Embed Size (px)
Citation preview
CONTENTS
ViewpointWarmer Days Are Fast Approaching! . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2
Industry NewsDr. Mark Thompson President and CEO . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .4
Semikron Reports Record Turnover . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .4
Rutronik and Osram extend Franchise to Italy . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .4
License Agreement for Double-Sided Cooling MOSFET Packaging Technology . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .6
eupec Website in Chinese . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .6
Murata Joins POLA . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .6
New 4-Switch Buck-Boost Controller Delivers High Power Density and High Efficiency . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .8
Tin Oxide Based Material to Replace Carbon Composition Resistors, Kirk Schwiebert, Director of Marketing, Ohmite
Manufacturing Company . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .12
Cover StoryPower Supply Control Goes Digital, Marcus Zimnik, Texas Instruments . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .14
FeaturesMarketwatch—APEC Marks Digital Turning Point, Chris Ambarian, iSuppli . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 20
Capacitors—Ultracapacitors Make Portable Products Go Further, Dr. Adrian Schneuwly, Maxwell . . . . . . . . . . . . . . . . . . . . . . .22
Energy Technology—Pulling the Plug on the Gremlin, Bernd Schwinn, Siemens AG . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .28
Design and Simulation—Design Tools Needed for Even the Smallest Device, Alexander Craig and Eric Hertz,
Fairchild Semiconductor . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .32
Power Management—Simlplifying Complex Power Sequencing, David Cooper, Applications Manager,
Potentia Semiconductor . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .35
Optoelectronics—Design Considerations When Using Optocouplers, Alexander Jaus, Agilent Technologies, Germany . . . . . . .38
Portable Power—Next Generation Power Chips, Tobias Buehler, austriamicrosystems AG . . . . . . . . . . . . . . . . . . . . . . . . . . .40
Digital Power—Why Digital Power Conversion?, Lou Pechi, Power One . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .43
New Products . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .46-48
Member Representing
Eric Carroll ABB SwitzerlandPaul Löser Analog DevicesArnold Alderman AnagenesisTodd Hendrix Artesyn TechnologiesHeinz Rüedi CT-Concept TechnologyDr. Reinhold Bayerer eupecChris Rexer Fairchild SemiconductorDr. Leo Lorenz Infineon TechnologiesEric Lidow International Rectifier
Member Representing
David Bell Linear TechnologyRüdiger Bürkel LEM ComponentsPaul Greenland National SemiconductorKirk Schwiebert OhmiteChristophe Basso On SemiconductorDr. Thomas Stockmeier SemikronDr. Martin Schlecht SynQorUwe Mengelkamp Texas InstrumentsPeter Sontheimer Tyco Electronics
Power Systems Design Europe Steering Committee Members
PowerSystems DesignE U R O P EPowerSystems DesignE U R O P E
2 Power Systems Design Europe April 2005
We have been busy attending many of thepower trade shows in Europe, the US andChina. It gives you a good picture into the glob-al world of “Power Electronics” and also a goodfeeling about how valuable our “niche” in powerplays into the many aspects of today and intothe future products of tomorrow.
What was important at Embedded inNuremberg? The overall size and weight of amotor drive or of a power supply is influencedby other circuit requirements beside the powerelectronics. Power management starts alwaysat the device level with the controller ICs thatconserve energy by putting non-required func-tions to sleep or by slowing down clock cycles,followed by the distributed power at the boardlevel. Additionally, embedded system solutionsare being incorporated to boost the design tohigher levels of performance.
What does embedded technology do forpower depends upon the application.Individuals working on automotive controllerssee the benefits quite clearly. The motor driveindustry is a fertile applications arena forembedded solutions, so complex conversionas well.
APEC in Austin Texas has the focus on energy efficient solutions including digital power.California legislation will drive the power supply manufacture to do more efficient solutions for active level of full power and thestandby operation.
EMV in Stuttgart gave a good view of EMCorientated design that includes all areas of electronics. The conclusion is that EMC designhas become part of the development processfrom the beginning on, instead of an add onprocess afterwards.
The PCIM China show gave us the clearmessage that the potential in China for powerelectronics designs and development is huge.
Our new publication PSD China under thedirection of our editor Liu Hong is receivingpositive feedback from the marketplace.
The June PCIM Nuremberg show is comingup soon. PCIM is a big Power Family Partythat no one should miss. The PCIM Nurembergconference is the platform to provide a futureview into technology. The strength of this tech-nology can conserve our rapidly depleting nat-ural resources. It can improve our lives and itcan be a means of ensuring future generationsa better world to live in.
I am again looking forward to seeing you atPCIM in Nuremberg the place to meet thepower experts.
Best regards,
Bodo ArltEditorial DirectorThe Power Systems Design Franchise
Warmer Days Are Fast Approaching!
VIEWPOINT
AGS Media Group
Katzbek 17a
D-24235 Laboe, Germany
Phone: +49 4343 421790
Fax: +49 4343 421789
www.powersystemsdesign.com
Editor-in-Chief
Bodo Arlt, Dipl.-Ing.
Publishing Director
Jim Graham
Associate Publisher
Julia Stocks
Creative Direction & Production
Eyemotive
www.eyemotive.com
Circulation Management
Advertising Rates, Sizes and Digital File
Requirements available at:
www.powersystemsdesign.com
Your Free Subscription Request available at:
www.powersystemsdesign.com/psd/
subslogn.htm
Registration of copyright: January 2004
ISSN number: 1613-6365
Issue print run: 20,000
Printing by: Central - Druck Trost GmbH & Co.
Heusenstamm, Germany
AGS Media Group and Power Systems Design Europe
magazine assume and hereby disclaim any liability to
any person for any loss or damage by errors or omis-
sions in the material contained herein regardless of
whether such errors result from negligence, accident
or any other cause whatsoever.
Send address changes to:[email protected]
Volume 2, Issue 3
4 Power Systems Design Europe April 2005
INDUSTRY NEWS
Fairchild Semiconductor announced theappointment of Dr. Mark Thompson asPresident and CEO. Thompson, who will also become a member of the FairchildSemiconductor board of directors, will succeedfounding President and CEO Kirk Pond effec-tive as of the company’s annual stockholders’meeting on May 4.
Pond will remain chairman of the board.Thompson currently serves as Fairchild’sexecutive vice president of its Manufacturingand Technology Group.
“I’m pleased the Board is moving forwardwith the final stage of our existing successionplan by appointing Mark Thompson to lead thecompany into its next phase of growth,” saidPond. “We’ve transformed the company intothe leading supplier of semiconductors to opti-mize system power while building one of themost cost-effective and skilled manufacturingoperations in the industry. Mark is the rightperson to now accelerate the company’s nextevolution, focusing on developing higher mar-gin, more complex products to increasegrowth and deliver superior financial perform-ance.
Since Mark joined Fairchild, I have beenconsistently impressed by his knowledge,integrity, judgment, people skills and commit-ment to exceptional performance. I’m confi-dent our customers, investors and employees
will be similarly impressed by his abilities,”concluded Pond.
“I have great respect for the work Kirk andhis team have accomplished establishingFairchild and transforming it from a standardcomponents manufacturer into a profitable,leading global supplier of solutions to optimizesystem power and I am honored to have beenchosen to be his successor,” said Thompson.“We will build on a company that is well posi-tioned in terms of process technology, accountposition, employee talent and financial healthby accelerating our move to higher value,more integrated IC solutions.”
Thompson holds a Ph.D. from theUniversity of North Carolina and a bachelor’sdegree from the State University of New York.He previously served as vice president andgeneral manager of Tyco Electronics PowerComponents Division.
Dr. Mark Thompson President and CEO
Semikron Reports Record TurnoverIn the business year 2004 Semikron
enjoyed a turnover increase that reacheddouble figures. This huge improvement com-pared to last year’s figures means that theSemikron Group registered the biggestturnover in its corporate history. Semikron,the European market leader in the field ofpower modules (with a market share of 33%according to the IMS study), benefited in theprevious business year not only from 40%growth increases on the Chinese and USmarkets. Japan and Korea are also fast-growth markets for Semikron. And thanks tothe recent initiative to form a global networkcomprising the top 9 Semikron companies in the field of power electronics systems
(see contemporaneous press release onSemikron Solution Centers), the companyhas managed to secure further growth poten-tial worldwide.
“These good figures give us the leeway weneed to actually implement our strategies,”commented Peter Frey, Managing DirectorSales/Marketing, Semikron International.“The years of investing in Asia are finallybeginning to pay off.”
www.fairchildsemi.com
Rutronik and Osram Extend Franchise to Italy
Europe Sales Offices: France 33-1-41079555 Italy 39-02-38093656 Germany 49-89-9624550 Sweden 46-8-623-1600 UK 44-1628-477066 Finland 358-9-88733699 Distributors:Australia Soanar 61-2-9741-0122 Belgium ACAL 32-0-2-7205983Finland Tech Data 358-9-88733382 France Arrow Electronique 33-1-49-784978, Tekelec Airtronic 33-1-56302425
Germany Insight 49-89-611080, Setron 49-531-80980 IrelandMEMEC 353-61-411842 Israel Avnet Components 972-9-778-0351Italy Silverstar 39-02-66125-1 Netherlands ACAL 31-0-402502602Spain Arrow 34-91-304-3040 Turkey Arrow Elektronik 90-216-4645090 UK Arrow Electronics 44-1234-791719, Insight Memec44-1296-330061
www.semikron.com
Rutronik and Osram Opto Semiconductors(OS) have expanded their franchise agree-ment: Effective immediately, Rutronikassumes responsibility for distribution in Italyas well. With that, Osram takes a decisivestep towards a pan-European franchise withRutronik. The extensive cooperation betweenthe two companies is rooted in a long-term,successful partnership: For four years nowRutronik has been positioning itself as themost successful Osram distributor in Europe.
Meanwhile, Osram OS has become one ofRutronik’s largest lines. Following Rutronik’smid-2004 nationwide launch in Italy, Europe’sthird-largest distribution market, the expan-sion of the franchise has been a logical con-sequence of the effective cooperative rela-tionship. Rutronik has a presence in Italy witheight branches and a 55-strong sales team.In addition to Germany, the Osram franchiseextends already to the Benelux countries,France, Spain, Switzerland, Portugal and
Eastern Europe. The portfolio includes alldesign-in products such as Golden DragonLEDs, OLEDs, high-power laser and intelligent displays.
www.rutronik.de
6 Power Systems Design Europe April 2005
INDUSTRY NEWS
Power Events
• SENSOR+TEST, May 10-12, Nuremberg,
www.sensor-test.de
• PCIM Europe 2005, June 7-9, Nuremberg,
www.pcim.de
• Automation Seminare, June, Germany,
www.mesago.de/automationseminare
• GEMV Energietechnisches Forum,
June 14-15, Kiel, www.gemv.de
• EPE 2005, September 11-14, Dresden,
www.epe2005.com
• H2Expo 2005, Aug. 31 - Sep. 1, Hamburg,
www.h2expo.de
• INTELEC2005, September 18-22, Berlin,
www.intelec.org
• ISPS/IPC/DRIVES, 22-24 November,
Nuremberg, www.mesago.de/sps
Murata and Texas Instruments announcedplans to expand a global effort to promotepin-compatible, non-isolated, point of loadplug-in power modules and provide a secondsource centered on leading semiconductortechnology.
Murata has joined the POLA alliance, agroup that includes TI, Emerson's AstecPower, Artesyn Technologies and EricssonPower Modules. Murata will offer telecom anddata communications power supply designerspin-compatible footprints that provide thesame functionality and form factors as othermembers of the POLA alliance. To date,POLA companies have manufactured andintroduced more than 70 products with thesame electrical designs to ensure full interop-erability and true second sourcing.
POLA plug-in power modules, availabletoday in volume production, are based on the PTHxx series with Auto-Track sequenc-ing, introduced by TI in 2003 (See:power.ti.com/sc03105). Artesyn, Emerson’sAstec Power and Ericsson Power Modulesare delivering POLA, pin-compatible modules
to hundreds of customers. Murata’s plug-inmodules are expected to be available in thesecond quarter of 2005.
Late last year, POLA members announcedtheir intention to join the PMBus digital proto-col initiative with other leading power supplymanufacturers and semiconductor makers.The coalition agreed to develop and supporta new communications standard defining anopen architecture for power systems controlusing an industry-standard I2C serial bus.POLA plans to implement the new digital protocol in future generations of plug-inpower modules.
Murata Joins POLA
www.murata.com
STMicroelectronics and Siliconix announcedthat the two companies have concluded anagreement whereby ST will license from
Siliconix a new power MOSFET packagingtechnology that provides superior thermalperformance via top and bottom heat dissipa-tion paths in systems using forced air cooling.
Offered by Siliconix under the PolarPAKname, the new package’s leadframe andplastic encapsulation are similar to thoseused for most standard power MOSFETpackages, ensuring good die protection andeasy handling in manufacturing. Yet com-pared with the standard SO-8, the PolarPAKpackage dissipates heat so efficiently that itcan handle twice the current within the samefootprint dimensions.
By delivering superior thermal performanceand reducing package-related losses, the5mm by 6mm PolarPAK package allowsdesigners to create smaller, more compactcircuit designs with a lower component count.With a height dimension of just 0.8 mm, half the height of the SO-8, the PolarPAKpackage enables end products that are thinner as well.
www.st.com
www.vishay.com
License Agreement for Double-Sided Cooling MOSFETPackaging Technology
eupec Website in Chinese Just in time for PCIM China 2005in
Shanghai in March, eupec GmbH, startedtheir Internet site to their Chinese customersin Chinese language.
With this, we meet our Chinese customers’wishes and underline our commitment to the Chinese market, summarises company
spokesman Jörg Malzon-Jessen the addition-al language version.
The amount of enquiries from China hasbeen increasing significantly and shows agrowing interest in eupec products. Apartfrom good contacts to Chinese partners overthe past few years, eupec contributes this to
the fact of having been one of the first manu-facturers of semiconductor products whohave been participating in PCIM Shanghaisince it first took place in the year 2002.
www.eupec.com
www.ti.com
8 Power Systems Design Europe April 2005
Acommon DC/DC converter prob-lem is to generate a regulatedoutput voltage which lies some-
where between a wide range of inputvoltages. To help put this problem into aclearer perspective, it is useful to thinkof two everyday applications. Firstly,consider a Telecommunications 3Gbase station. These base stations con-sist of multiple rack-mounted cards with-in a large cabinet enclosure. This cabi-net provides cooling for the cards via alarge bank of cooling fans. These fansoperate at a nominal 24V and take mul-tiple amps of current in this type of con-figuration. However, the power sourcefor this array of cooling fans can varydepending upon the design constraintsand whether the system has switched tobattery back-up power. As a result, thepower source could come from one ofeither a 24V or 12V back-plane voltage,or a bank of sealed lead-acid (SLA) bat-teries. These SLA batteries usually havean output voltage range of between 10Vto 13.6V. Another application is portablemedical equipment, such as a kidneydialysis machine or a patient monitoringdevice. In this instance, these machinesare usually powered via an AC adaptor– the output being 12V at a couple ofamps. But, in this case, the dialysis
machine can also be powered by anumber of NiMH batteries as a back-uppower source. Now the input voltagecan vary from 9V to 18V, but is stillrequired to give a fixed 12V output.
The traditional approach in solvingthis type of problem has been to useeither a single-ended primary inductanceconverter (SEPIC) or a buck/boost con-verter. These types of converters willdeliver a fixed output voltage whetherthe input voltage is above, below, orequal to the output voltage. However,there are some significant drawbackswhen using a SEPIC converters:• Complicated design due to the multiple
inductors or bulky transformer required.• The control loop is difficult to control
during the transition between operating modes.
• The solution footprint is large and alsohas a high height profile.
• The efficiency of conversion is low –usually mid-70s to low 80s percent.
• Thermal problems can arise at higher output power levels.
A more effective approach would be touse a single inductor-based controllerwhich can control four external switchesto perform the step-down, step-up and
LTC3780 Efficiency and Power LossCurve for a 12V output with a 5A load.
LTC3780 's Operating Modes versusDuty Cycle.
10 Power Systems Design Europe April 2005
100% duty cycle modes. This type of 4-switch buck-boost converter would havethe advantage of being easy to design,have high power density and also pro-vide high efficiency operation due to itssynchronous drive capability.
New ApproachA new 4-switch controller IC from
Linear Technology does all of the above.The LTC3780 offers a compact solutionfootprint, high efficiency operation in the95% range and a simple design using asingle off-the-shelf inductor and a currentsense resistor. Its constant frequencycurrent mode architecture allows aphase-lockable frequency of up to 400kHz.With a wide 4V to 30V input and outputrange and seamless transfers betweenoperating modes, the LTC3780 is idealin telecom, medical, industrial and auto-motive applications.
The LTC3780 is a current mode con-troller that provides an output voltageabove, equal to, or below the input volt-age. Utilizing a proprietary topology and
control architecture, the LTC3780employs a current-sensing resistor inBuck or Boost modes. The sensedinductor current is controlled by the volt-age on the ITH pin, which is the outputof the error amplifier. The VOSENSE pin(refer to figure 1, pin 6) receives thevoltage feedback signal, which is com-pared to the internal reference voltageby the EA.
The top MOSFET drivers are biasedfrom floating bootstrap capacitors CAand CB (refer to left side of schematic infigure 1), which are normally rechargedthrough an external diode when the topMOSFET is turned off. Schottky diodesacross the synchronous switch D andsynchronous switch B are not required,but provide a lower drop during deadtime. Low inductor ripple current and theuse of synchronous rectifiers allow theLTC3780 to achieve very high efficiencyover a wide input voltage range. Whenthe input and output are 12V, the 4-switch buck-boost has 98% efficiency ata 2A load and 97% at a 5A load. With its
current mode control architecture, theconverter has excellent load and linetransition response, minimizing therequired filter capacitance and simplify-ing loop compensation. As a result, verylittle filter capacitance is required. Thesingle sense resistor structure dissi-pates little power (compared with multi-ple resistor sensing schemes) and pro-vides consistent current information forshort circuit and overcurrent protection.
Figure 1. LTC3780 simplified layout diagram.
www.linear.com
12 Power Systems Design Europe April 2005
For years, carbon compositionresistors were the most versatileand widely used resistors in the
electronics world. Its low inductance, rel-ative precision, low cost, wide resist-ance range and good surge capabilitiesmade it a solid choice for almost anyapplication. Today, however, carboncomposition resistors have becomemore difficult to acquire, and the rangeof sizes and packaging are limited. Toaddress the growing need for a carboncomp replacement, Ohmite has devel-oped a new, Tin Oxide based, bulkceramic resistive material. I will explaincomposition of this resistive materialand its basic electrical properties.
Tin Oxide, bulk ceramic resistors aremade from tin oxide, antimony andglass. Tin Oxides (SnO2) have beenused in resistor manufacturing for years.Most thin film resistors are made bydepositing a thin layer of tin oxide onto aceramic core and spiral trimming todesired value. In pure SnO2 conductivityis directly proportional to free electronconcentration (_ = [e-] q _) thus the con-ductivity will depend on partial oxygenpressure and is expected to sharply risewith temperature (making TCR unac-ceptable for resistor application). Theaddition of antimony is a proven way oflimiting TCR effect of pure tin oxide.Conductivity model of SnO2 doped with Sb2O3 shows free electrons con-
centration independent of partial oxygenpressure and temperature.
Tin oxide – antimony compositionshowed good "resistor properties" how-ever by itself has limited resistancerange and is fairly brittle. A glass phaseis commonly used to bond tin oxides.We chose 2 commercial frits anddesigned a series of experiments toillustrate resistance and TCR range oftin-oxide/glass matrix. Samples wereprepared by grinding and mixing fritswith tin-oxide/antimony at given volume-to-volume ratios, pressed to 0.5 in. pellets and cintered.
As expected porosity decreases anddensity increases with firing tempera-
ture. Networking of tin-oxide particlescontrols conductivity in the matrix. Tin-oxide particles are better networkedafter 1200°C firing then 700°C; henceconductivity increases with cinteringtemperature.
TCR generally shows decrease withfiring temperature, which once againshould be credited to improved connec-tions between tin-oxide particles there-fore reducing glass’ contribution to TCR.
Composition of glass and manufactur-er of SnO2 selection had a bigger effecton overall electrical properties than originally expected. Glass compositionhad effects on conductivity mainly dueto the different types of structures itdeveloped.
The major advantage of tin-oxidetechnology is its bulkiness and high den-sity, which naturally lead to high pulsehandling capabilities. In surge applica-tion precision and TCR play a second-ary role, most popular resistance valuesare in the range of 1-100 Ohm _and lowinductance is desired in high frequencyswitching applications.
www.ohmite.com
By Kirk Schwiebert, Director of Marketing, Ohmite Manufacturing Coompany
14 Power Systems Design Europe April 2005
COVER STORY
Parameters can simply be changed by software
A long wearing dream finally became reality: The recent introduction of digital power supply
controllers by Texas Instruments enables the design of power supplies with capabilities
that have not been possible before with analog controllers.
By Marcus Zimnik, Texas Instruments
This article discusses the merits of digital power supply controland gives an overview of
Texas Instruments’ Fusion Digital Power products.
Most power supply applications todayare controlled by analog controllers.However, in the recent years the indus-try added more and more digital con-trolled architecture to power supply con-trollers. This silent change was hiddento most power supply designers as theystill were programming the functions ofthe control IC in an analog manner.Predictive gate drive or a simple func-tion like UVLO (under voltage look out )are a good examples for added digitalcontrol architecture to analog powersupply control ICs. Thus it has beenonly a matter of time until full digital con-trol will hit the street. So what is digitalpower supply control all about whenmany digital functions are already present in analog ICs?
What makes a controller a digitalcontroller? The principal differencebetween analog and digital control is theformat of the parameters that are usedin operating the power supply. Analogcomponents typically consist of com-parators and amplifiers and the signals
are kept analog as long as practical.The configuration values like frequencysetting, soft start time etc are set usingpassive analog components such asresistors and capacitors. In contrast digi-tal components typically consist of dataconverters and timers and the signalsare usually converted into digital data assoon as practical. Here the configurationvalues are set using data memory andmay be non-volatile or volatile.
What drives the digital decision?1) Flexibility
There are many aspects that drive thedigital decision. The major driver is flexi-bility. With digital control the power sup-ply designer is not anymore bound tothe hard coded decision trees buriedinside the control IC. For instance, theusual reaction to an over-current eventis to shut down the converter and initiatea restart. The power supply designerhas no option and changing the pre-pro-grammed behavior requires a significantamount of external circuitry in mostcases. With a digital controller the powersupply designer now has the optionwhat to do in such an event. Whether itis to accept a certain amount of over-current events to ride through a shortoverload condition or to shut down thepower supply immediately. The designer
has the ultimate freedom to decide theright course of action to a given stimuliallowing complex algorithms yielding tosophisticated functions. This flexibilityalso allows implementing adaptive con-trol law schemes. For example thepower supply can automatically set forhighest efficiency by calculatingPout/Pin or to switch compensationparameters when the power supplyenters discontinuous mode. As thetransfer function changes at the modeboundary, the control loop has to bechanged also in order to provide opti-mum performance under both operatingconditions. With digital control, changingthe compensation parameters can beaccommodated within one clock cycle.
2) CommunicationAnother driver is communication. A
digital communication interface allows ahost or system level processor to controland monitor the power supply. This digi-tal interface provides entry for other digi-tal tasks like on-the-fly programmability.This feature has been already welladopted in processor applications usingVID code (voltage identification code).
Beyond programming, communicationpermits remote data logging of the oper-ation conditions of the power supply.
16 Power Systems Design Europe April 2005 www.powersystemsdesign.com 17
COVER STORY
Shifts in the captured data (e.g. temper-ature rise, output voltage ripple increase)can be leading indicators of pending failures. These trends can be used toenhance system reliability by failure pre-diction and allow downtime avoidanceand intelligent fault management.
3) ProgrammabilityAll these options are possible due to
software programmability of the powersupply. This approach promotes plat-form design and offers a faster time tomarket window as parameters can simply be changed by software. Theproduct differentiation now lies in thehand of the software engineer and is notanymore limited by hard-coded analogcontrollers. Since the digital controllerallows the integration of many functionsthat else wise would have been imple-mented with additional components, the number of components in a complexpower supply can be reduced. This lowers manufacturing cost andenhances reliability.
4) AccuracyThe above mentioned features are
results of programmability and are themost significant differentiators in favor of the digital power supply controllers.Besides that, digital power supply
control is much more accurate com-pared to analog control. When designinga power supply in the analog world, thedesigner has to account for the compo-nent tolerances. For instance the switch-ing frequency in identical power suppliesis never equal (except they are synchro-nized) due to component and IC manu-facturing tolerances. The same is truefor the loop compensation values. Thedesigner has to take into account worstcase scenarios and consider this whenselecting the components. In contrast,the switching frequency in digital powersupply controllers is derived from a high-ly accurate crystal oscillator clock andcompensation values are just numbersstored in a memory. This yields to a veryaccurate control and allows tweakingthe design more to the theoretical limitsresulting in higher performance.
5) CostTolerances in current sense thresh-
olds usually require over-designing thepower stage to the worst case tolerance.A digital controlled power supply couldcalibrate itself and eliminate these toler-ances. Over designing the power stagedue to worst case tolerances isn’t nec-essary anymore. This results in smallercomponents, smaller form factor andlower cost for the power supply. Since
most of the cost in a power supply is inthe power stage, the cost saving couldbe quite significant.
Devices for digital power supply controlTo enable all this new features men-
tioned before, Texas Instruments recent-ly introduced its Fusion Digital PowerProducts. By combining expertise inanalog power management and digitalsignal processing, TI is able to offerleading edge solutions for both isolatedand non-isolated power supplies.Solutions include the whole range of acline to point of load applications likeuninterruptible power supplies (UPS),server, telecom, datacom, industrial andvoltage regulator modules (VRM) appli-cations. Fusion Digital Power productsare designated with the prefix UCD.
The UCD7K Fusion Digital PowerControl Drivers (Figure 1) interface thedigital power controller to the powerstage. These devices offer the digitalcontroller direct control of the powerstage. A programmable analog over-cur-rent limit with flag provides protection forthe power supply in the unlikely eventthat the digital controller does not detecta failure situation in time. They also sup-ply the bias voltage for the digital con-troller. The UCD7K family features TIsTrueDrive technology enabling high cur-rent drive in the MOSFETs Miller plateauregion for high speed switching.
The UCD8K controllers (Figure 2)integrate PWM circuitry and a FusionDigital Power control driver to close thefeedback loop in the analog domainunder digital supervision. They providemost of the benefits of digital power con-trol and can be used with more modestdigital controllers. These controllers arefor use in digitally assisted power sup-plies in which a standard micro-con-troller or a DSP is already present orrequired. The UCD8K family includescircuitry and features to greatly easeimplementing a converter with high levelcontrol features. Digitally assisted powersupplies provide programmability ofmany power supply parameters such asSwitching Frequency, Synchronization,Dmax / V*S Limit, Start/Stop voltage,Input OVP, Soft-start Profile, CurrentFigure 1. Full Digitally Controlled Isolated Forward Converter
18 Power Systems Design Europe April 2005
www.ti.com
Limit Operation, etc. and status report-ing of Input V/I/Power, Overload, andtemperature. The UCD8K devices canbe pin programmed to work either involtage or current mode applications.
UCD9K devices (Figure 3) are digitalpower controllers that provide full digitalpower management capabilities includ-
ing monitoring and supervision to clos-ing multiple feedback loops in the digitaldomain. This family integrates theresources that provide the user flexibilityto develop their own intellectual propertyin digital power management.
The UCD9501 is a DSP based digital power supply controller with an
enhanced digital PWM resolution of150ps. A 32bit, 100MIPS DSP engineenables complex, multi loop applicationsbeing controlled by a single UCD9501.For instance one UCD9501 can controla dual interleaved PFC stage and aphase shifted controlled full bridge.Besides closing the loop digitally, theUCD9501 offers a variety of interfacesto the outside world like CAN, I2C, SPI etc. TI’s UCD9k controllers are supported with Code Composer StudioSoftware. This is an integrated develop-ment environment (IDE) that provideskey development tools to reduce devel-opment time and effort.
Digital power supply controllersenhance the performance of power sup-plies and enable functions not possiblewith analog implementations. Thesecontrollers add flexibility and increasereliability of power supplies, while reducing total BOM and cost. TexasInstruments’ Fusion Digital PowerProducts represent the ideal platform for all different complexities of digitalpower supplies.
Bibliography:[1] Power Supply Seminar SEM–1600Topic 6: “A Practical Introduction toDigital Power Supply Control”, by LaszloBalogh, Texas Instruments LiteratureNo. SLUP224.[2] Fusion Digital Power ControlSolutions, Texas Instruments LiteratureNo. SLUB008
COVER STORY
Figure 2. Push-Pull Converter with UCD8620 and MSP430F1232.
Figure 3. Full Digitally Controlled Power Supply with UCD9K and UCD7K.
www.powersystemsdesign.com 2120 Power Systems Design Europe April 2005
Most agree digital technology is the future of power management
By Chris Ambarian, iSuppli
The 20th annual Applied PowerElectronics Conference andExhibition (APEC) in Austin this
year might well be remembered as theturning point for market recognition ofthe importance of power electronics.
Digital power issues took center stage at APEC. Texas Instruments Inc.,Potentia Semiconductor Corp. and iWatt Inc. all announced innovative new products, all taking a completelydigital approach to power, and all ofthem different.
The Digital Power RevolutionThe subject of digital power seemed
to be discussed by everyone at theshow. Whether you are at the vanguardof Digitally-Controlled Power (DCP), ora die- hard analog skeptic—or confusedand lost somewhere in between—youwere talking about it, and probably hadan opinion on it.
To iSuppli Corp., DCP is a foregoneconclusion; the only remaining ques-tions relate to how much of the industryit will penetrate, in which applications,and when. This process has alreadystarted. The real issue is the PowerOperating System (POS), iSupplibelieves. DCP concerns the making of apower supply using digital techniques tocontrol the conversion devices and func-tions. Digital Power Management (DPM)involves the use of digital techniquesand communication to control the powersupply in relationship to the local systemin which it is operating. The POS repre-sents the idea of using software to con-trol an entire system of power conver-sion, control and distribution devices—
a system that inherently is defined bythe user’s application needs.
Digital DiscordWhile most market participants agree
that digital represents the future ofpower management, there was somedebate over what digital technology canand cannot do.
iSuppli at APEC asked a representa-tive of Texas Instruments Inc.—theapparent 800-pound gorilla in the DCPspace—if an analogy to the sensorlessvector control of a motor drive could beimplemented in digital technology. Hereplied with a calm and authoritative“no”. Two days later, iWatt announcedjust such a chip, apparently not knowingthat it could not be done.
There also is a great deal of resist-ance among the analog design commu-nity to many of the digital concepts.Many of the analog designers wereevaluating the viability of DCP by imag-ining doing what they do today, but
using digital techniques. And many ofthem don’t see the merit of this. Forexample, the analog makers would saythat the control loops can be done effec-tively and much more cheaply in analog.But what if digitalization enables a loop-less topology? What if digital offers agreater level of control? The iWatt chipappeared to short-circuit the brains of alot of senior designers; they just couldn’tseem to comprehend the idea that aregulated power supply could beaccomplished without a control loop.
Analog designers also say that theDPM function can be accomplished bysmacking a port on an existing analogsolution. It remains to be seen how thisapproach compares on a cost basis withporting to a digital solution that hasmuch the same functionality—or evenbetter, porting to one of the totally newtopologies enabled by digital, ala iWatt.
The analog designers’ identities seemto be tied up with their black-art expert-ise in analog; it’s just too scary or painfulto imagine that such knowledge mightnot be valuable in the near future.
However, the analog suppliers findthemselves very much in the same posi-tion as mom-and-pop grocery stores inthe United States. iSuppli’s advice tothose too nostalgic to move on is: “You’dbetter get over it. Wal-Mart is coming.”
Christopher Ambarian is a senior analyst with iSuppli Corp. Contact himat [email protected]
MARKETWATCH
www.iSuppli.com
22 Power Systems Design Europe April 2005
The big power pulses required in portables overstress existing battery technologies,
severely limiting their practical life between charges. And if you need to be
charging frequently, that’s not exactly portable.
By Dr. Adrian Schneuwly, Maxwell
Store more energy than conventional capacitors
Now more popular than ever,portable electronics are marvelsof engineering, with more fea-
tures and functions than we could haveimagined even a few years ago. Theyallow us to put cameras, video, highfidelity audio, GPS navigation and wire-less communications capabilities intoour pockets. All of these features can befound today in commercially availableproducts, which are no larger than adeck of cards. Despite all these capabili-ties, they often become mere toysbecause their batteries are not up to thetask. The big power pulses required inmoving camera motors, burning imagesinto flashcards and sending bursts ofinformation over wireless systems, alloverstress existing battery technologies,severely limiting their practical lifebetween charges. And if you need to be charging frequently, that’s notexactly portable.
The combination makes the difference
The reality is that we’ve becomedependent on our new portable tools. It only takes a few weeks for a “nice to
have” handheld portable to become anecessity. But what’s the consumer todo when their tool is reduced to an
inconvenient toy because it runs out ofbattery life? One option is the next gen-eration energy source now being hypedto replace batteries; the eternally “rightaround the corner” fuel cell. Many com-panies are on the verge of demonstrat-ing micro fuel cells that power a note-book for hours and then instantly refillfrom a small methanol fuel cartridge,similar to refillable fountain pens.Although clever, these fuel cells are along way off from commercialization.Furthermore, they have the same challenge as batteries: very poor power performance.
In applications requiring only highamounts of energy and low amounts of power, for example calculators,clocks/watches and portable radios, abattery or set of batteries can be morethan sufficient to supply a small amountof current over a reasonable amount ofthe product’s lifetime. However, in appli-cations with an additional demand forhigh power, batteries have proven to beless than satisfactory. So how do we getenough power for all the features of ourportable products and still have the
CAPACITORS
www.powersystemsdesign.com 25
Wireless Systems using burst-modecommunications
Wireless systems that use burst-modecommunications techniques benefit inthe same way as cameras. Burst-modecommunications are used in local areassuch as stock and commodity exchangefloors, warehouses and even in restau-rants, where order takers and informa-tion processors use wireless terminalsand modified PDAs to communicatereal-time information to a central datacenter. High power loads are not theonly consumer applications to benefitfrom ultracapacitors. An entire range ofapplications are made possible byanother feature of the ultracapacitor; itcan be recharged as quickly as it can be discharged
Fast Charge OpportunitiesA number of children’s toys use this
“instant charge” design concept to maketoys. The most popular is a small racecar which has an ultracapacitor inside;you charge up the car in less than 10seconds from a set of alkalines in asmall charging pod. The play schemefits the short attention span of childrenvery well, and adds a real-world “pitstop” element to racing.
Ultracapacitors also enable the abilityto instant charge hand tools. The week-end handyman can also reap the bene-fits of ultracapacitors with instant chargemodules for cordless tools. Many timeswhen we just need to drill a couple ofholes, we find that the battery is deadand we have to wait for an hourrecharge to do a 2-minute job. With the ultracapacitor, this is no longer a problem.
New OpportunitiesIntegrated tools with cell phone,
camera, PDA and MP3 functionalities as well as beacons with GPS, personalalarm and emergency features have allbeen demonstrated and a flood of addi-tional devices are forthcoming. Stuffingall these features into a small packagerequires a lot of energy AND power. Ni-cads, currently used to supply thepower, have a finite life and would ulti-mately need to be replaced. By contrast,an ultracapacitor can cycle enough
times to last the life of the product.Another advantage that ultracapacitorshave over ni-cad batteries impacts prod-uct manufacturing. While ni-cad batter-ies need conditioning during the manu-facturing process, ultracapacitors donot. As a result, ultracapacitors canactually speed up manufacturing, trans-lating into decreased production timeand cost. In conclusion, you can get allthe power out of a battery and kill it, or you can carry a huge battery, but that defeats the whole pocket appliance concept.
The greatest opportunities for ultraca-pacitors in consumer electronics are allthose ideas that can only be accom-plished by separately addressing powerand energy needs. The right approach isto design intelligently and use two com-ponents to solve two different require-ments, a battery for energy and an ultra-capacitor for power.
ConclusionNew power-intent electronic products
such as wireless communicationdevices and digital cameras are creatingsignificant markets for energy storageand power delivery. Today, ultracapaci-tors are a viable component for produc-tion-intent designs in the power elec-tronics world. Numerous firms are wellinto the production cycle for ultracapaci-tor-based systems, recognizing theadvantages and availability of the ultra-capacitor to meet their business andtechnical requirements. Ultracapacitors
are available, cost effective, and perform well in industrial systems.
Engineers generally address peakpower needs by designing the primaryenergy source, such as an engine or abattery system, to the size needed tosatisfy peak demands, even if thosedemands occur for only a few seconds.Sizing an entire system for peak powerneeds, rather than for the averagepower requirement, is costly and ineffi-cient. Such systems can be significantlyimproved by storing electrical energyfrom a primary energy source such as abattery and then delivering that energyin controlled high power bursts whenrequired. Such high power delivery pro-vides electrical systems with dynamicpower range to meet peak powerdemands for periods of time rangingfrom fractions of a second to severalminutes. Batteries are not designed toprovide bursts of power over many hundreds of thousands of cycles.Ultracapacitors perfectly meet thisrequirement.
There are two primary uses for ultra-capacitors. The first is for temporarybackup power in electronic devices andsecondary short-term emergency powerwhen a primary power source is insuffi-cient. Here ultracapacitors have becomean alternative to batteries in applicationswhere the ultracapacitor is charged fromthe primary power supply, but functionsas a backup power source when the primary source fails.
Figure 3. Number of cycles per battery pack using the same U/C pack.
CAPACITORS
24 Power Systems Design Europe April 2005
energy to use them for a long time? Theanswer is really quite simple; use twodifferent components, one for power and one for energy. The solution forpower: ultracapacitors.
Today, the best ultracapacitors areextremely high power devices, offeringpower densities up to 20 kW/kg, whiletheir energy is still a fraction of that ofbatteries, Compact in size (small-cellultracapacitors are often the size of apostage stamp, or smaller), ultracapaci-tors can store a considerably greateramount of energy than conventionalcapacitors and can release that energyat both a high or low rate. They have anextremely long life and are designed tolast the lifetime of the product into whichthey are designed. When combined withthe newest technology in high-energybatteries, the best of both worlds is possible; high power features and longoperating life.
Portable productsAs the need for smaller and more
lightweight systems increases, designengineers require innovative designapproaches to reduce size and heavi-ness without sacrificing overall perform-ance and reliability.
There are two primary uses for ultra-capacitors. The first is for temporarybackup power in electronic devices forfunctions such as computer BIOS set-tings, telephone and camera configura-tion settings, and short-term emergencypower when the primary power source isinsufficient. Here the ultracapacitor ischarged from the primary power supply,but functions as a backup power sourcewhen the primary source fails.
The second use for ultracapacitors is supplying peak power in electronicdevices. In these applications, ultraca-pacitors are used in tandem with batter-ies for systems that require both con-stant low power discharges for continualfunction and a pulse of power for peakloads. Here, ultracapacitors relieve bat-teries of peak power functions, resultingin an extension of battery life and areduction of overall battery size which inturn can allow for product size reduction.
In fact, ultracapacitors have beenused in a variety of applications, rangingfrom portable scanners for factory bar-code reading to digital cameras. Anyapplication that requires the storage ofelectrical energy and the discharge ofhighly variable amounts of power is apotential market for ultracapacitors.Many of the end products into which our ultracapacitors have been designedare now beginning commercial production.
High Power Pulses OpportunitiesDigital cameras have a variety of fre-
quent pulse loads, from lens motors tomemory writing. Expensive recharge-able batteries (and the cost of the charg-ing circuitry) are being replaced withembedded ultracapacitors and conven-tional, replaceable, off-the-shelf alkalinebatteries. Normally, even though alkalinebatteries have a lot of energy, they can’tdeliver the power needed to operate thecamera. By pairing the alkalines with anultracapacitor, each does what they dobest; the alkalines deliver continuousenergy, and the ultracapacitor deliverspulses of power. As such, users do nothave to rely on rechargeable batteries,making the camera smaller, lighter, and
truly portable. The advantage of theultracapacitor is that it will last for the lifeof the product and allows the customerto use inexpensive basic alkaline’srather than the new expensive type.Even by using two ultracapacitorsacross four alkaline you do not jeopard-ize the life of the product.
The major high peak demands areobserved during the microprocessoractivity, writing to disk and LCD opera-tion. Graph 2 shows the voltage swingfor each cycle. As it can be seen thevoltage drop increases rapidly with batteries only, but when an ultracapaci-tors are placed in parallel in the systemthe entire voltage drop is decreased and maintained.
In conclusion the results show that byadding ultracapacitors to the product itallows the end customer to use inexpen-sive alkaline batteries, which in the longrun saves them money and allows easeof use As it can be seen from the chartbelow the performance of the capacitorhas not changed even after 16 sets ofbattery changes, which was equivalentto about 9000 cycles.
Figure 1. Ultracaps mounted in parallel to the alkaline batteries.
Figure 2. Voltage swing without ultracapacitors(left) and with (right)
CAPACITORS
www.powersystemsdesign.com 27
The second use for ultracapacitors isin supplying peak power in electronicdevices. In these applications, ultraca-pacitors are used in tandem with batter-
ies for systems that require both con-stant low power discharges for continualfunction and a pulse of power for peakloads. Here, ultracapacitors relieve bat-
teries of peak power functions, resultingin an extension of battery life and areduction of overall battery size.
26 Power Systems Design Europe April 2005
Figure 4. “application model”
www.maxwell.com
CAPACITORS
www.powersystemsdesign.com 29
When two tripping events occurred,the system electricians were able tolocalize the cause of the fault immedi-ately using the data lines. The parame-ter for the short-time delay of the short-circuit release (S release) was set to theminimum value (1.25 times the nominalcurrent) using the rotary coding switch.However, this value didn’t correspond tothe significantly higher setpoint valuethat had been calculated in the planningphase using the Simaris deSign soft-ware. One possible reason for the unex-pected tripping event had been found.
Simplified cause researchBy checking all of the characteristic
values, the technicians were also ableto quickly detect a second cause: A trip-ping current of over 1,400 A in onephase had overloaded the energy distri-bution system which has a 1,000 A limitset for continuous operation (Fig. 2).
Knowledge of the number of trippingevents and the timestamps of theevents immediately directed the systemelectricians onto the right track, so thatthey were able to deduce the exactseries of events that had lead up to thetripping of the circuit-breaker.
A generator set in the production witha large inrush current had automaticallyactivated itself acyclically. A rarelyoccurring current peak, combined withthe large inrush current, had a valuethat exceeded the tripping conditions of the circuit-breaker, causing thetripping event.
Although the setpoint values of thecircuit-breaker had been documentedand printed out using the Switch ESPower software during the acceptanceinspection, the parameters were laterchanged. Without the incorporation ofintelligent, communication-capable cir-cuit-breakers, the true causes of thetripping event would have remainedunknown for a long time (Fig. 3).
The typical example of the tasks of areliable energy management systemillustrates the importance of an intelli-gent circuit-breaker. Since system faultsor irregularities generally aren’t simplythe result of one single cause, andMurphy’s law states that in the majorityof cases the fault will be sought in thewrong place, data regarding operatingcycles, maintenance-relevant events,parameters as well as consumption andstatus information should be automati-cally saved to the greatest possibleextent, so that it is immediately avail-able, if required, when a fault needs tobe analyzed. In order to unify the com-munication within the energy distributionsystem, the communications profile3.122 for low-voltage controls and distri-bution switchgear devices has been
agreed upon by the Profibus-Nutzerorganisation (Profibus userorganization) in coordination with themanufacturers. Circuit-breakers are also described within this communica-tion profile.
As mentioned earlier, the Sentronfamily of Siemens circuit-breakers con-sists of two series, the molded-case VLcircuit-breakers and the air WL circuit-breakers. Both types of circuit-breakersare connected to the higher-level controlvia the Profibus DP. Furthermore, theyalso support the Profibus DPV1 functionextension. Parallel to enabling commu-nication between the switchgear devicesand SPS, this makes it possible toexchange data with a commissioning ordiagnostics tool as well as with an HMI(Human Machine Interface) system, sothat large amounts of data can be trans-ferred on demand via an acyclic channel.
The unique feature of the circuit-breakers, which fit ideally in modern,communication-capable systems, is thatSiemens offers a software tool for pro-gramming and data management ofboth the VL and the WL series. Thisaffords the user maximum support andthe Switch ES Power software makesthe work of the commissioning person-nel significantly easier.
Parameterization for ProfessionalsThe Switch ES Power software makes
it possible to output and display theswitchgear identification data. The pro-gram can load data from a computer orlaptop even when the Sentron circuit-breakers are simultaneously exchangingdata with other stations (for example anSPS) via Profibus. Software settingscan even be specified offline, and arethen transferred to the circuit-breakerswhen a line connection exists (Fig. 4).
Depending on the respective featuresof the circuit-breakers, a variety of set-tings are available. To ensure clarity, theappropriate parameters for the specificapplication are stored in the familiar treestructure. When a connection with thecircuit-breaker is established, the SwitchES Power software automatically recog-nizes which type of circuit-breaker is
Figure 2. The well-known tree struc-ture ensures that the Switch ESPower software is easy for the opera-tor to use. If a computer with the soft-ware running on it is connected to acircuit-breaker, it immediately recog-nizes the possible parametersand settings.
Figure 3. The main view of the param-eter/diagnostics software is the start-ing point for the online diagnosis. Themost important status information isdisplayed on just one screen.
ENERGY TECHNOLOGY
28 Power Systems Design Europe April 2005
Until a few years ago, the task oflow-voltage circuit breakers wassolely to protect the subsidiary
energy distribution systems. However,the situation has changed in recentyears due to an increase in the competi-tion dominating the production sector.Even in highly developed industrializedcountries, the pressure of rationalizationpermeates down to energy supply anddistribution. As a consequence, intelli-gent, communication-capable modulesmust be used which can be harmonious-ly integrated in existing system structures.
The VL and WL circuit-breakers fromthe Siemens low-voltage controls anddistribution Sentron product range aremodern components which enable thetargeted optimization of energy distribu-tion. An example situation illustrates thesituation which confronts productionoperation on a daily basis.
The energy distribution of a produc-tion site in a plastic processing companyin southern Germany is protected bySentron circuit-breakers, among others.The desired protection parameters canbe set on the front panel of the respec-tive ETU45B trip unit using the rotarycoding switch. Additional communicationvia the Profibus DP enables the valuesto be output.
Knowing about the circuit-breaker allowsprecisely maintenance
Modern energy distribution systems expect circuit-breakers to do more than simply protect.
Therefore, low-voltage controls and distribution from Siemens A&D also concentrate on
the standards of modern control technology: bus capability, software support,
diagnostics and documentation.
By Bernd Schwinn, Siemens AG
Figure 1. Even when the ETU55B tripping unit is used for the Sentron WL circuit-breaker, the parameters that were set once are automatically transferred to thedevice when the communication starts up or when a circuit-breaker is replaced.
ENERGY TECHNOLOGY
www.powersystemsdesign.com30 Power Systems Design Europe April 2005
Intelligent switchgear makes maintenance easier
The more information that is knownabout the processes in the circuit-breaker, the more precisely mainte-nance can be planned. This results insavings in both costs and in personnelmanagement.
The example of the Sentrol VL or WLcircuit-breakers connected to the SwitchES Power parameterization/diagnosticssoftware clearly illustrates the positiveeffects that intelligent low-voltage con-trols and distribution have in the dailyoperation of a plant.
However, whoever doesn’t find thesepositive effects sufficient can gain addedconvenience and safety in their plant byusing the ETU55B tripping unit to pre-vent undesired alterations to the operat-ing parameters. In the practical examplementioned at the beginning of this arti-cle, the incorrect parameters wereentered at a later point in time using therotary coding switch. When using theETU55B, however, values can only beentered using an input device on site, orby the controller using Profibus. Whenstarting up the communication system orwhen a circuit-breaker is replaced, theparameters that were originally set usingthe software in SPS are automaticallytransferred to the device. This ensuresthat the plug is pulled on the gremlinonce and for all.
www.siemens.com/sentron
connected, displays the relevant param-eters, and only allows the editable fieldsto be changed. This enhances the clarityof the program and also prevents theuser from entering information incorrectly.
After all of the settings have beenspecified, the values can be loaded intothe circuit-breaker, exported for use inother circuit-breakers, or be saved onthe hard drive for documentation. Duringthe saving procedure, the software cre-ates an additional HTML file which dis-plays the information in a clear manner.This file can be viewed using a standard browser.
Through the integration into theSimatic-Step-7 environment, all of thespecified circuit-breaker parameters canbe transferred in the S7 file format usingthe Object Manager of the Switch ESPower software. This has the advantageof ensuring that the correct protectionparameters will always be loaded eachtime the SPS starts up or each time acircuit-breaker is replaced, since the val-ues are only entered once. The energydistribution is thus fully integrated withinthe Siemens “Totally IntegratedAutomation” (TIA) system concept.
Diagnostics simplify energy management
Large amounts of diagnostics informa-tion is decisive for fast and effectivetroubleshooting and error elimination.The functional range depends on thetype of circuit-breaker used. The firstlevel of advance warning is attainedwhen one of the freely specified threshold values is reached.
If one of the monitored measured values is exceeded, a threshold warningsignal is triggered, which can be actedupon immediately. Exceeding a thresh-old value in itself does not lead to the circuit-breaker breaking the main circuit (Fig. 5).
The next safety level contains warninginformation. The user can see the timethat will elapse before the line is tripped(if no counter-measures are undertaken)as well as the cause of the warning (e.g.an overload). The circuit-breaker hard-ware stores all the events that occur(warnings, threshold warnings, controlinformation) in a logbook with the appro-priate time stamp. The clocks in the indi-vidual circuit-breakers can be synchro-nized to facilitate the analysis of consecutive events.
If a tripping event does occur, thecause of the trip can be quickly diag-nosed using the Switch ES Power soft-ware. The circuit-breaker supplies thetripping reason as well as the current
that led to the tripping event and theappropriate phase and enters the eventin the tripping logbook with the appropri-ate time stamp.
If the circuit-breaker is equipped withthe “Messfunktion Plus” (“MeasurementPlus”) option, it is also possible to usethe Switch ES Power software to recordthe change of the voltage and currentwith time for different specifiable events(e.g. a tripping event). In this way, theoperator of the energy distribution system receives detailed informationthat was previously not accessible.Comprehensive information makes itpossible to carry out an analysis to pre-vent the same or similar incidents fromoccurring again in the future.
As a further highlight, the “MessfunktionPlus” (“Measurement Plus”) option alsoenables harmonic analysis of the volt-age and current. All information relevantto maintenance is recorded and can beloaded. This includes, for example, thenumber of operating hours, trippingevents and operating cycles. In addition,the circuit-breaker also measures theJoule heating values flowing across thecontact at the moment the circuit is broken and empirically calculates thecontact erosion.
Figure 4. The most important settingsof the ETU45B tripping unit areentered using the rotary coding switchon the front. These can be read usingthe Switch ES Power software.
Figure 5. The circuit-breaker suppliesall of the information required foractive maintenance management andalso outputs maintenance informationindependently. Figure 6. The change of the voltage
and current with time during triggeredevents can be recorded using“Messfunktion Plus” (“MeasurementPlus”) and analyzed using the SwitchES Power software.
ENERGY TECHNOLOGY
www.powersystemsdesign.com 3332 Power Systems Design Europe April 2005
Manufacturers push the limits of size versus power dissipation
With all the press and design support effort given to the newest and most sophisticated devices, it
is important to remember that like the "O" rings in the space shuttle the smallest and simplest
device can cause catastrophic system failure if improperly selected for a design. For that reason it
is important to have simple design tools and methods to examine the thermal stresses of these
ubiquitous devices in systems to insure safe operation.
By Alexander Craig and Eric Hertz, Fairchild Semiconductor
As the cost of Printed CircuitBoard (PCB) board spacebecomes more and more valu-
able, circuit designers continue to pushfor smaller and less costly solutions fortheir component needs. The difficulty is the drive for “circuit miniaturization”versus the need for more power. Withadvances in new silicon technology, it is often found that the die has becomeso small that we are now primarily upagainst package size limitations in ourquest for smaller packages versuspower dissipation.
In the 1960s, we utilized thru-holepackaging where the leads are perpen-dicular to the device and suitable forinsertion mounting into a PCB. This wasa good solution that in some areas isstill being used today but it is limited todesigns that can accommodate largersize components with single-side boarddevice population. Since the thru-holecomponent mounts directly thru theboard, only 1 side of the PCB can be used.
As far back as the 1970s, we beganutilizing Surface Mount Technology
whereby the leads don’t penetrate thePCB surface but are mounted on thesurface using reflow techniques. Notonly can these devices be smaller insize but also allow the designer to useboth sides of the PCB more effectively.These packages types include the likesof the TO263 (D2pak) that has a foot-print of 155mm2 and other smallerpackages such as TO252 (Dpak) thattake up only 1/3 of the board space, or57mm2. Improved manufacturingprocesses and equipment in the 1980sand 1990s enabled us to again drasti-cally reduce the footprint of semicon-ductor packages. The most commonSmall Packages widely used today areSO8 (30mm2) *FDS6694, SSOT6 /SC59-6 (9mm2) *FMB5401, SOT23(7.4mm2) *BAT54 , SC70-6 (4mm2)*FFB2907A, and more recently releasednew packages such as SOT563F(2.56mm2) *FJYF2906, SOT323(2.5m2) *FJX597JBTF and SOT623F(1.68mm2) *FJZ594JBTF. Currently,there is work being done on packages<1mm2 footprint and chip scale pack-ages. (* Fairchild Semiconductor Part Number)
Semiconductor manufacturers contin-ue to respond to market needs by offer-ing a variety of new packages that con-tinue to push the limits of size versuspower dissipation. With the variety ofpackages being offered today, it isimportant for designers to utilize tools to properly select the "right" device forthe application.
Small Surface Mount packages(<30mm2) are used for a variety of dis-crete components such as Diodes andTransistors, including; MOSFETs, SmallSignal, Schottky, TVS, JFETs, etc.When a designer is selecting a smallpackage device, the first place that isoften researched is the datasheet. Insome cases, the datasheet is an excel-lent tool to help choose the properdevice but in more cases then not, thiscan often be a tedious and unrewardingexercise. Many datasheets today onlycontain limited data on key electricalspecifications pertaining to the deviceoperation but thermals/power dissipa-tion is often ignored.
Datasheets specifications are meantto be used as a guide for a device’s
performance under a particular circuit/environment. In most cases, this may ormay not represent the application thatthe designer is working on. Anotherissue in regard to datasheets is thatover time some datasheets havebecome more of a marketing tool andless of an engineering tool. A better way to determine if a device us properlysuited for an application is through pack-age modeling.
Since the silicon is small com-pared to the package and the sili-con thermal impedance resistanceis small compared to the packagethermal impedance. Figure 1below shows the assembly draw-ing for the BAW56. One can easilysee that the Silicon chip is smallenough to fit in the lead. So smallchanges in the size of the die havenegligible effect in the thermal path(Figure 1).
This model uses the electrical/thermal analogy in which tempera-tures are represented by voltagesand power or heat flow is repre-sented by current. Thermal resist-ances are represented by resistorsand thermal mass is representedby capacitors. Since the capacitorvalues are so large the absolutecharge tolerance at any point intime “CHGTOL” for SPICE mayneed to be adjusted to a largervalue SPICE default =10-14 C.The values for the Resistors andCapacitors are derived by curvefitting empirical data. Using theelectrical model for the device todetermine the power loss thoughnot required is suggested sincethe data sheets for these devicesgenerally have limit data. Theelectrical models also provide theuser the ability to examine com-plex in-circuit operation. PSPICEis used in these examples but anycircuit simulator is capable of per-forming the same analysis. Thisapproach allows the user to exam-ine the operation for differentambient temperatures.
Lets take the BAW56 an 85V 200mAsmall signal diode in a SOT-23 as anexample. The data sheet states that thedevice has a thermal resistance of 357oC/W. The device has a Maximum junc-tion temperature of 150 oC. So in 25 oCambient the BAW56 can dissipate P =(Tjmax-Ta)/RqJA = 350mW which is list-ed as PD on the data sheet. But it isimportant to note that this is the averagesteady state power and this, is whatdetermines the parameters like IF(AV),
(Average Rectified Forward Current)which is listed as 200mA But also listedon the data sheet is IFSM, (Non-repeti-tive Peak Forward Surge Current) whichlists 1A for 1s the Vf for 1A is not listedbut the Vf or 150mA is 1.25V so the Vffor 1A must well over 1.25V but for argu-ments sake assuming Vf at 1A is 1.25Vthat means a PD of 1A x 1.25V = 1.25Wfor 1s that is almost 3 times the PD onthe data sheet. This shows the impor-tance of understanding the transient
DESIGN AND SIMULATIONDESIGN AND SIMULATION
www.powersystemsdesign.com 35
Use POLs designed specifically for trackingIf requirements conflict, it can be necessary to use more than one power rail of
the same voltage, and start up in a different sequence.
By David Cooper, Applications Manager, Potentia Semiconductor
Most complex ICs use more thanone voltage rail, and to guaran-tee reliable operation and avoid
damage, many have very specificrequirements for the relationshipbetween the different rail voltages. Thepower system must not only control theirstartup sequence, it must also ensurethat the correct sequencing is followedduring shutdown. It has becomeincreasingly important to review themanufacturers' datasheets for alldevices used on a card, in order todetermine what power rails are needed and how they inter-relate during sequencing.
If requirements conflict, it can be nec-essary to use more than one power railof the same voltage, and start up in adifferent sequence. For best perform-ance, the shutdown sequence needs tobe maintained, as far as possible, underfault conditions, to limit damage to anyother ICs on the card. This articlereviews the challenges presented bycomplex power sequencing which arebeing addressed by a new breed of ded-icated power management controller.
Modular power systemsMost high-performance power sys-
tems use modular DC-DC point of loadconverters (POLs) as building blocks, asshown in Figure 1. In this type of powersystem each POL operates independ-
ently, and even identical devices startup and shut down at a slightly differentinput voltage level.
The time required for each voltage railto come within specification at startupdepends on the load, and at shutdowneach voltage rail decays at a differentrate depending on the hold-up capaci-tance. Consequently the voltage rela-tionships between the POL outputs areunpredictable, and an overall manage-ment function is necessary to ensureproper sequencing.
SequencingBecause of the wide variation
between POLs, a simple sequencing
control based on fixed time delays is notusually sufficient. To guarantee propersystem operation even under overloador fault conditions, an interlockedsequencing control gives the best per-formance, with a separate startup andshutdown threshold for each rail.Independent timers for each rail allowadditional flexibility to meet particularsystem needs. At startup, each powerrail must reach its pre-set startup inter-lock voltage before the next rail is start-ed. If any rail fails to meet its startupthreshold due to a fault, the power sys-tem is immediately shut down withoutattempting to start the subsequent rails.Similarly, during shutdown each railmust drop to its shutdown threshold
Figure 1. Typical high-performance power system using modular DC-DC point ofload converters as building blocks.
POWER MANAGEMEBT
34 Power Systems Design Europe April 2005
performance of a device. By creating asimple model for the thermal impedanceof the SOT-23 mounted to a PCB, onecan more accurately determine the safeoperation of a device. The model is asimple RC ladder network the R & C val-ues are determined by curve fitting theEquation 1 to the measured data. Z1(t)is the thermal impedance for a powerpulse of width equal to (t).
This equation is fitted to the data asseen in figure 2. It is important that thevalues allow the model to be well
behaved out side the range of the data.To insure that the values selected pro-vide a stable model two requirementsare met. First the sum of all the R val-
ues must equal the RqJA value357oC/W for the SOT 23. Secondextending the graph to a point that is in the range of the fastest the part willoperate. See Figure 3.
By simply applying the R & C valuesis a circuit simulator such as SPICE asis shown in Figure 4 Spice will calculatethe Junction temperature for any Powerpulse. By using a electrical model alongwith the voltage controlled voltagesource E1 and current controlled voltagesource H1 supplying the Voltage con-
trolled current sourceG1 the engineer caneasily model a complexwaveform or operatingcondition. By adjustingVTEMP different ambi-ent temperatures canbe examined.
Figure 5 Shows the1A 1s pulse current sit-uation listed in the datasheet this pulse resultsin a Tj rises to ~120Cfor the typical device a part.
This same type ofanalysis can easily beperformed for anydevice. All that’s requiresis the thermal imped-ance model for thedevice or the packagefor any small device.
In conclusion, thequest for “circuit minia-turization” continues topush the size versuspower envelope of sili-con & packaging capa-bilities. The result ofthis ongoing quest isthat it has becomeincreasingly more criti-
cal to insure that a chosen device isproperly suited for the intended applica-tion. Like the “O” rings in the spaceshuttle example, even the smallest,seemingly non-essential device cancause the entire system to break downand fail. When selecting and electronicsdevice, datasheets are a good start, but to truly determine the thermal stresslevels, package modeling which takesinto consideration electrical and thermalperformance is recommended. Onlythen, can one more confidently deter-mine that a device is properly suited for the application.
Figure 3. Equation 1 is well behavedout side the range of the data.
Figure 4. Spice will calculate the Junction tempera-ture for any Power pulse.
Figure 2. Equation 1 is fitted to the data.
Figure 1. Assembly drawing for the BAW56.
Figure 5. The 1A 1s pulse current situation.
Equation 1.
www.fairchild.com
DESIGN AND SIMULATION
www.powersystemsdesign.com 3736 Power Systems Design Europe April 2005
a highly logical and graphical format.The power converters to be used arefirst entered into a Topology diagram, as shown in Figure 5. Once the basicpower topology has been entered, thestartup sequence is selected through aseries of checkboxes, shown graphicallyon screen (Figure 6).
The required interlock thresholds andtimers for each rail are then entered intoa Parameters table. To further simplifythe task, all parameters are initially setas default values and in most applica-tions only a few may need to bechanged. Finally, the device GPIO pins are configured.
When the configuration is complete, it is downloaded into the device throughI2C or JTAG ports using a programmingcable during the development process,or through an in-circuit tester in produc-tion. If the order of sequencing or theparameters need to be changed,amendments are simply re-enteredusing the configuration software. Figure7 shows an actual plot of the startup andshutdown of three power rails controlledby a PS-2406, and closely follows theidealized waveforms of Figure 2.
SummaryAlthough the power sequencing
requirements for a card power systemcan be quite complex, the use of a con-figurable power management devicecan guarantee the sequence is correctlyachieved under all conditions, while dramatically simplifying the design.
For further information and readerenquiries:Ian McGill, Potentia Semiconductor,200-4043 Carling AvenueOttawa, ON K2K 2A4, Canada
Tel: +001 613 592 7270E-mail: [email protected]: +001 613 592 0027
Figure 5. Configuring the basic power topology. Figure 6. Selecting the startup sequence.
Figure 7. Scope plot of the startup and shutdown of three power rails controlled by a PS-2406.
www.potentiasemi.com
POWER MANAGEMEBT
before the next rail shuts down. Figure 2illustrates this type of interlockedsequencing between three voltage rails,with separate interlock thresholds andtime delays for each rail.
During normal operation, if a faultoccurs on any rail the complete powersystem is shut down. To preserve thesequence as far as possible, each railthat would normally have shut downbefore the failed rail is immediately shutdown at the instant the fault is detected.All remaining rails shut down in sequence.After a shutdown, all rails must drop totheir preset restart thresholds beforeany rails attempt to restart. This pre-vents a possible incorrect startupsequence when the holdup capacitorson all rails have not yet fully discharged.
TrackingIn some cases, two or more voltage
rails must precisely track during startupand shutdown. Essentially, both voltagesmust start up at the same instant andany voltage difference must be minimal.
It is possible to implement tracking byadding a MOSFET at the output of eachPOL and controlling the MOSFET turn-on so that the rails precisely track.However, this approach is not very prac-tical at low voltages and high currentssince the MOSFET degrades the effi-ciency and regulation. A much betteralternative is to use POLs designedspecifically for tracking. These are avail-able from several suppliers and includea tracking pin that directly controls theoutput voltage during startup. Simply by
interconnecting the tracking pins of sev-eral POLs as shown in Figure 3, theoutputs are made to track accuratelywithout reducing efficiency.
The sequence controller can simplytreat these two rails as a single entity,allowing them both to start after the previous rail reaches its startup inter-lock threshold. Figure 4 illustratessequencing and tracking for three rails.
Rail 1 uses an interlock voltage andtime delay, as in Figure 2. Rails 2 and 3track during startup and shutdown,using tracking POLs.
Sequencing using a power management controller
Several manufacturers have intro-duced dedicated power managementICs capable of very complex sequencingcontrol. For example, the PS-2406 fromPotentia Semiconductor controls startupand shutdown sequencing for up to fourPOL converters and the PS-2610 con-trols up to six POLs. These devices alsoprovide many other power managementfunctions such as voltage readout, out-put trimming, output voltage fault protec-tion and fault logging. All operatingparameters are fully configurable, allow-ing the sequencing and other functionsto be easily set up to meet the needs ofa specific application.
When using either device, therequired sequencing is simply enteredinto the PowerCenter Designer configu-ration software, where it is displayed in
Figure 2. Interlocked sequencing between three voltage rails, with separate interlock thresholds and time delays for each rail.
Figure 3. By interconnecting the track-ing pins of POLs outputs are made to track accurately without reducingefficiency.
Figure 4. Sequencing and tracking of three rails. Rails 2 and 3 track during start-up and shutdown, using tracking POLs.
POWER MANAGEMEBT
www.powersystemsdesign.com 3938 Power Systems Design Europe April 2005
Table 2 shows equipment-level speci-fications from regulatory organizations
for major categories of applications.Each equipment specification is a mas-ter document, with many subordinatespecifications referenced to completethe total regulatory requirements.
Table 3 shows the most significantspecifications for optoisolator compo-nents. The equipment level specificationcan reference the component levelspecification as a subordinate docu-ment, or there may be no direct connec-tion between equipment and componentlevel specifications.
For optocouplers the most relevantcomponent level standard today isIEC/EN/DIN EN 60747-5-2 . VDE 0884was the dominant worldwide standardfor optocouplers until it was supplantedby IEC/EN/DIN EN 60747-5-2 inJanuary, 2004. IEC/EN/DIN EN 60747-5-2 defines safety related parameterssuch as isolation voltages (maximumand working), clearance and creepagedistances and other critical optocouplerrelated parameters. The European regu-latory authority (EN) and related nationalauthorities such as German DKE/VDEfollow the IEC for new standards.
More detailed information on regula-tions affecting optocouplers is providedin the Agilent Optocoupler Designer’sGuide. A copy in Adobe Acrobat
format may be downloded at: www.agi-lent.com/view/optocouplerguide.
Although functional isolation or levelshifting can be achieved by varioustechnologies such as high voltage inte-grated circuits (HVICs), reinforced isola-tion, which is considered to be failsafewithin the maximum specifications, isprovided only by qualified pulse trans-formers and optocouplers. Emergingtechnologies such as magnetic isolatorsor magnetic couplers currently can onlybe considered to provide functional iso-lation, since there is no final worldwideaccepted standard that covers them.Designers need to be aware of what iso-lation quality is required for their endproduct and choose the right technologyto meet these requirements.
Any optocoupler that passesIEC/EN/DIN EN 60747-5-2 testing iscertified for reinforced isolation andtherefore is qualified as isolator for safe-ty critical applications. The maximumisolation voltages relate to packagedimensions (internal and external clear-ance, creepage), the ability of theirdielectric isolation to withstand high volt-age, and the mold compound character-istics (comparative tracking index orCTI). In addition, the optocoupler com-ponent standards are recognized byvarious equipment-level standards,which is helpful for the equipment quali-fication as well.
Common-mode rejection (CMR)Circuit designers often encounter the
adverse effects of common mode noise.CMR failures are related to high voltagetransients causing leakage currentacross the isolation boundary due toparasitic capacitances. Optocouplershave very low parasitic capacitances,generally less than 1 pF, which com-pares to the 40 pF or more common topulse transformers.
Leakage current is proportional to theparasitic capacitance and can be quitesignificant. A transient with 10 kV/_sslew rate (dV/dt) causes around 7 mA ofleakage current: more than 1000 timeshigher than the current of the photodi-ode. Because that 7 mA can interferewith the functionality of the photodiode,which is basically a high gain transim-pedance amplifier.
Therefore shielding technology isused to minimize the leakage currentand shunt away the remaining leakagecurrent from the critical nodes of theoutput IC, especially from the photo-detector area. The shielding perform-ance of the optocoupler is critical for theoverall CMR immunity. Agilent uses aproprietary shielding technology, whichprovides CMR performance of up to 15kV/_s (Figure 1). Further improvementscan be achieved by optimizing the inputcircuit, for example splitting the LED-biasing resistor and using a low imped-ance output stage which does notrequire a pull-up resistor.
Table 2. Application Categories Vs. Global/Regional Safety Standards.
Table 3. Optoisolator Component Level Specifications.
Figure 1. Agilent’s shielding technol-ogy significantly reduces the para-sitic capacitance between input leadframe and detector IC.
www.agilent.com
Shielding provides common-mode rejection of up to 15 kV/_s
Designing robust electronic systems is challenging. The selected components need to meet envi-
ronmental conditions that may include electromagnetic noise, electrostatic discharge, common
mode transients and high voltages. Optocouplers meet all these challenges by providing signal
isolation combined with high voltage insulation.
By Alexander Jaus, Agilent Technologies, Germany
Market trends such as the trendtowards lower operating voltagelevels to reduce power con-
sumption and smaller packages to fit inthe restricted space in smaller electronicassemblies present designers with newchallenges. Other considerations con-cerning printed circuit board layout,operating conditions, application specificparameters and regulatory requirementsare essential, too, and have to beaccounted for in early design stages.The designer often has choicesbetween different solutions and tech-nologies. This article will highlight some of the design considerations related to optocouplers.
Optocouplers are used for two primary reasons. First the signal isola-tion, to improve signal quality by elimi-nating group loop current, and blockingnoise signals and common mode tran-sients, and second the electrical insula-
tion, to protect both the optocoupler andsensitive circuitry from damage resultingfrom high voltage potentials, and tomake the component or equipment safefor users.
Regulatory aspects are criticalApplications calling for galvanic isola-
tion often take place in environmentswhere high voltages are present. Theability of the isolator to sustain and toisolate high voltages, both transient and working, is the reason why optocou-plers are required in many designs.Equipment operators and circuits withinequipment may need safe isolation andprotection from high voltages. Becauseof the potential dangers of high volt-ages, galvanic isolators and the equip-ment in which they are used are oftensubject to safety standard regulations.The intent of these regulations is to offer security to the user and to provideguidelines to the industry on the applica-
tion of high voltages. Given the largeassortment of regulating organizationsand the associated standards and spec-ifications, regulatory compliance foroptocoupler manufacturers and equip-ment suppliers can be confusing.
Various regions of the world deter-mine their individual standards and anorganization in that country issuesapprovals or certificates for equipmentand products. Since standards bodieshave often begun as national organiza-tions, many countries have their ownregulatory environment. As internationalcommerce grows there is a trend toward continental or international safe-ty regulations and standards. Table 1shows several of the standards bodiesinvolved with electrical/electronic sys-tems in general, and isolation compo-nents specifically.
Note that the DKE authors DIN speci-fications, and that for historical reasonssome DIN specifications are called VDE(Verband Deutscher Electrotechniker)standards. While VDE wrote the originalversion of these specifications, theresponsibility for maintaining and devel-oping German standards belongs toDKE. The DKE also represents Germaninterests within CENELEC and the IEC.
Table 1. Some of the most frequently-encountered standards bodies.
OPTOELECTRONICS OPTOELECTRONICS
www.powersystemsdesign.com 4140 Power Systems Design Europe April 2005
amplifier (GSM: 217 Hz) that is periodi-cally turned on and off. Peak currents ofup to 2 A and parasitic battery resist-ance of 0.3 Ohms lead to a 0.6 V ripple.The LDO has the task of properly isolat-ing this noise from the locked PLL cir-cuit. To achieve this analog perform-ance, the LDO is designed as a two-stage amplifier, with a high gain, lowbandwidth outer loop and a nested highbandwidth low gain inner loop.
Figure 2 displays a simplified symbol-ic schematic to show the two loops.(The realization on silicon is slightly different and much more complex.)Dynamic biasing reduces the currentconsumption down to 50 _A. The inter-nal 300mA current limitation protects thedevice in the event of short circuits. Theintegrated PCH 1 Ohm pass transistorguarantees low dropout voltages of 150mV at 150 mA. The resistor divider isprogrammable in 50 mV steps from 1.85 V to 3.4 V.
Digital circuits are less critical con-cerning noise and regulation character-istics. Therefore digital regulators canbe designed with a much lower powerdrain. Memories and baseband proces-sors also require lower supply voltagesdown to 0.8 V.
Depending on the applications runningon the CPU, the battery life for portabledevices can be increased by dynamicvoltage frequency scaling (DVFS). Theclock speed and the actual supply volt-age of the processor are dynamicallyadjusted to meet the required MIPS.The dependency power consumption tosupply voltage is quadratic; thus evensmall voltage reductions will lead to sig-nificantly lower power demand.
Therefore, the digital LDOs of thedevice can be programmed “on the fly”from 0.75 V to 2.5 V in 50 mV steps.This feature helps to run large applica-tions on mobile battery powered devicesand maintain an acceptable battery life.
To further increase efficiency, aDC/DC step down converter should beused to act as a pre-regulator for thedigital LDO. In this case, the input
Figure 1. Block Diagram PMU.
PORTABLE POWER
Next generation cell phones are the targetThe introduction of simple, dedicated Power Management Unit ICs (PMUs) offered significant
improvements, saved a lot of PCB space, and reduced overall costs. The technologies
used for these devices were 1.2µm CMOS, BiCMOS.
By Tobias Buehler, austriamicrosystems AG
Approximately ten years ago whenthe first cellular phones emerged,they contained a number of dis-
crete power components. The overallform of the first cell phones was verybulky, and their high power consumptiononly allowed for short standby and talk times. In response to increased consumer demand for smaller, higherperformance wireless electronics, engineers have developed advanceddesign techniques, that allow savingPCB (printed circuit board) real estateand provide a faster and more efficient performance.
The introduction of simple, dedicatedPower Management Unit ICs (PMUs)offered significant improvements, saveda lot of PCB space, and reduced overallcosts. The technologies used for thesedevices were 1.2µm CMOS, BiCMOS orin some cases simple bipolar technolo-gy. In these early designs, the pass tran-sistor of the high current LDOs had tobe external, due to weak transistor per-formance, resulting in poor regulationcharacteristics and higher power drain, compared to today's more efficient designs.
Only when submicron CMOS process-es were introduced could low imped-ance power transistors be integratedwith justifiable silicon area and digital
functionality could be increased withmore complex state machines.Additionally, simple DC/DC converters,chargers and backlight switchesbecame integrated into the PMUs.
Most vendors' preferred BiCMOSprocesses due to its better analog per-formance, but this came with much high-er manufacturing costs compared tostandard CMOS processes. Inresponse, dedicated engineering effortswere undertaken in CMOS designs toyield special design techniques withnearly equal performance as BiCMOS.
Today, the demands on PowerManagement Unit ICs increase everyday and new trends indicate that manyor all features not currently covered bythe chip set will eventually be integratedinto the PMU. Very often considered anASIC, these function-rich PMUs enablecell phone manufacturers to decide withvery short notice to add special function-alities, demonstrating great flexibility.The more modern 0.35 _m/0.25 _mtechnologies allow for complex digitalparts with up to 100k gates.
Another important issue is the 5 Vcompliance of the process, due to thedirect battery connection. Fully chargedLi-Ion Batteries exhibit 4.2 V, and evenregulated AC adapters output 5 V. This
will become one of the main challengesfor future power chips, because the lat-est 0.18 _m/0.13 _m CMOS technolo-gies are only available with 3.3 V options.
Current SolutionsHighly integrated PMUs are being
introduced to the market. One device ismanufactured in a 0.35 _m CMOS tech-nology and housed in a thermallyenhanced 6 _ 6 mm 48 pin QFN package.
Features of the device include 10 pro-grammable high performance linear reg-ulators, two highly efficient DC/DC stepup/ step down converters, a 2 _ 0.5 Wstereo audio power amplifier, a completechemistry independent battery chargerand eight programmable general pur-pose I/Os for interrupt, hardware enableand high current LED supply.
The LDOs offer unparalleled regula-tion characteristics, comparable with thebest available discrete components. Thetypical output noise is less than 30 _V(100 kHz bandwidth), which allowsdirect connection to ultra sensitive RFblocks such as VCOs and referenceoscillators. The line and load regulationis better than mV static and ± 10 mVtransient. Any battery noise will be sup-pressed effectively at the LDO output.Especially TDMA systems produce ahigh battery ripple due to the RF power
PORTABLE POWER
www.powersystemsdesign.com 4342 Power Systems Design Europe April 2005
DC/DC converters approaching the magical limit of 100%
Using digital PWM control opened the horizon to cost effective implementation of many additional
functions by being able to change the operation and the characteristics of the converter
with simple software programming.
By Lou Pechi, Power One
Starting with a look into the back-ground leading to the need forDigital Power Conversion. In the
last several years lot of buzz was creat-ed about the need for Digital PowerConversion (DPC). While all this talkwas going on, most of the manufactur-ers were trying to outdo each other byintroducing higher and higher efficiencyand power density DC/DC converters.With efficiencies approaching the magi-cal limit of 100%, the incrementalimprovements from manufacturer tomanufacturer were quite minimal. Withdifferences of efficiencies reduced tojust a few percent the only differentiationbetween the products that remainedwas price.
At the same time semiconductor man-ufacturers introduced a multitude of con-trol and monitoring ICs, filling the needto control the proliferation of voltages ona typical circuit board. Matching the con-trollers to the DC/DC converters was dif-ficult, since there were not enough con-nections provided to the converter’sinternal circuitry. Inhibit inputs to theconverters, could be used to sequenceseveral of them, while external circuitry,occupying large board space, wasrequired to perform any other functions.When considering all the external com-
ponents, the actual result was adecrease of the overall power densityfor the power conversion functions.
Additionally, customer’s system engi-neers were spending inordinate amountof design time trying to implement all thecircuitry required to support the DC/DCconverters. Troubleshooting and testingthe system added to the design timeand became an issue as well. With thequick and frequent improvements andthe resulting changes of customer’send-equipment, the design and develop-ment time increased to unacceptablelevels and needed to be reduced.
During the same time, arguments,about the merits of various systempower architectures on the circuit boardsuch as: Distributed Power Architecture(DPA), Intermediate Bus Architecture(IBA) and other high to low voltage con-version methods were going on as well.As the number of voltages increasedand exceeded three voltages per board,IBA became the architecture of choice.
Examining these customer’s needsand comparing them to what was avail-able on the market from both analogconverter and semiconductor manufac-turers, it became obvious that dramatic
improvements in performance could beachieved by combining intelligence andpower conversion. Since electronicbrains by their nature are digital, it madesense that the converter brains be digi-tal as well.
The building block of the whole archi-tecture had to be a digitally controlledPoint of Load (POL). While the switchingsection could use a traditional driver thatpowers two output FETs, (Figure 1) thecontrol circuitry required the use a digital Pulse Width Modulated (PWM) controller.
The PWM section of the controller(Figure 2) operates by monitoring thePOL output voltage. The output voltagein turn is summed with a programmableratio of output current and comparedwith a reference voltage. The sum of thesignals is amplified and converted bythe Analog to Digital Converter (ADC) to a digital value, filtered by a DigitalFilter and fed to the Digital Pulse WidthModulator (DPWM). The DPWM con-verts the filtered digital value to a PWMsignal which controls the FET drivers.The bandwidth of the circuit allows thecalculation of a new value of error atevery cycle for switching frequencies ofup to 2 MHZ.
DIGITAL POWER
voltage of the LDO will be only a few100 mV higher than the output voltage.
The DC/DC step down converterworks at 1MHz to allow the use of smalland inexpensive external components.Only a 4.7µH inductor and a 4.7µFcapacitor are required—no externalSchottky diode. The achievable efficien-cy with low ESR coils is 95% at 100 mA.
An increasingly popular feature in cel-lular phones is polyphonic ring tones.Users enjoy listening to music samplesand want to move beyond the simplebuzzer sounds of previous generationphones. To achieve this, it is necessaryto drive a small speaker at high power.The required 0.5 W to 1 W power in an8 Ohm speaker can only deliver by abattery powered audio amplifier, and fits perfectly into a PMU. High powersupply rejection, especially in TDMAsystems, is also important for audioamplifier (217 Hz).
More and more cell phones areequipped with color displays requiringwhite LEDs for backlights. The typicaldropout voltage of white LEDs is in the3.5 V to 4.2 V range and thereforeneeds a step up converter. Display manufacturers integrate this LED into
the display module and connect them inseries. Three to four LEDs require 10 Vto 16 V total supply voltage, typicallygenerated by a DC/DC step up convert-er. The output voltage is regulated to theactual voltage drop of the white LEDs inorder to achieve the best efficiency.
Another common function of PMUs isthe battery charger. The requirementscome in three stages of charging: tricklecharge for empty batteries, constantcurrent and finally constant voltagecharging. All parameters should be pro-grammable to meet different chargingalgorithms. Additionally, a real batterycurrent integrating “fuel gauge” helps toprovide an accurate and linear batteryindicator for the user.
Next GenerationsThe typical partitioning of a cellular
phone shows the digital basebandprocessor, the analog baseband, memo-ries, the radio and the power chip. Theanalog and digital baseband are alsomanufactured in CMOS as is the PMU.Using the latest technology (such as0.13 _m) shrinks digital components bya factor of two (compared to 0.18 _m),but analog components are reduced byonly a few percent. The engineeringeffort required for a digital shrink is
limited, but analog cells have to be completely redesigned. Therefore, theoptimum partitioning for future platformswill be the marriage of analog basebandand power management units. As men-tioned before, the 0.25 _m CMOS tech-nology with the high gates densityallows for increased digital functionalities.All baseband Sigma Delta Convertersneed significant digital pre- and post filtering, but it is feasible using this technology.
Necessity of increased battery lifetimewill demand additional DC/DC convert-ers (e.g. RF-power amplifier supplies).Combining the different blocks, highpower switching components, and lownoise A/D converters on one piece of sil-icon will be a tough challenge for PMUdesigners. Careful floor-planning andlayout, as well as unique on-chip isola-tion techniques, will enable design engi-neers to develop stable, reliable highlyintegrated PMUs.
Figure 2. Symbolic schematic to show the two loops.
www.austriamicrosystems.com
PORTABLE POWER
www.powersystemsdesign.com 4544 Power Systems Design Europe April 2005
Power Manager (DPM) and a Z-POLconverter. The DPM bi-directionallycommunicates with up to 32 Z-POLsthrough a single wire Z-One Digital Busand with the main system through astandard I2C Bus.
The DPM, through a GUI I2C inter-face, besides communicating with themain system, is capable of setting eachone of Z-POL’s parameters, such as:output voltage, sequencing, tracking,protection type and thresholds, frequen-cy, fault propagation, interleave, and
feedback loop compensation. The DPMconstantly monitors each Z-POL’s out-put voltage, current and temperature, as well as the status of: over and undervoltage protection (OVP & UVP), powergood high and low (PGH & PGL).
Benefits from the usage of Digital IBAvs. standard analog IBA are realized in:product performance, development time,material logistics management, andlower production costs. While the mostobvious overall benefit is cost saving,the reduction in component count,reduction of different part numbers,reduction of number of vendors, reduc-tion in the number of traces that needsto be laid out on a circuit board, and theoverall savings in board space are just afew additional benefits that can be real-ized. With frequent new end-equipmentreleases, shorter design times, quickerdevelopment time not only saves devel-opment cost but allows for increasedrevenues resulting from faster new endequipment production releases.
While many of the functionsdescribed, could be achieved by usinganalog DC/DC converters and externaldiscrete control and monitoring ICs,such systems, besides occupying largeboard space would be very complicatedand expensive. The simplicity andsophistication of the Z-One Digital IBAprovides a cost effective solution thataddresses all the requirements for newcircuit board power requirements. UsingDigital Power Conversion with the inte-grated power management supports thecustomer’s needs, by allowing quickerintroduction of competitive products atcompetitive costs.
www.power-one.com
Figure 4. Benefits of Z-One Digital System Implementation.
Figure 3. Z-One Digital System.
Figure 2. PWM Section of the Controller.
DIGITAL POWER
Using digital PWM control opened thehorizon to cost effective implementationof many additional functions by beingable to change the operation and thecharacteristics of the converter with simple software programming.
The addition of communications capa-bilities, allowed the Z-POL to communi-cate with other Z-POLs and through aDigital Power Manager (DPM) with thehost system.
The DPM, functions as a traffic con-troller, communicating bi-directionallyover a single serial bus with all the Z-POLs on the circuit board while interfac-ing with the host system through a stan-dard industry accepted I2C bus.
The Z-One Digital IBA systemachieves both logistics and performancebenefits by simply using only twosophisticated components: a Digital
Figure 1. Z-POL Digital PWM Converter.
DIGITAL POWER
www.powersystemsdesign.com 4746 Power Systems Design Europe April 2005
Unlike traditional packaging materials,AlSiC enables a tailored coefficient ofthermal expansion (CTE), offering com-patibility with various electronic devicesand assemblies. The isotropic CTE valueof AlSiC can be adjusted for specific appli-cations by modifying the Al-metal/SiC-par-ticulate ratio. AlSiC’s CTE matching capa-
bilities eliminate the need for thermalinterface stacking, increasing reliability in the field.
AlSiC also exhibits a high thermal con-ductivity that results in extremely efficientthermal dissipation. Coupled with itssuperior CTE matching, AlSiC’s high ther-mal conductivity prevents the bowing andflexing of packaging and substrate materi-al that can lead to failure. Traditionalpackaging materials with lower thermaldissipation can cause delamination, lead-ing to air gaps and poor reliability.
The CPS AlSiC near and net-shapefabrication process both produces thecomposite material and fabricates theproduct geometry, resulting in a cost-effective product and allowing rapid
prototyping for high volume advancedthermal management solutions. Theunique casting process enables integrationof very high thermal conductivity inserts(>1000 W/mK) or cooling tubes for moreadvanced thermal management solutions.
CPS Corporation is the worldwideleader in the design and production ofAlSiC (Aluminum Silicon Carbide), ametal matrix composite that provideshighly reliable and cost-effective thermalmanagement solutions for power elec-tronics including base plates for insulatedgate bipolar transistors (IGBT).
For more information about CPS’ AlSiC,visit booth # 12-125 at PCIM Europe orcall (508) 222-0614.
AlSiC Metal Matrix Composite
www.alsic.com
Bergquist’s new Sil-Pad 1500ST (SoftTack) electrically insulating thermalinterface material provides exceptionallyhigh thermal performance, even at contact pressures as low as 69 kN/m2(10 psi). The material is designed to provide an electrically insulating inter-face between a power device and itsheatsink. Sil-Pad 1500ST is inherently
tacky on both sides, allowing it to beautomatically dispensed and placedwithout the use of adhesives.
Sil-Pad 1500ST is an elastomericmaterial reinforced with fibreglass forresistance to cut-through. Supplied inthicknesses of 0.2mm and 0.3mm, it isdesigned for continuous use across awide temperature range (-60°C to+180°C), providing a thermal conductivi-ty of 1.8W/mK. With a dielectric con-stant of 6.1 (at 1kHz) and volume resistivity of 1011Ohm-meter, Sil-Pad1500ST provides electrical insulation inapplications such as power supplies,automotive and motor control.
The ability of Sil-Pad 1500ST to wetout the surfaces to which it is applied –even at very low pressure – ensures itsthermal performance is consistent
across a wide range of contact pres-sures. Bergquist supplies Sil-Pad1500ST 8 mil in roll form for auto dispensing. Custom die-cuts are also available.
Sil-Pad 1500ST is the latest inBergquist’s extensive range of thermallyconductive, electrically insulating Sil-Pad materials. Like all Sil-Pad materials,Sil-Pad 1500ST provides a clean andefficient alternative to the use of mica,ceramics or grease in thermal manage-ment for electronic assemblies. In par-ticular, the use of Sil-Pad materials caneliminate the risk of cracking or break-age caused by the brittle nature of mica,which must be used in combination withgrease to improve thermal conductivity.
NEW PRODUCTS
Low Pressure Thermal Management Material
www.bergquistcompany.com
There is a new magnetically shieldedSMD Power Choke of the series WE-PD: The Type XS. That is the smallestdimension in the group of WE-PD with asurface of only 6 mm_ and a height ofonly 3.3 mm.
Being available in the most commoninductance values, this componentsseries is perfect for different switch-con-troller applications. The very low DCR-
value allows high efficiency for DC/DC-converters. The WE-PD Type XS islead-free and is suitable for the mostpopular ICs like Linear Technology,National Semiconductor, OnSemiconductor, ST Microelectronics,Texas Instruments and Fairchild.
www.we-online.com
Very flat SMD Power Chokes for IC-Design
NEW PRODUCTS
LEM has introduced the ITB 300-S low-cost, high-accuracy current transducer.Specified for 300 ARMS nominal meas-urements, this new model joins the ITcurrent transducer at a lower price pointwhile offering a level of performance that
is only slightly lower than other membersof the family. Linearity is better than0.001% and overall accuracy at ambienttemperature is 0.05%. Thermal offset driftis very low, at only 1µA/K.
Featuring galvanic isolation, the ITB300-S can be used for current measure-ment of any type of waveforms (includingDC, AC, mixed and complex). It has beendesigned to operate from a bipolar +/-15V DC power supply, and will accommo-date a round primary conductor of up to21.5 mm diameter.
In addition to its normal current output(150 mA for 300 A primary), an additionaloutput indicating the transducer state(opened or closed contacts) is available.
The new model offers an extendedoperating temperature range of -40 to+85°C, compared to the +10 to +50°C ofthe IT models, allowing its use in a widerrange of applications, including high preci-sion power supplies, medical equipment,calibration units, precise and high stabilityinverters, power analysers and metering.
The transducer is CE marked, con-forms to the EN 50178 and EN 50155standards and is suitable for industrialand traction applications.
LEM Components offers these trans-ducers with a two-year warranty.
Visit LEM at PCIM 2005, Hall 12/401.
High-Accuracy Current Transducer
www.lem.com
Seiko Instruments’ S-1167 Series volt-age regulators are ideally suited aspower supplies for portable equipment,battery-powered devices, and cellularphones requiring ultra low current con-
sumption, a small package design andthe reliability and accuracy of CMOSbased technology.
Ultra low power consumption, highripple rejection, low dropout voltage(150 mV typical), and high output volt-age accuracy (±1%) are distinctive fea-tures that make this positive voltageregulator an extremely reliable and cost-effective solution.
With a current consumption level of9µA typical during operation, the S-1167realizes a 70dB ripple rejection rate;optimal for noise sensitive applications.The device boasts a wide input voltagerange of 2.0 to 6.5V while providing a1.5 to 5.5V selectable output voltage in0.1V steps.
Two compact package designs can bespecified to meet individual integrationrequirements (SNT-6A(H) and SOT-23-5).A built-in shutdown circuit ensures bat-tery longevity and energy management,while integral overcurrent protectionassures continuous performance.
Technical Contact:Jim Schlumpberger, Seiko InstrumentsUSA, Inc., 2990 Lomita Blvd.Torrance, CA 90505Phone: (310) 517-7884 E-mail: [email protected]
Voltage Regulator for Portable Equipment
www.sii-ic.com
Advanced Power Technologyannounced two new IGBT products forTIG Welding applications: APT200GN60Jand APT200GN60JDQ4. These discretedevices use the latest generation FieldStop Trench Gate Technology and areoffered in the industry standard SOT-227 package. These products arespecifically targeted for TIG weldingapplications, and are an addition toAPT’s new product line of discrete FieldStop IGBTs. APT200GN60J is a singleIGBT and APT200GN60JDQ4 includesa 100A anti-parallel diode. Other anti-
parallel diode options are available.These IGBTs are used in the output
inverter of aluminum TIG welders. Thedevice can also be used in any applica-tion requiring very low conduction loss-es at high currents and 0 to 30 kHzswitching speed.
Data sheets are available to assist thedesigner, describing the features andbenefits of this new 600V IGBT ingreater detail. These may be down-loaded from APT’s website.
www.advancedpower.com
IGBT for TIG Welder Applications
www.powersystemsdesign.com 4948 Power Systems Design Europe April 2005
Ideal Vicor subsdiary Picor annoncesthe introduction of three new additionsto the QPI family of active EMI filters.The new products, QPI-3, QPI-5 andQPI-6, support 24 and 48V DC-DC con-verter applications with higher current
ratings and lower costs. Providing atten-uation from 150kHz to 30MHz, the QPIproducts deliver over 40dB of common-mode and more than 70dB of differen-tial-mode noise attenuation at 500kHz.All products are pin and footprint com-patible with previous versions in com-pact (25.4 x 25.4 x 5.1mm) surface-mount packages, offering board spacesavings of 50% to 80% over passiveEMI filter solutions.
The QPI-6 meets the specifications forthe international 36 to 76Vdc telecombus, including the 100V, 100ms surge
requirement. Rated at 14A, the unit sup-ports single or multiple DC-DC convert-ers requiring up to 672W of input powerat 48 volts input. The QPI-3 and QPI-5operate over a 10 to 40Vdc range tosupport industrial bus and COTSdefence applications. The QPI-3 andQPI-5 are rated at 7 and 14A, respec-tively. The units support single or multi-ple DC-DC converters requiring up to360W of input power at 24V input.
E-mail:[email protected]
www.vicoreurope.com
NEW PRODUCTS
Companies in this issue
Company Page Company Page Company Page
ABB Switzerland . . . . . . . . . . . . . . . . . . .1Advanced Power Technology . . . . . . .41Advanced Power Technology . . . . . . . .47Agilent Technologies . . . . . . . . . . . . . . .38Allegro MicroSystems . . . . . . . . . . . . . . .1Anagenesis . . . . . . . . . . . . . . . . . . . . . . .1Analog Devices . . . . . . . . . . . . . . . . . . . .1Ansoft . . . . . . . . . . . . . . . . . . . . . . . . . .13Ansoft . . . . . . . . . . . . . . . . . . . . . . . . . . .1APEX Microtechnology . . . . . . . . . . . .11Artesyn Technologies . . . . . . . . . . . . . . .1Austria Microsystems . . . . . . . . . . . . . .40Bergquist . . . . . . . . . . . . . . . . . . . . . . . .47CT-Concept Technology . . . . . . . . . . .21CT-Concept Technology . . . . . . . . . . . . .1Danfoss . . . . . . . . . . . . . . . . . . . . . . . .26Epcos . . . . . . . . . . . . . . . . . . . . . . . . . .31eupec . . . . . . . . . . . . . . . . . . . . . . . . . .1,6Fairchild Semiconductor . . . . . . . . . .C2
Fairchild Semiconductor . . . . . . . . . . . . .1Fuji . . . . . . . . . . . . . . . . . . . . . . . . . . . .23Hitachi . . . . . . . . . . . . . . . . . . . . . . . . . .33Infineon . . . . . . . . . . . . . . . . . . . . . . . . . .1International Rectifier . . . . . . . . . . . . .C4International Rectifier . . . . . . . . . . . . . . . .1iSuppli . . . . . . . . . . . . . . . . . . . . . . . . . .20IXYS . . . . . . . . . . . . . . . . . . . . . . . . . . .44LEM Components . . . . . . . . . . . . . . . . .3LEM Components . . . . . . . . . . . . . . . .1,46Linear Technology . . . . . . . . . . . . . . . . .5Linear Technology . . . . . . . . . . . . . . . .1,8Maxwell . . . . . . . . . . . . . . . . . . . . . . . . .22Micrel . . . . . . . . . . . . . . . . . . . . . . . . . . .9Murata . . . . . . . . . . . . . . . . . . . . . . . . . . .6National Semiconductor . . . . . . . . . . . . .1Ohmite . . . . . . . . . . . . . . . . . . . . . . . . .19Ohmite . . . . . . . . . . . . . . . . . . . . . . . .1,12On Semiconductor . . . . . . . . . . . . . . . . . .1
PCIM Europe 2005 . . . . . . . . . . . . . . . .27Power Integration . . . . . . . . . . . . . . . .15Power One . . . . . . . . . . . . . . . . . . . .43,48Power Sources Manufacturing Assoc . . .4Power Systems Design Europe . . . . .C3Rutronik . . . . . . . . . . . . . . . . . . . . . . . . . .4Semikron . . . . . . . . . . . . . . . . . . . . . . .1,4Sieko Instruments . . . . . . . . . . . . . . . . .46Siemens . . . . . . . . . . . . . . . . . . . . . . . . .28ST Microelectronics . . . . . . . . . . . . . . . . .6SynQor . . . . . . . . . . . . . . . . . . . . . . . . . .1Texas Instruments . . . . . . . . . . . . . . . . .7Texas Instruments . . . . . . . . . . . . . . .1,14Tyco Electronics . . . . . . . . . . . . . . . . . . .1Vicor Europe . . . . . . . . . . . . . . . . . . . . .48Vishay . . . . . . . . . . . . . . . . . . . . . . . . . .17Vishay . . . . . . . . . . . . . . . . . . . . . . . . . . .6Würth Elektronik . . . . . . . . . . . . . . . . . .47
Please note: Bold—companies advertising in this issue
Active EMI Filters for DC-DC
Power-One introduces the BLP55-Series of 55W single-output ac-dc powersupplies. Benchmarked with all availableoutputs: 5, 12, and 24V, the cost-effec-tive BLP55 utilizes superior thermal-management techniques to deliver full-rated power with only 10 CFM cooling.BLP55 products are designed to meetthe rigorous requirements of networking,data communications, commercial, andindustrial applications.
Reduced airflow requirements facili-tate greater overall-system layout flexi-
bility as the BLP55 does not have to belocated directly adjacent to the systemcooling fan. Depending on system con-figuration, the BLP55 may also eliminatethe need for a dedicated power-supplycooling fan, especially in space-con-strained 1-U “pizza-box” applications.
A compact 3.00" x 5.00" x 1.25"(76.2mm x 127.0mm x 31.8mm) pack-age combines enhanced features andperformance with compliance to estab-lished industry-standard footprints andconnectors. Standard features include
remote sense, internal overvoltage pro-tection, and 85-264VAC wide-range input.
Regulatory agency approvals includeUL recognition to UL60950-1/CSA 22.2No. 60950-1 and TUV approval toEN60950-1. Onboard EMI filtering pro-vides Class B compliance to FCC CFRTitle 47, Part 15, Sub-Part B -Conducted; and EN55022/CISPR 22Conducted Class B.
www.power-one.com
55W AC-DC Power Supply Combines Superior ThermalManagement with Low Cost
Get your own copy of Get your own copy of
Sign up today online at www.powersystemsdesign.comSign up today online at www.powersystemsdesign.com
your trade journal for the power engineer!your trade journal for the power engineer!
FREE Subscription!FREE Subscription!
