thermal paste layer boasts an accu-

racy of +/-10 μm. The application process is monitored using SixSig-

ma quality control methods. The modules with thermal paste layer

are transported to the customer in purpose-developed, patent-pro-

tected packaging that ensures contact-free transportation of finished

modules containing a thermal paste layer. Modules containing a ther-

mal paste layer can be stored in this packaging for up to 18 months.

The thermal paste application service is available for SKiM 63 and

93, SEMIPACK 2, SEMITRANS 2, and MiniSKiiP modules.

PCIM Booth 12/411

www.semikron.com

Figure 3: Module showing poor (left) and optimum (right) thermalpaste application


Figure 2: Module underside showing problematic thermal paste layerapplication




For more information and your

free drill hole stencil please visit

www.biricha.com


Initially the range seemed to be sufficiently comprehensive, with a

metering capacity of up to 100A. However, it was soon found to be

too limited for the industrial or for the heavy-duty service domains

and this was without taking into account the fact that monitoring often

begins by measuring global consumption at the point of energy input

– requiring a capability to measure up to 2000A.

LEM therefore developed the RT current sensor adapted to these

EMN devices, which provides the same flexibility of installation as

split-core current transformers from the lower range, but with the

same class 1 precision required for the sub-metering field. The

Rogowski coil, which has long been noted for its ease of installation,

offered the right solution provided that its major drawback could be

overcome - that of inaccuracy caused by the sensitivity to the posi-

tion of the conductor inside the loop.

From theory to practice

A simple explanation of the Rogowski coil theory (“Die Messung der

magnetischen Spannung”, Archiv für Elektrotechnik, 1912), is that it is

a coil-winding that closes back on itself, wrapping the conductor to

be measured like any toroidal-type current intensity transformer, the

only – but major – difference being that there is no magnetic core.

While Ampère’s theorem still applies, the equations are slightly differ-

ent because at the sensor output we find that the voltage is in pro-

portion, not to the primary current, but rather to its derivative: U =

M*di/dt. M is the mutual inductance between the primary conductor

and the coil, which to some extent represents the coupling between

the primary and secondary circuits. All the difficulty in obtaining good

accuracy from this principle derive from the fact that the simplified

analytical expression of this equation implicitly supposes perfect sym-

metry of the coil (M must be constant). However this is never the

case in practice, and we shall illustrate this by looking at the three

critical points that cause M to be variable.

The density of the turns: the coil-winding must be perfectly regular to

ensure that the winding density is uniform throughout. Turns that are

not equidistant create asymmetry, the effect of which is to cause the

coefficient M to vary according to the position of the primary conduc-

tor. This induces a de facto error resulting from the position of the

cable or the busbar to be measured, an error which increases the

closer the conductor is located to the area where the density differs

from the average spread value.

The coil cross-section: in the same way as the turns density, if the

cross-section is not uniform all along the coil surrounding the conduc-

tor, term M will not be constant and an error is produced due to the

positioning of the conductor. In this case too, the closer the conductor

is located to a zone that significantly differs from the average spread

value, the greater is the error.

The coil clasp: a major advantage of the flexible Rogowski coil is that

it provides an extremity with no electrical connection, the return sig-

nal being wired back inside the coil. Here is a major source of asym-

metry caused by the discontinuity in the coil-winding which affects the

turns density, when the theory requires that the coil should be per-

fectly continuous and homogeneous. This is by far the most critical

point and generates the greatest errors.

The reality of the figures:

Until now Rogowski coils have delivered at best a 2% positioning

error. Added to this there are often restrictions, which exclude the

conductor from certain zones inside the loop, in particular at the clo-

sure of the clasp head. In reality this could even be quite catastroph-

M E A S U R E M E N T

A New Class of Rogowski CoilSplit-Core Current Transducers

To guarantee stability in time and temperature, the RT series coil is moulded integrally into a PU resin

The monitoring of electric power consumption has become a key element for managingelectrical installations in industrial and commercial sectors, such as manufacturing

facilities, data centres, food processing industries, retail outlets, hospitals or educationalestablishments. Three years ago LEM introduced a system called Wi-LEM onto the market which is based on wireless sub-metering components, the EMN, enabling

measurement of electricity segmented by activity (lighting, HVAC, motors, heating, etc.).

By Pierre TURPIN, Project Manager, LEM Energy & Automation

ic, resulting in errors close to the clasp head in the order of 6%. For

this reason it is easy to understand why manufacturers of energy

metering equipment have always avoided employing this technology.

However, LEM recognised the viability of this technology for energy

measurement but it depended on whether they could manufacture

coils which delivered a positioning error of less than 0.75% as a mini-

mum. In fact, the objective of developing class 1 energy meters is to

obtain an overall accuracy of better than 1% over the entire measure-

ment chain, including the current sensor, the voltage sensor and the

signal processing.

The challenge met by LEM

Multiple solutions based on electrical or mechanical concepts have

been developed for nearly 100 years in order to resolve, albeit with

very limited success, the main problem of the Rogowski coil current

sensor, i.e. the error caused by the imperfect sensor closure. Taking

this into consideration LEM engineers decided to revisit the theory in

greater depth in order to better understand the reason for these

unsuccessful attempts. Thanks to their expertise in magnetics they

have been able to develop a very simple but effective solution... a

sleeve made of magnetic material, making it possible to make an

entire zone around the coil invisible (magnetically), and thus to mask

the imperfections on the closing mechanism as well as the connec-

tions of the sensor’s secondary wires. The sleeve acts as a magnetic

short-circuit (or more precisely a reluctance short-circuit), “virtually”

bringing together the two sections of the coil located on each side.

Their approach was a complete success - the error associated with

the coil clasp has become almost negligible. Naturally enough, the

idea was the subject of a patent application in 2007.

The hidden challenge

While the major problem with the split-core Rogowski coil had finally

been solved, other problems became apparent which diminished the

success of the magnetic sleeve. The error associated with the

Figure 1: Measurement error according to the position of the conduc-tor within the loop:traditional Rogowski coil compared to the LEM RT

Figure 2: Sensor head clasp implementing a new “magnetic sleeve”


42 Bodo´s Power Systems® April 2010 www.bodospower.com

design of the coil clasp system had previously been so important that

it had, to some extent, masked the other causes of asymmetry. LEM

continued to work to improve this current sensor and after a total of 2

years has been able to develop the processes and machinery that

significantly reduce the symmetry faults, both with regard to the regu-

larity of the coil-winding and creating a uniform section over the

entire length of the loop.

The results

The graph below illustrates the improvements that LEM has been

able to produce in the split-core Rogowski coil, compared with the

level of accuracy of the other products on the market that are based

on this technology.

Today the error due to the positioning of the conductor is specified at

a maximum of 0.65% of the measured value for a 15mm diameter

conductor irrespective of where it is positioned, even if it is adjacent

to the coil clasp.

For a better appreciation of the results that have been obtained, here

is another graph showing the maximum value of the error over a

sample of 210 RT Rogowski coils. Typical value of the error due to

the positioning is 0.31% of the measurement for the new LEM sen-

sor.

What we should also know about Rogowski coil sensors.

External conductors

Performance of the Rogowski coil is generally expressed in terms of

error associated with the positioning of the conductor to be meas-

ured, but a good sensor must also remain insensitive to all other

external conductors located nearby. It happens that a relation exists

between these two characteristics a perfect loop is perfect for both,

and a bad loop will be bad for both. This is a result of Ampère’s theo-

rem, which states that any error associated with any form of asym-

metry is valid both inside and outside the loop. Let us take, for exam-

ple, a conductor on which a 100A current is circulating, situated

inside the Rogowski coil and which is in contact with a section of the

loop inducing an error of +0.5%. A measurement of 100.5A is there-

fore obtained. This same conductor at the contact of the same sec-

tion, but outside the loop, will likewise cause an error of 0.5A, but this

is added to the current measured inside the loop which results from

the principle of rejection of the external magnetic fields.

Absolute accuracy

In general, the absolute accuracy of Rogowski coil sensors is low

because their gain (expressed by the term M) depends on physical

parameters that are hard to control when it comes to mass produc-

tion. To summarise, it is not realistic to try to manufacture this type of

sensor with a gain dispersion of less than several percentage points

(say between 2 and 5% depending on the technology used). This

would mean designing coil-winding machines in which the pitch

would be regulated to an accuracy in the order of microns, as well as

being able to produce the coil-winding base with the same level of

accuracy. It is therefore customary to connect the Rogowski coil to an

active or passive electric circuit so that it can be calibrated, and a

good absolute accuracy can be obtained.

On the other hand it is essential to guarantee excellent stability of the

sensor characteristics, in particular with regard to temperature, to

prevent any drift from having to be corrected by recalibration to com-

pensate for changed conditions during use. For example, LEM’s RT

range has proved to be excellent in this respect at 30 ppm/°C.

No Limit!

It is a recurring issue when specifying measuring systems: will the

sensor saturate if the current goes above its nominal value? Of

course in the case of the Rogowski coil the answer to this is “no”

since it does not have a magnetic core, and as a result runs no risk

of saturation. In theory the limit of the current to be measured is infi-

nite! In practice it is the diameter of the closed loop that establishes

the nominal value of the current, not in relation to the measuring

capacity, but rather in relation to the dimensions of the primary con-

ductor. In the specific case of high di/dt (impulsion) the limit will be

fixed by the voltage developed at the coil terminals instead.

Linearity

Of course linearity is important when a sensor is intended for preci-

sion measurements. For the same reason that no saturation takes

place on the Rogowski coil, it is not possible for there to be a lack of

linearity, since the coil is intrinsically perfect in this respect. If differ-

ences were nevertheless observed, it would be necessary to ques-

tion the measurement methods and not the Rogowski coil!

Phase-shift

Phase-shift is an extremely important parameter with respect to ener-

gy that is calculated from measurements of currents and of voltages.

In the same way as for saturation and linearity, the Rogowski coil is

perfect with regard to the phase – which means it induces no phase-

shift. However it is worth bearing in mind that it is necessarily asso-

ciated with an amplification stage (as described below under the

heading “integrator”) which itself will generate a phase-shift. In con-

clusion, the phase error is intrinsically zero when the coil is not con-

nected, but it can reach high values as soon as a load is connected.

However this error can be easily quantified by calculation or by simu-

lation of the equivalent RLC circuit, and compensated for by an ad

hoc method.


Figure 3: Measurement error due to the conductor position within the loop: thenew LEM RT sensor compared to the traditional Rogowski coils

Figure 4: Distribution of the max. positioning error for a sample of210 RT transducers

The choices made by LEM

Today, the Rogowski coil sensors can compete against the best cur-

rent intensity transformers in the energy measurement sector. It

became very clear that LEM would need to exploit the properties of

this technology to the maximum which could create a net benefit

when measuring high currents, i.e. the weight, overall dimensions,

flexibility and manageability. With a cross-section measuring 5mm,

which could almost be classed as a “universal” size, the sensors in

the RT range are among the most slender Rogowski coil sensors on

the market.

The (patented) coil clasp device is also small (28 x 30 x 16 mm), and

provides a reliable connection of the loop to its coaxial signal cable.

Here, the choice of a coaxial-type cable is directly associated with

the low cross-section of the coil. In fact, since the gain is proportional

to the cross-section, a fine coil develops little voltage, and it is appro-

priate to control the signal-over-noise ratio by starting to eliminate all

risks of interference between the loop and the amplification stage.

Finally, to guarantee stability in time and temperature, the RT coil is

moulded integrally into a PU resin, using an original process devel-

oped by LEM engineers. This wrapping technique also helps to main-

tain the different sections firmly and imparts a robustness to the

assembly, as required for places where it is difficult to install.

So, current transformer (CT) or Rogowski coil (RT)? LEM has

already made its choice, but is prepared to share it with you!

Application note: design of an integrator for the Rogowski coil

The Rogowski coil supplies a voltage in proportion to the derivative of

the primary current at its terminals. An electrical integrator circuit is

therefore necessary to convert this signal into a signal that is propor-

tional to the value of the primary current.

The integrator is an essential component in current measurement

with the Rogowski coil, and the way this amplification stage is imple-

mented will have a major impact on the sensor’s electrical perform-

ance (linearity, phase-shift and frequency bandwidth). A list of the

various critical aspects of such an integrator, with some possible

solutions, is given below:

Very low signal level (for example 20 mV / kA for sensors in LEM’s

RT range)

→ The use of very low noise OpAmps is recommended to opti-

mise the signal/noise ratio

→ It is necessary to try to minimise the surface of the PCB or pos-

sibly to shield the amplification stage to reduce sensitivity to

external fields.

Low cut-off frequency

When an integrator is connected to a Rogowski coil the two form a

high-pass filter. Since it will reject very low frequencies it is necessary

to define the cut-off frequency in order to optimise performance at the

nominal operating frequencies, while still obtaining the best possible

response time.


Figure 5; Installation of an EMN energy meter with 3 RT Rogowskicoils in an electrical cabinet

Figure 6: Typical Frequency response of coil and Integrator

E M C C O M P O N E N T S I N D U C T O R ST R A N S F O R M E R S R F C O M P O N E N T SC I R C U I T P R O T E C T I O NC O N N E C T O R SP O W E R E L E M E N T S S W I T C H E SA S S E M B LY T E C H N I Q U E www.we-online.com

High Current Inductors with flat wire coilSMD Inductors WE-HC & WE-HCA

Current capability up to 65 A

No thermal aging

Available ex stockSamples free of charge

Design-In support included

New core material for lower core loss

Flat wire coil for lower losses at high frequencies

Reference design of all major IC manufacturers

PCIM Booth 12/649

Rejection of offset

The main problem of a pure integrator lies in the fact that it will inte-

grate the slightest parasitic offset (e.g. due to the AmpOp), with the

effect that the output will always be unstable and will drift sooner or

later to saturate at the upper or lower level. Consequently it is essen-

tial to limit this drift, using a static gain or an active compensation

stage:

Total offset rejection

It is possible to completely eliminate the residual offset by adding a

capacitive coupling device between the integrator and the measuring

stage:

Phase-shift

The offset rejection circuits described above will generate several

degrees of phase error which poses a major problem for the meas-

urement of power. In this type of application, it is therefore necessary

to add a phase-shift compensation stage, which generally consists of

a low-pass filter. Unfortunately, the correction will not be constant, but

will depend on the frequency, meaning it will be necessary to opti-

mise the design to minimise the phase difference at the fundamental

frequency, typically 16 2/3, 50, 60 or 400 Hz.

Calibration: active adjustment of gain

A Rogowski coil requires calibration against a reference signal in

order to fine-tune its gain which is never exact by construction, due to

inevitable imperfections in the manufacturing process. In general

engineers use the integrator stage to which an analogue device,

such as a potentiometer, is attached. The most recent digital calibra-

tion solutions are more likely to use a microcontroller or the combina-

tion of a microcontroller and a PGA (programmable gain amplifier). In

all cases calibration is specific to each individual Rogowski coil which

must always use the same circuit with which it has been calibrated.

Calibration: passive adjustment of gain

Historically, the Rogowski coil was used simply for measurement of

the current effective value (rms) without phase constraint. Many

loops offered factory calibration based on a purely resistive or a

resistive/capacitive circuit (RC circuit). While this method continues to

be straightforward and economical, unfortunately it does not lend

itself to power measurements due to the strong phase error that it

generates, and its possible dependence on the frequency (if an RC

circuit is used).

When developing the new Rogowski coils, LEM decided to offer a

simple and generic product, keeping in mind that the integrator tech-

nology leads to the best performances and is a well known method.

Therefore the transducers of the RT family are not calibrated in the

factory, do not use any additional electronic components or housings

and do not require power supply. Using an integrator specific to the

device connected to the Rogowski such as energy, power quality or

pulse power monitor, is a cost effective and high performance solu-

tion.

PCIM Booth 12/402

www.lem.com

D I G I T A L P O W E R


Figure 12: Example of Phase shift corrector within the Frequencyband 50/60 Hz

Figure 7: Output of the OpAmp is saturated, Signal is distorted

Figure 9: Offset compensation stage. Low Output offset optimizedoutput Signal dynamic

Figure 8: High Output Offset, manageable with low output Signaldynamic

Figure 10: AC coupling

Figure 11: Low-pass filter


This article gives an overview of a new fami-

ly of InPower intelligent IGBT gate drivers

and discusses the merits of digital technolo-

gy for driving high power IGBT modules.

An IGBT gate driver is a key element of any

power electronic system. Therefore, the

choice of driver is relevant for its reliability. It

is a well-known fact that market require-

ments for the gate drivers need to be taken

into consideration in a way that offers the

user advantages versus existing drivers in

order to succeed with a new driver concept.

Most applications using IGBT modules today

are still controlled by analogue drivers. How-

ever, more and more industrial customers

change to a digitally controlled architecture.

Thus especially applies to those who use

modules with 3.3kV to 6.5kV blocking volt-

age, as they mainly benefit from advantages

offered by digital technology. Hence, it will

be only a matter of time until intelligent digi-

tal IGBT gate drivers for high power applica-

tions will become common place.

Why go digital?

To handle and manipulate analogue signals,

such as timing and amplitude related param-

eters, requires very different methods in

electronic circuitry as compared to their digi-

tal counterparts. The latter may easily be

altered by changing some lines in a software

program. That makes it very attractive. How-

ever, the digital technology is not an end in

itself but a means to an end.

The switching-on process is optimised

through the gate current characteristics.

Hence this latest technology focuses on the

digital control of the value of the gate resis-

tors. The binary coded variance of theses

resistors forms the key in the precise control

of the gate current. The optimisation of the

gate current with adjustable gate resistors as

well as continuous acquisition of di/dt

enables better IGBT performance, gate

boosting, reduction of switch-on losses and

softer switching with decreased transient

emissions. Additionally, the digital input filter

for switching signals guarantees that

unwanted signals will not impact the whole

power electronic systems. All parameters

can easily be changed by software and the

user does not need any special knowledge

for this optimisation.

The advanced protection functions such as

four level advanced desaturation monitoring,

two level di/dt monitoring, feedback clamping

with active function and multi-step soft shut

down are also optimised owing to the digital

control.

The digital technology has a strong point in

particular with regard to reliability and high

flexibility. Rapid short circuit recognition or

limitation and, therefore, reliable protection

against over-current in all short circuit condi-

tions and over-voltage during short circuit

turn-off as well as simple tuning according to

the customer application are very promising

features for the power electronics system

solutions.

Furthermore, this technology supports spe-

cial topologies such as parallel connections,

two- or multi-level. Implementation of such a

multi-functional driver is reasonable espe-

cially in applications with very high power

and requirements for extreme reliability such

as traction application, large industrial drives

or renewable energy.

New digital drivers for 3.3kV IGBT modules

A new series of digital intelligent drivers is a

sophisticated “Plug & Play” solution which

can be implemented in two- and multilevel

topologies.

The 1IPSE1A33-60 is a single-channel driv-

er and is based on new digital driver technol-

ogy with smart switching using variable gate

resistors to control high power IGBT mod-

ules. The digital technology provides control

parameter settings which are to be adapted

to meet the individual needs.

The 1IPSE1A33-60 is designed for applica-

tions with high reliability and safety require-

ments and allows optimised switching

behaviour and safe protection of an expen-

sive IGBT module during the entire switching

period. It features a very sophisticated pro-

tection concept against short circuit and pro-

tects IGBT modules from a short circuit

directly at IGBT as well as from a low induc-

tive short circuit at the load side. Due to the

double monitoring during the entire switching

period - a four level desaturation monitoring

and two level di/dt – monitoring - the driver

provides reliable protection against over-cur-

rent in all short circuit conditions including

hard and soft short circuits. This driver is

completely equipped with further protection

functions: reliable protection against over-

voltage during turn-off due to a digital feed-

back clamping, supply voltage monitoring

and digital input filter against disturbance of

the switching signals.

The 1IPSE1A33-60 has an integrated

DC/DC converter which provides an isolated

power supply for the driver and is designed

such that it offers lowest coupling capaci-

tances and high isolation stability. The driver

is provided with fibre optics for transmission

of controls signals and status feedback sig-

nals.

According to the tests the drivers of this

series have achieved a very high perform-

ance in terms of smart switching, flexibility

and reliability. In particular, the digitally con-

trolled feedback clamping with transils

I G B T D R I V E R S


IPS Drivers Combine Highest Performance and Design Flexibility

First-class performance can simply be achieved by software

Using digitally controlled IGBT gate drivers opens the horizon to high reliability and lessswitch-on losses by changing the operation characteristics with software programming

By Robert Hemmer, Pavel Kviz and Marita Wendt, InPower Systems GmbH

Figure 1: The 1IPSE1A33-60 is designed forhigh reliability and safety

ensure exceptional robustness against voltage spikes which should

otherwise destroy the IGBT module. Driven by these drivers IGBT

modules benefit from the excellent switching performance and multi-

step soft shut down in short circuit and other fault conditions.

To avoid heat transfer from the power electronics circuitry, the driver

control unit is not mounted directly onto the module; instead only the

less sensitive adaptation board. This matched to every type of IGBT

module and is connected by cables to the driver board. This driver

family represents complete solutions matched to individual IGBT

types. Therefore, the customer has no development effort for dimen-

sioning, matching and integration of these drivers. This saves the

user considerable time and improves the reliability of the whole

power electronics system.

As a matter of fact the new drivers offer maximum utilisation of the

IGBT module thanks to a smart design. Easy paralleling was consid-

ered as well in this design. Good current sharing is realized if the

drivers are used in parallel-connected IGBT applications.

The parameters of this driver can be adjusted an endless number of

combinations and, therefore, increase the perfect adaptation of the

driver to any power electronic system solution.

Tests show that 3.3.kV IGBT modules driven by this driver withstand

short circuit currents up to 6000A (at 1500A nominal current) at

120nH stray inductance when fed from a 2000V DC bus. The dia-

grams show some waveforms of hard (low inductance to load, high

di/dt) and soft (high inductance to load, low di/dt) short circuit meas-

ured at the high side IGBT.

The 1IPSE1A33-60 provides an output voltage VON/VOFF of ±15V

and a switching frequency of fS max=120kHz. Further characteristics

of this drivers are: peak output current ±70A,peak output power

PDC/DC=3W, coupling capacitance primary / secondary side

1 to 2pF, turn-on

and turn-off delay

times 400nsec,

creepage distance

more than 30mm,

input supply voltage

range +14 to +30V,

isolation testing

voltage 6000V,

operating and stor-

age temperature -40

to +85°C.

Conclusion

At PCIM 2010 InPower Systems GmbH will also represent further

digital drivers such as a dual-channel 2IPSE1A33-100 driver for

3.3kV IGBT modules. These drivers ensure a stable operation,

decrease switch-on losses and increase reliability.

They are particularly suitable for all applications where the reliability

is an indispensable issue. Flexibility through a programmable control

unit makes the implementation of the 2IPSE1A33-100 perfect.

To make inverter designs more efficient and reliable InPower also

modified its standard single-channel drivers for 4.5kV and 6.5kV

blocking voltages IGBT modules as well as dual-channel driver for

1700V.

Visit our boot at PCIM 2010 in Nuremberg where we will demonstrate

the performance and the advantages of the new digital technology,

both in terms of reliability and final user flexibility.

PCIM Booth 12/602

www.inpower-sys.com

I G B T D R I V E R S

Figure 2: Hard short circuit waveforms with an IGBT module 3.3kV1500A

Figure 3: Soft short circuit waveforms with an IGBT module 3.3kV1500A

Figure 4: dual-channel digital drivers2IPSE1A33-100 for 3.3kV IGBT modules

WHEN WE IMPROVE OUR PHOTOCOUPLERSWE THINK BIG – AND SMALL.As a leading manufacturer of photocouplers, Toshiba’s product range continues to pioneerinnovation. Like lower power consumption and higher switching speed. Yet it’s all in newsmaller packages. Our latest SDIP package is 50% smaller than previous devices andis ideal for circuits that require the reinforced isolation demanded for international safetycertification.

Whether your application is for industrial or domestic appliances, drives or factoryautomation interfaces, when you want less, Toshiba gives you more.

Viisit us today at www.toshiba-components.com/photocouplers

PCIM Booth 12 / 301

Of the many possible topologies for a switch-mode power supply

(SMPS), the most popular in the sub-150 W category, is the flyback

converter. This, in spite of the fact that it places high stresses (volt-

age and current) on the primary side switch and secondary diode.

However, the flyback converter does offer major advantages. It can

operate over a wide input voltage range and provide multiple isolated

regulated output voltages from a single switching element.

The flyback topology

Figure 1 illustrates the basic implementation. The flyback converter is

derived from buck-boost topology with a mutually-coupled inductor

that forms the transformer. The transformer provides a voltage ratio

from input to output and the advantage of galvanic isolation.

In Figure 1, the control IC incorporates a high-voltage MOSFET

switch that switches the input (normally rectified AC mains) via the

transformer primary. Figure 2 shows the resulting voltages and cur-

rents seen by the MOSFET switch and output rectifier diode.

In the ON state, the voltage across the switch falls to zero and the

current ramps linearly in the primary winding. During this time, the

magnetizing inductance of the transformer accumulates energy from

the input. Meanwhile, the output diode is off (reverse biased) and

energy is supplied to the load from the output capacitor.

When the MOSFET turns off, the energy stored in the magnetizing

inductance transfers to the secondary winding. The voltage across

the secondary winding reverses, the diode turns on (forward biased),

and the magnetizing energy transfers to the output capacitor and

load.

To maintain isolation from input to output, an isolated feedback circuit

is required to provide either voltage or current mode control. In Fig-

ure 1, an optocoupler is used to provide the feedback control path.

There are two main operating modes for the flyback converter: con-

tinuous conduction mode (CCM) and discontinuous conduction mode

(DCM) (see Figure 3).

P O W E R S U P P LY


The Fundamentals of FlybackPower Supply Design

The development of small, light, low cost, and highly efficient switching power supplies is one of the major contributors to the reduction in power consumed by modern electronicproducts and has enabled the introduction of ever tighter industry standards for efficiency

and standby power consumption.

By Sameer Kelkar, Staff Applications Engineer, Power Integrations, Inc. (San Jose, CA)

Figure 1: Implementation of a flyback converter

Figure 3: Continuous and discontinuous conduction modes

Figure 2: Operation of a flyback converter (from top to bottom, MOS-FET current, MOSFET voltage, Output Diode Voltage and OutputDiode Current

With DCM, the energy stored in the transformer is delivered to the

load before the next switching cycle commences. With CCM, the next

switching cycle commences while magnetizing energy is still stored in

the transformer. DCM has the advantage that a lower inductance

transformer is required for a given output power, but at the cost of

higher peak primary and secondary currents, hence higher conduc-

tion losses and lower overall efficiency.

Irrespective of the operating mode employed, when the MOSFET

turns off, an extremely high voltage is generated across the device

(as shown in Figure 4). The voltage sustain capability of the MOS-

FET is the most critical parameter to consider in flyback power sup-

ply design because it affects the selection of all other major compo-

nents.

When the MOSFET turns off, the output voltage across the second-

ary is reflected back through the transformer and multiplied by the

turns ratio (VOR). This voltage appears in series with the (maximum)

input voltage VMAX. In addition, there is a spike due to the effect of

parasitic leakage inductance which appears also in series with the

input voltage VMAX and the VOR. This requires the inclusion of a

voltage limiting clamp network to prevent the voltage rating of the

MOSFET being exceeded.

Maximizing VOR enables a higher turns ratio to be used in the trans-

former. This provides the advantages of lower primary side peak and

RMS currents and lower peak inverse voltages across the secondary

diode. Higher VOR values are especially suited for higher output volt-

ages (12 V) where the corresponding turns ratio does not get too

large. Lower values of VOR are better suited for lower output volt-

ages (e.g., 5 V) or in multiple-output power supplies.

Minimizing the peak and RMS currents clearly reduces conduction

losses within the primary side switch. Another factor to consider is

switching losses within the MOSFET resulting from the charge and

discharge of parasitic capacitances. Power Integrations (PI) imple-

ments the integrated MOSFET using a technology that results in

much lower parasitic capacitances. Hence, the switching losses are

much lower . The technology enables 132 kHz designs to run as effi-

ciently as 66 kHz designs, providing the opportunity for significant

savings in the transformer design by reducing the core size required

for a given output power or increasing efficiency for the same core

size.

Magnetics design

After the controller/MOSFET device, the most critical component in a

flyback converter is the transformer.

The transformer in a flyback converter is always custom made for the

specific application. The design of a transformer involves the consid-

eration several variables – many of them obscure magnetics parame-

ters. Fortunately, smart design tools are available to assist the design

engineer, such as PI Expert? from Power Integrations.

The flyback transformer is different than a conventional iron-cored

line frequency transformer, being more like two coupled inductors. A

conventional transformer is not designed to store energy, whereas

energy storage is fundamental to the operation of a flyback trans-

former. A flyback transformer operates at frequencies above 50 kHz


Figure 4: Reflected output voltage

Figure 5: Flyback transformer construction



vs 50/60 Hz for an iron-cored transformer.

For these reasons, the core is commonly

made of ferrite rather than iron and includes

a non-magnetic air gap. Virtually all the ener-

gy is stored across this gap.

The design of the windings must take

account of DC losses and current density, as

well as AC effects such as the skin effect

(where the current tends to flow along the

surface of the conductor) and the proximity

effect (where the current crowds to one side

of the wire).

Key component considerations

Figure 6 is the schematic for a complete sin-

gle-output power supply employed as an

example to illustrate key component choices.

U1 is a member of the TOPSwitch-JX? con-

troller family from PI, incorporating a MOS-

FET rated at 725 V with low drain-to-source

capacitance (COSS). The low COSS

enables the circuit to operate efficiently at

132 kHz, allowing a smaller, lower cost

transformer core to be used.

The use of high VOR allows the use of 60 V

Schottky diodes (D8, D9) instead of 80 V or

100 V types. The 60 V diodes are more effi-

cient and less expensive because of their

lower forward voltage drop.

The input capacitor C3 must be chosen to

provide the minimum DC voltage to maintain

regulation at the lowest specified input volt-

age and maximum input power. The high

DCMAX limit and optimized dual slope line

feed forward for ripple rejection enable a

capacitance of only 82 μF to be used in this

30 W design.

A primary clamp is necessary to limit the

peak source-drain voltage across the MOS-

FET integrated in U1. An RCDZ (Zener

bleed) clamp (VR1, C4, R5, D5) is used to

give higher load efficiency and lower no-load

consumption. The RCDZ clamp provides a

tighter tolerance than a simple Zener clamp

and allows a VOR as high as 150 V.

The secondary snubber (C12, R17) attenu-

ates ringing generated by parasitics on the

secondary side. If not controlled, this ringing

would generate radiated EMI and could

damage the secondary diodes by exceeding

their voltage rating.

A post filter formed by inductor L2 and

capacitor C16 is often placed at the output of

a switching power supply. This second stage

LC filter reduces high frequency switching

ripple at the output of the power supply. For

lowest ripple, low ESR capacitors should be

used.

Multiple output designs

The flyback topology lends itself well to the

design of multiple output power supplies.

Figure 7 illustrates one example.

The primary choice to be made in designing

a multiple-output supply is whether to use

independent or stacked secondary trans-

former windings. In Figure 7, the -12 V wind-

ing is independent but the others are

stacked. The use of stacked windings

ensures better cross regulation, but a disad-

vantage is that the current of all three out-

puts passes through the lowest winding. In

this design, optimum cross regulation is

achieved on the 3.3 V and 5 V outputs by

the use of foil windings and by sum regulat-

ing (obtaining feedback from both outputs).

With multiple outputs, the design of the

transformer takes on an extra level of com-

plexity, and compromises often have to be

made in the selection of turns ratios. Fortu-

nately, PI Expert provides full support for

multiple output designs.

Conclusion

The flyback converter has become ubiqui-

tous in low-power switch-mode power sup-

plies and in low-cost multiple-output supplies

for a huge range of applications. The avail-

ability of sophisticated controller ICs, togeth-

er with high-voltage MOSFETs and design

support software, enable flyback supplies to

meet the most demanding performance and

environmental standards. It is anticipated

that further developments will push the

boundaries of performance and power con-

sumption even further in the years to come.

www.powerint.com

Figure 6. Schematic of high efficiency 12 V, 30 W, universal input flyback supply with very lowno-load

Figure 7: Schematic of a universal input, multiple-output DVD player power supply

wwwsmt-exhibition.com

Organizer: Mesago Messe Frankfurt GmbH, Rotebuehlstrasse 83–85, D-70178 Stuttgart, Tel. +49 711 61946-79, Fax +49 711 61946-93, [email protected]

Maximizing efficiency in a low-profile form factor is a non-trivial chal-

lenge for even the most experienced power supply designers. Some

examples of systems requiring low-profile power supply designs

include: flat panel displays, rack mounted computer equipment and

telecom and aerospace chassis-mounted assemblies. Equipment in

this class can require several hundred watts of power delivered to the

load at any given time. For example, a typical 12V, 300W power sup-

ply used in a 1U rack mounted application has a maximum height

restriction of 1.75 inches (44.45 mm) and would include forced air

cooling available from 1 or more fans. But for systems with height

restrictions less than 1U, forced air cooling may not be possible,

which means the heat dissipated must be managed using costly, low-

profile heat sinks with large surface area. Therefore, designing for the

highest efficiency is critical because it has a direct impact on reduc-

ing the size and cost of the heat sinks and increasing the overall reli-

ability of the design.

In most cases, AC-DC power supplies operating at these power lev-

els will require some type of active power factor correction (PFC).

The necessity for PFC can be driven by one or more criteria includ-

ing: power level, end application, equipment class and geographical

location and is usually guided by specifications such as EN6100-3-2

or IEEE 519. For an AC-DC power supply, a non-isolated, off-line,

boost pre-regulator is normally used as the PFCstage where its DC

output voltage is seen as the input to a downstream, isolated DC-DC

converter. Since two converters appear in series with each other, the

overall system efficiency, çSYS, is defined by the product of the indi-

vidual converter efficiencies.

(1)

From equation (1) it is apparent that careful consideration must be

given toward choosing the best power topologies

and control techniques for both converter stages. One system solu-

tion that has many interesting high efficiency characteristics is the

combination of an interleaved dual boundary conduction mode (BCM)

PFC followed by an asymmetrical half-bridge (AHB), isolated DC-DC

converter using a current doubler rectifier secondary with self-driven

synchronous rectifiers (SR).

For PFC converters in the 300W-1kW range, interleaved boundary

conduction mode (BCM) PFC should be considered due its higher

efficiency compared to continuous conduction mode (CCM) PFC con-

trol at similar power levels. Interleaved BCM PFC is based upon a

variable frequency control algorithm where two PFC boost power

stages, are synchronized 180 degrees out of phase with respect to

each other. The high peak currents normally seen by the EMI filter

and PFC output capacitor are thereby reduced due to the effective

inductor ripple current cancellation. The output PFC bulk capacitor

benefits from ripple current cancellation because the AC RMS current

flowing through the equivalent series resistance (ESR) is reduced.

Further efficiency benefits are realized since the boost MOSFETs

turn off under AC line-dependant zero voltage switching (ZVS) and

turn on under zero current switching (ZCS). For a 350W interleaved

BCM PFC design, MOSFET heat sinks can be eliminated as can be

seen in Figure 1. Conversely, the boost MOSFET used in a CCM

PFC design is subjected to frequency dependant switching losses

that are proportional to input current and line voltage. By switching

the interleaved BCM boost diodes off at zero current, there are no

reverse recovery losses. Eliminating reverse recovery losses allows

the use of less expensive, fast recovery rectifier diodes and can

remove the need for heat sinking in some cases. For a CCM PFC

design, reverse recovery losses are unavoidable and often dealt with

by applying an RC snubber across the diode, which will lower effi-

ciency or specifying higher performance, silicon carbide diodes,

which have higher associated costs.

For the isolated DC-DC converter design, the half-bridge is a good

topology choice since there are two complementary driven, primary

side MOSFETs and the maximum drain-to-source voltage is limited to

the applied DC input voltage. Two variations of the half-bridge, known

as the LLC and asymmetrical half-bridge (AHB), are widely used part-

ly due to the availability of power management control IC’s uniquely

dedicated to these topologies. The LLC takes advantage of the para-

sitic elements associated with the power stage design to achieve ZVS

using a variable frequency control technique. However, because the

regulated DC output only uses capacitive filtering, this topology is

best suited for lower output ripple, higher output voltage applications.

As a general guideline for off-line, DC-DC applications the LLC tends

to be favored when the output voltage is greater than 12VDC.

��



High Efficiency, Low-ProfileAC-DC Power Supply Design

There may be more than one ideal solution

By Steve Mappus, Principal Systems Engineer, Fairchild Semiconductor, Power Conversion America, PCIA, Bedford, NH

Figure 1: 12V, 300W, Low-Profile, Universal AC-DC Power Supply

The AHB is an efficient choice for a 300W, 12V DC-DC converter. A

fixed frequency control method is used, where the primary current

naturally lags the transformer primary voltage, providing the neces-

sary condition for ZVS of both primary MOSFETs. Similar to the LLC,

the ability to achieve ZVS with the AHB relies upon a thorough

understanding of circuit parasitic elements such as transformer leak-

age inductance, winding capacitance and junction capacitance of dis-

crete power devices. Compared to the variable frequency control

method used for LLC control, fixed frequency operation greatly sim-

plifies the task of secondary side, self-driven SR. The self-driven, SR

gate drive voltages are easily be derived from the transformer sec-

ondary. Adding a low-side MOSFET driver, such as the Dual 4A,

FAN3224 shown in Figure 2, provides accurate level shifting and high

peak drive current through the MOSFSTs Miller plateau region to

assure fast, efficient SR switching transitions.

The current doubler rectifier can be applied to any double-ended

power topology and for high DC current applications, it has several

noteworthy attributes. First, the secondary consists of a single wind-

ing, simplifying the transformer structure. Second, since the required

output inductance is divided between two inductors, the power dissi-

pated due to the high current flowing in the secondary is distributed

more efficiently. Third, the individual inductor ripple currents cancel

each other as a function of duty cycle (D). The cancelled sum of the

two inductor currents has an apparent frequency equal to twice the

switching frequency allowing higher frequency; lower peak current

flowing into the output capacitor. And finally, in a symmetrical convert-

er (push-pull, half-bridge, full-bridge), each current doubler inductor

would carry half the output current but for the AHB this is not exactly

the case.

If unaccounted for, the asymmetrical voltage applied to the secondary

side rectifiers can be one of the AHB drawbacks. When the AHB is

operated near its limit of D=0.5, the applied SR voltages are nearly

matched. However, it is more reasonable that the transformer turns

ratio be designed such that D is within the practical range of

0.25<D<0.35 during nominal operation. When D is within this range,

the voltage stress between Q1 and Q2 and the applied voltage

across L1 and L2 become imbalanced, resulting in an uneven current

distribution between L1 and L2. Similarly, the voltage ratings for each

SR MOSFET must also be considered. For this reason, it may be

acceptable to use inductors L1 and L2 that are not equal in value and

SR MOSFETs that have different voltage ratings. The transformer

turns ratio can also be wound asymmetrically but these techniques

require a detailed understanding of the circuit behavior under all

operating conditions.

To demonstrate the feasibly of the recommended solution, the speci-

fications shown in Table 1 were met using an interleaved dual BCM

PFC boost, pre-regulator followed by an asymmetrical half-bridge,

DC-DC converter with self-driven SR, as pictured in Figure 1.

The specifications shown in Table 1 are a simplified summary of the

full design requirements. The primary design goals are to:

• Maximize efficiency over the widest range possible

• Achieve lowest possible design profile

• Minimize size and use of heat sinks

Maximizing efficiency over the widest possible load range requires

careful consideration when choosing materials and components for

each power stage, particularly in the area of magnetics design.

Because the frequency for the interleaved BCM PFC can reach sev-

eral hundred kHz, and vary by as much as 10:1, the boost inductors

need to be custom designed. Using a properly rated, equivalent

gauge litz wire gives best results for minimizing AC losses that can

dominate copper loss in a BCM PFC boost inductor. A gapped ferrite

material suitable for high frequency operation should be used and for

this example, N87 material from EPCOS was chosen on a low-profile


Figure 2. FAN3224, Self-Drive SR with Current Doubler Rectifier

Table 1: Low-Profile, AC-DC Power Supply Design Specifications

��

��

�� !"� �� "� �� #�$��

EFD30 ferrite core set. Measured efficiency results for the PFC are

shown in Figure 3.

One solution for a 300W, low-profile, AHB transformer requires two

horizontal core structures, where the primary windings are connected

in series and the secondary windings connected in parallel. The use

of two transformers is necessary because the cross sectional area,

Ae, of each core is nearly half of the 150mm2 required to avoid satu-

ration. Finding a single, conventional core shape with a 150mm2

cross section would not be possible in a less than 20mm low-profile

component. Similar to the BCM PFC inductor design, litz wire and a

high frequency ferrite core material are used to maintain high efficien-

cy. A final important design step is controlling the amount of allowable

leakage inductance in the AHB transformer. Some value of leakage

inductance is required for ZVS and adjusting the timing delay for the

self-driven SR. For this design the effective leakage due to both

transformers was optimized to 7μH or 1.5% of the total effective mag-

netizing inductance. Measured efficiency results for the 300W AHB

DC-DC converter are shown in Figure 4.

Full load efficiency is dominated by conduction losses through the

converters power stage so there is little a controller can do to help

under these conditions. However, there are several controller tech-

nologies that should be considered for maintaining higher light load

efficiency. The FAN9612, an interleaved dual BCM PFC controller,

limits frequency-dependent Coss MOSFET switching losses at light

load and near the zero crossing of the AC input voltage by utilizing

an internal fixed maximum frequency clamp. During the portion of the

AC line voltage for VIN>VOUT/2, Coss capacitive switching losses are

reduced through a valley-switching technique used to sense the opti-

mal MOSFET turn-on time. Conversely, when VIN<VOUT/2, the PFC

boost MOSFETs always turn-on under ZVS conditions. Light load effi-

ciency improvements are further attained by introducing an automatic

phase management feature that reduces operation from dual channel

to single channel mode. The light load efficiency advantage from

phase management can be seen in Figure 3 for 10%<POUT<20%,

where the efficiency “curve” appears more flat. Operating in single

channel mode minimizes the impact of switching losses on light load

efficiency. The ability of the interleaved PFC to maintain synchroniza-

tion during phase management is shown in Figure 5. The left-sided

plot was recorded when transitioning from single channel to two

channel operation as the load is increased from zero to 19% (64W).

Similarly the right-sided plot was recorded when transitioning from

dual channel to single channel operation while the load is decreased

from full load to 12% (42W).

The implementation of the AHB isolated DC-DC converter is

achieved using the FSFA2100, AHB controller, which integrates the

pulse width modulation (PWM) control, gate drive functionality and

internal power MOSFETs into a single 9 pin SIP power package. This

advanced level of packaging and integration allows designers to

achieve very high efficiency up to 420W, using fewer external compo-

nents. Combining these three critical functions into a single package

eliminates the task of programming the dead time required for ZVS

and minimizes gate drive parasitic inductance between the internal

driver and MOSFETs. Most of the power dissipated within the SIP

power package is due to the switching internal MOSFETs, so a low

profile extruded heat sink is required, especially for a 300W design

with no available forced air cooling.

The total AC-DC system including, the input EMI filter, bridge rectifier,

interleaved BCM PFC and AHB DC-DC yields a measured overall

efficiency as shown in Figure 6. The design achieves 91% peak effi-

ciency for Vin=120VAC, 92% peak efficiency for Vin=230VAC and

greater than 90% for Vin=120VAC or 230VAC, POUT>38% (114W).

Magnetic component design, power semiconductor selection, pcb

layout, choice of heat sinks and controller features all must work per-

fectly together for a successful low-profile AC-DC power supply

design demonstrating high efficiency over a wide load range.

Depending upon system requirements, there may be more than one

ideal solution best suited for a particular application. The design dis-

cussed herein is just one example for achieving high efficiency from

a universal AC input to 12V, 300W design requiring PFC and a low

profile of only 18mm total height.

PCIM Booth 12/601

www.Fairchildsemi.com



Figure 4: AHB 390V to 12V/25A, DC-DC Measured Efficiency(100%=300W)

Figure 5: PFC Phase Management (1→2, 19%=64W and 2→1,12%=42W)

Figure 6: Total Measured System Efficiency (EMI Filter Included)

Figure 3: Interleaved BCM PFC Measured Efficiency (100%=330W)

Register now: www.sensor-test.com

Organiser: AMA Service GmbH, P.O.Box 2352, 31515 Wunstorf/Germany, phone +49 5033 96390, [email protected]

2010

18 – 20 May 2010Nürnberg, Germany

17th International Trade Fair for Sensorics, Measuring and Testing Technologies with concurrent conferences

The most comprehensiveindustrial trade fairranging from innovativesensors to highlysophisticated analysis

The following article is a preview summary of ST’s App notes AN

3105 and 3106 appearing in summer 2010 on ST’s homepage.

Abstract

As a result of their features such as high efficiency and very long life-

time, LEDs are becoming increasingly popular. They are driving inno-

vation of current lamp types and make a substantial contribution to

energy reduction for internal or external lighting. This is also happen-

ing in street lighting applications, where the higher efficiency and life-

time are vital for reducing total costs (including maintenance) and

energy consumption. For these reasons, a street lighting power sup-

ply designed to power an LED lamp has to have high efficiency and

at least similar lifetime to the LED, in order to guarantee the mainte-

nance-free operation required by this kind of application during the

LED’s useful lifetime.

This article describes the characteristics and features of a 130 W ref-

erence design board adapted to a LED power supply specifically

designed for street lighting.

Introduction

The circuit is composed of two stages: a front-end PFC using ST’s

L6562AT and an LLC resonant converter based on ST’s L6599AT.

The main features of this design are very high efficiency (more than

90%), a wide input mains range (85-305 VAC) operation and long

term reliability as well.

Because reliability (MTBF - mean time between failures) in power

supplies is typically affected by electrolytic capacitors and their high

failure rate, unless very expensive types are used, this board shows

a very innovative design approach. This board uses film capacitors

(from Epcos) instead of electrolytic capacitors. Component derating

has also been carefully applied during the design phase, so decreas-

ing the component stress as recommended by MIL-HDBK-217D.

With the use of ST’s new L6562AT and L6599AT devices, the num-

ber of components used for this solution has also been minimized,

thus increasing the MTBF and optimizing the total component cost.

As a result of the high efficiency achieved, only a small heatsink for

the PFC stage is needed, while the other power components are

SMT like most passive components, thus decreasing the production

labour cost.

The board is also protected against overload or short circuit, open

loop of each stage or input overvoltage because of the particular

application, after intervention the system auto-restarts.

Main characteristics and main functional block description

The main features of the SMPS are listed here below:

• Extended European input mains range: 85 to 305 VAC - Frequency

45 to 55 Hz

• Output voltage: 48 V at 2.7 A

• Long lifetime by film capacitors from EPCOS

• Mains harmonics: according to EN61000-3-2 Class-C

• Efficiency at nominal load: better than 90%

L I G H T I N G


High-Efficiency Converter withPFC for LED Street Lighting

at 48 V and 130 WThe Symbiosis of active and passives IC reference design

For an advanced PCB-design, developers want to make the right choice by knowing the application and the specific requirements. Next to the design assistance this is the most

important reason for App-Notes issued by the IC makers. In this early stage, IC makers such asST needs the support of a broad liner in passive components like TDK-EPC. The wide portfolio

of TDK-EPC, using brand components both TDK and EPCOS, allows the designer to choose theright components out of this portfolio or to realize a design to fit component.

By Claudio Spini, Senior Engineer in the Application Laboratory of STMicroelectronicsAgrate Brianza, Italy and Davide Giavarini EPCOS AG – A Group Company of TDK-EPC

IC Reference Design, Milano, Italy and Wolfgang Dreipelcher EPCOS AG – A Group Companyof TDK-EPC Senior Director IC Reference Design,Munich, Germany

Figure 1: Main Board


• EMI: according to EN55022-Class-B, EN55015

• Safety: double insulation, according to EN60950, SELV

PFC circuit

The PFC stage, working in transition mode, acts as pre-regulator and

powers the resonant stage with the output voltage of 450 V. The PFC

power stage is a conventional boost converter, connected to the out-

put of the rectifier bridge. It is completed by the boost coil, the rectifi-

er diode and the output capacitors. The PFC output capacitors are

film type, 5 μF, 800 V from Epcos.

The boost switch uses a MOSFET. The board is equipped with an

input EMI filter required to filter the commutation noise coming from

the boost stage. The PFC implements the controller L6562AT, a small

and inexpensive controller guaranteed for operation over a wide tem-

perature range necessary for outdoor operation.

Resonant stage

The downstream converter implements the ST L6599AT, incorporat-

ing all the functions necessary to correctly control the resonant con-

verter and working with 50 percent fixed duty cycle and variable fre-

quency. The transformer uses the integrated magnetic approach,

incorporating the resonant series inductance. Thus, no additional

external coil is needed for the resonance. The transformer configura-

tion chosen for the secondary winding is the typical center tap, using

a couple of power Schottky rectifiers, type STPS10150CG. The out-

put capacitors are film type, 4.7 μF, 63 V from Epcos. A small LC fil-

ter completes the output section in order to filter the high frequency

ripple. A feedback network guarantees the required stability of the

output voltage.

Efficiency measurement

Table 1 shows the overall efficiency, measured at 230 Vac, 50 Hz

and 115 V, 60 Hz with different loads also. At 115 Vac and full load,

the overall efficiency is 90.96% and it increases up to 93.39% at 230

Vac. This makes this design suitable for applications requiring high

efficiency.

Measuring the efficiency at 25%, 50%, 75% and 100% according to

the ES-2 standard and calculating the average efficiency, this is

91.04% at 230 Vac and 89.52% at 115 Vac. This shows that the con-

verter can operate with a high efficiency not only at full load but also

at lower loads such as in the case of LED deep dimming.

Conclusion

A 48 V to 130 W power supply for street lighting applications has

been designed and the prototype has been tested. The adopted solu-

tions meet the LED street lighting specifications for wide input voltage

range, wide temperature range operation and high efficiency.

Additionally, the design guidelines and solution implemented meet

the required MTBF target.

Reference

[1] Energy Star Program requirements for EPS: Version 2.0 available

at www.energystar.gov

[2] S. De Simone; C. Adragna; C. Spini; G. Gattavari: Design-orient-

ed steady-state analysis of LLC resonant converters based on

FHA. SPEEDAM 2006. International Symposium, May, 23rd -

26th, 2006

[3] AN2321 - Reference design high performance, L6599-based HB-

LLC - available at www.st.com

[4] AN2761 - Solution for designing a transition mode PFC preregu-

lator with the L6562A - available at www.st.com

PCIM Booth 12/535

www.tdk-cpc.com

www.epcos.com

www.st.com

Figure 2: Board with the line DC/DC-Modules

Table 1. Efficiency by load percentage

A brief History of RT Simulation

Interestingly, the first real time (RT) simulator was built, in fact, for

power, (albeit for power transmission, not PE) in 1927 at MIT. The

first RT digital simulator, Whirlwind, was built also at MIT in 1951 for

air defence with data being sent to it, in RT, via telephone lines. In

the 1980s, digital simulation was employed to emulate aeronautical

systems and is today extensively used in the increasingly complex

world of automobiles which contain up to 80 micro processors and

the accompanying communication network between them. However,

these simulators can only emulate relatively slow, mechanical sys-

tems and any PE response is effectively averaged since a typical

computation cycle takes about 50μs (admittedly, with a very large

number of simultaneous floating point operations - FLOPS).

Faster processors are bringing RT time steps down to about 30μs but

semiconductor switching events occur within one microsecond, for

the most part and that is about the time it takes to fail a semiconduc-

tor, so until now, RT simulation of converter circuits has remained the

domain of analogue simulators: in 2010, this changes with the arrival

of the first dedicated PE real time digital simulator (RTDS).

Digital versus analogue simulation

As mentioned, analogue systems have been successfully used for

decades, so what is the reason for going digital?

The analogue system is usually a scaled down version of the real

converter-under-test, which is still a real converter in its own right. It

has to be built and verified before it can serve as a test vehicle.

Moreover, it has to be tested in a power lab by trained technicians,

with variable loads, variable power-supplies, instrumentation (oscillo-

scopes etc) while respecting the applicable safety regulations. When

something goes wrong (and if it didn't there would be no need to test)

there is often a loud “bang” and the simulator has to be repaired or

rebuilt. In companies where several converters are in design or being

upgraded with new controls, these test facilities and their personnel

become bottle-necks in the development process.

A digital simulator, by contrast, is essentially a fast, dedicated desk-

top computer which can be used in the same office environment as

the off-line simulator which supported the original design. Thus a

recently designed controller, its software and firmware can be fully

tested in the "Hardware in the Loop" (HIL) configuration before ever

going into a power lab.

HIL testing allows:

• reduced development cycles

• safer and lower-cost testing

• better code coverage during tests

• easy testing of parameter changes

• elimination of costly failures and down-time

• automated test cycles which permit all operating conditions to be

scanned with any malfunctions recorded and flagged

• comprehensive testing for improved safety and reliability (reduced

commissioning times, fewer on-site repairs or recalls).

D E S I G N A N D S I M U L A T I O N


Ultra low latency HIL simulatorfor Power Electronics Applications

Software testing moves from the laboratory floor to the office desk

Digital simulation in power electronics (PE) is a standard tool of design but it has traditionally referred to "off-line" simulation in which simulation time on a PC may beorders of magnitude greater than the simulated events would be. For this reason, power

electronic circuits are today still real-time (RT) tested with analogue simulators – basically scaled down versions of the real converter in a real power lab, because PCs are

simply not as fast as power electronic switches. All this is about to change thanks to custom processors with one microsecond latency and their accompanying software.

By Eric Carroll and Ivan Celanovic, Typhoon RTDS GmbH

Figure 1: RTDS150 schematic editor and graphic user interface


Testing in the power lab is not eliminated, as a final verification of a

complete system remains necessary but it is now limited to type test-

ing and out-going inspection rather than being an integral part of the

development process.

Other applications of PE RTDS include customer and service person-

nel training as well as teaching of controls and power electronics.

The Typhoon RTDS Breakthrough

Today's RT digital simulators successfully emulate complex systems

such as power system networks, trains, planes and automobiles in

which massive computational power is required but the I/O (or loop-

back) latency is generally in the range of 50 to 100μs. PE represents

a different challenge in that even a complex switching matrix such as,

say, a back-to-back three-level inverter with braking choppers contain

relatively little data (78 states, voltages and currents) but which

change very quickly. Thus the processor required to emulate such a

system needs to be very fast but does not necessarily need to han-

dle many GFLOPS.

Where commercial processor development strives to achieve high

levels of computing power, the Typhoon RTDS processor aims to be

very fast operating at a fixed cycle time of 1μs by which it is under-

stood that the combined I/O latency and calculation time is 1μs, irre-

spective of the circuit complexity. With such a short latency, the

switches respond as quickly as in a real converter (turn-on and off

times for a 1700V IGBT are about 1 and 2μs respectively).

T-RTDS Modelling Approach

To achieve “hard” real time digital simulation of switched dynamic

systems, not only the processor has to be fast but the PE system

representation has to be minimalist i.e. the algorithms must be "lean

and mean" and the input/output boards (I/O), very fast. It is the com-

bination of a custom processor, ideal switches and optimised algo-

rithms which allows a completely deterministic simulation time step –

the key to real-time simulation. As opposed to off-line simulations,

where variable time steps are possible and time-reversals are used

to correct zero-current crossings; with RT, there is no going back!

Typhoon Simulators

Currently, one standard product is available for simulating a 2-level/3-

phase inverter with a diode rectifier input, according to the circuit of

Fig. 1, in which all the passive components and motor parameters

can be programmed and the supply voltage set and perturbed by

harmonics, flicker or glitches. Fig. 1 also shows the graphic user

interface (GUI) for selecting and scaling the o/p signals

The RTDS150 has 8 BNC outputs which can be scaled and pro-

grammed to display any state, voltage or current in the circuit of

Fig. 1. Fig. 2 shows a typical output of four channels displayed on

a Tektronix oscilloscope.

The RTDS800 is a more versatile machine scheduled for release in

the autumn of 2010. It is based on the circuit of Fig. 3 and allows a

degree of topology configuration achieving 16 variants of the circuit

(single-phase, 3-phase, brake and boost choppers etc.) and allows

the addition of up to 50 passive components.

Beyond these two standard simulators, other semi-fixed topologies as

well as multi-converter and/or multi-level configurations can be

realised on a custom basis.

Outlook

The high speed platform having been firmly established, two major

developments will follow in the next 2 – 3 years.

Firstly, Typhoon's present custom-processor technology allows a fur-

ther reduction of latency below 1μs which will be needed for the emu-

lation of switches in very high frequency converters (say, to 1MHz

PWM). Secondly, Typhoon application-SW will be expanded with a

suit of design-evaluation tools such as test-automation, oscilloscope-

functionality, junction-temperature calculation as well as efficiency

and EMC evaluation.

Conclusions

The growth of PE in recent years has been remarkable, driven by the

need for energy savings (e.g. with motor drives) and energy produc-

tion from renewables (e.g. wind and solar) as well as by the rapid

introduction of electric drives in automobiles. This increasing develop-

ment speed puts new time-to-market constraints on both develop-

ment and test engineers.

HIL testing makes controls development and upgrades faster, more

thorough and bug fixes are far less costly when caught early in the

process, which is the reason why HIL testing methodology is so

widely adopted for testing complex systems, other than PE. With the

latency barrier finally broken, full RTDS for stand-alone PE simulators

and the embedding of PE RTDS into existing system simulators, is

now not only, possible but also very cost-effective.

www.typhoon-RTDS.ch

www.bodospower.com May 2010 Bodo´s Power Systems®

D E S I G N A N D S I M U L A T I O N

Figure 2: Typical simulator output

Figure 4: The RTDS150 which simulates the system of Figure 2

Figure 3: The versatile RTDS800 has semi-variable topology


Since the introduction of the IGBT, the concept of a cold plate to cool

and serve as a mechanical base for that subassembly has imposed

itself as a reference for many reasons:

- excellent cooling performances thanks to forced circulation of a

fluid (water or water + antifreeze);

- use of a fluid with little environmental impact;

- the small footprint of the cold plate;

- simplicity of maintenance, since the component can be dismounted

with no loss of cooling fluid;

- reasonable cost, weight and robustness;

- the prospect of standardizing the cooling units used, since the

power electrician can use the same cold plate for many electrical

solutions and diagrams;

- an interface between the product and the need that matches the

industrial model and the skills of the three players involved: the

designer of the power electronics function, the maker of the power

electronics component, and the cold plate manufacturer.

Our goal here is to describe some cooling solutions (shape) and cold

plate concepts, and explain the reasons for choosing Ferraz Shaw-

mut products. A cold plate (picture 1 gives an example) is basically

the component's mechanical base, with heat exchange surfaces and

circulation of a cooling fluid. To improve performances, larger heat

exchange surfaces are provided on the base supporting the compo-

nent (on the same principle as fins in air cooling devices). Another

important "detail" is the cover, which serves to seal the product.

Aluminium is the best material available, because of its good thermal

conductivity, its density and its cost, and because it is easy to

machine and form into complex geometrical shapes.

As a young engineer, once I had estimated the thermal requirements

(heat exchange surface area, etc.) for my first cooling unit, I was

stumped by another question: how would I be able to seal my cold

plate effectively with a cover? Contemporary engineering being well

versed in automotive applications, a simple "cylinder head gasket"

seemed an easy solution. But power electronics products have spe-

cific requirements, like outstanding durability and the total exclusion

of any leakage (of water!) in a high voltage environment. So... back

to the drawing board.

Vacuum brazing of aluminium is a technology Ferraz Shawmut mas-

tered more than 15 years ago and is a simple, reliable solution. Braz-

ing is similar to soldering but without melting the base part. When

aluminium is brazed, an aluminium-silicon eutectic alloy is heated to

the melting point between the parts. This guarantees both durability

and an excellent contact between the cover and the base plate over

the entire surface of the cold plate.

T H E R M A L M A N A G E M E N T

Cold Plates for Water CoolingElectronic Components

The thermal and engineering solutions

Cooling electronic components, especially power electronics, is a field where many skills converge: thermal design of the cooling unit, familiarity with the components and

with all the requirements pertaining to the power diagram, and mechanical design and construction of the cooling unit.

By L Dubois/ JL Dubelloy, Ferraz Shawmut Thermal Management, La Mure, France

Figure 1: Heat extraction cross section

Figure 2: Principle of vacuum brazing

Figure 3: Vacuum brazing furnace


There are other technologies available, but brazing with flux cannot

guarantee the parts will remain as clean, and electron beam or laser

welding, despite the cost aspects, is best used for straight welds and

is unsuitable for large flat areas.

With a perfected technology to make a closed, watertight cold plate,

we can go on to optimize it thermally. The "improved" heat exchange

surfaces we were familiar with in the field of air cooling led us to offer

our customers some optimized cooling shapes to meet requirements

such as:

- affordable cost;

- a reasonably "open" shape to avoid any risk of plugging in the cir-

cuit;

- surefire solutions that guarantee watertightness and high perform-

ance for mass produced products with a long life span.

Machining those shapes offers a control of geometry that is not pos-

sible with a technology involving the attachment of fins. So Ferraz

Shawmut proposes cooling shapes machined directly in the cold

plate for optimal product quality. Indeed, brazing filler (the material

that is melted) is generally only about 100 microns thick, and any

lack of sufficient thickness when a fin is attached can degrade cool-

ing of the component's chip in places (over a few square centime-

ters) if the contact at that point is not good enough.

Optimizing heat exchanges led us to develop cooling shapes in either

pin or wave forms. Heat exchange performances are summarized in

the following chart for a typical application (pure water) with these 3

types of shape. The extra heat exchange provided by pin or wave

shapes cuts thermal resistance in half compared to conventional

straight fins (marked "straight" in the chart).

In recent developments proposed by FSTM, the concept of a counter

current, combined with straight or wavy fins, offers a further step for-

ward. How the cooling fluid heats as it circulates is directly related to

fluid flow and power dissipation. For instance, a flow of 30 l/min in a

cold plate cooling 8 components, each with power of 2500 W, means

the fluid itself (water + 40% EG) will heat by 11°C. So the last com-

ponent on the cold plate is cooled by fluid at a much higher tempera-

ture. That effect can be limited by a dual design with one hot and one

cold channel for the cooling fluid, restricting heating to 5.5°C and

guaranteeing a uniform temperature under each component, which is

an important factor in the electrical balance between components.

Conclusion

FSTM's efforts at optimization in the field of cooling units for power

electronics have constantly driven us to improve our cooling topolo-

gies, in order to offer the best possible performances and high quali-

ty, durable products. New shapes are being tested to further improve

both performance and manufacturing costs, so that we can continue

to offer our customers the best products on the market. Ferraz Shaw-

mut Thermal Management is one of the foremost players on that

market.

PCIM Booth 12/510

www.ferrazshawmut.com

T H E R M A L M A N A G E M E N T

Figure 4: Example of a cold plate

Figure 5: Thermal performance






www.biricha.com

PCIM Booth 12/422

The key features of the new IGBT module

include an integrated copper base and pin-

fin technology. This allows the implementa-

tion of greaseless direct liquid-cooling which

enables an extremely low thermal resist-

ance. The module also adopts a Si3N4 insu-

lated substrate that is highly fracture resist-

ant and enhances the long term reliability of

the module. RoHS compliant lead free solder

and ultrasonic bonding technology are used

for the main terminals.

The footprint size of the cooling jacket is the

same size as the number of IGBT modules

required for the inverter or converter, up to a

maximum of six. Depending on the ability of

the coolant pump, users can choose a cool-

ing jacket with serial flow channel, 2-parallel

flow channels, 3-parallel flow channels or 6-

parallel flow channels. Each jacket has inter-

nal flow junctions and two connections; one

inlet and one outlet. The power unit is 37%

lighter and 45% smaller than a conventional

power unit of equivalent power capacity con-

sisting of indirect liquid-cooling IGBT mod-

ules and heatsink.

The thermal resistance of the new IGBT

module is reduced by direct liquid-cooling

technology. In an indirect liquid-cooling IGBT

module, heat from the IGBT die flows

through the solder layer, metal layer, ceramic

layer and thermal grease layer to the

heatsink. The thermal conductivity of grease

mounting compounds is almost similar in

value to the thermal conductivity of copper.

Therefore the thermal resistance of indirect

liquid-cooling IGBT is high. Conversely, the

direct liquid-cooling IGBT module does not

use thermal grease and the thermal resist-

ance of a direct liquid-cooling IGBT is small.

The temperature distributions under switch-

ing operation in the IGBT die are shown in

Figure 2. The temperature distribution is pre-

sented using thermal-liquid simulation.

The pressure drop in the coolant channel is

measured with a coolant jacket of a serial

flow channel for 2 IGBT modules. The

coolant used in the experiment was com-

prised of 50% ethylene glycol and 50%

water. Coolant temperatures are 0°C, 10°C

and 50°C. The newly designed pin-fin base

plate and channel cover jacket enabled a

reduced pressure drop compared to that of a

conventional indirect liquid-cooling heatsink.

The reliability of the new IGBT module is

highly affected by any coolant leakage. In

order to predict the risk of leakage under

operating coolant pressure, stress simulation

testing was performed (see Figure. 3). The

maximum warpage deformation at the O-ring

contact surface points under 500kPa coolant

pressure, which is regarded as the typical

discharge pressure of the coolant pump, is

approximately 0.04mm. This value is smaller

than the accepted deformation to avoid

coolant leakage. Therefore the module can

endure coolant pressure under operation

and avoid any coolant leakage. A channel

jacket for 2 IGBT modules was used in the

test.



Direct Liquid Cooling IGBT Modulefor Wind Power Applications

Compact size, low thermal resistance and high reliability packaging technology

The market for renewable wind power generation devices continues to increase rapidly.Power electronics for wind power systems should be small, lightweight and highly

reliable in order to minimise maintenance and enhance reliability throughout the productlifetime. In order to satisfy these stringent needs Hitachi has developed a new 600A,

1700V direct liquid cooling IGBT module.

By Neil Markham, Hitachi

June 9–11, 2010

The World´s Largest

Exhibition for the Solar Industry

New Munich Trade Fair Centre | Germany

www.intersolar.de

1,500 Exhibitors

130,000 m2 Exhibition Area

60,000+ Visitors

64 Bodo´s Power Systems® May 2010 www.bodospower.comBodo´s Power Systems® May 2010 www.bodospower.comBodo´s Power Systems® May 2010 www.bodospower.com

The thermal fatigue life of the solder layer is

improved by using a coefficient of thermal

expansion (CTE) matching technology. In a

conventional IGBT module a thick AlN (alu-

minium nitride) substrate with a thin alumini-

um layer is used to insulate the IGBT circuit.

The new Hitachi IGBT module uses a thin

Si3N4 (silicon nitride) substrate with a thick

copper layer. The high fracture resistance of

Si3N4 compared to AlN allows the use of a

thick copper layer. The thin Si3N4 substrate

and thick copper layer increases the equiva-

lent coefficient of thermal expansion of

Si3N4 and copper laminate. The difference

between the CTE of the Si3N4 and copper

laminate and CTE of copper pin-fin base

plate is small. Thus, the thermal stress of

substrate/base plate connecting solder layer

is reduced and the thermal fatigue life of the

solder layer is improved. Hitachi’s bespoke

lead-free solder also improves the fatigue life

of the solder. Thermal fatigue life diagram of

new IGBT module is shown in Figure 4.

Power cycling life diagram of new IGBT is

shown in Figure 5.

Junction temperatures under several operat-

ing conditions are estimated. In the estima-

tion a 500kW electric power conditioning

system with six IGBT modules (two IGBT

modules per phase) is considered. Coolant

flow rate is set to 8 litres per minute/channel.

There are 3 flow channels and a serial chan-

nel for 2 IGBT modules. Total flow rate is 24

litres per minute. Coolant temperature is

assumed as 50°C Tj. Estimated results show

that according to its low thermal resistance,

the new Hitachi direct liquid-cooling IGBT

module can operate over 5 kHz switching

frequency with the proposed flow channel

diagram.

As a result of extensive research and devel-

opment a new Hitachi direct liquid-cooling

IGBT module has been developed. The new

module uses an integrated pin-fin base plate

to reduce the thermal resistance from IGBT

chip to coolant without the use of thermal

grease. The pin-fin layout and channel cover

jacket design have been optimised by fluid-

thermal simulation. Coolant leakage reliabili-

ty is also optimally designed by stress simu-

lation. The Si3N4 substrate and RoHS bond-

ing technology are developed to ensure high

reliability and long lifetime of the IGBT mod-

ule. Due to its low thermal resistance,

approximately 65% of the thermal resistance

of conventional indirect liquid-cooling IGBT

module, a 500kW power conditioner with 6

IGBT modules can operate at 5 kHz or dou-

ble typical application frequencies of 2.5

kHz. Additionally to offer support for many

wind and solar power applications Hitachi

will also be launching 600v and 1200v ver-

sions of the new liquid cooled direct cooling

IGBT module in the near future.

PCIM Booth 12/355

www.hitachi.eu/pdd


Figure 1: Outline of the new direct liquid-cooling IGBT module

Figure 2: A comparison of direct liquid-cool-ing and indirect cooling (with thermal grease)showing the temperature distribution aroundthe IGBT die. The simulation results indicatethat the thermal resistance of the direct cool-ing module is 25% to 80% lower than theindirect cooling module.

Figure 4: Thermal cycling diagram of newIGBT module

Figure 5: Power cycling diagram of newIGBT

Figure 3: Warp deformation of direct coolingpin-fin under coolant pressure. Reliability ofcoolant sealing structure is guaranteed byoptimised pin-fin design with simulation tech-nologies. Maximum warpage under coolantpressure 500kPa < 0.04mm

JULY 13–15 Moscone Center, San Francisco, California

Get back tothe business of innovation.

It’s time to get back to work on the ideas and technologies

that advance microelectronics design and manufacturing.

It’s time to discover new opportunities and markets that are

changing the future of the industry. Most of all, it’s time for

SEMICON West.

Register Now:www.semiconwest.org

It’s elemental—plan now to be part of SEMICON West 2010!

the elements of innovation

Brought to You by:

66 Bodo´s Power Systems® May 2010 www.bodospower.comBodo´s Power Systems® May 2010 www.bodospower.com

Haefely Technology’s new ONYX electrostatic discharge simulator is

the most ergonomic 30kV ESD gun without an additional base con-

trol unit that can be battery or mains operated. It is available in 16kV

or 30kV versions. The easy to use touch screen, integrated “SMART-

KEY”, ergonomic design, modular RC units, multilingual interface,

remote control software, built-in LED light, temperature and humidity

display allows for trouble-free use of the ONYX in all types of test

sites.

www.haefely-onyx.com

ESD Simulator ONYX

N E W P O R D U C T S

Microchip announced the expansion of its 16-bit dsPIC® Digital Sig-

nal Controller (DSC) portfolio for digital power-conversion applica-

tions. Microchip’s 16-bit dsPIC33F ‘GS’ Series DSCs provide on-chip

peripherals specifically designed for high-performance, digital power

supplies. On-chip digital power peripherals include high-speed

Pulse-Width-Modulators (PWMs), ADCs and analogue comparators.

The newly expanded dsPIC33F ‘GS’ family supports applications

such as induction cooking, uninterruptable power supplies, solar and

pure sine-wave inverters, intelligent battery chargers, power factor

correction, HID lighting, fluorescent lighting, LED lighting, and AC-DC

and DC-DC power converters.

These new DSCs provide up to four times the memory, compared to

Microchip’s existing SMPS & Digital Power Conversion families.

Additionally, these flexible DSCs can be configured for a variety of

topologies, giving power-supply designers the complete freedom to

optimise for specific product applications. The eight new DSCs offer

up to 18 channels of PWMs with 1nS resolution, enabling an

unprecedented number of completely independent digital control

loops.

The eight new dsPIC33F ‘GS’ series digital-power DSCs enable digi-

tal control loops with 12 to

18 high-speed, 1ns resolution PWMs and one or two 10-bit, on-chip

ADCs, providing 2 to 4 Million samples per second (MSPS) for low

latency and high-resolution control. They range from 64 to

100 pins and 32 to 64 KB Flash memory. These DSCs feature inter-

active peripherals that both minimize the intervention of the proces-

sor and are able to handle the real-time needs of high-speed current-

mode control.

PCIM Booth 12/363

www.microchip.com

Digital Signal Controller for Digital Power Applications

Isabellenhütte is presenting the IPC series of current measurement

modules for the motor phase current on measurement shunt basis at

the PCIM and Sensor+Test trade fairs. The modules were designed

to control electric high-performance motion control units of all types.

The data acquisition runs at sampling rates from 10 to 300 kHz. A

programmable sampling delay time triggered with an external signal

is supported. The modules also supply a fast over current signal

(response time < 2 μsec).

The IPC series features a modular structure. Currents from 20 A up

to several thousand amperes can be measured by matching the

shunt value. For example, a continuous current measurement of 5 kA

can be realised by using a 2 ìOhm shunt. All IPC modules are gal-

vanically isolated, highly dynamic, an extremely accurate measure-

ment and with a resolution between 12 and 16 bit. An integrated

modulation unit processes the current value signals under considera-

tion of the calibrated parameters. The resulting digital signal is trans-

mitted, potential-free, over various interfaces. It can be processed

directly by motor controller via a digital input.

Further advantages of the current measurement modules are the low

ohmic shunts that reduce power loss, low offset and drift as well as

the extremely reduced noise level.

PCIM Europe, Hall 12 / 629 and

Sensor+Test, Hall 12 / 246.

www.isabellenhuette.de

Isolated Current Measurement Modules


N E W P R O D U C T S

Microsemi Corporation extended its family of DC-to-DC controllers

and switching regulators with the introduction of a synchronous buck

switching regulator in a compact multi-chip module with built-in power

functionality. The NX9415(TM) is designed for step-down DC-to-DC

converter applications in LCD TVs, set-top boxes, DSL modems, net-

books, telecom, networking and other point-of-load DC-to-DC appli-

cations.

Microsemi's use of advanced multi-chip module (MCM) packaging

enables the NX9415 to convert 8- to 22-volt input voltages down to

as low as 0.8 volts with an output current of up to 5A amps, while

occupying a small, 4mm x 4mm footprint and maintaining the

device's high-frequency capability. Alternative, monolithic silicon solu-

tions require a die size with a significantly larger footprint in order to

achieve the same high frequency and low drain to source on-resist-

ance (RDS(on)) as the NX9415.

PCIM Booth 12 / 422

www.microsemi.com

Smallest High-Efficiency Switch-

ing Regulator for Consumer

Texas Instruments introduced two highly efficient 2-A step-down

DC/DC converters for portable electronics applications. The

TPS62065 and TPS62067 3-MHz synchronous converters reduces

board space and helps extend battery life by providing up to 95 per-

cent power efficiency from a tiny 2-mm x 2-mm x 0.75-mm SON

package. For product details, see www.ti.com/tps62065-preu.

The new TPS6206x devices operate from a 2.7-V to 6-V input volt-

age, allowing them to support lithium-based batteries with extended

voltage ranges or a 3.3-V or 5-V system supply rail. A unique Power

Save Mode with excellent PWM to PFM transition allows the convert-

ers to operate at light load currents to maintain high efficiency over

the entire load range.

.

www.ti.com

3-MHz, 2-A DC/DC

Converters for Portables

PCIM Booth12/602

PCIM Booth12/648




International Rectifier has introduced the AUIRS2301S 600V IC for

automotive motor drives, micro inverter drives and general purpose

three-phase inverter applications.

The AUIRS2301S is a rugged, general purpose driver IC with inde-

pendent high- and low-side referenced output channels, offering a

gate drive supply range from 5V to 20V. The device’s output drivers

feature a high-pulse current buffer stage designed for minimum driver

cross-conduction while the floating channel can be used to drive N-

channel power MOSFETs or IGBTs in the high-side configuration

operating up to 600V. Additionally, the IC features negative voltage

spike (-Vs) immunity to protect the system against catastrophic

events during high-current switching and short circuit conditions.

The new device, which is 3.3V, 5V and 15V logic compatible with

standard CMOS or LSTTL output, also features output current capa-

bility of 120mA/250mA, under-voltage lockout and matched propaga-

tion delays for both channels, outputs in phase with inputs, and lower

di/dt gate driver for better noise immunity.

The AUIRS2301S utilizes IR’s advanced high-voltage IC process

which incorporates next-generation high-voltage level-shifting and ter-

mination technology to deliver superior electrical over-stress protec-

tion and higher field reliability.

The IC is qualified according to AEC-Q100 standards, features an

environmentally friendly, lead-free and RoHS compliant bill of materi-

als, and is part of IR’s automotive quality initiative targeting zero

defects.

PCIM Europe, Hall 12 / 202

www.irf.com

Rugged, Reliable 600V IC for Automotive Drive Applications

CUI Inc’s power line, V-Infinity, announced

the release of the VOF-80, a low cost open

frame ac-dc power supply. The VOF-80 has

a low no-load power consumption of <0.5 W.

Its combination of efficiency and competitive

pricing makes the series ideally suited for

use in ITE, industrial, and consumer elec-

tronics applications.

The VOF-80 provides 80 W of continuous

output power, universal input (85-264 Vac),

and is offered in 3.3, 5, 12, 15, 24, and 48

Vdc output voltages. The series has a 2 x 4”

industry standard footprint with efficiencies of

up to 89%. Protections for over voltage and

short-circuit conditions are included. The

units operate up to +60°C with derating.

“The VOF-80 expands our offerings of effi-

cient, green power supplies that consume

low power in no-load conditions,” stated

Kraig Kawada, CUI’s V-Infinity Product Man-

ager. The VOF-80 is available immediately

through Digi-Key with prices starting at

$30.34 per unit.

www.cui.com

80 W Power Supply is Energy Efficient

LEM will be featuring products from their recent acquisition, Danish

company Danfysik ACP A/S, at PCIM this year. The Danfysik portfolio

of very high precision current transducers further strengthens LEM’s

position as the world’s leading supplier of current and voltage trans-

ducers.

The LEM Danfysik range of transducers offers nominal current meas-

urement from 12.5 A to 25 kA, providing overall accuracy at +25°C

from 1 ppm. Thermal offset drift is extremely low, from only 0.1 to 2.5

ppm/K. Models from 12.5 A to 60 A nominal can be used for PCB

mounting, models from 60 A to 25 kA are for panel or rack mounting.

The exceptional performance is obtained using Closed Loop Fluxgate

technology, enabling high accuracy, dynamic performance and a

wide measuring range. Featuring galvanic isolation, all components

can measure the current of any waveform (including DC, AC, mixed

and complex).


www.lem.com

High Precision Current Transducers

PCIM Booth 12/649



CUI Inc announced a new addition to its high

resolution, low-cost AMT encoder line with

the AMT303 series. The encoder generates

standard U/V/W commutation signals for

vectoring current to brushless motors. Res-

olutions can be set through the AMT303’s

SPI (Serial Peripheral Interface). Commuta-

tion output can accommodate brushless

motors with 2, 4, 6, 8 or 10 pole pairs, and is

also selectable via the SPI. The AMT303

offers supplementary incremental A, B, and

Z channels for various servo positioning and

startup sequences. Additional options

include nine mounting patterns and ten bore

sizes, creating a flexible platform that is able

to mate with many industry standard motors.

The AMT303 generates position information

using CUI’s patented, capacitive code gener-

ation system coupled with a proprietary

ASIC. This technology is immune to envi-

ronmental particulates and magnetic interfer-

ence, creating a reliable, economical and

stable control and positioning solution. The

encoder consumes a maximum 10 mA at 5

Vdc, making it ideal for any application

where power consumption is a concern.

The AMT303’s SPI affords higher throughput

and simpler hardware interfacing than I²C or

SMBus. Utilizing TCL code, the on-board

PIC 16F690 MCU operates at up to 10MHz

providing for high speed applications. Zero

position may be set by SPI command or

ground trigger, removing the need for time

consuming mechanical alignment in the

mounting process. Additionally, an onboard

EEPROM can store up to 128 bytes of cus-

tomer data. A demo board is available for

stand-alone demonstration, PC access to

SPI interface and example TCL code.

“We believe the AMT303 series represents a

breakthrough in commutation encoders,”

stated James Seiler, CUI’s Encoder Product

Manager. “Users will really be able to lever-

age the benefits of our ASIC-based system.

When compared to optical encoders typically

used in commutation applications, the

AMT303 will bring a level of ruggedness and

ease of use unrivaled in the industry.”

The AMT303 and AMT303 demo kit will be

available through Digi-Key in Q2 of 2010

with prices starting at $46 per unit. Please

contact CUI for OEM pricing.

www.cui.com

Commutation Encoder Offers Flexibility

Microsemi Corporation and Tyco Electronics

announced the delivery of the industry's

most advanced RJ-45 connector with built-in

power over Ethernet (PoE) technology.

Until now, Enhanced capabilities - the ability

to deliver power together with data and voice

over a standard Ethernet cable with power

management - were typically provided as an

ASIC, an integrated module or as a midspan

device. This latest innovation enables

Enhanced PoE functionality to be embedded

within the RJ-45 connector, reducing the

cost, footprint and design-cycle time associ-

ated with integrating PoE into Ethernet

switches.

Tyco Electronics' RJ-45 connector,

enhanced by Microsemi's advanced

PD69012 device, is expected to reduce the

number of components and assembly cost

for manufacturers who are designing next-

generation Ethernet switches with IEEE

802.3af-compliant or IEEE802.3at-compliant

PoE functionality. It is the first RJ-45 connec-

tor in the market capable of delivering

IEEE802.3at power with built-in high end

power management capabilities including

dynamic power management, emergency

power management, backplane power man-

agement and resilient power management.

An integrated PoE RJ-45 connector will be

critical for vendors who are designing com-

plex Gigabit Ethernet switches that have

very little room on the printed circuit board

for additional electronic circuitry. The RJ-45

connector comes in two pin-out configura-

tions, one fully compatible with the PoETec

2.0 industry standard, and another with the

additional power management capabilities.


www.tycoelectronics.com

www.microsemi.com

LAN Connector with Power over Ethernet Plus

UltraVolt, Inc. announced today that the V

Series and M Series of microsize, micropow-

er products have been enhanced. Power lev-

els within the series have increased up to

50%.

V Series and M Series modules are now

offered at 500mW with 12V input, 800mW

with 15V input, or 1W with 24V input for all

output voltage ranges. Output voltages for

these product lines range from 0 to 600V

through 0 to 1.5kV with output current from

330μA to 1.66mA. V Series and M Series

modules offer programmable regulated out-

put, high accuracy, and low ripple (0.01%

peak to peak). The volume for the V Series

is just 0.84in3 [13.8cc], and the volume for

the M Series is 1.28in3 [20.9cc].

“UltraVolt’s continuous product-improvement

efforts have yielded these valuable specifica-

tion enhancements, which are important to

our customer base,” said Scott Wilson,

Director of Business Development. “The

increase in output power offers our cus-

tomers expanded opportunities in additional

applications.”

The V Series and M Series are optimal for

handheld devices and lightweight systems.

Typical applications include avalanche photo

diodes (APD), photomultiplier tubes (PMT),

and micro-channel plates.


www.ultravolt.com

Increased Power Levels up to 50 Percent


Strip Wound Cores

Powder Cores

Ferrite Cores

PCIM Booth 12/251




ABB France 37

ABB semi C3

Bicron 25

Biricha 19+31+38+61

Centraldruck C3/2

Cierre 41

CPS 57

Cree 7

CT Concept Technologie 9+29

CUI Europe 25

Danfoss Silicon Power 35

Darnell Asia 45

Electronica C3/1

EPE 39

Ferraz 21

Fischer 25

Fuji Electric C2

GVA 3

Hitachi 33

Infineon 17

InPower 67

International Rectifier C4

Intersolar 63

Isabellenhütte 11

IXYS 1

Lem 5

Magnetics 71

Microsemi 61

Payton 37

PCIM China 69

pem uk 67

Powersem 23

Richardson 71

Rogers 15

Semikron 13

Sensor Test 55

SMT 51

Toshiba 47

Transic 53

Tyco Electronics 19

VMI 49

Würth Electronic 43

ADVERTISING INDEX

TDK-EPC, a Group Company of TDK Corpo-

ration, presents the new PCC™ (Power

Capacitor Chip) from EPCOS as an ideal

DC-link circuit solution for the electric drives

of motor vehicles. The B25655J4307K*1 and

B25655J4507K*5 types were developed

specifically for reference designs of the IGBT

modules HybridPACK™1 (up to 20 kW) and

Hybrid-PACK™2 (up to 90 kW) from Infineon

Technologies. The PCCs are also contained

in the new evaluation kits from Infineon.

Depending on the power output, the inverter

modules are suited for mild hybrid drives or

for full electric drives. These modules are the

only solutions currently in volume production.

The rated voltage of these PCCs is 450 V

DC, their capacitances are 300 and 500 μF.

They also feature an especially low ESL of

no more than 15 and 25 nH, respectively.

The ESR is a maximum of 1 mΩ for all

types. The capacitors are designed for a

temperature range of

-40 °C to +110 °C and can also be briefly

operated at 125 °C. Their average operating

life is 15 000 hours. They are self-healing,

meaning that breakdowns of the film at over-

load do not lead to short circuit or destruc-

tion of the capacitor. Despite their high per-

formance, these DC-link circuit capacitors

have dimensions of only 140 x 72 x 50 mm³

and 237 x 72 x 50 mm³, respectively.

PCIM Booth 12/535

www.epcos.com/pec

Compact DC-link Circuit PCCs for Electromobility

Mitsubishi Electric is introducing an Evalua-

tion Board for its Intelligent Power Modules

(IPMs) of the entire L1/S1 Series. The new

board named EVBL1S1XX enable the

design engineer to functionally test the fea-

tures and the performance of these IPMs

from Mitsubishi Electric. The board can be

used as a reference design regarding layout

and selected components.

The Board

Basically, the EVBL1S1XX’s circuit is based

on the interface and supply circuit recom-

mendations specified in the L1/S1 Series

IPM datasheets.

Two different IPM connectors are available

onboard to match both L1 and S1 devices.

In terms of control signals high-active or low-

active signals may be applied through the

standard 2.54mm pin connectors. Only one

single 24V supply is required to power the

EVBL1S1XX including control and driver part

of the IPM. A safety insulation barrier

between the control input of the evaluation

board and the IPM is achieved by DC-DC

converters in conjunction with photo cou-

plers.

The IPMs

Mitsubishi Electric’s IPMs of the L1-Series

are equipped with low-loss IGBT chips

based on full gate CSTBT™ (Carrier-Stored

Trench Gate Bipolar Transistor) technology.

Protection functions against short circuit,

over-temperature and supply under voltage

are included. The L1-Series IPMs are

mechanically compatible to the existing L-

Series IPMs, however, they provide a better

trade off between on-state and switching

losses at higher switching frequencies, bet-

ter IGBT temperature monitoring and an

improved power cycling capability.

The L1-Series IPMs are available for voltage

ratings of 600V/1200V in three different

packages for rated currents between 25A

and 300A.

PCIM Booth 12 / 421

www.mitsubishichips.com

Evaluation Board for IPMs

Mouser Electronics, Inc announced it is

stocking solid state thin film thermoelectric

device evaluation kits from Micropelt, an

innovator in Peltier cooler and thermogener-

ator devices.

Mouser’s stock of Micropelt products

includes the TE-Power NODE Evaluation Kit for thermal energy har-

vesting. Highly modular, the kit shows how free excess heat can pro-

vide a continuous power source for low-power wireless sensing appli-

cations. The kit provides customers with an easy-to-handle plug and

play wireless sensor system which allows the user to explore and

understand thermal harvesting and comes complete with the thermo-

electric generator TE-Power Base, various power and power mod-

ules, a low-power wireless sensor module and Texas Instrument’s

USB Wireless Receiver. This simple-to-use evaluation kit lets you

demonstrate thermoelectric generation within minutes of unpacking it.

www.mouser.com/micropelt

Energy Harvesting Thermogenerator Evaluation Kits

Central-Druckprinting with all the bits and pieces

Central-Druck is a committed service-provider for highly demanding customers. Our customerappreciates the close working cooperation and perfect results.

ZKZ 64717

05-10ISSN: 1863-5598

Electronics in Motion and Conversion May 2010

www.centraldruck.de

Central-Druck Trost GmbH & Co. KGIndustriestr. 2, 63150 Heusenstamm, GermanyPhone +49 6104 606-205, Fax +49 6104 606-400Email [email protected]

Brochures

Books

Catalogs

Technical magazines

Flyers

Business reports

Business equipment

Calendar

Mailings

Staff magazines

Newsletter

Placards and posters

Presentation folders

Prospectus

Wall scheduler

cd a4_bodospowersystem_engl510:Briefbogen_2006 20.04.2010 13:56 Uhr Seite 1

electronica 2010components | systems | applications

24th International Trade Fair

New Munich Trade Fair Centre

09–12 November 2010

www.eelectronica.de/en

get the whole picture

focus on the future.

change

100210 e2010_BodoPowSys_210x297_E.indd 1 13.04.10 11:09

Lean on me

4 – 6 May 2010PCIM NurembergHall 12, Stand 408

ABB Switzerland Ltd SemiconductorsTel: +41 58 586 1419www.abb.com/semiconductors

Reliablewith HiPak modulesfrom ABB

Power and productivityfor a better world™

Part Number Package Input VoltageMaximum

Current

Maximum

Frequency

Additional

Features

IR3843WM 5x6 PQFN 1.5V to 16V 2A 1.5MHz OV detection, Sequencing




IR3832WM 5x6 PQFN 1.5V to 16V 4A 1.5MHz OV detection,DDR tracking

IR3831WM 5x6 PQFN 1.5V to 16V 8A 1.5MHz OV detection,DDR tracking

FEATURES:

Wide input voltage range (1.5V to 16V

with 5V bias)

Small footprint (5x6mm) and low height

(0.9mm)

Programmable frequency up to 1.5MHz

1% accurate 0.7V reference voltage

Programmable hiccup current limit and

soft start

Enhanced pre-bias start up

Thermal protection

Enable pin with voltage monitoring

capability

Power Good output for over-voltage and

under-voltage detection

Optimized solutions for sequencing

(IR3840/1/2/3W) and DDR memory

tracking (IR3831/32W)

-40˚C to 125˚C operating junction

temperature (Tj)

Pin compatible with Gen2 SupIRBuck

products

For more information call +49 (0) 6102 884 311 or visit us at www.irf.com

IR’s Gen2.1 SupIRBuck™ devices save time, space and energy for your POL design

Exceed 96% Peak Efficiency With Gen2.1

SupIRBuck™ Integrated Voltage Regulators

SupIRBuck™ is a trademark of International Rectifi er

828384858687

Eff

icie

nc

y (%

)

8889909192939495

9697

21 3

1.8V 3.3V 5.0V

4 5 6 7

Load Current (A)

IR3840W Efficiency vs. Load Current at 600kHz fs, 12Vin

8 9 10 11 12

IR384XW/ 3XW SupIRBuck™

FAMILY IMPROVEMENTS

Added Over Voltage Detection

(PGOOD

window comparator)

Shorter dead-time to reduce power loss

DS(on) maximum

value for better over-current limit accuracy

REDAELTNEMEGANAMREWOPEHT

Documents

Electronics in Motion and Conversion May 2010 · Lighting High-Efficiency Converters with PFC for LED Street Lighting at 48 V and 130 W; By Claudio Spini, STMicroelectronics, Davide