Electronics in Motion and Conversion November · PDF fileElectronics in Motion and Conversion...

ZKZ 6471711-10

ISSN: 1863-5598

Electronics in Motion and Conversion November 2010

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

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

ILLUMINATINGYOUR 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

www.bodospower.com November 2010

Viewpoint

It´s Show Time! . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4

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

News . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6-10

Blue Product of the Month

Looking to the Stars and Beyond . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12

Guest Editorial

Re-Defining Power ManagementBy Mansour Izadinia, Chief Technology Officer, IDT . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14Electronica Hall A4, Booth 169

Market

Electronics Industry DigestBy Aubrey Dunford, Europartners . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 16

Market

Microgrids Redefine power DeliveryBy Linnea Brush, Senior Analyst, Darnell . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 18-19

Cover Story

Bringing GaN on Si Based Power Devices to Market By Michael A. Briere, ACOO Enterprises LLC, under contract by International Rectifier . . . 20-24Electronica Hall A5, Booth 320

IGBTs

650V IGBT4 the Optimized Device for Reduced EMI and Low ÄV By Wilhelm Rusche, Dr. Andreas Härtl, Marco Bässler, Infineon Technologies AG . . . . . . . 26-29 Electronica Hall A5, Booth 506

High Power Switch

A 10kV HPT IGCT with Improved Switching CapabilityBy Tobias Wikström, ABB Switzerland Ltd, Semiconductors and Iulian Nistor, ABB Switzerland Ltd, Corporate Research . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 30-32

Technology

Review of the ECPE Workshop on Advanced Multilevel Converter SystemsBy Prof. Thierry Meynard (University of Toulouse, ENSEEIHT – LAPLACE), Technical Chairman of the ECPE Workshop . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 34-35

Power Management

Using Microprocessor Supervisory DevicesBy Eric Schlaepfer, Senior Member of the Technical Staff, Applications Maxim Integrated Products Inc., Sunnyvale, CA . . . . . . . . . . . . . . . . . . . . . . . . 36-38Electronica Hall A5, Booth 324

Automotive

Automotive Using Ethernet as Physical Layer Data BusBy Mike Jones, Senior FAE, Micrel . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 40-44Electronica Hall A4, Booth 125

Protection

Enhanced Over-Voltage Protection of Solar InstallationsBy David Connett, Director IC Reference Design, EPCOS AG . . . . . . . . . . . . . . . . . . . . . . .46-48Electronica Hall B5, Booth 506

Technology

Driving eGaNTM FETs By Johan Strydom PhD, Director of Application Engineering, Efficient Power Conversion Corporation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 50-52

Power Modules

17 mm technology: Rectifiers, IGBTs and drivers for motor controlBy Wolf-Dieter Roth, HY-LINE . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 54-55Electronica Hall A6, Booth 606

Smart Power

Dual High Side Switches in Smart Power Technology By Giuseppe Di Stefano and Michelangelo Marchese STmicroelectronics . . . . . . . . . . . . . . 56-58

Power Supply

Low Profile AC/DC Power SuppliesBy Alexander Goncharov, P. h.D., Konstantin Stepnev and Oleg Negreba, AEPS Group . . 60-62Electronica Hall B2, Booth 450

New Products . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 64-72

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

The Gallery

Does your digital power-supply design require high performance flexible on-chip peripherals?...Control complex Digital Power applications and save power

Mem

or yAnalog

DigitalSignal

ControllersMicrocontrollers

dsPIC33F ‘GS’ Series Plug-In Module – MA330024

GET STARTED IN 3 EASY STEPS

1. Purchase a new ‘GS’ Series Plug-In Module2. Download Digital Power Reference

Design3. Order samples and start designing! www.microchip.com/SMPS

Microchip’s new dsPIC33F ‘GS’ Series DSCs provide on-chip peripherals includinghigh-speed Pulse-Width-Modulators (PWMs), ADCs and analogue comparators,specifically designed for high performance, digital power supplies.

The powerful dsPIC33F ‘GS’ series is specifically aimed at power control applications and can beconfigured for a variety of topologies, giving power-supply designers the complete freedom tooptimise for specific product applications. Multiple independent power control channels enablean unprecedented number of completely independent digital control loops. The dsPIC33F ‘GS’series offers the optimal digital power solution supported by royalty free reference designs andadvanced power design tools.

Typical applications of the new ‘GS’ series DSC include: Lighting (HID, LED, fluorescent),uninterruptable power supplies, intelligent battery chargers, AC-DC and DC-DC powerconverters, solar and pure sine-wave inverters, induction cooking, and power factor correction.

Digital control loops with 12 to 18 high-speed, 1 ns resolution PWMsUp to 24 channels 10-bit on-chip ADCs2 to 4 Million samples per second (MSPS) ADC for low latency and

high-resolution controlPin range – 64 to 100Memory – 32 to 64KB Flash memory

The Microchip name and logo are registered trademarks of Microchip Technology Incorporated in the USA and other countries. All other trademarks are the property of their respective owners.

© 2010 Microchip Technology Inc. All rights reserved. ME252B-Eng07.10

www.microchip.com/smps

Intelligent Electronics start with Microchip

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

For electronics companies November in Ger-

many is dominated by two major shows.

First we’ll have the party of all parties in

Munich in the second week of November at

the electronica - an electronica year with a

strong upturn in business. The world’s most

important electronics trade show will provide

us with a mid-term indicator of future devel-

opments in general, and a short-term fore-

cast of what the next year will look like busi-

ness-wise. Later in November the

SPS/IPC/DRIVES will showcase the industri-

al side of electronic power design in drives

and control applications. This show has

been growing consistently since popping up

on the radar in the 90s. A great many Euro-

pean and German companies who operate

globally will again exhibit.

It’s becoming increasingly clear that the

world has become a single marketplace.

Industrial customers have their contacts

worldwide and thus, supporting companies

have to be positioned globally. We’re contin-

ually seeing more and more companies

restructure themselves to serve these

demands. One example is Mersen, formerly

Ferraz, who recently consolidated all of their

resources under the new name.

The Russian market is now also focussing

on power electronics with the Power Elec-

tronics, Energy and Energy Savings show

taking place in Moscow at the end of

November. Both Russian and international

companies will support the show strongly. It’s

great to see the world getting together to talk

about technology that will help secure the

future for upcoming generations.

Whereever it is that we may live and work,

we must put the best solutions in place for

energy efficient design. Every country and

every government must develop laws to

make use of most efficient designs and we

the people have to be careful with our con-

sumption to preserve resources for those

who come after us – like the school children

in Los Angeles who were thrilled when

schools were closed due to the heat in the

last week of September and the fact that the

officials were worried that the power supply

might fail.

Friends told me that during that week Los

Angeles registered the highest temperatures

ever measured. At the time I was visiting

Maxim up north in Sunnyvale and not only

was the temperature moderate but I also

learned a lot about efficient design in every

direction of new electronics.

Silicon Valley is a very special place. The

fascinating fact is that you constantly see

new companies starting up and pushing new

and advanced design ideas. Here you feel

the pioneering freedom of progress in semi-

conductors. Beside that you find the best

steaks on Earth at the Black Angus in Sun-

nyvale!

Including this November issue - delivered, as

always, on time – we will have produced a

total of 674 pages this year: strong perform-

ance thanks to strong support.

My Green Power Tip for November:

Clean your solar panels. The more dirt on

the surface, the less efficient the panels are.

Maintenance is the key to any technical

solution running properly and efficiently.

See you in Munich or Nuremberg

Best regards

It´s Show Time!

V I E W P O I N T

4

A MediaKatzbek 17aD-24235 Laboe, GermanyPhone: +49 4343 42 17 90Fax: +49 4343 42 17 [email protected]

Publishing EditorBodo Arlt, [email protected]

Creative Direction & ProductionRepro Studio [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 wideSingle issue is 18 €[email protected]

circulation

printrun

25000

Printing by: Central-Druck Trost GmbH & CoHeusenstamm, Germany

A Media and Bodos Power Systemsassume and hereby disclaim any liability to any person for any loss ordamage by errors or omissions in thematerial contained herein regardless ofwhether such errors result from negligence accident or any other causewhatsoever.

Events

Electronica

Munich Ger. Nov 9-12 http://www.electronica.de/en

SPS/IPC/DRIVES

Nuremberg Ger. Nov 23-25http://www.mesago.de/en/SPS/main.htm

Power electronics

Moscow Nov.30-Dec.2 http://www.powerelectronics.ru

Embedded World

Nuremberg, Ger. March 1st-3rdhttp://www.embedded-world.eu/

APEC 2011 Ft. Worth,

TX, USA March 6th -10th http://www.apec-conf.org/

EMC2011,

Stuttgart/Ger. March.15th – 17thhttp://www.mesago.de/de/EMV/home.htm

New Energy

Husum Ger. March17th-20thhttp://www.new-energy.de

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

Future precision. Future performance.Now available.

SPS/IPC/

Drives

Hall 1.525

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

N E W S

Integrated Device Technology, Inc.,

the Analog and Digital Company

delivering essential mixed-signal

semiconductor solutions, announced

that it will have a major presence at

Electronica 2010, which will take

place between November 9-12 in

Munich, Germany. Visitors to the

IDT booth will see how the company

has rapidly evolved to become a

leading supplier of analog and digi-

tal solutions for a wide range of

leading-edge technologies and prod-

ucts in sectors that include industri-

al, consumer, entertainment and medical electronics, along with wired

and wireless telecommunications.

IDT will be demonstrating several products from its portfolio of mixed-

signal solutions, including high-performance signal conditioning

repeaters for multi-gigabit, serial-differential protocols, and industry-

leading solutions for Serial RapidIO®

and PCI Express® protocols.

In addition, IDT will feature products

from its video, enterprise computing,

capacitive touch and high perform-

ance timing product lines. One of the

key demonstrations will be the indus-

try’s first Enterprise Non-Volatile

Memory Host Controller Interface

(NVMHCI) Flash Controller. The goal

of the Enterprise NVMHCI standard

is to drive the adoption of PCI

Express-based Solid State Drives

(SSDs), which will offer reduced

power consumption and a significant improvement in storage per-

formance when compared to SAS/SATA-based SSDs.

Electronica Hall A4, Booth 169

www.idt.com

IDT to Showcase at Electronica 2010

Maxim Integrated Products announced its

inaugural visit to electronica 2010. Focusing

on the medical, industrial, and automotive

markets, Maxim will present demos covering

blood glucose meters, automotive infotain-

ment, smart-grid solutions, HBLEDs, wire-

less HD video, and other related areas.

"Maxim has many new and exciting tech-

nologies for the medical, industrial, and auto-

motive markets. We are extremely happy to

make our first visit to electronica and share

these developments with everyone," said

Walter Sangalli, Managing Director, Euro-

pean Sales and Applications. "As the largest

electronics industry trade show in the world,

electronica 2010 is a perfect fit for us to

showcase our innovative products and solu-

tions," Sangalli added.


www.maxim-ic.com

Maxim to Highlight Products at electronica 2010

International Rectifier announced it will

showcase the company’s industry-leading

power management solutions at electronica

2010.

IR’s innovative energy saving technologies

and products will be on display in Hall A5,

Booth 320 including demonstrations of the

company’s GaN-based power device plat-

form, GaNpowIR™. IR’s SupIRBuck™ inte-

grated voltage regulators, and benchmark

MOSFETs and DirectFET® MOSFETs,

IGBTs and high-voltage ICs for a diverse

range of applications including appliances,

automotive, lighting, computing and Class D

audio will also be on display as well as IR’s

DC-DC converters and modules for high reli-

ability applications.


www.irf.com

Showcase Industry-Leading Power Management Solutions

Fairchild Semiconductor will demo its latest

technological advancements for mobile

applications, as well as power solutions that

focus on the smart grid at electronica 2010.

Fairchild will feature technology demonstra-

tions that enable mobile connectivity and

optimize energy usage in power supplies

(AC/DC and DC/DC), mobile, LED lighting,

motor, solar, computing, consumer and auto-

motive applications.

Fairchild solves difficult power management

and signal path problems for leading-edge,

top tier customers around the world. Our

commitment to energy savings and meeting

the most stringent regulations has lead to

the development of innovative power and

mobile solutions that maximize performance

while reducing board space, design com-

plexity and system costs.

Please join us at electronica 2010 to see the

extensive portfolio of power and mobile

products that enable the right technology for

your success.


www.fairchildsemi.com

Focus on Smart Grid Technology at electronica 2010

The search for the most productive turbine

has once again led ENERTRAG, one of the

biggest independent renewable energy pro-

ducers in Europe, to Vestas.

ENERTRAG decided to order seven units of

the V112-3.0 MW – Vestas’ latest and most

technologically-advanced wind turbine – to

most effectively optimise an existing wind

farm.

The seven turbines will be installed at the

Uckermark wind farm, next to ENERTRAG’s

head office in Dauerthal. For this inland site

ENERTRAG needed a productive turbine

with an ideal tower height to ensure maxi-

mum energy production from Uckermark’s

wind conditions and which presents a very

attractive business case through high full

load hours. The V112-3.0 MW can generate

even more electricity than other turbines in

the 3 MW class and delivers industry-leading

reliability, serviceability and availability.

www.vestas.com

ENERTRAG Orders Seven Vestas V112-3.0 MW Turbine

Please read the article on page 54


N E W S

From November 30th to December 2nd 2010

the show will take place in Pavilion 2, Hall 6,

Crocus Expo, Moscow, Russia. Every year,

the event brings together key distributors,

suppliers and producers of new develop-

ments in power electronics, energy and

energy saving, and shows the potential of

the Russian power electronics market.

To date, the exhibition has grown by 23%

compared with last year’s results, and there

are still two and a half months left before the

exhibition, which means the overall growth

will be even greater.

The 2nd International Conference 'Power

Electronics – a Key Technology for Russian

Industry in the 21st Century' will be the main

event of the exhibition. The focus of the con-

ference is to provide a platform for discus-

sion of the most important aspects of power

electronics development at a professional

level, with the participation of key represen-

tatives from science, government agencies,

the business community and public organi-

sations.

http://power.primexpo.com/

Power Electronics, Energy and Energy Saving in Moscow

At Electronica 2010 Toshiba Electronics Europe (TEE) will be show-

casing a selection of advanced semiconductor solutions for automo-

tive applications and presenting a paper on new microcontroller tech-

nologies for automotive safety systems.

Visitors to the Toshiba stand in the Hall 6 automotive sector will have

the opportunity to see the latest ‘Capricorn’ system-on-chip (SoC)

solutions for driving and managing high-quality automotive displays;

new ARM Cortex™-M3 microcontroller technologies that address

ISO26262 ASIL ( Automotive Safety Integrity Level) requirements for

safety-critical systems; and automotive-qualified ASSP (application

specific standard products) driver ICs that simplify the implementation

of brushless DC (BLDC) motor control. The company will also be

unveiling a new BiCDMOS semiconductor process platform for next-

generation automotive ICs and showcasing automotive LEDs and

power MOSFETs.

Electronica Hall A6, Booth A21

www.toshiba-components.com

Advanced Semiconductor Solutions for Automotive

Power Your Recognition InstantlyBased in Munich, Germany, ITPR Information-Travels Public Relations is a full-service consultancy

with over a decade of experience in the electronics sector.

As a small exclusive agency, we offer extremely high ROI,

no-nonsense flexibility and highest priority to only a handful of companies.

Strategical SupportCorporate/Product Positioning, Market/Competitive Analysis, PR Programs, Roadmaps,

Media Training, Business Development, Partnerships, Channel Marketing, Online Marketing

Tactical PRWriting: Press Releases, Feature Articles, Commentaries, Case Studies, White Papers

Organizing: Media Briefings, Road Shows, Product Placements in Reviews and Market Overviews,

Exhibitions, Press Conferences

Monitoring and Research: Speaking Opportunities, Editorial Calendars, Feature Placement,

Media Coverage, Competitive Analysis

Translations: Releases, By-Lined Articles, Websites, etc.

Call or contact us today for a free consultation on how PR

can dramatically affect your company’s bottom line.

ITPR Information-Travels Public Relations

Stefanusstrasse 6a, 82166 Gräfelfing-Munich, Germany

Tel ++49 (89) 898687-20, Fax ++49 (89) 898687-21,

[email protected]

w w w . i n f o r m a t i o n - t r a v e l s . c o m

www.bodospower.com Novewww.bodospower.com

Demonstrating its commitment to provide the most comprehensive

products and tools for power line communications (PLC) develop-

ment, Texas Instruments announced the PLC Development Kit (TMD-

SPLCKIT-V2) based on the industry’s only PLC modem solution

capable of supporting multiple modulation and protocol standards on

a single hardware platform. The kit provides everything developers

need to network systems and implement monitoring capabilities and

other new services that reduce device maintenance cost while

increasing system reliability to create greener, more efficient prod-

ucts. Developers will now be able to quickly evaluate the suitability

of using PLC-based communications and then jumpstart develop-

ment for Smart Grid applications ranging from smart electrical meters

to intelligently controlled industrial applications, including lighting,

solar, home automation, building control, plug-in electrical vehicle

and energy-managed appliances.

Electronica Hall A4, Stand 420

www.ti.com/plc-pr

Entry for Power Line Com-

munications with Develop-

ment Kit

Despite extreme shifts in pricing, demand and governmental subsi-

dies, the global photovoltaic market in 2011 will experience robust

growth, with installations rising by 42.3 percent for the year, accord-

ing to the market research firm iSuppli Corp.

iSuppli forecasts that worldwide solar installations will reach 20.2

Gigawatts (GW) next year, up from 14.2GW at the end of 2010. Ger-

many, the world’s leading Photovoltaic (PV) market, will continue to

play a key role and account for half of the total installations, at

9.5GW.

While an impressive growth total for the year, the expansion will be

down significantly from the 97.9 percent increase in 2009.

The attached figure shows iSuppli’s forecast of global PV installations

by region from 2009 to 2014.

In the near term, the nuclear reprieve in Germany will have no effect

on the PV markets, even if passage might have sent the wrong signal

to PV global markets for the time being, iSuppli maintains. And with

German polls suggesting overwhelming support—80 percent by one

count—among voters in favor of renewable energy generation, the

forecasts for a strong German PV market in 2011 continue to hold

and remain unchanged.

To learn more about this topic, see iSuppli’s new report, entitled:

Global PV Market to Double in 2010, Germany Leads the Way.

www.isuppli.com

Solar Market Keeps Shining

in 2011

0

5.000

10.000

15.000

20.000

25.000

30.000

35.000

40.000

2009 2010 2011 2012 2013 2014

Me

ga

wa

tts

Rest of the Wor

Asia*

North America*

Europe*

Germany

�� !�� "�� #�� $��%&�"�'�� ()�*�� #�� +�� ,+��

�� +�� -�.�� /�� 0�� &)��

��

� ��

��

� �� !� �"#��$��"#�#�%


N E W S

To counteract the growing problem of counterfeit and substandard

semiconductors entering the global marketplace, the Semiconductor

Industry Association

(SIA) and Rochester

Electronics have

joined to develop a

comprehensive,

worldwide directory

of companies that

are authorized by original semiconductor manufacturers to distribute

their products. Only authorized distributors can guarantee traceability

and authenticity of components; buying authorized eliminates the

potential risk of purchasing counterfeit or substandard parts. The

Authorized Directory provides two quick and easy worldwide search

tool options to help buyers find authorized distributors: search by

semiconductor manufacturer, or search by part number.

http://www.authorizeddirectory.com

Safe Connection to Authorized Suppliers

NEC Corporation announced it is working

with the Electric Power Research Institute,

Inc. (EPRI) to conduct joint field trials of an

electricity storage system using NEC’s lithi-

um-ion battery.

In the first phase of the testing, a 25 kW sys-

tem provided to EPRI by NEC will be tested

as an initial step toward future smart grid

applications. Follow-up electric utility demon-

strations of larger 1 MW systems could be

possible as part of an EPRI – U.S. electric

utility industry research collaborative.

Initial testing will be carried out at EPRI’s

Knoxville, Tenn., laboratory using NEC’s

integrated lithium-ion battery system to

examine operational performance and func-

tionality for U.S. grid compliance. NEC will

provide a fully integrated lithium-ion storage

system which EPRI will evaluate. It will

include power electronics and NEC’s IT net-

work technology, necessary for power con-

trol and energy management.

EPRI is engaged in the testing and evalua-

tion of a variety of electric utility scale energy

storage systems in support of the electric

industry’s transition towards smarter electric

grids. In the United States, energy storage

systems are expected to be key assets for a

wide range of applications, including integra-

tion support for wind and solar power gener-

ators, distribution grid asset and operational

management as well as energy manage-

ment for commercial buildings and resi-

dences.

www.nec.com

Test Lithium-Ion Storage Systems for Smart Grid

Rogers has showcased its high perform-

ance, cost-effective laminate materials for

antenna applications at the Antenna Sys-

tems 2010 conference and expo (October

19-20, 2010, Gaylord Texan Resort & Con-

vention Center, Dallas, TX).

Representatives from Rogers Advanced Cir-

cuit Materials (ACM) Division had been pres-

ent to explain the optimal use of Rogers

RO3730™, RO4730™, and RO4500™ lami-

nate materials for a wide range of printed-

circuit antennas, including in third-generation

(3G) and 4G wireless base stations and in

broadband WiMAX systems.

Antenna Systems 2010 (www.antennason-

line.com) is a leading international confer-

ence and exhibition devoted to antenna

designers, manufacturers, and system inte-

grators focusing on the latest advances in

antenna systems and technology.

www.rogerscorp.com

Highlighted Materials at Antenna Systems 2010

Nextreme Thermal Solutions announced that

it has been awarded a United States patent

for the design of an innovative solar thermo-

electric generator (solar TEG) for high-tem-

perature solar thermoelectric energy conver-

sion.

Patent #7,638,705 - Thermoelectric Genera-

tors for Solar Conversion and Related Sys-

tems and Methods describes a method of

using thermoelectric generators in combina-

tion with thermally conductive plates to gen-

erate power in response to solar radiation.

Thermoelectric devices generate electricity

via the Seebeck Effect, where voltage is pro-

duced from a temperature differential applied

across the device.

High-temperature solar thermal systems that

incorporate solar concentrators can operate

between 600° and 700°C. At those tempera-

tures, a multi-stage cascade thermoelectric

power generator, as depicted in the above

illustration, may provide a design efficiency

of well over 15%. Design efficiencies in this

range permit flexibility and adaptability to

new and cost-effective real-world solar ther-

mal systems.

The invention was developed in collabora-

tion with Dr. Rama Venkatasubramanian,

director of the Center for Solid State Ener-

getics at RTI International in Research Trian-

gle Park, North Carolina.

Nextreme is seeking commercial partners to

further develop the technology for large

scale solar thermal energy harvesting solu-

tions.

www.nextreme.com

High-Temperature Solar Thermoelectric Power Generator

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

==


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

Looking to the Stars and Beyond

TDK-Lambda configurable power supplies drive Back-end Electronics

in Giant Radio Astronomy telescope in Chile. Earlier this year

astronomers and engineers successfully positioned and linked three

antennas used in a pioneering telescope based at high altitude in the

Atacama region of northern Chile. When fully-commissioned with 66

antennas, the giant telescope will be used to observe the universe

with pin-point accuracy and help astronomers answer important

questions about our cosmic origins. Inside these antennas are two

electronic equipment racks, each powered by one of TDK-Lambda’s

Vega configurable AC-DC power supplies.

The ALMA (Atacama Large Millimetre/Sub-millimetre Array) antennas

use an advanced technology, called the interferometric technique,

and are the most sophisticated antennas in the world. The ALMA

operations site is based on the Chajnantor plateau in the Andes,

which is some 5100 metres above sea-level. Here the Atacama

Desert is considered as one of the driest places on Earth and the rar-

efied air is ideal for ALMA’s observations. In these conditions, howev-

er, remote control of the antenna array from a site based at lower alti-

tude some 20km away was necessary so system reliability was criti-

cal.

Hank Newton, Integration Electronics Engineer in the Back-end Elec-

tronics Group (BEND) of the ALMA project, describes his involve-

ment: “Surviving strong winds and temperatures fluctuating between -

20°C and +20°C, is quite a challenge even for the two electronic

equipment racks inside each antenna. In these racks, we process the

signal from the output of the cryogenically cooled antenna front-end

and produce a digital output on fibre optic cable.”

To power the racks, the BEND team selected Vega configurable

power supplies from TDK-Lambda, one of the world’s leading power

supply manufacturers, in preference to developing and testing their

own solution as it was more economical and specific outputs require-

ments could be met. In addition to the Vega’s well-renowned reliabili-

ty, the ‘smart’ communication capability helped facilitate remote con-

trol.

“We use the RS232 port on the Vegas to determine the health of the

system, as well as to control the voltage and current settings,”

explains Newton. “This remote control capability is another important

aspect for the team’s choice in power supply, as each antenna in the

array operates fully automatically.”

Having three antennas observing in unison, provides the missing link

to correct errors that arise when only two antennas are used, thus

paving the way for precise images of the Universe at unprecedented

resolution. Ultimately, ALMA will have at least 66 antennas – that’s

132 Vega power supplies from TDK-Lambda - which can be placed

on any of about 200 pads, spread over distances of up to 18.5km

and operating as a single, giant telescope. When fully-functional,

astronomers will study cold clouds of gas and dust, where new stars

are being born, and remote galaxies towards the edge of the observ-

able universe.

www.emea.tdk-lambda.com

© ALMA (ESO/NAOJ/NRAO

For electric, hybrid and battery vehicles

MOSFET inverter: up to 55 kVA

Vbattery: 24V - 160V

IGBT inverter: up to 55 kVA

VDC :150V - 850V

IP67 enclosure

SKAITM

Most compact inverter systems: 20 kVA/l

Australia +61 3-85 61 56 00 Brasil +55 11-41 86 95 00 Cesko +420 37 80 51 400 China +852 34 26 33 66 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

3-phase IGBT inverter system up to 250 kVA


It is obvious that our

society is consum-

ing more power than

ever before. With

the plethora of

portable electronics

available today and

our seemingly inces-

sant need to power

these and other

devices that we “can’t live without,” our natu-

ral resources are becoming depleted. Fur-

thermore, consumer handheld devices need

to do more while, at the same time, operate

on longer battery life. And, companies are

required to implement power savings into

their latest device designs.

At Integrated Device Technology (IDT),

power management is a critical component

in the design of all of our devices. The IDT

approach to power management is unique

as it is focused around end applications.

Rather than viewing power management as

an end market in itself, we see it as an

enabler for application-optimized, system-

level solutions. Our power management

technology touches all of our key end mar-

kets, including Wireless Infrastructure, Per-

sonal and Enterprise Computing, Video and

Displays, and Portable Consumer Electron-

ics. Our power management and system

experts work closely with our customers to

discover new ways of reducing power and

prolonging battery life in their next-genera-

tion devices.

Classic Power Management Solution

Traditionally, companies have been focused

on reducing power dissipation by optimizing

the electrical parameters on individual power

management devices. With each new

design, companies attempted to shrink the

size of each individual chip on the board, as

well as make them faster and more power

efficient — all at a lower cost. However, this

approach has amounted to incremental sav-

ings. This chip-level approach also only

addresses part of the problem. Even though

the chips use less power, there is little atten-

tion paid to the power efficiency of the entire

system. In recent years, we have seen the

deployment of serial busses to create com-

munication between the individual chips or

subsystems on a board. These inter-chip

communication standards are important in

reducing system power, however, often

require costly components.

One of the reasons that companies don’t

take a system-level approach to power man-

agement is that the company needs a diver-

sity of technologies to successfully and opti-

mally address all of the issues related to

integrating the entire system power manage-

ment onto a single chip. Most companies

specialize on only a few technologies and do

not have the breadth of knowledge and

expertise to do this.

But, IDT does. IDT used its 30-year heritage

and leadership in digital technologies, and

married it with our in-house analog talents

and capabilities to develop our integrated

power management approach.

The IDT Approach: Integrated Power Man-

agement Examples

IDT looked at power management from a

system level, and discovered that by control-

ling the power that flows through the entire

system, greater power savings and system

efficiency can be realized.

One example of the IDT integrated power

management approach is our newly intro-

duced IDT P95020. This device is an Intelli-

gent System Power Management Solution

targeted for portable consumer products,

such as Smartphones, portable navigation

devices, mobile Internet devices and

eBooks.

The IDT P95020 incorporates a best-in-class

high-fidelity audio subsystem, clock genera-

tion, resistive touch controller, backlight LED

driver, Li+/Polymer battery charger, multi-

channel DC-to-DC converters and a high

resolution analog-to-digital converter (ADC)

onto a single chip along with an embedded

microcontroller.

Using our digital and analog expertise, IDT

integrated an intelligent microcontroller onto

a chip along with all of the associated sub-

systems. This single-system solution actually

manages the flow of power around the entire

system. For example, the microcontroller

can, in real time, monitor the functions of all

the subsystems, and power up or power

down each individual subsystem depending

on what is going on in real time. If a particu-

lar system is not being used, the microcon-

troller can power it down, saving power. In

essence, the microcontroller acts as a brain

within the device. Traditionally, implementing

this kind of functionality required implanta-

tion of serial busses, such as I2C, SMBus,

SPI or PMBus, with all their associated over-

head.

The innovative IDT power management

architecture allows the microcontroller to

manage all on-chip resources and also

offload general housekeeping and I/O pro-

cessing tasks from the application processor.

This unique feature, along with programma-

ble system power regulation blocks and an

on-chip power management scheme, results

in higher system performance and longer

battery life.

Another example of our integrated power

management approach is the IDT PowerS-

mart™ technology for the notebook and net-

book displays. IDT introduced the industry’s

first single-chip power management solution

that integrates a timing controller (TCON),

power management and LED driver onto a

single chip, reducing bill of materials and

footprint in netbooks, tablet PCs and note-

books. The solution from IDT integrates a

full-function, low-voltage differential signaling

(LVDS) input and mini-LVDS output timing

controller with fully integrated power man-

agement, and a four-channel LED driver for

LED backlighting. The IDT PowerSmart solu-

tion helps display engineers save money

and results in a much faster time to market.

Conclusion

Many companies are trying to get power

management onto their chips. But, IDT is uti-

lizing power management as only one tech-

nology to bring application-optimized, sys-

tem-level solutions to market.

By utilizing our digital and analog capabili-

ties, IDT is developing products that are not

classical power management products.

Instead, IDT is developing system power

management solutions that control the flow

of power throughout the entire system.

www.idt.com

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

Re-Defining Power ManagementBy Mansour Izadinia, Chief Technology Officer, Integrated Device Technology

Figure: PND Application Diagram

Cirrus Logic power amplifiers, featuring the technology of Apex Precision Power®,

simplify the circuit design process. The PA98 swings up to ±215V dual supply with

speed as high as 1,000 V/μs in a space-saving PowerSIP packaging to drive piezo

electronics positioning elements. The PA85 mirrors the PA98 in performance, but

provides a hermetically sealed metal package for applications that demand

ruggedness.

· High voltage supply ±215 V dual supply, ±440 V single supply

· High speed rail-to-rail 1,000 V/μs

· High current output 200 mA continuous

· PowerSIP or hermetic packaging option

The superior quality and reliability of these off-the-shelf solutions can eliminate

design cycle headaches and speed your time to market.

THE PA98, PA85 HIGH VOLTAGE, HIGH SPEED POWERAMPLIFIERS DRIVE PIEZO ELECTRONICS

When youneed simple.

Off-the-Shelf.

Why complicate your design needlessly?

Choose a proven, highly reliable,

single-package solution that is

guaranteed to cut your time to market.

LEARN MORE AT:

www.cirrus.com

© 2010 Cirrus Logic, Inc. All rights reserved. Cirrus Logic, Cirrus, the Cirrus Logic logo designs, Apex Precision Power, Apex and the Apex Precision Power logo

designs are trademarks of Cirrus Logic, Inc. All other brands and product names may be trademarks or service marks of their respective owners. BPS112010

AAAAAAAAAAAAAAAPPPPPPPPPPPPPPPEEEEEEEEEEEEEEEXXXXXXXXXXXXXXX VVVVVVVVVVVVVVV11111111111115555555555555DDDDDDDDDDDDAAAAAAAAAAATTTTTTTTTTTAAAAAAAAAAAA BBBBBBBBBBBBBOOOOOOOOOOOOOOOOOOOOOOOOKKKKKKKKKKKKKrequest a copy atwww.cirrus.com/98bps


GENERAL

The MEMS market

reached $ 6.9 billion in

2009 and will be

around $ 8 billion in

2010, so Yole

Développement. A

restart of the growth is

expected after 2010,

with a CAGR of 13

percent in the next 5 years. Growth is back,

but the growth has changed: only a few

companies have 200 mm production infra-

structure in place and it provides them a

strong cost benefit, helping them to target

lower price consumer electronics applica-

tions.

SEMICONDUCTORS

The World Fab Forecast released by SEMI

at the end of August indicates a 133 percent

increase in equipment spending for front end

fabs this year and about 18 percent growth

in 2011. Worldwide installed fab capacity

(without discretes) is expected to grow by 7

percent to 14.4 million 200 mm equivalent

wafers per month in 2010, and by another 8

percent in 2011.

Maxim Integrated Products has acquired pri-

vately held Phyworks for approximately $

72.5 M in cash. Founded in 2001 and based

in Bristol, UK, Phyworks is a developer of

high-speed communications chips designed

to significantly cut the cost of 10 Gbps and

below copper and optical interconnects. Phy-

works' products for fiber-to-the-home (FTTH)

applications complement Maxim's datacom

and telecom portfolio.

Members of Sitelesc reported French semi-

conductor market revenues in Q2 2010 up

2.9 percent (on a euro basis) compared to

the previous quarter (+4.5 percent in inte-

grated circuits but –9.2 percent in discretes).

´

Showa Denko (SDK) has increased its pro-

duction capacity of blue LED chips in Japan

to 340 million units per month, from 200 mil-

lion units per month. Demand for blue LEDs

is expected to grow around 10 percent a

year in coming years due to increased use in

such applications as backlight for LCD TVs

and general lighting.

Microsemi, a manufacturer of high perform-

ance analog mixed signal integrated circuits,

high reliability semiconductors, and radio fre-

quency (RF) subsystems, has acquired all of

the assets of VT Silicon. Located in Atlanta,

VT Silicon designs and manufactures multi-

band radio frequency integrated circuit

(RFIC) solutions for the mobile wireless

broadband market.

OPTOELECTRONICS

Global shipments of small/medium TFT LCD

panels, which are advanced types of dis-

plays used in mobile devices like smart

phones and tablet PCs are set to rise by

28.1 percent in 2010 to reach 2.3 billion

units, so iSuppli.

Global smart phone shipments are set to

rise by 35.5 percent in 2010. Meanwhile,

tablet PC shipments will grow by a stunning

787.3 percent, driven almost entirely by

Apple’s iPad.

PASSIVE COMPONENTS

June revenues for Germany's PCB manufac-

turers increased by 9 percent sequentially

and 40 percent compared to June last year,

so the ZVEI. The first half ended with an

increase of 33 percent, bolstered by the

weakness of the euro against the U.S. dollar,

which promotes exports in the major seg-

ments of automotive industry and mechani-

cal engineering.

Tyco Electronics has entered into a definitive

agreement to sell its mechatronics business

located in Niefern, Germany to L. Possehl &

Co. The business designs and manufactures

customer-specific components, primarily for

the automotive industry, and is expected to

generate sales of approximately $ 100 M in

the current fiscal year.

OTHER COMPONENTS

Martek Power, a French supplier of power

supplies and power converters, announced

the acquisition of Laser Drive, a US-based

company specializing in power supplies for

various laser and light sources. As a result of

this acquisition, Martek Power will have a

greater presence in the laser and lighting

power supply market.

DISTRIBUTION

Silica, an Avnet company, pan-European

semiconductor distribution specialist, has

launched a designers’ community providing

open access to local in-house engineering

expertise. The Silica Designers’ Community

is integrated into the company’s new web

site, www.silica.com, and provides a portal to

forums, video content and technical

resources supported by the distributors own

front-line engineers. Silica is a $ 1 billion

company, serving over 15,000 customers

across Europe.

Farnell has also announced an exclusive

distribution agreement with Würth Elektronik:

EBV Elektronik, an Avnet company, and

SIMCom Wireless Solutions, announced a

distribution agreement to deliver SIMCom

M2M products to customers within the

EMEA region.

Avnet Memec has signed a new pan-Euro-

pean distribution agreement with Intersil to

deliver their entire line of analogue and

power management products in EMEA.

Avnet Embedded announced that its fran-

chise agreement with Sharp Microelectronics

Europe for display products has been

extended to cover the UK.

Link Microtek, a specialist supplier of

microwave and RF components and sys-

tems, has been appointed as UK representa-

tive for HBH Microwave, a German manufac-

turer of power amplifiers and custom speci-

fied components.

Maxim, a manufacturer of analog and mixed-

signal semiconductors, has added the Sym-

metron Group as a new franchised distribu-

tor in Russia, Ukraine, and Belarus.

Since 1993 Symmetron is one of the leading

distribution companies in Russian electronic

industry.

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


The microgrid market could reach $1.8 billion by 2015, according to

projections by the NanoMarkets Smart Grid Analysis (SGA). Micro-

grids are distributed resources (DR) island systems, according to the

Institute of Electrical and Electronics Engineers (IEEE). The IEEE

created the term “DR island systems” to generically call all intentional

island systems that could include local and/or area electric power

systems (EPSs). DR island systems, sometimes referred to as micro-

grids, are used for these intentional islands. DR island systems are

EPSs that: (1) have DR and load; (2) have the ability to disconnect

from and parallel with the area EPS; (3) include the local EPS and

may include portions of the area EPS; and (4) are intentionally

planned. DR island systems can be either local EPS islands or area

EPS islands.

Interestingly, over 40% of the market opportunity in the microgrid

space is represented by one application: institutional/campus installa-

tions. According to SGA’s projections, this application alone will gen-

erate almost $775 million in revenue by 2015. In addition, the SGA

predicts that the cost per megawatt for campus/institutional networks

will decline about 15% by 2015, making microgrids economically

viable for smaller institutions including colleges, hospitals and mili-

tary/police facilities.

Over half of the microgrid market is expected to come from North

America over the next decade. One reason for this is that some large

US universities have had primitive microgrids in place for some time,

so the concept is well-established. In fact, microgrid companies, still

finding their footing, have already turned to campuses – where

research and interested residents could help refine the concept.

Existing microgrids are serving about 322MW to institutional campus-

es, and this number is predicted to soar as high as 1.2GW by 2015 if

the right policies are implemented.

For example, EDSA Micro Corp. and Viridity Energy recently

announced a collaboration to technically support what is described as

a groundbreaking microgrid project, called RESCO, being deployed

at the University of California, San Diego. When operational, the

effort will result in what is said to be the world’s first use of real-time

software systems serving as the “Master Controller” in a live cus-

tomer installation – an achievement that the companies say industry

experts predicted would not be technologically feasible for at least

five more years.

RESCO stands for “Renewable Energy Secure Communities,” a proj-

ect funded by the California Energy Commission (CEC). The project

consists of UC San Diego demonstrating integration of on-site renew-

able energy production. UC San Diego’s campus-wide microgrid is

said to be recognized as one of the most technologically advanced in

the world. The microgrid serves a 1,200-acre, 450-building campus

with a daily population of 45,000, running two 13.5MW gas turbines,

one 3MW steam turbine and a 1.2MW solar-cell installation that

together supply 82% of the campus’s annual power.

If the technology can be proven in these locales, it might have a bet-

ter shot at residential deployment – with whole neighborhoods oper-

ating on the same microgrid. The growing demand for power quality

in North America will be more economically provided by microgrids

than by installing more generating capacity. In addition, the SGA

believes that the US will experience robust military microgrid growth

as part of the military’s Energy Surety and Net Zero Carbon Footprint

program.

But Europe is also active in microgrid development. Serious work on

microgrids in Europe actually started earlier than in the US, due to

political pressure to explore power solutions with a lower carbon foot-

print, along with EU legislation that removed the barriers to entry for

distributed resources.

Currently, 11 European countries are operating microgrid projects,

with Denmark in the lead. The best known of these microgrid demon-

strations in Denmark is the Bornholm Island microgrid. It provides

over 55MW of peak power and incorporates 30MW of wind power.

The microgrid is connected to a high-power node in Sweden and is

able to successfully island off from the overall grid when power quali-

ty is low.

At present, the technological immaturity of the microgrid concept has

resulted in a high value being placed on certain specialized micro-

grid-related products and services. Microgrids are novel concepts

with several distinct advantages: They are more suitable for the inte-

gration of renewable energy systems like rooftop solar panels, waste

heat generators and fuel cells. On a smaller scale, it is easier to track

not only how much energy is actually being produced from these

sources, but also how it is being used and distributed for more con-

sistent service. Right now, the majority of the approximately 455MW

being circulated in microgrids is still generated by traditional coal and

natural gas operations – but this will probably change rapidly.

Emerging standards will also help support the deployment of micro-

grids. IEEE P1547.4™/D10.0 Draft Guide for Design, Operation, and

Integration of Distributed Resource Island Systems, along with Elec-

tric Power Systems IEEE Std P1547.4, is part of the IEEE Std 1547

series of standards. This series was created to develop a national

consensus on using DRs in electric power systems. IEEE Std

P1547.4 was specifically developed to address the lack of informa-

tion included in IEEE Std 1547-2008 regarding intentional islands.

This document covers intentional islands in EPSs that contain distrib-

uted resources.

The SGA has identified a number of specialist microgrid firms as suc-

cessfully playing to this opportunity. These include: Balance Energy,

BPL Global, Encorp, NSEE, Pareto Energy, Valence Energy and

Viridity Energy. According to the SGA, specific offerings that the

microgrid market needs are: automation of power resources, energy

management, modeling and energy simulations, demand/response

M A R K E T

Microgrids Redefine Power Delivery

By Linnea Brush, Senior Research Analyst, Darnell Group

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

management and energy trading platforms. In other words, the

opportunities in the growing microgrid market are similar to those

found in the Smart Grid as a whole, including smart meters with

sophisticated communication capabilities to monitor energy usage

and allow residential and business consumers to make informed

choices about how much energy to use. Smart meters include a

microcontroller with onboard ADC and DAC, a sense component for

both voltage and current, an ac-dc power converter, battery back-up,

and wireless or wired communication capability.

Future building electric systems could be based on a dc-powered

microgrid system. The Center for Power Electronic Systems (CPES)

has coined the term, “dc nanogrid,” to describe this architecture,

which brings advantages such as fewer power converters, higher

overall system efficiency, and easier interface of renewable energy

sources to a dc system. The consumer electronics, electronic bal-

lasts, light-emitting diode (LED) lighting, and variable speed motor

drives can be conveniently powered by dc.

Basically, the nanogrid of the building is seen by the utility grid as a

single electronic load/source, dynamically independent of the grid but

dispatchable by the utility operator. The energy management center

(ECC) is entrusted with the operation of the local renewable genera-

tion, load shedding, utilization of the static or mobile battery, energy

and other power management functions, as well as nanogrid stabi-

lization and advanced, active islanding in the event of outages or

other low-frequency disturbances on the utility side.

This approach could then be extended hierarchically so that a num-

ber of such semi-autonomous nanogrids are combined to form a big-

ger microgrid system which, in turn, is interfaced to a minigrid

through a higher (substation) level ECC with high-power bidirectional

converter, and so on. In the proposed hierarchical grid architecture,

the nanogrids are fully dynamically decoupled from the microgrid

through the ECC, so that their internal architecture is completely

independent and can have different voltage, phase, and even fre-

quency, from dc to kilohertz.

The future home dc nanogrid is envisioned to have two dc voltage

levels: a high-voltage (380V) dc bus powering HVAC, kitchen loads,

and other major home appliances, and a multitude of eight low-volt-

age (48V) dc buses powering small tabletop appliances, computer

and entertainment systems, and LED lighting. Similar 380V/48Vdc

power distribution systems are currently being considered for data-

com centers in Japan, Europe, and the US, and are also being con-

templated for plug-in hybrid vehicles and aircraft power systems.

Several manufacturers already have on the market high power-densi-

ty bus control modules that supply 48V from 380V and are intended

for these applications.

http://greenbuildingpower.darnell.com/

http://dcbuildingpowerjapan.darnell.com/

M A R K E T

Electrical Engineers (m/f) ABB is a global leader in power and automation technologies that enableutility and industry customers to improve their performance while loweringenvironmental impact. ABB operates in more than 100 countries andemploys about 117,000 people, whereof 6,400 in Switzerland.

ABB Corporate Research is developing the foundations for the nextgeneration of ABB products in close collaboration with ABB businessareas. In Switzerland, the ABB Corporate Research Center is locatedin Baden-Dättwil in the proximity of Zurich and employs around 200 scientists from more than 25 countries.

In the area of Power Electronics we are looking for high-level electricalengineers (m/f) with strong theoretical skills to join our R&D team.

We offer an inspiring R&D environment, self-determined scientificworking in motivated teams with a wide range of experience andcompetence, the possibility to contribute in emerging R&D areasrelevant for future technologies, and excellent opportunities for furthercareer development.

Your tasks: perform applied research in the field of advanced powerelectronics and integration technologies for industrial components

and systems • contribute to and later acquire and lead R&D projects• conduct research with focus on multi-domain and electromagneticmodeling, simulation, and measurement methodologies • investigateapproaches for improving electromagnetic compatibility (EMC) andpower density while reducing electromagnetic interferences (EMI)

The requirements: PhD in electrical engineering • deep knowledgeof power electronics, electromagnetics, and manufacturing technol -ogies • willingness and ability to learn quickly new scientific areasand broad technical interest • readiness to work in an industrial R&Denvironment in collaboration with business units • motivation forinnovation and independent thinking • fluency in English • strongcommunication and interpersonal skills

Your contact:Andrea Kuhn, Recruiting Consultant

Apply online – or find other exciting jobs: www.abb.ch/careersJob ID: CH4495

Cities that consume 30% less energy? Certainly.

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

It is well established that due to increases in standard of living

throughout the world, total energy consumption is expected to

increase by at least 35 % over the next 20 years [1].

It is less well known that a significant reduction in worldwide energy

consumption can be achieved through the wide spread adoption of

improved load architectures [2,3]. In total, over 25 % of worldwide

annual energy consumption can be saved through widespread (i.e.

>90 %) adoption of these efficient load technologies enabled by

advanced power electronics. This energy conservation represent

over $ 2 Trillion/year in cost savings (at $ 45/barrel oil prices), far

greater than the approximately $ 50 Billion/year market for power

electronics today.

The energy savings are, for the most part, achieved through the

nature of the working load, though the performance of the loads

requires substantial, optimized and intelligent power electronics.

Even though both the required loads and the necessary power elec-

tronic architectures are, in principle, presently available to implement

these energy saving solutions, adoption is expected to remain rela-

tively anemic for at least another decade. This is due to the price pre-

mium which is passed to the end consumer of the complete systems

incorporating these energy efficient solutions. Only when this premi-

um is substantially reduced or eliminated, will the adoption of energy

efficient systems approach dominance, a necessary requirement for

substantial worldwide energy savings. The reduction of total system

costs can be substantially enabled by intelligent power electronics

which optimize performance/cost.

Modern power electronics solutions provide an array of system level

enhancements such as communication protocols, load condition

reporting, as well as optimal balancing and coordination and protec-

tion of power conversion sub-systems and loads. As important as

these advances have been, it is the continued progress in the per-

formance of the power converter sub-systems themselves that have

enabled increasingly dense and efficient working loads.

Value in Power

It can be argued that the intrinsic value proposition of the power con-

version sub-systems is density*efficiency/cost. This

performance/cost figure of merit (FOM) for power processing is the

equivalent driving force behind innovation as the logic unit/ $ FOM is

to the well known Moore’s law of the data processing industry. There

have been significant advancements in both FOMs over the past 40

years. It can be argued that the most significant advances in energy

conversion efficiency* density/cost have been achieved through req-

uisite improvements in the power devices used. Generally, advances

through improved circuit architectures, from linear to switching regu-

lation, hard to soft switching, passive to synchronous rectification,

etc., have all been accomplished by leveraging the inherent capabili-

ties and avoiding the inherent limitation of the power switch compo-

nents used. It can therefore be expected that radically improved

power switch performance might well drive a revolution in power

electronic architectures and systems.

The ability of power semiconductor devices to enhance the power

electronics performance/cost figure of merit can be simplified by its

own price/ performance figure of merit, namely switching power loss*

ohmic power loss *cost, where the switching power loss reflects the

thermal limitation of density, most often achieved through increasing

switching frequency and subsequent reduction in output filter compo-

nents. For inverter circuits this can be referred to by Ets* Vceon*cost,

for silicon based IGBT switch/ diode pairs. For dc-dc converter cir-

cuits such as common buck regulators, the FOM is

R(ds)on*Qsw*cost.

Since the advent of commercially viable silicon power FETs, intro-

duced some 30 years ago, enabled the widespread adoption of

switch-mode power supplies, replacing the linear regulator as the

dominant power architecture, the silicon power FET has become the

dominant power device. The silicon based IGBT, combining the ease

of charge control with the benefits of conductivity modulated drift

resistivity, has been another mainstay, especially in the lower fre-

quency conversion systems, e.g. motor drive inverters. Of course,

the same minority carrier injection that provides for lower ohmic loss-

es also increases switching losses through the effects of subsequent

tail currents. Over the last 3 decades significant engineering efforts

have driven the improvement in the performance figure of merit of

these silicon power devices by more than an order of magnitude.

However, as this technology approaches maturity, it becomes

increasingly expensive to achieve even modest improvements in the

device FOM. It is estimated that less than a factor of two improve-

ment will be economically feasible to achieve for 30 V FETs [4], with

perhaps a factor of five possible for 600 – 1200 V silicon IGBTs [5].

C O V E R S T O R Y

Bringing GaN on Si BasedPower Devices to Market

The Status of the GaNpowIR™ platform at International Rectifier

The availability of new power electronics based on commercially viable wide band gapsemiconductors such as GaN on silicon power devices fabricated in silicon foundries,provides the required performance to cost value proposition to enable lower economic

barrier to adoption for energy efficient power delivery architectures needed to significantly reduce global energy consumption in the coming decades.

By Michael A. Briere, ACOO Enterprises LLC, under contract by International Rectifier

21www.bodospower.com November 2010 Bodo´s Power Systems®

If further advances in power device performance are required by

future electronic loads, as is currently apparent, then these advances

must be achieved through the use of alternative materials.

One of the most promising alternatives to silicon is gallium nitride

based power devices.

Even though the basic GaN HEMT transistor was first invented over

15 years ago by M. Asif Khan [6], significant development efforts on

practical power devices using GaN-on-Si technology have been fairly

recent, predominantly in the past 5-7 years. GaN based power

devices are expected to improve rapidly over the next 10 to 20 years.

In fact, it is expected that an order of magnitude in improvement in

the key device performance FOMs will be achieved over the next five

years.

In addition to efficiency improvements, the use of wide band gap

semiconductors instead of state of the art silicon based devices for

power electronic systems allows the reduction of size/weight of the

conversion subsystems by between 2 and 10 fold, due to significant

reduction in cooling system requirements, further promoting adoption.

Commercialisation Barriers Overcome

There have been however, several significant barriers to the commer-

cialization of GaN based power devices. Chief amongst these is the

cost of production. The production of power devices includes the

costs of substrate, epitaxy, device fabrication, packaging, support

electronics and development.

The viable economic based limit of about $ 3 / cm2 for substrate and

epitaxy cost set by the power device marketplace is exceeded by all

substrate choices except silicon wafers.

Next to the cost of substrate and epitaxial layers, device fabrication

costs are the most critical. In fact, currently, substrate diameters of at

least 150 mm are required to achieve widespread commercial viabil-

ity for power device fabrication. To gain broad adoption of alternative

material based power devices, fabrication costs must approach that of

silicon based power devices. Such device fabrication costs are only

achievable if high volume (> 10,000 wafers/week), high yielding stan-

dard (silicon compatible) semiconductor fabrication lines are used.

Similarly, the volume necessary to support the broad power device

market (10 million 150 mm wafer equivalents per year) requires scala-

bility in device manufacture provided most readily by existing silicon

device fabrication facilities and silicon substrate supply.

One example of a technology program that has been developed to

address these issues is the GaNpowIR platform of International Rec-

tifier [3]. This technology platform uses GaN-on Si hetero-epitaxy and

device fabrication processing that can be performed in a standard

modern silicon CMOS manufacturing line with little modification to

equipment or process discipline. Therefore, this technology platform

is able to provide power devices with compellingly superior perform-

ance/cost FOMs compared to silicon which will promote widespread

adoption.

One of the most fundamental challenges to the commercialization of

GaN based power devices is the development of cost effective, high

yielding, high throughput III-Nitride epitaxial processes on large di-

ameter silicon wafers. The intrinsic mismatch in both lattice constant

and thermal coefficient of expansion with the requisite III-Nitride epi-

taxial films causes threading dislocations, as well as significant ma-

croscopic film stresses, which result in excessive wafer bow and

plastic deformation (cracks) in the films. These issues have been

addressed by engineering the proprietary epitaxial film growth on

standard thickness (625 um) 150 mm (111) silicon wafers to both

eliminate most of the threading dislocations, resulting in 109 cm-2,

predominately edge dislocations for 2 um thick films (comparable to

similar thickness films grown on SiC), as well as compensating for

the stresses due to thermal coefficient mis-matches. These result in

a high quality device layer. In addition, the resulting wafer bow of

< 20 um (3 sigma), is well within the required limit for device fabrica-

tion of < 60 um. It should be noted that truly crack free material to

within 0.5 mm of the wafer edge are consistently produced by this

process in manufacturing volμme.

Figure 1: Reverse bias drain leakage behavior for LV GaNpowIRdevice ( Lg=0.3 um) at room tem-perature.

[email protected]+49(0)2871 7374

BICRONElectronics

Gate Drive • Switch Mode • Pulse • Load-Leveling • Sensing

Custom design matches your unique electrical characteristics for unequaled performance

Isolation for operating voltages up to 20KVprotects your system from sudden failure

High efficiency achieved through minimum need for compensation components

Each design rated to a specified Partial Discharge and Corona-free level verified by 100% testing

Design and manufacturing techniques safeguardagainst thermal instability

. . . Get the full details at www.bicron-magnetics.us

�

�

�

�

�

Bicron® High Frequency Transformers for Electronic Power Control Systems

Reliability for the Life of Your System!

Hybrid Vehicles • Rail Drive Controls Wind/Solar Power Controls

GaNpowIR devices

Much of the reported constructions for GaN devices to date utilize

Schottky gates and subsequently exhibit device leakage in operation

of mA/mm of gate width. For a power device, which often has an

effective gate width on the order of 1 meter, such gate leakage would

result in an unacceptable power loss/heating. Similarly, the maximμm

operating voltage has often been specified at reverse bias source-

drain current densities of mA/mm of gate width. Another challenge,

therefore, is the reduction of these leakage currents to less than 1

uA/mm. This has been achieved through the combined use of a

proprietary insulated gate construction and improved III-Nitride epi-

taxial film quality. This has resulted in gate and drain-source leak-

ages of 10 pA/mm, as shown in Figure 1. A punch through limited

S-D breakdown of > 40 V is seen for Vg= -20V, for these devices,

with Lg=0.3 μm and gate-drain and gate-source spacing of 1 μm.

The first product release to production in early 2010 on the IR GaN-

powIR technology platform is a 30 A capable 12V buck converter

power stage product, the iP2010. It incorporates the control and syn-

chronous rectifying switches together with the intelligent gate driver in

a low parasitic LGA package. Figure 2 shows the measured power

conversion efficiency for this first generation low voltage GaN product

compared to competitive silicon based solutions. As can be seen, the

GaN based power device solutions offer significant advantages over

silicon based alternatives. The devices from this first generation GaN-

powIR low voltage platform achieved the targeted performance figure

of merit (Ron*Qg) of 30 mohm-nC (packaged). Next generation low

voltage devices are expected to exhibit less than 20 mohm-nC with

comparable state of the art silicon devices still above 40 mohm-nC.

These devices are very rugged in their intended application of 12V to

1 V buck regulators, as can been seen in Figure 3, where the forward

biased safe operating area (FBSOA) is shown for such low voltage

power devices, far exceeding the requirements of the application.

The resulting ratio, for these 850 mm gate width device, of Ion/Ioff of

> 10 10 is substantially better than reported elsewhere for GaN based

devices.

Similarly, early 600 V GaNpowIR devices exhibit off-state leakage

currents less than 50 nA/mm (with Vg=-10V), far better than the 100

to 1000 uA/mm reported elsewhere, providing an Ion/Ioff ratio of

> 10 7, where Ioff is measured at 600 V.

As is the case in SiC based unipolar devices, GaN based HEMT

exhibit negligible minority carrier induced reverse recovery charge.

The resulting transient reverse recovery current is determined essen-

tially by capacitive components. This leads to much more desirable

characteristics as shown in Figure 4, where the GaN based device

exhibits nearly an order of magnitude better performance than sili-

con based alternatives. In this way, the greatest advantages

achieved through the use of expensive SiC diodes in the removal of

harmonic filtering snubber circuits in applications such as power fac-

tor correction AC-DC converters can be likewise achieved through

the use of much less expensive GaNpowIR rectifier products.

This advantage of low switching losses can further be seen in the on-

off transition induced losses for 600 V GaNpowIR HEMTs (Eoff) at

24 uJ, as compared to that of state-of-the-art silicon based super-

junction FETs, 38 uJ, and best in class, low loss silicon IGBTs at

144 uJ (tested at 300V and 6A). In fact, even in this early stage of

development, GaNpowIR switches exhibit at least a factor of 4

improvement in the Vceon*Esw figure of merit vs state of the art sili-

con based alternatives.

One approach to provide GaN based products with drop in replace-

ment capability in existing power electronic systems is the cascade of

a low voltage silicon device and a high voltage GaN HEMT. This pro-

vides normally off behavior with a well established, robust drive inter-

face. The transfer characteristics of such a prototype is shown in

Figure 5, exhibiting a Vt of about + 3V, consistent with today’s HV

switch applications. Figure 6 shows the output characteristics for this

same pair, providing well behaved on-state behavior.


C O V E R S T O R Y

Figure 2: Measured power conversion efficiency for initial GaNpowIRproduct,iP2010 and planned product iP2011, 12 Vin to 1.2 Vout POLconverter power stages operating at 1200 kHz compared to estimat-ed performance of two silicon based alternatives

Figure 4: Transient reverse current measurements for 600 V GaN-powIR HEMT, Si Fast Recovery di-ode and Si superjunction FETbody diode.

Figure 3: Forward biased SOA for low voltage GaN based powerdevices intended for 12Vin power conversion applications.

Conserve battery life with Maxim’sadjustable-timeout, 125nA watchdog solution

Industry’s lowest powersupervisory IC…AGAIN!

• Reset thresholds from 1.575V to 4.625V in100mV increments

• Fully specified over the -40°C to +125°Cautomotive temperature range

• Capacitor-adjustable reset watchdog timeouts

Part Watchdog Timer Reset Output Package

MAX16056 ✓Push-pull

8-TDFNMAX16057 — 6-TDFNMAX16058 ✓

Open drain8-TDFN

MAX16059 — 6-TDFN

94% lower supply current than competition

MAX16056–59 MAX6854–69COMPETITOR

ACOMPETITOR

B

125nA 170nA

2000nA

7000nA

TDFN

3mm x 3m

m

SmallPackage

Lowest

For free samples or technical support, visit our website or call +3531 2235500.Innovation Delivered and Maxim are registered trademarks of Maxim Integrated Products, Inc. © 2010 Maxim Integrated Products, Inc. All rights reserved.

www.avnet-memec.eu

™DIRECTwww.maxim-ic.com/shop www.farnell.com

www.silica.com

www.maxim-ic.com/MAX16056-9-info


Device yield is an important challenge for the commercialization of

large area power devices. It is economically imperative that yields

> 80 % are commonly achieved for large devices (e.g. > 10 mm2).

Such yields have been demonstrated achievable using this technolo-

gy platform, demonstrating the necessary level of process maturity

for commercialization.

Finally, the stability of device in-circuit performance is a prerequisite

to commercialisation. The critical FOM, Rdson shows excellent sta-

bility under accelerated conditions for > 6000 hrs. In fact, over

7,000,000 device hrs of reliability testing, with up to 9000 hours per

device, has shown performance in line with silicon based device

specifications. Figure 7, shows the excellent stability of the gate

dielectric, measured at Vg=-7.5V, rated at -8.5 V max for low voltage

devices, under extreme accelerated stress conditions of Vg=-50 V at

150 C for over 3000 hrs.

Drain leakage current has also proven very stable under 26 V

reverse bias stress with Vg=-7V at 175 C for over 3000 hrs. Impor-

tantly, Figure 8 shows that no physical degradation in the AlGaN bar-

rier layer is found at the gate edge under all applied stress condi-

tions. In addition to conditions already identified, this includes (a)

Vd=26 V, Vg=-14 V at 150 C for > 3000 Hours, (b) Vd= 34 V, Vg=-

22 V at 150 C > 600 hrs, (c) forward conduction of I=200 mA/mm

with Vd= 25 V. This is significantly better than results reported else-

where for GaN based HEMTs [7]. This is expected due to the signifi-

cantly reduced gate leakage currents found when using a gate

dielectric instead of a metal-semiconductor gate construction.

Conclusion

A great opportunity exists to significantly impact future global energy

consumption, with its many sociological, environmental and economic

consequences. A cost effective means of producing GaN based

power devices will help achieve the necessary adoption rates to meet

this challenge. International Rectifier’s GaNpowIR platform is such a

technology platform, demonstrating required performance from 20 to

600 V devices. Excellent device stability and long term reliability per-

formance has been shown for initial low voltage power devices.

References

www.eia.doe.gov/iea

Lidow, A., APEC 2005 Planery Talk and Briere, M.A., S2k Conference

2005

M.A. Briere, Proceedings of PCIM Europe 2009 and Briere, M.A.,

Power Electronics Europe (7), October / November 2008 pp. 29-31

Ikeda et.al. ISPSD 2008 p. 289

Nakagawa, A., ISPSD 2006 p.1

Khan, M.A. et.al, Appl. Phys. Lett (63) p.3470, 1993.

J. Joh and J del Alamo, IEDM 2006 p1-4

www.irf.com

C O V E R S T O R Y

Figure 5, Transfer characteristics for a proto-type cascade pairing ofa low voltage silicon FET and an early 600 V GaNpowIR HEMT.

Figure 6, On-state output characteristics for a proto-type cascadepairing of a low voltage silicon FET and a 600 V GaNpowIR HEMT

Figure 7: Stability of gate leakage current over 3000 hrs with -50 Vapplied to the -8.5 V rated gates at 150 °C. Lg= 0.3 um, Wg= 2600mm.

Figure 8: TEM cross section of low voltage device gate region, show-ing no degradation at gate edge of AlGaN barrier after 3000 hrsunder stress at 26 Vds and -7Vgs condition at 175 C.

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!

26

I G B T S


The device was designed especially for medium and high current

applications. In comparison with the 600V IGBT3 the new chip offers

a better softness during switch-off and a higher blocking voltage

capability. As an add-on the short-circuit robustness is significantly

improved. In contrast the 600V IGBT3 has been optimized for lower

power applications, or higher power in very low stray inductance

applications.

Design and Technology of the 650V IGBT4

The 650V IGBT4 [1] utilizes a trench MOS-top-cell, thin wafer tech-

nology and a field-stop concept as seen in Figure 1. The combination

of trench cell and field-stop enables comparatively low on-state and

turn-off losses. Compared to the 600V IGBT3, the chip thickness was

increased by about 15% and the width of the MOS channel was

decreased by about 20% as indicated in Figure 1. Thereby, the soft-

ness during switch-off is improved to reduce the EMI effort. However,

of course these measures also cause additional losses. So in order

to compensate for these side effects, the efficiency of the backside

emitter was increased by 50%. In addition to the optimization of the

dynamic behaviour the blocking voltage was grew by 50V to 650V.

Results of the 650V IGBT4 dynamic characterization

The stray inductance in combination with the current gradient has an

influence on the voltage characteristic during turn on and turn off as

ΔV=L*di/dt. Thus the over voltage increases when switching off with

larger Lσ. The turn off behaviour is quite insensitive to the gate

resistance. This behaviour is well known for trench field-stop IGBT

[6]. A consequence of this inherent IGBT behaviour would be in such

a case a special driver stage with integrated IGBT protection func-

tionalities and/or additional components like snubber capacitors. All

these functionalities and components create design effort and cost.

High current levels need, due to the high di/dt level, a DC-link design

with very small parasitic inductances. As an alternative specially

designed IGBTs with a soft switching characteristic like the new 650V

IGBT4, can be utilized.

The difference of the fast 600V IGBT3 and the soft 650V IGBT4 in

the switching behaviour becomes obvious in Figure 2, where the

switch-off behaviour of high current 600A EconoDUAL™3 modules

are compared.

For the investigations a standard DC-link design with Ls=60nH was

used. This setup is not ideally suited for a high current setup with the

600V IGBT3 [5]. Consequently, the switch-off of a current of 50%

Switching in silence650V IGBT4 the optimized device for reduced EMI and low ΔV

The trend of the last years of all power semiconductor manufacturers to increase theswitching speed of the devices offers the benefit of reduced switching losses and the

possibility to improve the efficiency of the system. These power devices require optimizedparasitic inductances (Lσ) of the DC link circuit. With respect of the needs of high powerapplications with its larger currents in various setups, now a new chip, the 650V IGBT4,has been designed to provide an additional degree of freedom. The IGBT4 device features

an improved softness during switch-off and a lower overshoot voltage as the result of areduced turn-off current slope di/dt.

By Wilhelm Rusche, Dr. Andreas Härtl, Marco Bässler, Infineon Technologies AG

Figure 1: Schematic cross section of the new 650V IGBT4, and thechanges implemented compared to the 600V IGBT3: increased chipthickness (y), decreased channel width (z), and increased backsidep-emitter.

Figure 2: Comparison of the softness during switch-off of a 600VIGBT3 left picture and the new 650V IGBT4 right picture, measuredin an EconoDUAL™3 module. Displayed are the voltage VCE as(black traces), the collector current IC as (red curves), and the gate-emitter voltage VGE as green traces during switching off a current of300A at 25°C.

World's Most Powerful

1700V Dual IGBT Modulefor High Power

Energy Conversion

World's Most Powerful

1700V Dual IGBT Modulefor High Power

Energy Conversion

Please visit us:

SPS 2010

Hall 1, Stand 639

www.mitsubishichips.eu [email protected]


Inom, IC=300A, and a DC link voltage of 300V at 25°C effects a quite

high overshoot voltage VCE,max and a snap-off with oscillations. In

contrast, the new 650V IGBT4, especially designed for such high cur-

rent applications, shows a smooth switch-off with a much lower

VCE,max, even at the typical DC link voltage of 300V in specific high

current setup.

In the given test setup the 600V IGBT3 device reaches the limit of

600V. While the 650V IGBT4 shows a smaller overvoltage of 530V. In

addition to the reduced overvoltage shoot the increased blocking

capability VCE_max, comes as a real surplus and offers the advantage

of an increased safety margin during turn-off.

Not only the turn off characteristics but also the softness of the IGBT

is quite insensitive to the gate resistance. The softness during switch-

off is improved to reduce the EMI effort. In Figure 3 the Fourier

Transformation spectra of a soft and a not soft turn-off waveform are

given. The oscillation leads to a 5 times higher level around the oscil-

lation frequency of roughly f=20…25 MHz, a frequency which is quite

typical for chip DC link oscillations at the given parasitic inductance.

Even though such a procedure is not able to predict passing or failing

of an EMI qualification, it obviously demonstrates the sensitivity of

EMI to snap-off phenomena.

The most important aspect in all designs is the improvement of the

DC-link design in order to be able to prevent additionally any kind of

oscillations.

For the inductance the lower the better is a simple rule for high effi-

ciency designs.

On the other hand, the softer switching behaviour has to be paid for

with higher losses during switch-off, Eoff, and with a slightly increased

saturation voltage ΔVCEsat≈100mV@T=25°C. Taking into account

common switching frequencies, this increase does not play a major

role. This fact is visualized in Fig. 4 showing a simulation with

IPOSIM. This tool, the Infineon Power Simulation program, can be

found on the Infineon homepage (www.infineon.com). It performs a

calculation of switching and conduction losses for all components,

taking into account conduction and switching losses as well as ther-

mal ratings. As can be seen in Fig. 4, the reduction of the RMS mod-

ule current due to increased losses of the 650V IGBT4 is only moder-

ate, between 4 and 9% at switching frequencies of 2kHz up to

10kHz, a typical range for common applications.

Besides this standard operating the design must be robust and has

also to withstand a case of failure. The established value in power

semiconductor datasheets is the specification of a hard short circuit

current (ISC).

Short circuit robustness

Despite the considerably reduced silicon thickness of field-stop

devices as compared to non-punch-through (NPT) designs, field-stop

IGBTs are known to feature a good short-circuit robustness [3, 4].

With the new 650V IGBT4, the short-circuit robustness is significantly

enhanced compared to the 600V IGBT3. The increased thickness of

the chip offers a larger thermal budget due to the thermal capacity of

the silicon volume. In addition, the decreased channel width reduces

the level of the short-circuit current, this effect is shown vice versa in

[5]. In sum, the 650V IGBT4 can resist a higher short-circuit energy,

and therefore the device is able to withstand a longer short-circuit

pulse time without getting destroyed. In Fig. 5, a typically hard short-

circuit pulse measurement of the 650V IGBT4 is displayed. As the

graph shows, the pulse time short-circuit event was 10 μs, and the

short-circuit current typically is about 4 times the nominal current of

the FF600R07ME4 device.

Conclusion

Infineon’s new 650V IGBT4 permits the development of inverter

design especially for large current applications, to be employed in the

corresponding modules. The device features reduced EMI effort as

the result of an improved softness during turn-off, a lower overshoot

voltage as the result of a reduced turn-off current slope di/dt, a higher

blocking capability of VCE_max=650V, an operation range of increased

DC-link voltages and/or higher stray inductances, an enhanced short

circuit robustness with 10μs pulse time @Tvjop=150°C, and an ideal

flexibility between highest output power at elevated junction tempera-

ture of up to Tvjop=150°C or highest power cycling capabilities at

lower junction temperatures.

I G B T S

Figure 3: Influence of a current snap off to the EMI; FFT of the volt-age curves of a FF600R07ME4 (black trace) with its soft turn-off anda snappy switching event of the FF600R06ME3 (red trace)

Figure 4: Calculation of the RMS current as function of the switchingfrequency of the 600V IGBT3 (black line for Tvjop=150°C, red line forTvjop=125°C) and the new 650V IGBT4 (blue line for Tvjop=150°C,green line for Tvjop=125°C), calculated in 600A EconoDUAL™ 3 mod-ules. Calculations were performed with IPOSIM tool can be found atwww.infineon.com; simulation conditions: Rthheatsink=0.09K/W, Tam-bient=40°C, Tvj,op=150°C resp. Tvjop=125°C, cos(ϕ)=1.

www.bodospower.com

In sum the 650V IGBT4 provides design engineers effective

degrees of freedom in their applications.

Literature

A.Härtl, M.Bässler, M.Knecht, P.Kanschat: “650V IGBT4: The opti-

mized device for large current modules with 10μs short-circuit with-

stand time”, Proc. PCIM Europe, (2010).

H. Rüthing et al.: "600V-IGBT3: Trench Field Stop Technology in

70μm Ultra Thin Wafer Technology", Proc. 15 th ISPSD, 66 (2003).

M. Otsuki et al.: “Investigation on the Short-Circuit Capability of

1200V Trench Gate Field-Stop IGBTs“, Proc. 14th ISPSD, 281

(2002).

T. Laska et. al.: “Short Circuit Properties of Trench-/Field-Stop-

IGBTs – Design Aspects for a Superior Robustness”, Proc. 15th

ISPSD, 152 (2003).

P. Kanschat, H. Rüthing, F. Umbach, F. Hille: “600V-IGBT3: A

detailed Analysis of Outstanding Static and Dynamic Properties”,

Proc. PCIM Europe, 436 (2004).

W.Rusche: Infineon Application Note “AN2003-03, Switching

behaviour and optimal driving of IGBT3 modules” (2003)

www.infineon.com

Figure 5: Measurement of a short-circuit pulse event of the 650VIGBT4. Shown are the collector-emitter voltage VCE (black trace)and collector current IC (red trace), and the gate-emitter voltageVGE (green trace). Test conditions were VCE =360V, VGE =±15V, Tvj

=150°C.The device shows an excellent switching and short circuit robust-ness with the specified short circuit time having been adjusted from6μs from the 600V IGBT3 to 10μs for the 650V IGBT4.

Your partner for high-tech competence in energy efficiency

desi

gnho

use

curamik electronics GmbH Am Stadtwald 2 D-92676 Eschenbach Phone +49 9645 92 22 0 [email protected] www.curamik.com

DBC COOLER

High thermalconductivity

Excellent for chipon board

Optimized heat spreading

Highly integrated cooler

Outstanding thermal performance

Customized design

DBC SUBSTRATE

Numerous modern power electronic applica-

tions, e.g. in MV Drives, Flexible AC Trans-

mission Systems (FACTS) are operating at

increasing line voltages, thus demanding

semiconductors with higher blocking volt-

ages. This allows a reduction of the current

and losses in the system, and avoids series

connection of semiconductor switches which

requires additional snubber components.

The 10kV IGCT devices, see figure 1, will

allow the operation of drives to voltages up

to 7.2kV rms and power of 12MW without

series or parallel connection of semiconduc-

tor switches (e.g. using a classical 3L-NPC

topology). Another possible application field

are high voltage applications with multilevel

configurations where the low switching fre-

quencies can make the 10 kV IGCT a viable

solution compared to lower rated IGBTs.

For the design of such high voltage IGCTs,

minute attention had to be given to the

resilience against cosmic ray events and its

impact on device design. In addition, losses

and maintaining competitive turn-off current

ratings comparable with today’s ratings of

6.5 kV IGCTs were significant technology

challenges that had to be addressed. ABB

has already presented a 10 kV – 2kA IGCT

that demonstrated the feasibility of the 10kV

technology on Silicon. However the current

ratings were not high enough for the target-

ed application. This article reports on

increased IGCT SOA capability using a

novel device design. The High Power Tech-

nology or HPT platform is the latest genera-

tion of ABB’s high power high current IGCTs

with enhanced Safe Operating Area (SOA).

These features ensure an increased margin

for the IGCT ruggedness, further expanding

the power capability per device. The cooling

of the semiconductor switch and not the

IGCT SOA is now the limiting factor in the

power system design. The 10kV HPT IGCT

is expected to be first commercially available

in 2011.

HPT IGCT Technology

The next generation HPT IGCT platform has

been designed to substantially increase the

Safe Operating Area of the device. The

“High Power Technology” (HPT) IGCT struc-

ture is based on a corrugated p-base doping

profile as seen in Figure 2.

The application of the HPT technology plat-

form has enabled ABB to establish a new

benchmark in the IGCT technology over the

whole voltage range. The concept has been

shown to increase the SOA of the previous

generation of commercial 4.5kV IGCTs by as

much as 40%. A peak power density of 700

kW/cm2 is reached for large area HPT

IGCTs, significantly higher than the capability

of previous standard devices.

Static 10kV IGCT Characteristics:

The required static blocking capability of a

10kV IGCT has been reached using a wafer

termination based on Variation of Lateral

Doping. This design of the termination is

uniquely suitable for large area power semi-

conductor devices being very efficient with

regards to the ratio active area/ termination

area. Figure 3 shows that the voltage block-

ing capability of these large area Si power

semiconductors exceeds 11kV with a leak-

age current lower than 20mA at 125 °C. Vari-

ations in the silicon design to optimise com-

ponents for lower machine rating, as 6 kV

rms, are in consideration.

H I G H P O W E R S W I T C H


A 10kV HPT IGCT withImproved Switching Capability

Developing a new platform for high voltage switching

A major issue for the 10 kV IGCT has been the limited turn-off capability. By introducingthe new HPT technology a 10 kV IGCT is in development that has a switching capability

comparable with the capability of former IGCTs with half the voltage rating.

By Tobias Wikström and Arnost Kopta, ABB Switzerland Ltd, Semiconductors and Iulian Nistor ABB Switzerland Ltd, Corporate Research

Figure 1: 10kV IGCT module using the HPTIGCT technology

Figure 2: The HPT GCT wafer structure

Figure 3: Forward blocking of the 10kV IGCTat 125°C

Figure 4: On-state characteristics of the10kV HPT IGCT at Tj=125°C

SIMPLY SMARTER


The on-state characteristic of a 91-mm IGCT

at Tj=125°C and after carrier lifetime engi-

neering is shown in Figure 4. The on-state

voltage drop is 5.2V at a current density of

50 A/cm2

Dynamic 10kV IGCT Characteristics &

SOA Performance:

The switching of the IGCT prototypes was

measured in the test circuit showed in Figure

5 in single-pulse & multi-pulse operation.

The turn off waveform is shown in Fig. 6.

The IGCT can turn off safely a current of at

least 3.2kA at both Tj=25°C and 130°C, limit-

ed by the capability of the test setup. The

peak power in the 10kV HPT IGCT reaches

19.77 MW, see figure 7, corresponding to a

power density of 460 kW/cm2. This is signifi-

cantly higher then standard 10kV IGCT

designs which are limited to values of about

12MW at these voltage levels, correspon-

ding to a power density of 300 kW/cm2.

However, the power density has decreased

from the value of 700 kW/cm2 for the 4500

V device due to the increased voltage rat-

ings. Nonetheless, the power handling capa-

bility reached with the 10kV HPT IGCT

demonstrates the excellent potential of the

HPT platform for high voltage IGCTs.

10kV Soft recovery Diode

The topology of a modern VSI converter

requires the use of freewheeling and clamp-

ing diodes with similar voltage rating as the

main semiconductor switch. Especially the

fast recovery freewheeling diodes must meet

certain technological criteria among which

soft reverse recovery is of main interest. The

technological challenges are related to the

trade-off between diode recovery losses and

snappiness. In general, designing the diode

for minimal losses leads to snappy reverse

recovery. In a standard HV diode design, the

reverse recovery current decreases to zero

with a very high di/dt. Taking into account

the presence of stray inductances in the cir-

cuit, high frequency & high amplitude oscilla-

tions can be noticed in the current and volt-

age waveforms. This can impact significantly

the level of electromagnetic noise, negatively

affecting the EMI compatibility of the system.

In addition, the overvoltage can generate

additional electrical stress on the compo-

nents. As this behaviour is not acceptable in

an industrial application, while low losses are

of highest importance, the “Field Charge

Extraction” (FCE) concept has been applied

as a mean to provide the optimal trade-off

between the apparently contradictory HV

diode design requirements.

Figure 8 shows the reverse recovery of a

standard 10kV diode under strong snap-off

conditions (current 5% of nominal value).

The on-state voltage drop of the standard

diode is 5.5V at Ion-state=1.7kA. As expected,

high frequency current oscillations as well as

the over-voltages are observed. Next, FCE

10kV diode prototypes have been manufac-

tured and the reverse recovery is shown in

Figure 9. The on-state voltage drop of the

FCE diode is 7.3V at Ion-state=1.7kA, there-

fore the life time of the minority carriers is

further reduced compared to the standard

diode. However, the FCE concept can com-

pensate for the reduced level of on-state

plasma yielding a soft recovery process.

In order to demonstrate the ruggedness of

the FCE concept at higher nominal currents

and temperatures, we have measured the

reverse recovery at values up to 2000A and

125°C (shown in Figure 11).

In addition, we show that the SOA of the

FCE 10kV diode is comparable with stan-

dard technology. The peak power during

reverse recovery of a standard 10kV diode

at Vdc-link=6kV, Inom=1.7kA is 6.8 MW and the

losses are 30 J at a temperature of 125°C.

For the FCE diode at Vdc-link=5.5kV,

Inom=2kA, the peak power during reverse

recovery is 5.8 MW, and losses are 20.35J

at a temperature of 125°C.

www.abb.com/semiconductors

H I G H P O W E R S W I T C H

Figure 5: The circuit used for measuring thedynamic performance of 10kV IGCTs Parameters: Li=10.3μH, Ls=350nH,CCL=3μF, RS=2Ω, CSn=3μF

Figure 8: Snappy reverse recovery wave-forms for 91-mm standard 10kV diode in typ-ical snap-off conditions: VDC=6kV,di/dt=500A/μs, Tj=25°C

Figure 10: Snappy reverse recovery wave-forms for 91-mm standard 10kV diode innominal conditions: VDC=6kV, di/dt=500A/μs,Tj=125°C

Figure 11: Soft reverse recovery waveformsfor 91-mm FCE 10kV diode in nominal con-ditions: VDC=5.5kV, di/dt=500A/μs, Tj=125°C

Figure 6: Turn-off waveform for a 91-mm10kV HPT IGCT at VDC=6kV, IT=2kA,Tj=130°C

Figure 7: Peak power and losses waveformsduring turn off of 10kV IGCT at 130°C

Figure 9: Soft reverse recovery waveformsfor 91-mm FCE 10kV diode in typical snap-off conditions: VDC=6kV, di/dt=1kA/μs,Tj=25°C

MAKING MODERN LIVING POSSIBLE

D A N F O S S S I L I C O N P O W E R SILICONPOWER.DANFOSS.COM

Wind power the cool wayCustomized power modules made for high performance

extended life and very low life cycle costs. In short, when you chooseDanfoss Silicon Power as your supplier you choose a thoroughly tested solution with unsurpassed power density. Day in and day out.

Please go to siliconpower.danfoss.com for more information.

It cannot be stressed enough: Efficient cooling is the most important feature in power modules. Danfoss Silicon Power’s cutting-edge ShowerPower® solution is designed to secure an even cooling across base plates. In addition, our modules can be customized to meet your wind power requirements in detail, offering: High quality, high performance,

Power Stacks withShowerPower®

With more than 165 participants, the ECPE Workshop held in

Västerås (Sweden) has confirmed the growing attention paid to multi-

level conversion systems (Figure 1). Initiated by ABB Corporate

Research (Dr. Demetriades and Dr. Tenca) the seminar has

addressed the many faces of multilevel conversion from topology to

control and applications, from very high power to low power.

Topologies

Thirty years after Baker’s patent on the Neutral Point Clamped Invert-

er (NPC), topologies allowing multilevel conversion form a large fami-

ly and most of these topologies have been presented and compared

during the seminar. The most striking topic of the seminar has cer-

tainly been the breakthrough of the Modular MultiLevel Converter

(M²C) and its variants (Figure 2).

Highly modular with a high number of levels, this topology introduced

by Prof. Marquardt (Univ. BW Munich) some ten years ago, seems

now clearly recognized as the best topology for HVDC applications.

As shown in the presentations by Prof. Clare (Univ. Nottingham), Dr.

Gambach (Siemens Energy), Prof. Nee (KTH Stockholm) and Dr.

Hasler (ABB Schweden), the major companies in the field (ABB,

Alstom and Siemens) have developed slightly different products

using this concept and allowing impressive figures such as power up

to 200MW and voltages of +/-200kV on the DC bus. The application

of the concept in the field of drives is now investigated, but in this

field the competition between topologies is more open because the

M²C requires roughly twice the amount of semiconductors (sum of

Volts.amps of all semiconductors) and 2 to 3 times the amount of

energy stored in the capacitors of a 3-level NPC. The problem of

stored energy is even more striking when the modulation frequency

is low or very low, a situation that definitely needs to be handled in

drives. As explained by Dr. Hiller (Siemens Large Drives), some dedi-

cated modulation strategies can reduce the requirement on stored

energy, but the more realistic application of M²C in the field of drives

is probably on the line side.

A quite comprehensive review of topologies used in MV drives has

been presented by Prof. Bernet (TU Dresden) (Figure 3) showing in

particular how the NPC and Flying Capacitor (FC) families gave birth

to the newest 5L-ANPC topology. This topology has now been suc-

cessfully introduced by ABB with 10kVdc in its ACS2000 drive of

which details have been presented by Dr. Schlapbach (ABB Switzer-

land). Today, a clear majority of MV drives now use two back-to-back

voltage source inverters multilevel (VSI) converters with semiconduc-

tors in series, but we also heard that this is only one part of a bigger

picture.

Series multicell VSIs generate multilevel voltage waveforms on the

AC side thus improving the harmonic content on the line and on the

machine side and this is what makes them so attractive; THD

requirements are very high on the line side and on the machine side

when long cables are used, and series multicell is a good match for

these applications. However, there are cases when the THD and

EMC requirements are stronger on the DC side, and onboard net-

works with a DC bus distributed along the vehicle are such a growing

market (Figure 4). Series multicell converters generally do not help in

T E C H N O L O G Y


Review of the ECPE Workshopon Advanced Multilevel

Converter SystemsBy Dr. Thierry Meynard (University of Toulouse,

LAPLACE - CIRTEM), Technical Chairman of the ECPE Workshop

Figure 1: ECPE Workshop held in Västerås (Sweden)

Figure 2: Modular Multi-Level Converter (M2C)

Slides from Prof S BernetSlides from Prof. S. BernetTU DresdenTU Dresden

Figure 3: Classification of Converter Topologies on the MVD - Market

Slides from Prof. S. BernetTU Dresden

these applications because the current on the DC bus is still a 2-level

current waveform. It has been shown that switching to parallel multi-

cell gives multilevel current waveforms on the DC side and maintains

voltage multilevel waveforms on the AC side, another advantage is

that it helps handling the high currents imposed by the demand for

increasing powers and with a voltage that is limited for safety rea-

sons.

In the field of very low voltages, this is already used in Voltage Regu-

lator Modules supplying processors with 1V/100A using typically 5

interleaved buck converters and InterCell Transformers to suppress

electrolytic capacitors. The concept has been further investigated by

Prof. Gateau (Univ. Toulouse) who presented three-phase parallel

voltage source inverters and Prof. Laboure (SUPELEC), who derived

isolated interleaved converters with reduced filters and a resulting

high power density. The potentially high power density of parallel

multicell converters has also been confirmed by Prof. Mertens (Univ.

Hannover) who showed how the choice of the topology impacts the

size of the filters. It has also been shown by Mr. Fritsch (Vincotech)

and Mr. Rizet (G2ELab / APC by Schneider Electric) that using paral-

lel multicell conversion and soft switching is a way to reach very high

efficiency in low voltage applications (230Vac) which is the key figure

of merit in photovoltaic applications and Uninterruptible Power Sup-

plies.

Modulation Strategies

Multilevel converters also offer many degrees of freedom in term of

control, and various aspects of the modulation of multilevel convert-

ers have been discussed. Dr. Tenca (ABB Sweden) has shown how

the practical requirements can be expressed in terms of energy you

cannot cheat with, and how some theoretical concepts and methods

can help solving some questions related to multilevel converters

(modulation strategies, but also safety of operation)! It has also been

shown by Mr. Thielemans (Ghent Univ.) how FC converters can be

modulated to stabilize the balancing time constant over a wide modu-

lation range, and Mr. Videt (Schneider Toshiba Inverter) described 3-

level modulation strategies reducing the Common Mode voltage and

the peak voltage generated at the end of long cables. As the number

of voltages increases, the need for a generic approach to the control

of multilevel three phase VSIs is needed. Such a generic approach

using topology independent carrier based strategies and topology

specific state machine decoders has been developed in the literature

and should be used as a basis for comparison each time a new

topology or a new application is investigated. The 5L-ANPC and par-

allel multilevel converters are no exception to this rule; Dr. Schlap-

bach applied the method to Direct Torque Control with a five level

state machine and Prof. Gateau showed how this generic approach

can be applied and adapted to solve some side effects inherent in

these configurations.

Optimized and dedicated compo-

nents

An analysis of existing topologies has

been presented by Mr. Schweizer

(ETH Zurich) who showed how this

analysis can guide the choice of the

type of semiconductor, and even its

surface to optimize the overall trade-

offs. Such an evolution towards opti-

mized components is also noticeable in

terms of commercially available mod-

ules; Prof. Kaminski (Univ. Bremen)

and Mr. Frisch (Vincotech) have out-

lined the existence of several dedicat-

ed multilevel modules, some of them

mixing different technologies inside the

same module to optimize performance.

They also tried to evaluate the poten-

tial of wide-bandgap semiconductors in

the field of multilevel conversion to

push the limits of efficiency and specif-

ic mass even further.

Safety issues

Reliability and continuity of operation

are critical issues and counting the

number of devices does not give all the

answer. Various aspects of fault-han-

dling have been discussed by Dr. Pou

(TU Catalonia) and Mr. Billmann

(Fraunhofer IISB Nuremberg), and it

has been shown that if properly han-

dled, faults in multilevel converters can

have a limited impact and multilevel

converters can be designed with fault

tolerance in mind.

Conclusion

Multilevel converters have grown into a

mature technology invading many

fields of application such as 100MW

Flexible AC Transmission Systems, 1-

10MW Medium Voltage Drives, 100kW-

1MW low voltage Drives and Uninter-

ruptible Power Supplies, 5-50kW PV

systems, 1-10kW DC-DC converters

for onboard applications and 100-

200W Voltage Regulator Modules for

processors. The related know-how in

terms of modulation, control and

design is immense, specific compo-

nents such as dedicated power mod-

ules and specific passive components

are available, so now is the time to

take advantage of the benefits of multi-

level conversion.

www.ecpe.org


Figure 4: Filtering Requirements for Different Types of Applications

St ti li tiStationnary applications

To be filtered To be filteredTo be filtered To be filtered

On-board applicationspp

To be filteredTo be filtered

up to 400 x 1200 mm possible

due to innovative internal construction extremely low Rth

values are possible

Any dimensions possible

Superior thermal performance

Design and production

in house

up to 400 x 1200 mm possible

Any dim

nce

due to innovative internal

S p

absolutely pure and homogenous leak free joints

VACUUM BRAZED COLDPLATES FROM DAU

no flux agent for brazing necessary

NEW TECHNOLOGY

Visit us inHall A2, Booth 349

- AUSTRIA DAU Ges.m.b.H. & Co. KG

Tel: +43 (0) 3143 / 23 51- 0 [email protected]

- USA DAU Thermal Solutions Inc.Phone: +1 519 954 0255

[email protected]


A typical reset IC contains a comparator with an integrated voltage

reference and a time-delay circuit. The reset circuit prevents a

processor from operating during undervoltage conditions and pro-

vides a reset pulse after power-up to initialize the processor’s regis-

ters and prepare it for operation.

Multiple vendors offer reset circuits, and the MAX809 is a typical

example of an industry-standard reset IC (Figure 1). The three-lead

circuit consumes just 12 μA and is housed in a tiny SOT23 package.

Some reset ICs also include a watchdog timer that can help initiate a

reset when the system is stuck in a loop or frozen, thus further

improving system reliability.

To choose a reset IC, several important parameters need to be con-

sidered. The most important are the reset threshold, reset timeout,

and output type. Other considerations include additional features that

may be integrated, such as a manual reset input, power-fail compara-

tor, or watchdog timer. Integrated features help reduce system cost

and give designers more options.

Selecting the reset threshold

To pick the correct reset threshold, the supply voltage tolerances

must be taken into account in addition to the tolerances of the reset

threshold itself. Figure 2 illustrates an example using a 3.3V ±5%

power supply. The maximum power supply voltage is 3.465V and the

minimum power supply voltage is 3.135V.

The figure of merit for a reset is typically the voltage listed in the

maximum value column of the electrical characteristics table, repre-

senting the highest threshold voltage possible over the full operating

range of the reset IC. When choosing a reset threshold maximum

value, make sure it is just below the minimum possible voltage of the

power supply. If it were above the minimum possible power supply

voltage, in some cases the circuit would never come out of reset.

Of course, the minimum reset threshold should also be above the

minimum operating voltage of the processor and any other devices

that depend on the reset signal for proper operation. For the example

in Figure 2, the maximum reset threshold was chosen at 10% below

the nominal power supply voltage, which is a commonly available

threshold. Reset ICs are also available with thresholds that are 5%

below the nominal power supply voltage. Today’s reset ICs are avail-

able with a wide variety of threshold voltages, such as the MAX6381-

MAX6390 family, which offers thresholds from 1.58V to 4.63V in

100mV increments. For lower voltages (such as microprocessor core

voltages), the MAX6841-MAX6845 can monitor voltages down to

0.9V.

The reset threshold chosen above is the falling threshold. The rising

threshold is higher than the falling threshold, typically by about 0.5%

of the actual threshold voltage, but reset ICs can be obtained with

more hysteresis, which is good for applications that require extra

noise rejection. If a particular reset IC has a measured reset thresh-

old of 2.9V (which is inbetween the minimum and maximum toler-

ance band in Figure 2), then the corresponding rising threshold volt-

age is 2.9 + 2.9 * 0.5% = 2.9 + 0.0145 = 2.9145V. This is enough

hysteresis for most applications. Even though typical supervisor

P O W E R M A N A G E M E N T

Using MicroprocessorSupervisory Devices

Several important parameters need to be considered

When microprocessor systems have to restart, a dedicated reset IC can improve systemreliability by ensuring proper initialization of the processor regardless of the input

voltage conditions.

By Eric Schlaepfer, Senior Member of the Technical Staff, ApplicationsMaxim Integrated Products Inc., Sunnyvale, CA

Figure 1: A reset chip such as the MAX809 requires just three pins –one to sense the supply voltage, one for ground, and the third todeliver the reset signal to the processor.

VCC VCC

GND GND

RESET RESET IN

MAX803/MAX809

VCC

RPU*

*MAX803 ONLY

MICRO-PROCESSOR

Figure 2: Reset threshold tolerances

datasheets do not provide minimum and

maximum limits on the hysteresis, an actual

IC will come very close (within about 5%) to

the expected hysteresis percentage.

It is important to ensure that the rising

threshold is below the minimum power sup-

ply voltage. In this case, 2.9145V is less

than 3.135V, so this requirement is satisfied.

If this was not the case, it would be possible

for the device to remain in reset if the actual

reset threshold was at the high end of its tol-

erance band and the power supply voltage

was at the low end of its tolerance band.

Glitch rejection is an important consideration

with any reset IC. This is a major difference

between a reset IC and an ordinary com-

parator, such as an LM339. A reset IC has

relatively slow comparator (generally 1/100th

the speed of an LM339) with predictable

glitch performance. To characterize glitch

performance, most reset IC datasheets

include a “Transient Duration vs. Reset

Threshold Overdrive” graph (Figure 3).

The graph in Figure 3 is generated by put-

ting a negative pulse on VCC with a particu-

lar threshold overdrive and transient duration

then checking the reset output to see if the

pulse triggers a reset (see Figure 4). Pulses

with transient duration and overdrive combi-

nations that cause a reset occur in the area

above the curve in Figure 3, and pulses that

do not cause a reset occur in the area below

the curve. A typical reset IC has glitch per-

formance optimized to reduce nuisance

glitches while remaining sensitive to fault

conditions that require action.

Monitoring multiple voltages

Often it is necessary to monitor more than

one voltage in a system. Instead of using

multiple reset ICs that each monitor one volt-

age, you can select a single reset IC that

can monitor many voltages. Devices such as

the MAX6710 and MAX16055, for example,

can reduce system cost by integrating all the

voltage monitoring into one device. The

MAX6715-MAX6729, MAX6734-MAX6735,

and the MAX16000-MAX16007 also inte-

grate a watchdog timer and additional fea-

tures (manual reset, power fail comparator,

and others) into a single space-saving pack-

age. These devices can monitor from four to

eight voltages and come in packages as

small as a 5-lead SOT23 to a 24-contact

QFN package.

Choosing the reset output

The reset output of a reset IC is available in

either a push-pull or open-drain style, and

also with either active-high or active-low

polarity. The push-pull reset output can sink

and source current, with the logic-high volt-

age relative to the input VCC. The open-

drain output can only sink current and

requires an external pullup resistor. Use an

open-drain output if the microprocessor has

a bidirectional reset pin or if multiple outputs

need to be connected to the master reset

signal (wired-OR configuration). The pullup

resistor can be connected to a voltage high-

er than the VCC of the reset IC in case level

shifting is required.

The push-pull output does not need the

pullup resistor but also does not support

bidirectional reset pins, level shifting, or

wired-ORing. The output structure generally

can sink much more current than it can

source: sink currents can approach several

milliamps while source currents generally are

limited to several hundred microamps.

The pullup resistor value (for open-drain out-

puts) is limited by the ability of the reset IC

to sink the current and the connected input’s

parasitic capacitance, which acts with the

resistor to create an RC response that may

be unacceptably slow. Another factor is the

leakage current of all the connected inputs

and open-drain outputs which can cause a

voltage drop across the pullup resistor. This

voltage drop may prevent the voltage from

exceeding the VIH of a connected input. A

good compromise for most applications is


Figure 3: Transient duration vs. reset thresh-old overdrive

Figure 4: Negative pulse used to measurecomparator response

VARSI, d.o.o., Stegne 35, 1521 Ljubljana, Slovenia, EUTel + 386 1 500 31 80, Fax + 386 1 500 32 [email protected], www.varsi.si

manufactured in EU

VARSI – EU metal oxide varistor (MOV)manufacturer, specialized in High EnergyVaristor segment, is offering standard and custom tailored solutions. Our more than30 years of experience give us the abil-ity to create solutions for the overvoltage protection of sensitive electronic systems in the variety of applications – wind and solar energy sources, transportation, surge protection devices (SPDs)…

80 mm varistor disc (MOV) with thermal disconnection and remote signalisation –your overvoltage solution for thewindmill applications.

Low voltage (40V, 50V, 60V) square shaped high energy varistors S40 (33mm x 33mm), 40kA, withstanding high surge impulses for Class I and Class II (IEC 61643-1) – your overvoltage solution for surge protection devices (SPDs) in renewable energy sources, especially in small size photovoltaic systems (micro inverters).

VARSI – solution provider with a personal touch

World's premier presentation

80mm VARISTOR discs WITH thermal DISCONNECTION and remote SIGNALISATION and LOW VOLTAGE 40kA varistors

10kΩ. Applications that need to drive a large

number of inputs or need a fast edge benefit

from decreased resistance, and applications

that need to reduce the power dissipation

benefit from increased resistance.

Power-up behavior of a reset IC is quite

important. With an active-low reset, the IC

must be able to keep the output low during

power up, and with an active high reset, the

output must be pulled to VCC. Active-low

outputs typically sink current when VCC

exceeds the VTH of the low-side MOSFET.

This current is in the low microamp range

but increases as VCC exceeds approximate-

ly 0.7V. Figure 5 shows the waveform that

can be observed on power-up: the small

“hump” on reset occurs because the reset

output cannot sink much current at that time

and does not overpower the pullup resistor

until VCC increases beyond 0.7V. To remove

the “hump” on reset, use a reset IC with a

push-pull output and connect a pulldown

resistor.

Selecting the reset timeout

The reset timeout is generally not critical.

However, some microprocessors require a

minimum pulse width to generate a proper

reset. Check the minimum reset timeout

period on the reset IC datasheet electrical

characteristics to make sure that this

requirement is satisfied.

Some reset ICs integrate a manual reset

(MR) input. This logic input (often with a

built-in pullup resistor) allows an external

switch or logic to trigger a reset pulse. Typi-

cal of such circuits, the Maxim MAX6443-

MAX6452 have an extended setup delay on

the MR input. This requires that MR be

asserted for a minimum period of time (usu-

ally about 6 seconds) before generating a

reset pulse (see Figure 6).

Such an approach allows designers to “hide”

a hard reset function in some other button,

such as a power button or other front panel

control (thus eliminating the need for a sepa-

rate button). Technical support can instruct a

customer to push and hold the control to

generate a hard reset. This approach avoids

unsightly and easily damaged “pinhole” style

hard reset buttons. Variations of this basic

part include the MAX6453-MAX6456 family,

have two separate outputs: one which trig-

gers immediately, and another which triggers

after the setup delay.

Power fail comparator

The power fail comparator is available in

some reset ICs and acts as an auxiliary volt-

age monitor. The output of the comparator

connects to a separate pin and does not trig-

ger a hard reset. This comparator usually

requires an external resistive divider to set

the comparator threshold. It is useful for

monitoring batteries or other supply voltages

but it can be used for any circuit that needs

a comparator. Hysteresis of the comparator

is usually quite small but this can be con-

trolled with an external feedback resistor.

Watchdog timer

Many reset ICs integrate a watchdog timer

function. The watchdog timer acts as a “last

resort” software recovery feature by generat-

ing a hard reset if an unrecoverable software

error occurs. A processor I/O line is typically

connected to the reset chip’s watchdog input

(WDI) and clears the internal watchdog timer

on every transition. If no pulses from the

processor are received by the IC, the inter-

nal watchdog timer eventually expires and

the reset output asserts, causing the proces-

sor to reset. Many modern processors and

microcontrollers have a built-in watchdog

timer, but it is possible for runaway software

to disable built-in watchdog timers. An exter-

nal watchdog timer is more difficult to disable

accidentally, and thus can improve system

reliability.

Some watchdog timers provide an initial

timeout, like the MAX6730-MAX6735, that is

much longer than the normal watchdog time-

out. This is useful to allow long boot routines

or flash upgrade procedures to complete

before automatically turning on watchdog

functionality. Other watchdog timers provide

a windowed function, like the

MAX6752/MAX6753, which trips reset if the

pulses from the processor come in too

quickly.

Watchdog timers can often be disabled to

aid debugging or to prevent spurious resets

during routines that do not clear the watch-

dog timer. One common mechanism is to

allow WDI to go to a high-impedance state

(as in the MAX6316-MAX6322). Other ICs

employ a separate logic-level input or inputs

to perform this function (including the

MAX6369-MAX6374). However, this feature

must be used with caution—it must be diffi-

cult or impossible for runaway software to

disable the watchdog timer, otherwise sys-

tem reliability will suffer.

It can be useful to trigger a nonmaskable

interrupt (NMI) when the watchdog times out

instead of generating a hard reset. Some

watchdog timer ICs provide a watchdog out-

put (WDO) which can be directly connected

to the NMI input on the processor. This is dif-

ferent from a standard watchdog timer

because the reset output does not assert

when the watchdog timer expires. An IC with

this feature, such as the MAX706, can be

converted back to a regular watchdog timer

by connecting the WDO to the manual reset

input (MR).

Conclusion

Microprocessor supervisory devices improve

system reliability in a number of different

ways: by monitoring supply voltages for fault

conditions, by providing predictable and

repeatable system reset signals, and by

detecting out-of-control software.

www.maxim-ic.com

P O W E R M A N A G E M E N T


Figure 6: Manual reset with extended setupdelay

Figure 5: Active-low open-drain reset power-up waveform

means exploring new paths.

change

SoftwareBoards Microcontrollers DSPs ForumVirtualisation

Time for embedded. Time for the future.

The entire range of key technologies and forward-looking innovations. Visit electronica

and the embedded Forum and experience their incomparable signifi cance to the entire

industry—with today’s solutions for tomorrow’s applications.

Parallel event: hybridica. Trade fair for hybrid-component production. www.hybridica.de

get the whole picture

24th International Trade Fair

New Munich Trade Fair Centre

09–12 November 2010

Register online and enjoy the benefi ts: www.electronica.de/en/tickets

electronica 2010embeddedwww.electronica.de/embedded

100378 e2010Emb-st_BodosPowSysE_210x297_E.indd 1 30.06.10 10:52


Ethernet currently dominates the world of office networking and is also

the choice for both factory and home. Simplicity and field-proven

open standardisation have significantly lowered the Cost of Ownership

which Automotive manufacturers are keen to exploit. Volumes of

scale in the office and consumer market, supported by a magnitude of

Ethernet vendors, have driven pricing levels far lower than any ‘cus-

tom’ designed protocol.

Initial applications for automotive Ethernet now routinely include On-

Board-Diagnostics (OBD); for Diagnostics and Software Download of

internal ECU static memory and hard disk(s). The adoption of Ether-

net will accelerate with the introduction of standardised IP Diagnos-

tics interface, as specified in ISO 13400.

The choice of a standard Ethernet CAT5 cable to interface to the

OBD port will allow service centres to seamlessly interface to either a

lap top or the Intranet when performing vehicle management diag-

nostics. Memory downloads for software updates also benefit from

the increased speed of 100Mbps (or 1000Mbps Gigabit Ethernet), full

duplex bandwidth available by the network. As the demand for

increased processor intensive functionality in a car grows, so does

the required memory. If present methods continue to be employed,

software download times will significantly increase. Off road time

results in direct additional servicing costs, due to the need to supply

the owner with a temporary replacement vehicle and the additional

time associated with workers on the clock.

Implementing Ethernet at the OBD port now allows the car to inter-

face to the World Wide Web, creating endless possibilities. For

example, the port can easily be plugged into a wireless unit for

remote diagnostics or downloads for in-car navigation, video or

music, all from the comfort of the owner’s home!

Moving forward, new ‘real time’ Ethernet AVB (Audio-Video Bridging)

can also offer high performance infotainment network solutions, again

with nearly endless possibilities.

The challenges for Ethernet to also become the de-facto automotive

bus are not necessarily unique. By combining Ethernet’s proven abili-

ty to reliably transfer high bandwidths of data (home/office) with real-

time performance in extreme environments (industrial control), the

basis of many automotive needs is formed.

Thermal and EMC Performance

The Industrial Control market has helped to prove that Ethernet net-

works are able to deliver robust performance in extreme conditions.

Extended temperature ranges, heavy vibrations, high EMC radiation

and dusty or wet surroundings are typical in many of these applica-

tions. Raising the ambient temperatures ‘under the hood’ over the

common +85°C will not cause thermal issues for Ethernet devices,

due to low power consumption and package selection. For example,

Micrel’s KSZ8041NL AM, AECQ-100 Single-Port Fast Ethernet PHY

solution, consumes a mere 175mW inside a thermally enhanced

5mm x5mm MLF® package. The KSZ8041NL family also offers a

Military specification variant which supports ambient temperatures of

up to 125°C.

Due to the demands of the industrial and automotive markets, many

newer Ethernet devices offer significantly improved ESD (Electro-Sta-

tic Discharge) performance. This is a major shift in emphasis from

original office applications where ESD rating was not considered of

major concern. For example, Micrel’s KSZ8041 PHY and KSZ8851

Controller families have a HBM (Human Body Model) ESD rating of >

6KV. The product’s evaluation board has also been shown to provide

> 9kV contact and >16.5kV air ESD ratings, without the need for any

external over voltage protection devices. This surpasses general

automotive vendor Electromagnetic Compatibility (EMC) require-

ments such as those set forth by BMW Group Standard GS 95002.

Cables & Connectors

Currently no standard Automotive Ethernet connector or cable exists.

The standard Ethernet RJ45 connector and CAT5 cable has proved

very robust and remains extremely popular in other applications,

including industrial. However, existing vendor specific connectors

and wiring looms are likely to be adopted in automotive applications,

at least initially. The Ethernet PHY (transceiver) is flexible enough to

utilise such connectors and cabling without any significant degrada-

tion to performance. Standard CAN cable exhibits similar character-

istics to unshielded, twisted pair CAT5. Thorough testing has proven

long term error-free transmission of Ethernet over in excess of 100m

CAN cable. The major difference between the two is that CAN cable

is only partially specified and does not provide a controlled imped-

ance or twist ratio. As a consequence, EMC behaviour and signal

integrity cannot be guaranteed, thus making CAN cable generally

unsuitable for high speed data transfer. CAN cable is currently used

for Ethernet On Board Diagnostics (OBD) and flash updating. Here,

the lines can be disabled during normal driving and only active in the

repair shop or production plant.

A cable example for high speed data transfer such as LVDS, USB

and Ethernet in automotive applications is the Leoni cable, for exam-

ple the twisted pair Dacar 503. This cable is shielded with controlled

100ohm impedance and qualified up to 1Gbps, giving a performance

similar to CAT6 rather than CAT5. It is not actually twisted pair but a

four-wire twist known as ‘Stern-Vierer’ (translated as Star Four Wire).

To enhance reliability Cable diagnostics technology, such as Micrel

A U T O M O T I V E

Ethernet In Fast ForwardAutomotive Using Ethernet as Physical Layer Data Bus

Ethernet has been officially be added to the list of automotive networks, such as CAN, LIN,FlexRay and MOST. But why did Ethernet make the list and for what specific automotive

applications? Most important of all, can it really meet the challenges of Automotive?

By Mike Jones, Senior FAE, Micrel Inc.

www.bodospower.com Novewww.bodospower.com

LinkMD®, goes beyond Ethernet-defined standards to provide a solu-

tion to such problems. LinkMD® cable diagnostics utilize time

domain Reflectrometry (TDR) to analyze each twisted pair for com-

mon cable problems, such as open circuits, short circuits and imped-

ance mismatches.

There is an alternative to copper cabling that comes in the form of

Plastic Optical Fibre (POF). Car manufacturers are already very

familiar with this physical media as it is deployed in MOST networks.

The same 1mm LED POF technology from MOST (including new

MOST-150) can also be used for 100Mbps Fast Ethernet transmis-

sion with reach of 100metres. POF is extremely robust, lightweight

and like all fibre, totally immune to electro-magnetic noise as it emits

no radiation.

Power Consumption

Power consumption of automobile electronics is becoming increas-

ingly more critical and significant factor in fuel efficiencies. It is

important to understand both how and where the power is dissipated

in Ethernet circuits to ensure optimum design. In any Ethernet

device, the major power dissipation is from the PHY transceiver.

Typically, most PHY designs are current-mode drivers and power is

consumed both internally to the PHY and externally in the trans-

former, as shown in Figure 1.

Ethernet datasheets commonly publish the device only current con-

sumption, Iphy. In which case, to calculate the total circuit current

consumption the designer must add typically around 40mA per

100Base-TX or 70mA per 10Base-T PHY for dissipation in the trans-

formers, Itrans. This power consumed externally in the transformer is

significant and typically accounts for around 30 to 50 percent of the

total PHY circuit current consumption.

Micrel’s new generation KSZ8051 Fast Ethernet PHY families differs

by utilizing a voltage-mode driver, along with patented DSP-

enhanced mixed signal architecture, to offer the lowest power con-

sumption in the industry. Device only power consumption is similar to

other leading Ethernet PHYs at sub 50mA. However, no current is

consumed externally in the transformer, due to the voltage-mode

driver design. Hence, a saving of up to 50 percent total circuit power

consumption is achieved over competing solutions.

PCB layout design is also simplified and real estate minimised by the

unique integrated line termination offered by the KSZ8051 Ethernet

PHY as demonstrated in Figure 2.

Figure 1: Ethernet PHY Circuit Depicting Power Dissipation in Cur-rent-Mode

You don’t believein poltergeist...


AECQ-100 grade qualified KSZ8051 parts are expected to be avail-

able from Micrel in the first half of 2011.

To investigate further how to reduce power consumption, one needs

to understand how an Ethernet link operates.

When analyzing a network one realizes that there are long quiet peri-

ods followed by relatively short bursts of traffic. During these quiet

periods, one may expect the Ethernet power consumption to signifi-

cantly drop, however, this turns out to not necessarily be accurate.

Both 1000Base-TX and 100Base-TX are designed so that the link

partners are continually ‘synchronized’ to each other. To enable this,

when no traffic is being transmitted, the PHY will automatically send

out IDLE symbols (11111 5B code), as shown in Figure 3 below.

As a consequence, during any quiet period, the PHY transmitter is

still operating in a manner similar to full traffic and will therefore con-

sume a similar amount of power. The IEEE recognises this ineffi-

ciency and formed a task force whose mission it is to reduce power

consumption during periods of low utilization. This task force IEEE

802.3az is commonly known as Energy Efficient Energy. The tech-

nique, known as Low Power Idle (LPI), will disable parts of the PHY

transceiver that are not necessary, whilst still maintaining the link

integrity.

If the Ethernet PHY is, ‘not in use’ then a software or hardware power

down mode is usually available. However, even in this low power

state the device will still consume ‘in the order of a mA’, which for

automotive applications is unacceptable. Hence, it is advised to fully

power down the circuit during such periods. For example the diag-

nostics and software download circuit only need be powered when

being serviced by the garage.

Another area where current consumption can be reduced for automo-

tive applications is found in the transmit current drive strength. The

IEEE802.3 specification is designed to always provide the capability

of operating up to a minimum 100m of CAT5 grade cable. As a con-

sequence, the PHY output drive strength is fixed at this criterion, con-

suming maximum power, independent of the actual length of cable

connected. Automotive networks will never require the capability of

100m cable reach and can guarantee a much shorter length. Con-

sider that one can reduce the PHY transmitter current drive from the

standard +/-1V amplitude of the 100Base-TX signal down by up to 50

percent and still operate error free over a 20m reach. The transmitter

current drive on Micrel Ethernet devices can be varied either via

internal software registers or by modifying the recommended resist-

ance to ground at pin ‘REXT’ (see datasheet for specific value). The

output drive strength will vary inversely proportional to the resistance.

This method yields significant reduction of both current consumption

and EMI (Electro-Magnetic Interference) Radiated Emissions.

Topology – Ring or Star?

Traditional Ethernet networks usually implement a ‘star’ configuration,

where a multi-port switch provides point-to-point links to other nodes.

However, industrial networks are usually based on a ‘ring’ configura-

tion, which eases the logistics of cable installation. Reductions in the

required cabling of an Ethernet ‘ring’ network provides benefits wel-

come to the automotive industry. Less cabling means weight reduc-

tions which has a direct impact on fuel efficiency and hence, overall

costs. The basic building block in a ‘ring’ network is the 3-Port

switch, for example the automotive AECQ-100 qualified

KSZ8873MLL AM from Micrel

The introduction of Ethernet into the car will most likely favour a ‘ring’

and ‘star’ hybrid topology approach. Figure 4 illustrates a possible

automotive Ethernet network.

Each physical media interface can independently be implemented

using copper cabling or POF. Here, a central Gateway will interface

to the On-Board Diagnostics (OBD) port and other available networks

within the car, such as MOST or FlexRay. The Gateway unit can

then act as bridge to the Head Unit, further connecting other systems

such as Navigation, Vehicle Computer and Rear Seat Entertainment

over a ‘ring’ or ‘star’ topology.

Figure 4: Depiction of a Basic Automotive Ethernet Ring Network

Figure 3: 100Base-TX Idle Pattern

A U T O M O T I V E

Figure 2: Integrated Line Termination of the KSZ8031/51 PHY Family

www.bodospower.com

Managing an Ethernet Ring

Unlike a MOST networks, it is in fact usually forbidden to configure

Ethernet as a true ‘ring’. Any loops within an Ethernet network will

result in the duplication of packets that are forwarded in endless

loops, quickly degrading the bandwidth and efficiency of a network.

To manage the ring, usually protocols such as Spanning Tree are

implemented to ‘break’ one of the links and enable again if a link fault

is detected elsewhere in the ring.

However Micrel’s Automotive switches; both the 3-Port KSZ8873MLL

AM and 5-Port KSZ8895MLU offer a unique feature to allow a true

Ethernet ring to be implemented without the need for management;

MAC Address Source Filtering.

A hardware mechanism of ‘Learning’ and ‘Forwarding’ is utilized by

all common Ethernet switches today. A switch will ‘learn’ and then

store the ingress packet MAC Source Address and the associated

port in a ‘Forwarding’ Table. A port forwarding decision is then made

by looking up the packets MAC Destination Address in the ‘Forward-

ing’ Table. If a match is found, then the packet is forwarded to the

associated port in the table entry. Failure to find a match results in

Figure 5: Illustration of a True Ethernet Ring Network

International Exhibitionwith Workshopson ElectromagneticCompatibility (EMC)15-17 March 2011Messe Stuttgart

EMV 2011Stuttgart

Your chance to shine......at the EMV marketplace

Further Information: +49-711-61946-0 or [email protected]

wwwe-emc.com

the packet being broadcast to all egress ports except the port it

arrived at. With this mechanism, the MAC Source Address is only

ever learnt and never used in the decision making when forwarding

the packet. MAC Address Source filtering enables the filtering of

packets based on the MAC Source Address (instead of the MAC

Destination Address). Now the switch can detect and filter (drop) any

packet that arrives with a MAC Source Address matched to the local

processor MAC Address. As a consequence, packets are always

removed from the ring following one complete loop. Figure 5 depicts

how this can work.

Switch #1 receives broadcast packet at port 3 (processor) with

Source Address 1

Packet is forwarded along the ring until it arrives back at Switch #1

Switch #1 drops packet, using MAC Source Address filtering feature

MAC Source Address Filtering also offers the additional benefit of

single fault redundant switchover, by exploiting packet forwarding in

both a clockwise and anticlockwise direction around the ring. For

more details see: ‘Unmanaged Redundant Ring – A White Paper’.

The Future

Ethernet’s unquestionable success in the Industrial networking sector

has proven reliability and quality in an extreme environment. This

marriage of industrial strength and consumer technology drive pro-

vides the perfect physical layer solution for automotive. Ethernet can

successfully bridge the gap between lengthy vehicle design cycles

and today’s fast moving IP world.

There is nothing complex about Ethernet technology overall; it is sim-

ple, proven and open ? the ver reason for its success. Cost is a cru-

cial factor in any market and Ethernet has consistently demonstrated

the lowest cost of ownership of any network.

Today, Ethernet has already emerged inside the car to provide an IP-

based standard interface for diagnostics and software downloading.

The next step is for Ethernet to form the backbone of the next gener-

ation automotive multi-media networks, carrying ‘live’ traffic. New

standards such as IEEE 802.3AVB (Audi-Video Bridging) initially

defined for Digital AV Home networking are being adapted to support

the same real-time services in the car. Following this, the ultimate

goal would to converge other bus systems inside the car into a single

common bus; Ethernet.

The following Automotive Qualified Ethernet devices are currently

available from Micrel.

For further details on Micrel Ethernet Solutions go to:

http://www.micrel.com/page.do?page=product-info/ether_over.jsp

Note: MOST is a registered trademark of Standard Microsystems

Corporation. LinkMD is a registered trademark of Micrel Inc.

www.micrel.com

A U T O M O T I V E

Table 1: Micrel Automotive Qualified Ethernet devices

KSZ8041NL AM Single-Port Fast Ethernet Physical Layer Transceiver KSZ8893MQL AM 3-Port Managed Fast Ethernet Switch with MII / RMIIKS8873MLL AM Enhanced 3-Port Managed Fast Ethernet Switch with MII KSZ8842-PMBL AM 3-Port Managed Fast Ethernet Switch with PCI KSZ8895MLU Enhanced 5-Port Managed Fast Ethernet Switch

You receive more information at Tel. +49 711 61946-828 or [email protected]

Products and Solutions,

Innovations and Trends

y Control Technology

y IPCs

y Drive Systems and Components

y Human-Machine-Interface Devices

y Electromechanical Components and Peripheral Equipment

y Industrial Communication

y Industrial Software

y Interface Technology

y Sensor Technology

ElectricAutomationSystems and ComponentsExhibition & Conference23 – 25 Nov. 2010Nuremberg

SPS/IPC/DRIVES/

Experience at Europes # 1 platform for electric automation...

www.mesago.com/sps

International Exhibition

& Conference for

POWER ELECTRONICS

INTELLIGENT MOTION

POWER QUALITY

17 – 19 May 2011

Exhibition Centre Nuremberg

2011

Mesago PCIM GmbH – 0711 61946-56 – [email protected]

PowerfulThis is the right place for you!

Power for Efficiency!


With what seems to be sustainable growth rates in excess of 40%

per annum, the expansion of solar energy systems remains phenom-

enal.

However, insufficient long term data is available to show whether the

sensitive electronics in the solar inverter, are able to survive typical

exposure to atmospheric influences and voltage fluctuations on both

the DC input and the AC output stage.

This article will attempt to show the current protection concepts cur-

rently used, the typical aging effects of the individual devices and

how these can be optimized to exceed their working lifetime.

Basic Inverter Block Diagram

On the left the DC Input from the Solar Module (Panel) enters the

inverter. Overvoltage Protection (OVP) between the module terminals

protects the Maximum Power Point Tracker (MPPT) and the Solar

Panel itself. The DC/AC inverter needed to convert the DC voltage

from the module into AC voltage for feeding into the mains supply

and filtering to reduce electro-magnetic interference (EMI) are them-

selves protected against disturbances originating from the mains by

the OVPs.

The simplified diagram does not show coarse protection devices that

could be installed externally to the inverter. However, the coordination

between fine and coarse protection should be considered in the

design / installation stage.

Component Selection

Typically Metal Oxide Varistors (Varistor) with a DC rating upto 1000V

are used for the DC input protection occasionally in combination for

Gas Discharge Tubes (GDTs). The AC output protection is also Varis-

tor based, however optimized for the network voltage (i.e. 300Vrms)

again with a possibility of a combination with GDTs.

Metal Oxide Varistors

A Metal Oxide Varistor is a voltage dependant resistor. The clamp

voltage of a varistor is defined by its voltage rating and its current

handling capability. Through-out its voltage / current characteristic,

the Varistor is actually conducting. In its normal, high resistance

mode, a leakage current that can be measured in the μA range is

always present. In the over-voltage, low resistance, mode in which

the Varistor is conductive, currents, measured in amps or for short

durations in kilo-amps, pass through the Varistor.

P R O T E C T I O N

Enhanced Over-Voltage Protection of Solar Installations

Through-out its voltage / current characteristic, the Varistor is actually conducting

Varistors offer a cost effective solution for the protection over Solar Invertors againstover-voltages. Thermally Protected Varistors from TDK-EPC can help to reduce down

times and to optimize returns on investment.

By David Connett, Director IC Reference Design, EPCOS AG. A Group Company of TDK-EPC Corporation

Figure 1: Solar panel

Figure 2: Basic Inverter Block Diagram

P R O T E C T I O N


Gas Discharge Tubes

As its name suggests, the gas tube is tube

filled with a gas. If a voltage exceeds the

breakdown characteristics of the gas, the

gas itself will ionize and will form a conduc-

tive path across which an arc is formed

between the charged terminals. However, as

the gas has a finite ionisation time, as the

voltage rise time increases so does the

breakdown voltage of the gas. For example,

a typical Gas Tube with a DC sparkover volt-

age of 230V at 100V/s, its maximum firing

voltage at 1.000V/μs could be closer to

600V. This firing voltage is commonly

referred to as the impulse sparkover voltage

or dynamic response.

Considerations

Varistors due to their high current handling

capabilities and cost-performance ratio make

ideal protection components. However, as

with most semi-conductor based technolo-

gies, Varistors are subject to degradation

(ageing) when exposed to repetitive pulse of

low amplitude. The degradation takes the

form of an increase in leakage current of the

Figure 3: Metal Oxide Varistors

Figure 4: Gas Discharge Tubes

Figure 5: Common Protection Concepts

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.

Connect with us:

P R O T E C T I O N


Varistor which can result in a phenomena

called “thermal runaway”. In extreme cases,

the thermal overload can result in a short cir-

cuit and rupture of the Varistor. This point

has also been reviewed in a number of inter-

national standards (UL, IEC) and the net

result is that thermal surveillance of varistors

will need to be considered in the future.

Common Protection Concepts DC Input

For the DC Input Varistors, remain the most

favoured primary protection. How they are

deployed however is varied. A few examples

follow – taking as an example an input upto

1000Vdc.

In Figure 5/1 a single Varistor is placed

between the PV + and PV - terminals from

the module. Rated at 1000Vrms, this 20mm

varistor would have a maximum DC rated

voltage of 1414Vdc and a clamp voltage of

2970V at 100A.

Figure 5/2 utilizes two varistors placed in

series. Typically two 550Vrms (745Vdc)

rated Varistors are deployed hereby achiev-

ing the same rating as Figure 5/1, with the

advantage of a lower maximum clamp volt-

age of 2710V at 100A.

Figure 5/3 deploys the same concept as Fig-

ure 5/2 but with a separation to system earth

provided by a GDT. This solution can assist

with the compensation of ageing of the

Varistors mentioned above in the normal

operating conditions. However, the extin-

guish characteristics of the GDT must be

taken seriously in account to avoid that the

GDT remains in a conductive mode (arc or

glow mode) after firing.

AC Output

Although, in this application, we are supply-

ing AC power to the Grid, the connection

provides a source of over-voltage and over-

current, not to mentioned other disturbances

such as EMI Noise. In this as such the pro-

tection of the Inverter can be compared to

the input stage of a standard power supply.

The typical concepts are similar to those

shown above. The main difference being the

selection of Varistor ratings to suit the AC

network voltage. Here as a general rule

300Vrms or 320Vrms rated Varistors are

common for European line voltages of up to

240Vrms.

Specific Requirements

All of these solutions fulfil the intended

requirements of providing over-voltage pro-

tection to the inverter, Figures 1 and 2 do

not address the previously mentioned prob-

lems of ageing. The newly published IEC

62109-1 “Safety of power converters for use

in photovoltaic power systems – Part 1: Gen-

eral requirements“, does not specifically

address this point. However, to draw some

analogies from other IEC standards, the lat-

est revision of IEC 60950-1 (Information

Technology Equipment) provides provision

that only Varistors, qualified against IEC

61051-2-2 and which meet the requirements

of Annex Q (IEC 60950-1), may be used as

the primary protector. In addition, for protec-

tion of Varistors, overcurrent protection or a

similar interrupting device / means is

required to be supplied in series with the

Varistors to ensure that in the case of ther-

mal runaway that the Varistor itself does not

become a safety issue.

Installation of External Overvoltage Pro-

tection

In general, the insurers of solar installations

demand that Overvoltage Category 2

(coarse protection) SPDs are installed when

the power of an insured installation exceeds

50kW. Below this rating, there are no clear

guidelines and hence it would seem that the

cost of protection versus the cost of replace-

ment means does not justify additional exter-

nal protection and that the fine protection

inside the inverter should be sufficient.

Enhanced Protection in Inverters

As a result of the statements from insurance

companies and the trend to extend warranty

periods, the demand for improved protection

which covers the previously mentioned prob-

lems of ageing coupled with a means of indi-

cating failure will increase.

Thermally Protected Varistors

Awareness of the problems associated with

ageing and thermal runaway have been

addressed by the major producers of Varis-

tors through so-called thermally protected

varistors. These devices feature a thermal

surveillance of the Varistor which can result

in the disconnection of the Varistor to the

supply when a threshold temperature is

exceeded. Through these devices, e.g. the

ETFV20K1000 from TDK-EPC, (ETFV –

EPCOS Thermal Fuse Varistor) some of the

harmful effects of aging of varistors have

been eliminated while still providing a cost

effective solution to Inverter designers. In

addition, since these devices feature an

external monitoring of the status of the pro-

tection through a LED. If the LED is no

longer illuminated, then the user should be

instructed to contact the service department

for a prompt replacement of the thermally

protected varistor otherwise the warranty is

invalidated and so that the returns on invest-

ment can be optimised. In the case of

replacement of the “opened” Varistor, the

thermally protected ETFV Varistor could be

hard-wired via screw terminals on the pcb

and not soldered onto the pcb using conven-

tional through-hole processes – this will

leads to simple and effective replacement of

the ETFV.

Summary

Varistors offer a cost effec-

tive solution for the protec-

tion over Solar Invertors

against over-voltages but

they themselves are the

subject of ageing that can

reduce their effective life-

time and make them a

safety hazard. Thermally

Protected Varistors from

TDK-EPC can help to

address this point and can

be easily replaced follow-

ing operation helping to

reduce down times and to optimize returns

on investment.

[email protected]

www.epcos.com

Bodo´s Power Systems® November 2010 www.bodospower.com48

Figure 6: Thermal protected Varistor Block Diagram

Figure 6: Thermal protected Varistor Device


In this article we discuss several key considerations for designers

wanting to quickly get eGaN-based systems to market.

Gate drive requirements

To determine the gate drive circuit requirements, and how they differ

from silicon MOSFET drivers, it is necessary to compare their device

parameters (see table 1). The three most important parameters for

eGaN FETs are, (1) the maximum allowable voltage, (2) the threshold

voltage and, (3) the “body diode” voltage drop. The maximum allow-

able gate-source voltage of 6V is low in comparison with silicon. Sec-

ondly, the threshold is also low compared to most power MOSFETs,

but does not suffer from as strong a negative temperature coefficient.

Thirdly, the “body diode” forward drop can be a volt higher than com-

parable silicon MOSFETs.

Gate pull-down resistance

A great advantage offered by eGaN FETS is their fast switching

speed. However, the higher di/dts and dV/dts that accompany this not

only require a layout with less parasitic capacitance, resistance, and

inductance, but also cause some new considerations for the gate

driver. Let’s consider a half- bridge with a high dV/dt turn-on of a

complementary device as shown in Figure 1. The ‘Miller’ charge cur-

rent flows from the drain (switching node) through CGD and CGS to

source as well as through CGD to RG (internal gate resistance) and

RSink (gate driver sink resistance) to source. The requirement for

avoiding dV/dt (Miller) turn-on is given by:

CGD x dV/dt x (RG + RSink) x (1 – e- dt/α) <VTH

Where α is the passive network time constant (RG + RSink) x (CGD +

CGS) and dt is the dV/dt switching time. Thus to avoid Miller turn-on,

it is necessary to limit the total resistance path (internal gate resist-

ance RG and external gate drive sink resist-

ance RSink) between the device gate and its

source. To be safe, a gate drive pull-down

resistance of 0.5Ω or less is recommended

for higher voltage eGaN devices.

Gate pull-up resistance

Because the total Miller charge (QGD) is

much lower for an eGaN FET than for a sim-

ilar on-resistance power MOSFET, it is pos-

sible to turn on much faster. As stated

above, too high of a dV/dt can actually

reduce efficiency by creating shoot-through

during the ‘hard’ switching transition. It would therefore be advisable

to have the ability to adjust the gate drive pull-up resistance to mini-

mize transition time without inducing other unwanted losses. This

also allows adjustment of the switch node voltage overshoot and

ringing for improved EMI. The simplest solution is to split the gate

pull-up and pull-down connections in driver and allow the insertion of

a discrete resistor.

Gate drive dead-time matching

eGaN FET reverse bias or “body diode” operation has the benefit of

no reverse-recovery losses. This advantage, however, can be offset

by the higher “body diode” forward voltage drop. The diode conduc-

tion losses can be significant, especially at low voltages and high fre-

quencies. Unlike diode reverse recovery losses; these conduction

losses can be minimized through proper dead-time management that

minimizes the “body diode” conduction interval.

T E C H N O L O G Y

Driving eGaNTM FETs Both gate and Miller capacitances are significantly lower

As enhancement mode gallium-nitride-on-silicon transistors (eGaNTM) gain wider acceptance as the successor to the venerable - but aged - power MOSFET, designers havebeen able to improve power conversion efficiency, size, and cost. eGaN FETs, however,

are based on a relatively new and immature technology with limited design infrastructureto quickly design and implement products.

By Johan Strydom PhD, Director of Application Engineering, Efficient Power Conversion Corporation

Table 1: Comparison between 100V Si MOSFETs and eGaN™ FETs

FET type Typical 100 V Silicon 100 V eGaN™ Maximum gate-source voltage +/-20 V +6 V /-5 VReverse ‘body diode’ voltage ~1 V ~1.5-2.5 V Gate threshold 2 V – 4 V 0.7 V - 2.5 VdV/dt capacitance (Miller) ratio QGD(50 V)/QGS(VTH) 0.5-0.8 1.1 Internal gate resistance >1 <0.6 Change in RDS(ON) from 25°C to 125°C >+70% <+50% Change in VTH from 25°C to 125°C -33% -3%Gate to source leakage few nA few mA Body diode reverse recovery charge high none Avalanche capable Yes not rated

Figure 1: Effect of dV/dt and requirements for avoiding Miller turn-on

Silicon gate drivers / controllers tend to have effective minimum

dead-time around 20ns (+/-10ns) for low voltages, increasing with

bus voltage to around 400ns (+/- 100ns) for 600V drivers. With eGaN

FETs, both gate and Miller capacitances are significantly lower than

equivalent silicon devices, leading to shorter delay and switching

times. These allow for much tighter dead-time control which would be

beneficial in reducing “body diode” conduction loss. A reduction of

dead-time between half and one-fourth the above values, with a simi-

lar reduction in the variation, would be preferred.

Gate drive supply regulation

The current maximum gate voltage limitation of 6V on the eGaN FET

restricts the allowable gate drive supply range, and requires at least

some form of supply regulation – especially on the high side. A post

bootstrap diode regulator eliminates the effect of changes in the sup-

T E C H N O L O G Y

Figure 2: Discrete gate-driver solution showing method for comple-mentary high-side and low-side supply voltage matching.

ABB FranceCurrent & Voltage Sensors Departement

e-mail: [email protected]

Improve drives

efficiency?

Certainly.

ABB takes environmental protection very seriously. This is a priority for every ABB company

and not least, for us at ABB Sensors, where we are proud of our ISO 14001 certification.

We also believe that wind generators represent the future of renewable energy, so we, at ABB

Sensors have, developed the ESM range of sensors especially for wind power applications.

Working closely with our wind power customers, we have optimised the ESM range in terms of

noise immunity and dynamic response and adapted other features to the requirements of

these demanding applications. www.abb.com

ply due to the dead time variation with a higher “body diode” voltage

drop. For complementary driven half-bridge with minimal dead-time, a

diode matching network shown in Figure 2 can be used.

Layout considerations

Gate drive loop inductance

The maximum allowable gate voltage of 6V is only one volt above the

recommended 5V drive voltage. This requires an accurate gate drive

supply, as well as a limited inductance between the eGaN device and

gate driver as the inductance can cause an overshoot on the gate.

Effect of common source inductance (CSI)

The addition of CSI effectively reduces efficiency by inducing a volt-

age across the CSI that opposes the gate drive voltage, increasing

switching times. It is therefore critical to minimize common source

inductance for optimum switching performance. Increase in CSI will

actually decrease the possibility of Miller turn-on if one accepts its

increased switching loss. This is because at the ‘hard’ turn-on of the

complementary device, the current commutation di/dt across the CSI

induces a negative voltage across the gate to help keep the device

off during part of the voltage transition. However, CSI, gate capaci-

tance, and gate drive pull-down loop now form an LCR resonance

that must be damped to avoid an equivalent positive voltage ringing

across the gate. This ringing could turn the device on again near the

end, or even after the voltage transition. Although increasing the gate

drive sink resistance can help damp this resonance, the addition of a

ferrite bead that is resistive at the resonant frequency can achieve

the same result with less increase in Miller turn-on sensitivity (Figure

3 shows the equivalent circuit and Figure 4 the conceptual wave-

forms). In short, CSI is much more important to eGaN FETs than sili-

con due to higher di/dt and dV/dt and should be minimized through

careful layout.

Suggested Layout

Given the considerations listed above, it is possible to develop a rec-

ommended layout. The layout presented depicts a half-bridge config-

uration, but following the above requirements can be applied to other

applications as well

A simple four layer PCB is presented in Figure 5. It should be noted

that the copper thickness must be maximized to limit resistive losses

and improve thermal spreading (2 oz copper on outer layers is rec-

ommended). In this example the source connection of each part is

brought underneath to act as shield (especially under the gate area)

and minimize additional parasitic CGD. The gate return connection is

made on the smaller source pad to separate gate return current and

power source current paths,– thus minimizing CSI.

Summary

EPC’s eGaN FETs give the engineer a new spectrum of performance

compared with silicon power MOSFETs. In order to extract full advan-

tage from this new, game-changing technology, designers must learn

some new techniques on how to design cost-effective eGaN drive cir-

cuitry that works on a cost-effective PCB.

www.ecp-co.com


Figure 3: Equivalent circuit showing di/dt effect of ‘hard’ turn-on ofcomplementary device

Figure 5: Suggested half bridge layout using 4-layer PCB

Figure 4: Conceptual waveforms for circuit in Figure 2 during ‘hard’turn-on of complementary device showing effect of CSI ringing

T E C H N O L O G Y

Chip Beadsfor RF applications

� WE-CBF HF serie� For fast data lines: CPU, high speed bus systems, HDD � Increased effective suppression range� 3 times higher impedance @ 1 GHz� Manufactured to stock� samples free of charge

www.we-online.com

electronica Hall B6, Stand 404


An IGBT (Insulated Gate Bipolar Transistor)

can be controlled with a high impedance,

like a tube, like the Thyratrons in the early

days of power electronics. But in contrast to

a Thyristor the IGBT can be switched off any

time like a MOSFET. At the same time the

forward voltage of the IGBT is limited like the

forward voltage of a bipolar-transistor. On

the other hand it does not need a permanent

control current to stay in conductive mode.

So it combines the best of both technologies

and is a robust component, which can be

used in power electronics in many different

ways, e.g. for motor-control.

Like rectifier diodes power IGBTs are put in

robust modular housings, which can be

screwed on heat sinks and can be connect-

ed with bus bars. Up to now the standard

height of these modules was 30 mm. Thus

the creeping distance in air between the bus

bars and the cover (or the door) of the con-

trol box was relatively small, if boxes had

been designed for flat modules.

Standardised construction height of 17

mm

The newest generation of IGBTs in the

Econo-Dual-Housing needs only a total

height of 17 mm – a reduction of 13 mm.

Now this space is additionally available for

clearance distance. At the same time this is

the 6th generation of the IGBT-manufacturer

Mitsubishi [1], who is represented by HY-

LINE Power Components [2]: After Planar-,

Trench- and lastly CSTBT-technology (Carri-

er Stored Trench Bipolar Transistor) now

Advanced CSTBT is state of the art.

The 17 mm IGBT modules of Mitsubishi are

available in single-version up to sevenfold

version for 600 V, 1200 V and 1700 V

reverse bias and for a current range of 50 A

to 1000 A.

Thereto the rectifier bridges of Powersem [3]

in 17 mm height can be used, which are also

provided by HY-LINE Power Components.

Hence the rectifiers in 17 mm height and the

IGBTs in 17 mm height can be connected in

one level with bus bars (Figure 1).

The driver is essential

An IGBT can be controlled easier than other

components, but the controlling is not trivial:

It is not enough to connect a simple signal

line – to build up a soundly controllable con-

verter with IGBTs you need specific control

logic.

Besides the driver has a big influence to the

efficiency of the power semiconductor: small

inaccuracy in switching already leads to

higher power dissipation and minor degrees

in efficiency as well as transient characteris-

tics caused by too long or -worse- too short

dead times, by switching too quick or too

slow. Especially three phase bridge circuits

are dependent on exact driving to avoid per-

formance-losses or even switching-failures,

which put the costly module at risk and could

lead to fatal consequences due to the high

power ratings.

Small errors in controlling could produce

big damages

Besides simple IGBT modules you will also

find modern IPMs (Intelligent Power Mod-

ules) with their own integrated protection cir-

cuit. IPMs do not provide galvanic isolation

though and limit potential performance char-

acteristics by the constant protection circuit.

Consequently, additional discrete, external

components are still necessary, having bad

influence on reliability and costs. Further-

more self-contained shutdown of power

semiconductors in equipment with multilevel

handling makes no sense and accordingly

can destroy the semiconductor and the com-

plete construction – a controlled shutdown is

absolutely necessary in this case. Finally the

driver should only become active with the

correct supply voltage being available to

avoid inaccurate switching operations.

Thus it makes more sense to integrate the

protection circuit in the control logic that is

needed anyway and to place it in the driver.

At failures like short-circuit, overload, con-

trolling error, over- or subvoltage ordinary

circuits fail quickly. In the case of high priced

power components that is unpleasant –

aside from the uncontrollable effects of high

powers being out of control.

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

Drive Engineering and Circuitry17 mm technology: Rectifiers, IGBTs and drivers for motor control

The IGBT is seen as the power semiconductor solving problems, combining the advantages of bipolar and field effect technology, thus making it easy to control evenlarge power converters. Modern solutions are realized in 17 mm stack height, which

increases the clearance distance in electric control cabinets. But only the adequate drivergets all advantages of an IGBT.

By Wolf-Dieter Roth, HY-LINE

Figure 1: Powersem rectifier modules andMitsubishi IGBT modules

Figure 2: Concept IGBT driver 2SP0115Tassembled ready to go on Mitsubishi CM200 DX-24S 17 mm IGBT module

Galvanic Isolation

Moreover galvanic isolation is necessary in most cases, which can

be realised inductively or optically. Optical fibres have not only the

advantage of being adequate for high potential differences, but also

of providing the transmitting medium at the same time. So optical

fibres are the favoured solution for high voltages and cascaded IGBT

circles, which for example are used for high-voltage direct-current

transmission (HVDC-transmission). In addition they are resistant

against transient characteristics, which can couple into the circuit

though stray capacities in solutions with transformers or also opto-

electronic couplers.

Contrary to optoelectronic couplers, which are too slow and do not

offer enough isolation voltage for many applications, the transmission

time of transformer-solutions is in the range of ns. Besides the trans-

former-solution is long time stable and therefore interesting for high-

er-frequency circuitry. Both coupling types can be integrated in a driv-

er circuit quite well, whereas it would be quite complicated to realize

galvanic isolation at the last moment, at the gate of the IGBT.

Integration in time

The common user could not and should not take care of these items,

but these things could bring incomprehensible problems in the circuit,

if they were not taken into account in the construction. At first view

cost-saving self-made developments, which supply only the basic

functions of driver circuits, could therefore not keep up with highly

integrated, intelligent drivers in the long time, which are adapted to

possible shortcomings of high-performance-IGBT-systems and avoid

potential breakdown of the costly components by well-timed integrat-

ed protection. This is featured by the IGBT-drivers of Concept Tech-

nologie [4] (Figure 2).

In addition the Scale Plug-and-play IGBT-drivers of Concept are sim-

ple to mount: In their PCB design the drivers are tailored to the IGBT

modules and are united with the IGBT module by soldering. So after-

wards only one module has to be mounted, which contains the IGBT

power module of Mitsubishi and the IGBT control circuit of Concept

(Figure 3).

With Concept Scale-Drivers of the second generation paralleling of

IGBT modules can be done more accurately than with standard cir-

cuits, because drivers can be decentralised and asymmetries of the

IGBT-modules have no influence on controlling. Even in 6.5 kV sys-

tems with optical coupling paralleling of several modules via their

own drivers and a common bus is no problem.

The drivers are available for the normal commercial (0°C to 70°C)

and for the industrial temperature range (-40°C to 85°C). They also

take care of aspects like the necessary creeping distance in air and

on surfaces and take the required testing of partial discharge into

account. The delay time is around 100 ns. The units come with trans-

formers and DC-converters to switch the IGBT correctly and to con-

trol it – even on the High-Side.

In the co-operation of three different products a reliable and cost-effi-

cient solution for converters and motor control units comes to your

hand, which may be directly screwed into the equipment in the field

without high development work.

[1] Start Page HY-LINE Power Components: www.hy-line.de/power

[2] Mitsubishi IGBT Modules of 6th generation: www.hy-line.de/nx6

[3] Powersem Rectifiers: www.hy-line.de/powersem

[4] Concept IGBT Drivers: www.hy-line.de/concept ;

www.igbt-driver.com

www.hy-line.de

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

Figure 4: View on a highly integrated ASIC in a Concept IGBT driver

Figure 3: Direct Master/Slave paralleling with optic coupling avoidsproblems with multiple optic coupling and differing delay times result-ing from this.

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


This article presents a dual high side switch

able to drive any type of load (resistive,

inductive and capacitive) with one side con-

nected to the ground. The device uses

STMicroelectronics’ VIPower technology, a

proprietary Smart Power technology that

allows integration of the control part and the

power stage on the same chip.

The block diagram in Figure 1 shows that

each channel is fully protected. Junction

over-temperature protection - thermally inde-

pendent for each channel, current limitation

(>1A, typically 1.6 A at 25 degrees Celsius),

and an inductive clamp (typically – 50V) are

built-in on silicon.

Thanks to current limitation and thermal pro-

tection, each channel is self-protected

against load short-circuit and over-current.

Due to the clamping chain at -50V, a demag-

netization circuitry is realized; the device is

able to manage very large inductive loads,

discharging the inductive energy quickly

without the need for an external freewheel-

ing diode. Under-voltage protection prevents

abnormal operations with very low supply

voltages, while loss of ground protection initi-

ates a switch-OFF of the power stages as

soon as the ground references are lost for

any reason; thus preventing the destruction

of the device.

The junction shut-down temperature for each

channel has a typical value of 175 °C; it pro-

tects the channel against a generic over-

load. The case over-temperature protection

has a double thermal protection integrated

on-chip, to avoid high temperature on the

PCB where the part is assembled.

The input blocks of the device are

TTL/CMOS compatible; they are designed in

order to minimize input switching times, and

to allow the direct connection of an optocou-

pler with a dark current of 10μA maximum.

The channels are switched ON with a mini-

mum level input voltage > 2.20V.

Open drain status pins are able to drive

directly a light emitting diode (LED); they

give indications of both junction over-temper-

ature shut-down and open load in OFF state

or short to Vcc.

Open Load Detection in Off State

In order to detect the open load fault in OFF

state, a pull-up resistor must be connected

between the Vcc line and the output pin (see

Figure 2). In normal conditions, the current

flows through the network comprised of the

pull-up resistor and the load. The voltage

across the load is less than the minimum

open load voltage; so the diagnostic pin is

kept at a high level.

When an open load event occurs, the volt-

age on the output pin rises to a value higher

than the maximum open load voltage and

the diagnostic pin goes low level, thus sig-

naling the open load.

Application Tests

Figure 3 shows a typical application circuit of

the device VNI2140J; it represents the out-

put stage of a programmable logic controller

designed for industrial automation or

process control.

In order to protect the device in high-side

configuration from the harsh industrial condi-

tions of power supply lines, optocouplers

diodes are typically used to separate the

application control circuits from the power

supply, the inputs and in the diagnostic pins.

A transil diode protects the High Side Switch

(HSS) against both positive and negative

surge pulses to make the device compliant

with IEC 61000-4-5.


S M A R T P O W E R

Dual High Side Switches inSmart Power Technology

Integrated solution for two output channels simplifies design and enhances reliability

The device, the VNI2140J, integrates on-chip two 45V Power MOSFETs channels (80mOhmtypical Rds(on) at 25 degrees Celsius) together with logic, driver, protection, and diagnos-

tic. The VNI2140J is housed in the tiny Jedec standard PSSO-12 lead power package.

By Giuseppe Di Stefano and Michelangelo Marchese STMicroelectronics

Figure 1: VNI2140J Block diagram


An electrolytic capacitor must be placed on the bus line (Vcc) in order

to filter bus inductance effect making the supply voltage stable and

avoiding under voltage shut-down. The size of the electrolytic capaci-

tor is selected based on the slope of the output current, the imped-

ance of the complex power supply cables, as well as the maximum

allowed voltage drop across the device. A low ESR capacitor is sug-

gested, as close as possible to the HSS, in order to filter the power

supply line for electromagnetic compatibility concerns. In our exam-

ple, a 47uF capacitor has been selected. To comply with IEC 61000-

4-6 (Current injection test), a 10nF capacitor is added to the output

pins.

The toughest loads to be driven in a factory automation/process con-

trol are the inductive ones; it is common to drive a 1.15 Henry nomi-

nal load. The associated energy to manage such inductive loads is

appreciable, carrying out a sensible power dissipation and a very

high junction temperature.

Behavior with shorted load

Over-current and short circuit of the load to ground are the harshest

events we must face during the digital output operation. Under these

demanding circumstances, output stages must survive the dissipation

of all the associated energy. Additionally, the loads, connected to the

output stages, must be protected from the peak of current that could

reach unexpected values.

In order to safely manage very high peaks of currents during short

circuit of outputs to ground, a current limitation block is integrated on-

chip. As a result, only a current spike for a short time is allowed: just

the time needed to intervene the current limitation circuitry, thus trim-

ming the maximum output current to an internally set value (typically

1.6A).

It is the same during a hard over-load. Internally limited output cur-

rent is not enough; however, in fact, if short circuit or over-load dura-

tion lasts throughout the time, the power dissipated into the device as

well as into the load becomes important, thus causing over-heating

enough to destroy the device and/or the load involved.

For that reason, thermal sensors have been built-in on-chip thus

switching OFF the over-loaded channels as soon as junction temper-

ature reaches an internally set value (>150°C).

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

S M A R T P O W E R

Figure 2: Open load detection in OFF state network

Behavior with capacitive load

The VNI2140J can also drive a capacitive load without problems; it is

able to drive capacitors with very high capacitance. In figure 4, wave-

forms are reported driving a 4mF/50V capacitor. Due to the high

capacitance, the output current during capacitor charge is in current

limitation, so that we do not see the real charging current but the limi-

tation current internally set in the VNI2140J. When the capacitor is

almost completely charged, the current goes below the internally set

current limiting.

Conclusion

A smart monolithic dual high side switch has been presented. The

new intelligent power switch (IPS) provides improved accuracy to

minimize energy losses and prevent system errors when faults occur.

These advantages are achieved using ST’s latest generation VIPow-

er™ technology, which allows a lower over-load current limit to main-

tain stable power conditions while the system is recovering.

By providing an integrated solution for two output channels, the

VNI2140J also simplifies design, enhances reliability, and saves PC-

Board space. This new two-channel IC is an important addition to

ST’s VIPower portfolio of industrial IPS, which already includes sin-

gle-, quad-, and octal-channel devices.

REFERENCES

[1] “VNI2140J Dual high side smart power solid state relay,”

Datasheet, www.st.com.

[2] G.Di Stefano, M.Marchese, “A single switch quad high side

switches with minimized power dissipation”, PCIM Nurnberg

November 2008

www.st.com

Figure 3: Typical Application Schematic Figure 4: Waveforms with capacitive load 4mF / 50V (yellow Vout,blu Vin, green Iout, red Vdiag)

S M A R T P O W E R

Power Your Recognition InstantlyBased in Munich, Germany, ITPR Information-Travels Public Relations is a full-service consultancy

with over a decade of experience in the electronics sector.

As a small exclusive agency, we offer extremely high ROI,

no-nonsense flexibility and highest priority to only a handful of companies.

Strategical SupportCorporate/Product Positioning, Market/Competitive Analysis, PR Programs, Roadmaps,

Media Training, Business Development, Partnerships, Channel Marketing, Online Marketing

Tactical PRWriting: Press Releases, Feature Articles, Commentaries, Case Studies, White Papers

Organizing: Media Briefings, Road Shows, Product Placements in Reviews and Market Overviews,

Exhibitions, Press Conferences

Monitoring and Research: Speaking Opportunities, Editorial Calendars, Feature Placement,

Media Coverage, Competitive Analysis

Translations: Releases, By-Lined Articles, Websites, etc.

Call or contact us today for a free consultation on how PR

can dramatically affect your company’s bottom line.

ITPR Information-Travels Public Relations

Stefanusstrasse 6a, 82166 Gräfelfing-Munich, Germany

Tel ++49 (89) 898687-20, Fax ++49 (89) 898687-21,

[email protected]

w w w . i n f o r m a t i o n - t r a v e l s . c o m

Toll-Free: 1-800-245-3984Phone: +1 412-696-1333

International SalesPhone: +852-3102-9337

www.mag-inc.com

Magnetics Large Powder Core Toroids

Kool Mμ® MPP High Flux

Visit Us at electronicaHall B6, Booth 512

Five new sizes, up to 165 mm, are now available. Large toroids are a cost-effective option for high current applications.

Uninterruptible power supplies

Power factor correction chokes

Inverters for renewable energy


There are operating conditions for industrial and special purpose

machinery with special requirements to power supplies for compact-

ness and resistance to intense influence of mechanical, climatic and

chemical factors (vibration, wide temperature range, dust, salt fog,

etc.).

AEPS Group has developed universal AC / DC power supplies that

are designed to power industrial and special equipments in all areas

of critical application with respect to combination of wide temperature

range, resistance to external influencing factors and low profile com-

pact design. Below, you will get a detailed overview of AEPS Group

power supplies and their functionality in three main area: Industrial,

Standard and Military.

Radio electronic equipments for «Military» and «Industrial» category

products typically require rated stabilized DC voltage of two to ten

from the range 3.3 V, 5 V, 9 V, 12 V, 15 V, 24 V, 27 V, 36 V, 48 V, 60

V for powering, which are generated from AC voltage of 220-230 V,

47 ... 440 Hz. Rated voltage of 14 V, 28 V and 56 V (taking into

account drop in ORing diode) are also used while operating with

buffer battery.

Currently, two structures, shown in Figures 1 and 2, are widely used

for distributed power systems.

Each of the given structure of power distribution has its own advan-

tages and disadvantages.

The disadvantage of the first structure is the remoteness of power

supplies from supply loads due to the need to exclude a large num-

ber of high voltage wires of input mains.

However, maximum efficiency and minimum heat loss is achieved

under this construction of power supply, which is often the determin-

ing factor. The second structure has greater flexibility and unlimited

functional capabilities. For example, one can separately remote con-

trol each DC / DC module. Only in such structure, power supplies

can be brought maximum closer to the electronic devices being pow-

ered.

Power Supply Category

The range of AC / DC power supplies is categorized as «Standard»,

«Industrial» and «Military»

The low cost AC / DC power supplies of «Standard» category are

designed to power a variety of industrial equipment for general use.

They are produced on standard component base, supplied without

sealing/potting and operate in temperature range of minus 10 ... + 70

° C. The optimum configuration for most applications reduces the

cost while creating power supply system.

AC / DC power supplies of «Industrial» category with range of oper-

ating temperature minus 40 ... + 85 ° C is designed to power industri-

al equipment of different climatic versions. Power supplies are made

on component base, tested in wide range of temperatures and filled

with heat conducting compound that protects the components from

adverse external influencing factors.

P O W E R S U P P LY

Low Profile AC/DC Power Supplies

Focused for industrial and special equipments

The modern market of AC / DC power supplies is widely represented by products withoperating temperature range of minus 10 to +70 ° C and of relatively larger size.

In addition, power supplies of power more than 150 watts often has in its compositionsuch potentially unreliable unit, such as built-in fan.

By Alexander Goncharov, P. h.D., Konstantin Stepnev and Oleg Negreba, AEPS Group

Figure 1: Distributed power supply system without intermediate con-versation

Figure 2: Distributed power supply system with intermediate conver-sation

AC / DC power supplies of «Military» category have operating tem-

perature range of minus 50 ... + 85 ° C and produced on custom-

made component base. They have polymer thermal conducting seal-

ing and are designed for powering industrial and special equipments

in most severe climatic conditions. These modules undergo special

types of temperature and limit tests, including butn-in test with

extreme conditions modes of on and off.

Modules of «Industrial» and «Military» category are produced as per

design document coordinated with instances determining the require-

ments towards quality and parameters of article for defense applica-

tion.

AC/DC power supplies are designed for using in various configura-

tions of power supply systems. Hence, modules of power up to 200

W inclusive is optimum for use as power source in power supply sys-

tem without intermediate conversion (figure 1). They can have up to

three output channels and built on the basis of flyback convertor with

galvanic isolation between the input and output (figure 3).

Modules of output power above 200 watts usually have one output

channel and can be used as a centralized stabilizer in the structure

shown in Figure 2. They are a half bridge asymmetrical forward con-

verter with galvanic isolation between input and output, shown

schematically in Figure 4.

Power supplies modules not only have high specific power/energy

characteristics, but have a special design for common heat-removal

base thus reducing the volume required for power supply system.

Figure 5 shows schematically the design of modules in section.

There is an aluminum casing in the module base on which heat is

removed from all thermal loaded components of the circuit - power

transistors, diodes, transformers, chokes. As a result, the module

casing also acts as a radiator having good heat dissipation. Thus,

KS500A-230WS24-SCN series power supplies without additional

heat sink under normal climatic conditions are capable to supply load

of power around 200 watts, and with radiator and forced cooling –

power load of 500 W up to ambient temperature of + 85 °C.

The casing has mounting holes for installation of modules on the

heat sink; there is also version for mounting on DIN-rack. Module

PCB is protected against mechanical and climatic influences by a

thin-walled steel cover. Such a design of casing with four sides clos-

ing the power supply components further improves its electromagnet-

ic compatibility (EMC) with other equipments. Power supplies ver-

sions with polymer potting (heat conducting sealing) are produced for

better protection against external influencing factors in industrial and

special-purpose equipments, which eliminates damage to power sup-

plies due to vibration or dust, moisture and salt fog.

There are input and output screw terminal blocks on the module

PCB’s, modification with soldering pin leads or flexible mounting lead

is also possible.

The described implementation of technical solution allows to confi-

dently compete AC/DC power supplies of AEPS Group with similar

products from other manufacturers.

www.aeps-group.com


Figure 3: Block diagram of modules of output power up to 200 Winclusive

Figure 5: Schematic design of module in cross section

Figure 4: Block diagram of modules of output power above 200 W

Table 1: Composition, basic characteristics and service functions of AC/DC AEPS Group modules

P O W E R S U P P LY

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 6471711-10

ISSN: 1863-5598

Electronics in Motion and Conversion November 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

64 Bodo´s Power Systems® October 2010 www.bodospower.com

N E W P R O D U C T S

Linear Technology Corporation introduces the LTM4609MP, a buck-

boost DC/DC uModule® system-in-a-package that is guaranteed and

tested for the wide -55ºC to 125ºC temperature range. The

LTM4609MP is designed to deliver precision regulation in demanding

environments in applications such as military, avionics, heavy indus-

trial machinery, and harsh environment sensors.

The LTM4609MP is a high voltage and high efficiency DC/DC system

in a surface mount 15mm x 15mm x2.8mm LGA (land grid array)

package. The device regulates an output voltage from a variable

input voltage greater than, less than or equal to the output voltage.

The LTM4609MP operates from 4.5VIN to 36VIN, regulates an output

voltage from 0.8V to 34V and can deliver an output power up to

100W. Included in this uModule regulator is a synchronous buck-

boost DC/DC controller, four N-channel MOSFETs, input and output

bypass capacitors and compensation circuitry in a small plastic mold-

ed package. Only an inductor, feedback and sense resistors, and

bulk capacitors are required to implement a very low profile, compact

and high efficiency design. Application examples are point-of-load

and intermediate-bus regulation in networking, industrial and automo-

tive systems and high-power battery-operated devices.


www.linear.com

Buck-Boost DC/DC up to 98% Efficiency

TDK-Lambda EMEA will be showcasing its latest offering to industrial

equipment manufacturers visiting this year’s electronica. With a

range of 1.5W to 100kW AC-DC power supplies, DC-DC converters,

Programmable (Lab) DC power supplies and EMC/EMI noise filters,

callers to the TDK-Lambda booth will find a reliable and efficient

product that will fit their applications.

Among the many new product introductions to be presented during

the show is the TDK-Lambda HFE1600 series of 1U high, single out-

put AC-DC hot swap front end power supplies that have an industry

leading power density for a 1.6kW front-end of 25.2W/in3. Intended

for equipment requiring reliable 12V, 24V and 48V bulk power, typical

applications include communications, broadcast, military (COTS),

laser and process control. Up to 20% output voltage adjustment is

possible, enabling the HFE1600 to be customer set according to spe-

cific needs. Operating from a universal 85 to 265Vac input, the high

efficiency of up to 92% minimises heat dissipation and power con-

sumption thus meeting Climate Savers Computing efficiency stan-

dards.

Electronica Hall B2 Stand 205

www.emea.tdk-lambda.com

Introducing Innovative and Efficient Power Supply Solutions

International Rectifier has introduced the industry’s first 8-pin reso-

nant half-bridge control ICs for energy efficient Switch Mode Power

Supplies (SMPS) used in LCD televisions and monitors, home the-

ater systems, desktop computers, printers and game console appli-

cations. Available in an 8-pin SO-8 package, the IRS2795(1,2)S res-

onant half-bridge control ICs offer a high level of programmability and

protection. Features include programmable switching frequency up to

500 kHz with 50 percent fixed duty cycle, programmable soft start

frequency and soft start time, and programmable deadtime for opti-

mized Zero Voltage Switching (ZVS) under all load conditions to

achieve high efficiency and low switching noise. The IRS2795(1,2)S

devices offer over-current protection using the on-state resistance

(RDS(on)) of the low-side MOSFET, eliminating the need for an addi-

tional current sense resistor.


www.irf.com

8-Pin High Efficiency Resonant Half-Bridge Control ICs

65www.bodospower.com November 2010 Bodo´s Power Systems®www.bodospower.com September 2010 Bodo´s Power Systems®


Micrel, Inc. rolled out its SuperSwitcher IITM family of integrated

MOSFET buck regulators for high power density DC-DC applications.

The MIC26xxx SuperSwitcher IITM family is comprised of three DC-

DC buck regulators featuring Micrel’s proprietary Hyper Speed Con-

trol TM architecture. The MIC26400, MIC26600 and MIC26950

devices operate with an input supply voltage range from 4.5V to 26V

and deliver an output current of 5A, 7A and 12A respectively. The

SuperSwitcher IITM family has been tailored to be Any CapacitorTM

stable and independent of output ESR, thus solving the perennial

problem of stability that power designers face with distributed output

capacitance.

Electronica Hall A4 Stand 125

www.micrel.com

Point-of-Load Power Designs

With New SuperSwitcher IITM

Toshiba Electronics Europe (TEE) has launched a digital output mag-

netic sensor series that provides high sensitivity and low power oper-

ation in small surface mount packages. These new sensor ICs sup-

port demand for power saving in portable, battery operated equip-

ment as well as home appliance or industrial applications where

switching off individual functions contributes to higher overall system

efficiency.

The TCS20Dxx series offers dual pole detection and comes in push-

pull (TCS20DPx) or open drain (TCS20DLx) output configurations.

The smallest package option is the ultra compact CST6C package

(TCS20DxC). Measuring just 1.5mm x 1.15mm with a height of only

0.38mm, this version is ideal for non-contact switching applications

such as open/close sensing in portable, battery-powered devices. As

home appliance and industrial applications are typically less sensitive

to board space requirements, Toshiba also offers the TCS20DxR

devices in SOT-23F package options measuring 2.9mm x 2.4mm x

0.8mm.

Electronica Hall A6 Stand A21

www.toshiba-components.com

Ultra-Thin Digital

Output Hall ICs

Efficiency Through Technology

USAIXYS [email protected]+1 408 457 9004

ASIAIXYS [email protected]+886 2 2523 6368

EUROPEIXYS [email protected]+41 (0)32 37 44 020

PartNumber

Vdss(V)

ID(A)

RDS(on)(mΩ)

Qg(nC)

Trr(ns)

RthJC(οC/W)

PD(W)

PackageType

IXTK600N04T2 40 600 1.5 590 100 0.12 1250 TO-264IXTX600N04T2 40 600 1.5 590 100 0.12 1250 PLUS247IXTK550N055T2 55 550 1.6 595 100 0.12 1250 TO-264IXTN550N055T2 55 550 1.3 595 100 0.16 940 SOT227IXFK520N075T2 75 520 2.2 545 150 0.12 1250 TO-264IXFX520N075T2 75 520 2.2 545 150 0.12 1250 PLUS247IXFN240N15T2 150 240 5.2 460 140 0.18 830 SOT-227IXFX240N15T2 150 240 5.2 460 140 0.12 1250 PLUS247IXFN320N17T2 170 260 5.2 640 150 0.14 1070 SOT-227IXFX320N17T2 170 320 5.2 640 150 0.09 1670 PLUS247

THINK POWER

FEATURESHigh current capability (up to�600A)Low Rds(on)�HiPerFET� TM versions available for fast power switching performanceAvalanches capabili�es�

APPLICATIONSSynchronous rec�fica�on�DC-DC converters�Ba�ery chargers�Switch-mode and Resonant-�mode power suppliesDC choppers�AC motor drives�Uninterrup�ble power supplies�High speed power switching �applica�ons

www.apec-conf.orgwww.apec-conf.org

2011March 6–10, 2011

Ft. Worth, Texas

THE PREMIER

GLOBAL EVENT

IN POWER

ELECTRONICSTM

THE PREMIER

GLOBAL EVENT

IN POWER

ELECTRONICSTM

SPONSORED BY

Visit the Apec 2011

web site for the latest

information!


Texas Instruments introduced a high-efficien-

cy, ultra-low power step-down converter for

energy harvesting and low-power applica-

tions. The new TPS62120 achieves 96 per-

cent efficiency, and can generate a 75 mA

output current from an input voltage of 2 V to

15 V. The high-performance device supports

energy harvesting and battery-powered

applications, as well as 9-V and 12-V line-

powered systems.

The TPS62120 synchronous converter fea-

tures a power save mode to provide high

efficiency over the entire load current range,

reaching 75 percent efficiency at loads down

to 100 uA. During light load operation, the

device operates in a pulse frequency modu-

lation (PFM) mode, consuming only 11 uA of

quiescent current. The TPS62120 also main-

tains smooth, efficient operation at higher

currents by transitioning automatically from

its power save mode to a fixed-frequency

pulse width modulation (PWM) mode.

Electronica Hall A4, Stand 420

www.ti.com

High-Efficiency Power Converter for Energy Harvesting

Actel Corporation announced the availability of SmartFusion intelli-

gent mixed signal FPGA reference designs targeting motor control

applications.. The reference designs, implemented in a single Smart-

Fusion device, illustrate Field Oriented Control (FOC) using various

feedback methods for permanent magnet synchronous motors

(PMSMs). SmartFusion devices, which integrate an FPGA, a hard

ARM® Cortex-M3microncontroller and programmable analog, are

uniquely suited for motor control applications and enable the design-

er to optimize the hardware/software partitioning for optimum motor

efficiency and performance.

The reference designs showcase a single A2F500 device controlling

up to four axes of PMSMs simultaneously using the complex FOC

algorithm with sufficient FPGA resources and bandwidth remaining

for additional custom logic.


www.actel.com

FPGA Reference Designs Targeting Motor Control

www.bodospower.comwww.bodospower.com

National Semiconductor Corp. introduced the LM5119, a high volt-

age, dual-channel, dual-phase, synchronous buck controller with

emulated current-mode (ECM) control. The LM5119 enables regu-

lation of single or dual high-current voltages directly from inputs as

high as 65V, simplifying high voltage DC-DC conversion and reduc-

ing PCB footprint by up to 50 percent versus alternative solutions

requiring two conversion stages. The LM5119 features a unique

combination of high performance, flexibility and ease-of-use for

demanding telecommunication, automotive and industrial control

applications that require accurate voltage regulation. The device

expands National’s portfolio of high voltage synchronous buck con-

trollers to address higher load currents.

Today’s telecommunication, automotive and industrial control appli-

cations must regulate low output voltages at increasing load cur-

rents from a high input voltage that can change widely while still

meeting stringent PCB space and efficiency requirements. Nation-

al’s LM5119 enables direct regulation of two output voltages from

an input.

www.national.com/analog/power

65V Dual-Channel, Dual-

Phase Synchronous Buck

Controller

Cree, Inc. announced that it has achieved a major breakthrough in

the development and wide scale commercialization of silicon car-

bide (SiC) technology with the demonstration of high quality, 150-

mm SiC substrates with micropipe densities of less than 10/cm2.

The current Cree standard for SiC substrates is 100-mm diameter

material.

SiC is a high-performance semiconductor material used in the pro-

duction of a broad range of lighting, power and communication

components, including light-emitting diodes (LEDs), power switching

devices and RF power transistors for wireless communications.

The significant size advancement of single crystal SiC substrates to

150-mm can enable cost reduction and increased throughput, while

bolstering the continued growth of the SiC industry.


www.cree.com

High Quality 150-mm Sili-

con Carbide Substrates

national.com/ultrasound© N

atio

nal S

emic

ondu

ctor

Cor

pora

tion,

201

0. N

atio

nal S

emic

ondu

ctor

, ,

Mic

row

ire, a

nd P

ower

Wis

e ar

e re

gist

ered

trad

emar

ks. A

ll rig

hts

rese

rved

.

Mobile Medicine. Powerful Diagnostics.

Shown: Portable Ultrasound System Diagram

Tx Beamformer

DVGA

PCI/USB

T/R Switch

Analog Front End

CW

Doppler

CT∑ΔADC

FPGAClock

Power

Micro- processor

Microwire

Pulser

DVGALNA

High Image Quality with Low Power ConsumptionWith increasing demand for accessible medical care, the need for portable diagnostic imaging equipment is growing in clinics, ambulances, and remote point-of-care facilities. Ultrasound is the least invasive, most mobile imaging solution and, with a much lower per scan cost, is positioned for the fastest growth. National Semiconductor’s eight-channel transmit/receive chipset allows designers to deliver high diagnostic image quality in portable systems with very low power consumption.



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 canuse 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

UK capacitor manufacturer, Syfer Technolo-

gy has addressed the issue of understanding

precisely how a de-rated DC capacitor will

behave in normal AC operational conditions.

“This has the great advantage of enabling

designers to specify these components with

much greater confidence,” explained

Matthew Ellis, Senior Engineer at Syfer.

A key feature of Syfer’s stand at Electronica

2010, is the firm’s latest range of fully char-

acterised X7R and C0G dielectric multilayer

chip capacitors (MLCCs). “These devices

are ideal for de-rating in a range of AC appli-

cations,” Ellis added. They offer capaci-

tances of up to 120nF, for continuous use at

up to 250Vac 60Hz. “And we’ll have a num-

ber of technical experts on hand at Munich

to advise engineers and specifiers.”

This range is complementary to Syfer’s certi-

fied ranges of surge and safety capacitors.

Y2/X1, Y3/X2 and X2 rated components are

available in case sizes 1808, 1812, 2211,

2215 and 2220 with certifications from TÜV

and UL for standard terminations, together

with the FlexiCap™ flexible polymer termina-

tion option.


www.syfer.com

Boosting Confidence

in de-rated DC Caps

for AC Circuits

CUI Inc’s power line, V-Infinity, announced

the addition of a 1.5 A model to their V78XX

switching regulator series. With this latest

addition, CUI now offers 0.5 A, 1 A, 1.5 A,

and 2 A switching regulators to meet a wide

variety of power needs. The compact

V78XX-1500 series has been designed to be

a high performance alternative to linear reg-

ulators. Unlike linear regulators, this series

does not require a heat sink, making it ideal

for applications where board space is at a

premium and energy efficiency is a concern.

The V78XX-1500 series has efficiencies up

to 95% in a compact SIP package measur-

ing 11.50 x 9.00 x 17.50 mm. Units are pin

out compatible with industry standard

LM78XX linear regulators and come in both

straight and right angle pin configurations.

This series has a wide input range available

from 4.75 to 18 Vdc and regulated output

voltages of 2.5, 3.3, 5, and 6.5 Vdc. Operat-

ing temperature range of -40 to +85°C at

100% load. The non-isolated converters

offer short circuit protection, thermal shut-

down, very low ripple and noise, and an

MTBF of 2 million hours. The V78XX-1500

series switching regulators are available

now.

www.cui.com

Efficient 1.5 A Switching Regulator


Microchip announces the 8-bit PIC18F87J72 microcontroller (MCU)

family optimised for single-phase, multi-function smart-metering and

energy-monitoring applications. Featuring a dual-channel, high-per-

formance 16-/24-bit Analogue Front End (AFE), the new MCUs pro-

vide an accurate, reliable, easy-to-use and cost-effective solution for

developing meters that exceed International Electrotechnical Com-

mission (IEC) class 0.5 performance.

The MCUs integrate 64 or 128 KB Flash programme memory and 4

KB RAM, enabling time-of-use and multi-tariff functions, as well as a

high level of peripheral integration, including a LCD driver, hardware

Real-Time Clock/Calendar (RTCC) and a Charge-Time Measurement

Unit (CTMU) for implementing a capacitive-touch user interface.

Energy-calculation firmware, a development board and a reference

design are available, providing a complete solution that lowers costs

and shortens time to market for a variety of smart-metering and ener-

gy-monitoring applications.

As an addition to Microchip’s existing energy-metering and power-

monitoring portfolio, the PIC18F87J72 MCU family addresses market

demands for an integrated smart energy-metering and power-moni-

toring MCU.


www.microchip.com

Microcontrollers Support Multi-Function Smart-Metering

The demand for low profile, high efficiency

magnetics , is driving the need for high per-

formance planar transformers to be used in

extreme low profile power converters such

as in the flat screen industry.

The low profile restriction for components

needed in the flat screen industry has chal-

lenged Payton to develop a standard line of

under 8mm height transformers for natural

cooling. Our 200Watt (55mm x 40mm x

8mm) transformer is designed for a resonant

half bridge operation. Payton can design the

low profile transformers for many switch

mode topologies with operating frequencies

of over 200khz. The input voltage is 400V

with over 3000Vrms isolation and full 8mm

creepage and clearance. The total power

losses are under 3.8 watts with 40°C tem-

perature rise with no additional cooling. The

efficiency of this magnetic is in the 98%

range. The parts can be used without a

heatsink.

www.paytongroup.com

Low Profile Resonant Transformers with Improved Cooling


TDK-EPC, a group company of the TDK Corporation, presents a new

sample kit of EPCOS current-compensated ring core power chokes.

These EMC components are designed for a voltage of 250 V AC and

offer current capabilities of between 0.4 and 6.0 A. Their inductances

are between 0.2 and 39 mH.

The power chokes of the series B82721* are designed for the sup-

pression of common-mode interference in compact switch-mode

power supplies and converters of all types.

Thanks to stray inductances of about 1 percent of the rated induc-

tance, symmetrical interference can also be suppressed. Depending

on the type, the DC resistance values range from 30 to 2000 mOhm.

The design of the chokes corresponds to EN 60938-2 (VDE 0565-2).

The entire series is approved to UL 1283 (up to 300 V) and/or ENEC/

VDE and is RoHS-compatible.

The sample kit contains a selection of available types in horizontal

and vertical versions.


www.epcos.com/power_chokes

Current-Compensated Ring Core Power Chokes

Mitsubishi Electric is introducing three models of gallium nitride

(GaN) high electron mobility transistors (HEMTs) with 10W, 20W and

40W output powers, for L to C band (0.5~ 6.0 GHz) amplifiers. The

three devices are designed for use in base stations for mobile

phones, very small aperture terminals and other transmission equip-

ment.

The new models feature high output power, high efficiency and high

voltage operation with output amplifiers of 10W, 20W and 40W. They

have a power added efficiency of about 50% (at 2.6 GHz) and a high

voltage operation of 47W. All three devices are integrated into a

4.4mm by 14.0mm small package which helps to reduce the required

mounting surface in amplifiers.

Gallium nitride is gathering attention due to its high breakdown volt-

age and high saturated electron speed. In March 2010, Mitsubishi

Electric became the first company in the world to manufacture fully

space qualified GaN HEMTs for C band space applications. HEMT

devices that use GaN have higher power density, which helps to

save energy and contributes to making transmitters more compact

and light weight. Furthermore, they offer an increased operating life-

time.

www.mitsubishichips.eu

GaN HEMTs for L to C Band Amplifiers


National Semiconductor Corp. announced the LM3492 LED driver

with dynamic headroom control that accurately and efficiently drives

current to two independently dimmable strings of LEDs. The LM3492,

a member of National’s PowerWise® energy-efficient product family,

maximizes system efficiency and reduces system complexity and

cost in automotive LCD backlight applications.

The LM3492’s dynamic headroom control feature dynamically adjusts

the LED supply voltage through the boost converter feedback to the

lowest level required to provide optimal system efficiency. Three

embedded MOSFETs reduce system complexity and cost.

National’s LM3492 integrates a boost converter and a two-channel

current regulator to efficiently and cost-effectively drive two independ-

ently dimmable LED strings with a maximum power of 15W and an

output voltage of up to 65V. Integrated fast slew rate current regula-

tors allow high frequency and narrow pulse width dimming signals to

achieve a very high contrast ratio of 1000:1. The LED current is pro-

grammable from 50 mA to 200 mA by a single resistor.

www.national.com

Two Independently Dimmable LED Strings

Tired of your existing opto+driver solution? Take the ISOdriver Challenge: www.silabs.com/ISOdriverChallenge©2010 Silicon Laboratories Inc. All rights reserved.

Say hello to the Silicon Labs’ family of digital isolator and ISOdriver solutions, and say goodbye to the limitationsof optocouplers. Silicon Labs’ isolators feature ultra low power consumption even at incredibly fast datarates, robust multi-channel and bi-directional communications and reliability unachievable with optocouplers.ISOdrivers combine our digital isolator technology with gate drivers, delivering up to 4 A peak output current.

DIGITAL ISOLATORS REPLACE OPTOCOUPLERS

ROBUST AND RELIABLEOPERATION THAT YOURAPPLICATIONS DEMAND

Silicon Labs’ isolators and ISOdriverslead the industry in data rate, lowpropagation delay, RF immunity, ESD and jitter performance. And they excel in even the harshest environments.

THE LOWEST POWERCONSUMPTION EVEN ATVERY HIGH DATA RATES

Based on our patented RF isolationarchitecture, Silicon Labs’ digital isolatorsoffer the lowest power consumption at datarates up to 150 Mbps. Power consumptionstays low even as data rates increase.

MULTI-CHANNEL AND BI-DIRECTIONAL COMMUNICATIONS—IT’S ALL IN THE FAMILY

Silicon Labs’ digital isolators are designed for a wide range of demanding applications. With a small footprint, up to 5 kV isolation and up to6 channels, we’ve got a solution for all of yourisolation needs.

Embracing our founding philosophy of harmony, sincerty, and pioneering spirit, HITACHI introduces the new line up of high efficiency E2 series IGBTs for high power, environmentally friendly, energy generation systems.

2 and 3 level MW inverter systems using E2 series modules may offer you 15% better efficiency, 20% higher operating

temperatures, 25% higher power density as well as customary HITACHI quality and service.

Whether you are designing for a wind turbine or solar array grid connection application, with HITACHI E2 IGBTs you are

one step closer to making your contribution to a world with lower emissions.

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

Power Device Division



ABB France 51ABB semi C3+19AEPS 57APEC 66Bicron 21Centraldruck 63Cierre 59cirrus 15CT Concept Technologie 25Curamik 29CUI 47+53Danfoss Silicon Power 33Dau 35electronica 39EMV 2011 41+43Fuji 17

GVA C2Hitachi 71infineon 11InPower 43International Rectifier C4Intersil 31ITPR 8+58IXYS 61+65KCC 1Lem 5LS Industries 69Magnetics 59Maxim 23Microchip 3Microsemi 51Mitsubishi 27

National 67Payton 61PCIM 45PEM UK 72Power E Moskow 49Powersem 7Silicon Labs 71Semikron 13sps ipc drives 44TDK-EPCOS 9Toshiba 55Tyco 68Varsi 37VMI 53Würth Electronic 53

ADVERTISING INDEX

Transducer specialist LEM has created a measurement technology

that enables unprecedented levels of accuracy in on-board monitor-

ing of train power consumption.

The matched and optimised devices comprise transducers for current

and voltage, and a new energy meter. All three units provide levels

of accuracy not previously seen in the sector, and come with the spe-

cific certification recognised in the electric rail traction industry. Used

together, they enable designers to meet or exceed existing and

planned specifications: they also offer the same levels of accuracy to

system designers working in other areas of high-power electrical sup-

ply.

Current, in the high-accuracy power transducer suite, is monitored by

introductions to LEM’s ITC 4000 or ITC 2000/1000 ranges. Certified

to Class 0.5R, the ITC 4000 employs an advanced closed-loop (com-

pensated) current measurement design based on the Fluxgate princi-

ple. Nominally rated at 4000A, the ITC 4000 will measure +/-6000A,

consuming less than 80 mA (at zero primary current) to under +/-340

mA (at 4000A primary current) from a supply voltage of +/-24V to its

measurement (secondary) circuit. Fluxgate technology is noted for its

high levels of both accuracy and linearity; the ITC 4000’s linearity

error is under 0.05%. The device’s offset current is less than +/-

10?A and it also exhibits extremely very low temperature drift. The

ITC 4000 operates over –40 to +85 deg oC, and meets or exceeds

all relevant standards for safety and operating environment.

www.lem.com

Measure Traction Energy with

Unprecedented Accuracy

Intersil Corporation introduced an efficient,

dual channel step-down regulator that

reduces component count and optimizes

design flexibility for high power-density

industrial, communications and consumer

electronics applications.

The new ISL85033 integrates two high-side

MOSFETs that can source 3A per channel or

current-share 6A on a single two-phase out-

put. A wide 4.5V to 28V input voltage range

makes it an ideal solution for intermediate

bus generation and point-of-load (POL) reg-

ulation.

The ISL85033 includes features that maxi-

mize performance and efficiency while

reducing external component count and

improving design flexibility. These include

low resistance power MOSFETs optimized

for thermal performance up to 3A of output

current per channel. In addition, a precision

internal reference supports output voltages

down to 0.8V. The PWM regulator switches

at a default frequency of 500kHz and can be

user-programmed or synchronized over a

300kHz to 2MHz range, allowing user opti-

mization of conversion efficiency versus

inductor size.


www.intersil.com

Efficient Dual Step-Down Regulator

Let there be light

ABB Switzerland Ltd SemiconductorsTel: +41 58 586 1419www.abb.com/semiconductors

Economicallywith ABBsemiconductors

Power and productivityfor a better world™

Part NumberV

DS

(V)

RDS(on)

Max.

VGS=10V

(mΩ)

ID

(A)

QG

(nC)Package

IRFS3004 40 1.75 195 160 D2PAK

IRFB3004 40 1.75 195 160 T0-220

IRFH5004 40 2.6 100 73 PQFN 5x6 mm

IRF7739L2 40 1 270 220 DirectFET-L8

IRFS3006-7 60 2.1 240 200 D2PAK-7

IRFS3006 60 2.5 195 200 D2PAK

IRFH5006 60 4.1 100 67 PQFN 5x6 mm

IRF7749L2 60 1.3 108 220 DirectFET-L8

IRFB3077 75 3.3 210 160 TO-220

IRFH5007 75 5.9 100 65 PQFN 5x6 mm

IRF7759L2 75 2.2 83 220 DirectFET-L8

IRFP4468 100 2.6 195 360 T0-247

IRFH5010 100 9 100 65 PQFN 5x6 mm

IRF7769L3 100 3.5 124 200 DirectFET-L8

IRFP4568 150 5.9 171 151 D2PAK

IRFH5015 150 31 56 33 PQFN 5x6 mm

IRF7799L3 150 11 67 97 DirectFET-L8

IRFP4668 200 9.7 130 161 T0-247

IRFH5020 200 59 41 36 PQFN 5x6 mm

IRFP4768 250 17.5 93 180 T0-247

IRFH5025 250 100 32 37 PQFN 5x6 mm

IRF7779L4 250 38 35 110 DirectFET-L8

Your FIRST CHOICE

for Performance

Features

• Low on resistance per silicon area

• Optimized for both fast switching and

low gate charge

• Excellent gate, avalanche and

dynamic dv/dt ruggedness

The IR Advantage

• Best die to footprint ratio

• Large range of packages

• Available from 40 V to 250 V

Applications

• DC Motor Drives

• Uninterruptible Power Supplies (UPS)

• DC-DC Converters

• Power Tools

• Electric Bikes

Rugged, Reliable MOSFETs

for Industrial Applications

THE POWER MANAGEMENT LEADER

For more information call +49 (0) 6102 884 311

or visit us at www.irf.com

Visit us in Hall A5, Booth 320

Electronics in Motion and Conversion November · PDF fileElectronics in Motion and Conversion...

Documents

Transcript of Electronics in Motion and Conversion November · PDF fileElectronics in Motion and Conversion...