EP&Dee no 3 - March 2013

EP&DeeDESIGN & MANUFACTURING MARCH, 2013 ISSUE NO. 3, VOL. 11

E L E C T R O N I C S P R O D U C T S & D E S I G N E A S T E R N E U R O P E

Bridging the skills gapin embeddedconsumer design

p16

THE EAST EUROPEAN RESOURCEFOR EMBEDDED APPLICATIONS

EP&Dee | March, 2013 | www.epd-ee.eu2

Table of Contents

MARCH 2013

EUROSTANDARD PRESS 2000

Tel.: +40 31 805 9955

Fax: +40 31 805 9887

[email protected]

www.esp2000.ro

VAT Registration: RO3998003

Company number: J03/1371/1993

EP&Dee (Electronics Products & Design Eastern Europe) is published 10 times per year in

2013 by Euro Standard Press 2000 s.r.l.

It is a free to qualified electronics engineers and managers involved in engineering

decisions. Starting on 2010, this magazine is published only in digital format.

Copyright 2013 by Euro Standard Press 2000 s.r.l. All rights reserved.

© 2013 by Eurostandard Press 2000

EP&DeeSubscriptions:

office@epdee.eu

EP&DeeWeb page:

www.epdee.eu

Group Publisher DirectorGabriel Neagu

Managing DirectorIonela Ganea

AccountingIoana Paraschiv

AdvertisementIrina Ganea

WEBEugen Vărzaru

Contributing editorsRadu Andrei Ross Bannatyne

ConsultingMarian BlejanBogdan GrămescuMihai Savu

Asian Reprezentative Taiwan Charles YangTel: +88643223633

[email protected]

EP&Dee is offering its readers the chance to win aMicrochip F1 Evaluation Kit, containing an F1Evaluation Platform and PICkit3. The F1 EvaluationPlatform is a simple development tool for EnhancedMid-range PIC microcontrollers(PIC12F1XXX/PIC16F1XXX) and demonstrates thecapabilities & low power enhancements of these new

PIC microcontrollers. Included with a PICkit3 for quickprogramming, this kit provides a platform for generalpurpose development and gives you the ability todevelop code for any PIC12F1XXX/PIC16F1XXXmicrocontroller. Quick & easy development is ensuredwith the integrated functionality including: prototyping area, LCD control, system current moni-toring, temperature sensing, Real-Time-Clock, LEDdrive, button control, and BLDC motor control.

The kit includes a F1 Evaluation Platform, PICkit 3 In-Circuit Debugger, USB Cable, Platform &Demonstration Guide, Source code for includeddemonstrations and Board schematics.

Win a Microchip F1 Evaluation Kit

For the chance to win an F1 Evaluation Kit, please visit:http://www.microchip-comps.com/epdee-f1 and enteryour details in the online entry form.

EDITORIAL

EMBEDDED WORLD 2013 - INDUSTRY NEWS (p 6, 7, 8, 9, 10, 11)

Distributors distribute, right? Common sense dictates this to be, well obvious and true. But the past 5 years have seen a transformation of the high-service electronics distribution industry where business models and service levels have been redrawn to become more relevant, more competitive and more attractive.

DESIGN FEATURES

1 2 From Visual Studio to FPGA HardwareA snapshot on the current state-of-the-art in FPGA software to hardware compilation.

20 Microcontrollers for human/machine interfaces Please touch!Today, touch functionality is absolutely indispensable for a modern human-machine-interface. All leading micro-controller manufacturers offer sophisticated solutions here. This means that developers are spoilt for choice.

22 It's easy to over-specify your AC/DC supply but you shouldn't.You can buy more supply than you need, yet less is actually often better, says Don Knowles, VP Engineering at N2Power.

24 Agilent E363xA Series - Programmable DC Power SuppliesOne of the product groups offered by TME consists of so-called programmable DC power supplies.

26 Driver Assistance Systems with the Power of FPGAsFPGAs can be leveraged to quickly bring new driver assistance innovations to market.

30 Measuring energy flows in cross-border rail traction operationsNew standards impose accuracy requirements for energy metering that represent a massive step-change relative to prior industry practice.

PRODUCT NEWS

Embedded Systems(p 14, 34, 35, 36, 37)Active Components(p 38, 39)Display (p 40)Sensors (p 41)

COVER STORY

Bridging the skills gap in embedded consumer designThe effective use of COTS solutions and manufacturer development kits reduces the learning curve for implementing new technologies and enables embedded products to get to market faster, explains Martin Hill of Microchip Technology.16

3836

10

7 6

4 The changing face of distribution: 3D isn’t just for the TV…or the printer


INDUSTRY NEWS EMBEDDED SYSTEMS

Distribution is 1DSo distributors have gotten smarter; they knowthe 1D of the traditional Distribution modeldoesn’t cut it anymore. At the heart of high-service distribution is hard core asset manage-ment of a vast array of relevant products, allavailable for fast delivery – giving designerswhat they want, when they want it and at acompetitive price. On top of this core offering,new services have been developed and lay-ered by the distributor: free technical support,packaging options such as full and half reels,re-reeling, drilled tubes and guarantees formoisture sensitive parts. Products can besearched for and found by architecture, tech-

nology type, application or end market (oh,and part #); customers with affinity for a pre-ferred distribution partner can be rewardedthrough loyalty programs; and, alliances withspecialist third-party providers such as PCB

fabricators allow distributors to broaden theirservices and move further to the left of theproduct development continuum. (Figure 1)

Development and Distribution is 2DHigh-service distributors no longer just pro-vide the parts needed to assemble and pro-duce a board level product or test equipmentneeded to verify the design; they are becom-ing more involved in Development, the 2Doffer they have already adopted. A number ofdistributors for example provide their own PCBlayout and editing tools such as EAGLE. Theyinvest in the software and IP, they deve lop theroadmaps and they support the engineer.

This board-levelintegration is alsomanifest in the waydistributors havewoven their parts

databases into other mainstream PCB tools.From within the UI of most PCB/ CAD tools,engineers can access millions of orderableparts in real time, check on detailed informa-tion such as parameters, stocking, pricing,

compliance and download symbols, footprintsand datasheets. They can also, referencing thecomment above, directly connect with a PCBfabricator and get prototype PCBs deliveredto their door…all without leaving their familiardevelopment environment.The focus on development can also be seenin the increasing levels of embedded soft-ware tools that distributors provide and thetechnical support they wrap around it.Development Boards and Kits (includingexclusives), RTOS, JTAG, debuggers, Flashers,EDA tools and IDEs to name but a few, can allbe supplied. A rich pool of engineeringknowhow has been made available online as

distributors exploit their extensive network ofglobal supplier relationships to aggregatecontent such as app notes, datasheets, videotutorials, blogs, manuals, white papers, codesnippets and design examples.This aggregation of engineering knowhow bythe distributor is also embracing ‘developer’generated content through active participa-tion in bespoke communities and socialmedia. The largest design engineering

Distributors distribute, right? Common sense dictates this to be, well obvious andtrue. But the past 5 years have seen a transformation of the high-service electronicsdistribution industry where business models and service levels have been redrawn tobecome more relevant, more competitive and more attractive. Relevant in the senseof keeping pace with shrinking development cycles and rapid advances in technology,competitiveness in a global economy that transcends geography and traditionalallegiances, and attractiveness by offering more value to the engineer.

by Chris SullivanTechnical Marketing Farnell element14

The changing face of distribution:3D isn’t just for the TV…or the printer.

Figure 1: The product development continuum, where the level of abstractionincreases as you move to left from the physical prototype or manufacturing pilot.

It’s no Fluke as Farnell element14 AnnouncesStock of RevolutionaryThermometer ProductFarnell element14 announces it will stock theVT02 Visual IR Thermometer from Fluke whichis revolutionising temperature measurementswith thermal blending.The product exists in the new tool category forVisual IR Thermometers and works by taking avisual image and applying a thermal heat mapover the image thus displaying an image withboth temperature and visual contextual content. The product offers five key benefits over othertemperature measurement devices:• Ability to view visual image of exactly what isbeing measured giving an accurate temperaturemeasurements.• Eliminates the task of taking multiple grid temperature readings manually by providingone blended image.• Near mode allows the user to get as close as 6 inches to allow close up blended images supporting diagnosis. Blended images allowsimultaneous tracking of both hot and cold.

• Little or no training isrequired due to compact andintelligent, focus free andbuilt in intelligence design.• The thermometer simulta-neously saves thermal heatmap data and visual imagesto an included MicroSD cardwhich can be viewed inSmartView software whichcan then produce reports foryour customer highlightingareas of concern.

The thermometer will be use-ful in the following industries:• Electrical – inspection ofwiring and potential over-loaded circuits;• HVAC Refrigeration –inspections on ducts, blowersand motor assemblies;• Industrial – preventativemaintenance on motors,bearings or electrical systems;• Automotive – troubleshoot-ing engines, brakes and heat-ing / cooling systems;• Property Management andremodelers – installing infloor heating systems andbasic temperature inspections.

To view the Google Hangouton the Fluke VT02 clickwww.youtube.com/watch?v=G68VXIy-iPI&feature=plcp

www.epd-ee.eu | March, 2013 | EP&Dee 5

community, Newark’s element14 Community, is a space forpeer review, discussion, debate,fun (!) and the sharing of ideasand experiences. It’s a channelthat semiconductor companiesand component suppliers arebecoming increasingly tunedinto as it gives extra voice to thedeveloper and a deeper insightand understanding into whatdrives them.

INDUSTRY NEWS EMBEDDED SYSTEMS

Using conventional wisdom,this is not what you wouldexpect from a 1D high-servicedistribution business model.

Design, Development andDistribution is 3DSo as this model continues toevolve, so distributors will con-tinue their journey to the leftof the product developmentcontinuum. To enable this,adding a true Design capabilityto the high-service distributionoffer is a logical next step. This offers an even greaterlevel of depth about thedesign process that buildsupon a distributor’s core com-petence in asset management,the global logistics and supplychain muscle, and the develop-ment solutions that support theengineer from end to end.

By providing newsolutions that areincreasingly moreabstract than board-level components,distributors areincreasingly placingthe designer, the developer and theengineer at the heartof their strategies,

(Figure 2) turningthe myth of

single source dis-tribution into a

very real pos-sibility. n

Figure 2: Design, Development and Distribution brings together the essential services and products designers need to understand, evaluate and access new technology quickly.


INDUSTRY NEWS EMBEDDED WORLD 2013

At the 2013 Embedded WorldExhibition, Maxim IntegratedProducts, Inc. presented theCupertino (MAXREFDES5#) sub-system reference design, a pre-cise drop-in-ready isolated 16-bit analog front-end (AFE). Thishighly integratedsubsystem refer-ence designmeets the needsof industrial sen-sors, process con-trol, and pro-grammable logiccontrollers (PLCs).Its analog -10V to+10V, 0 to 10V,and 4-20mAinputs support the ubiquitousanalog outputs of industrial sen-sors. In addition, both powerand data are fully isolated.Maxim provides all the hardwaredesign files, example drivercode, and test results needed tospeed design development. Industrial control and automa-tion applications often requireisolation, high resolution, and awide range of input voltages.Many industrial sensors have

high or wide-ranging analog out-put voltages. Field-programma-ble gate arrays (FPGAs) andmicrocontrollers often cannotaccept these high analog volt-ages directly. The Cupertino

design solves this problem byproviding an interface solutionbetween the sensor and thecontroller while also integratingisolated power and data, all in asmall 96.52mm x 20.32mm formfactor. The Cupertino AFE con-nects directly to FPGA/CPUdevelopment kit expansionports that conform to thePmod™ standard from Digilent.MAXIM INTEGRATEDwww.maximintegrated.com

Maxim Integrated Demonstrates a Drop-InAnalog Front-End (AFE) Solution for IndustrialSensors and Equipment at Embedded WorldThe Standardization Group for

Embedded Technologies (SGET)has officially announced the ratifi-cation of the new SMARC™Computer-on-Modules specifica-tion. Kontron played the leadingrole in the development of thespecification which had the work-ing title ULP-COM.This new SmartMobility ARChitecturefor ARM/SoC-basedextremely compressedComputer-on-Moduleswas approved within ashort time frame. Thespeedy ratification goesto underline the mar-ket’s need for a newARM/SoC-based form factor stan-dard, as well as, the agility of thenewly created SGET body.Coinciding with the adoption ofthis new standard, three designlines have already become avail-able at Kontron. Developers cannow begin engineering innovativeultra-low power devices based onthe new SMARC™ standard. Theratification of the SMARC™ stan-dard underlines Kontron's powerto innovate in its role as an inter-national technology leader and asa 'standardizer' of Computer-on-Modules. With over 15 years of

experience in the development ofCOMs, Kontron provides exten-sive support along with notablylong-term product availability.Industrial customers, partners andmodule manufacturers profit fromthe high level of investment secu-rity which Kontron offers.

Kontron's success story began in1998 with DIMM-PC® which, in2000, led to the ETX® standardbeing licensed worldwide. TheX-board® specification followedin 2002 and ETX Express® as thetechnological basis for PICMG’s®COM Express® technology waslaunched in 2003. COM Express®was finally developed by mem-bers of the PICMG® standardsconsortium, and Kontron was a major contributor to the technology. KONTRON www.kontron.com

Atmel® Corporation announcedthe company has expanded itsthird-party ecosystem of toolsand software partners for itsrecently launched ARM® Cortex®-A5 processor-based family ofproducts. Partner companiesinclude ARM, Timesys, ExpressLogic and IAR. According to the2012 Embedded Study conduct-ed by UBM, 61 percent of itsrespondents said the microproces-sor’s (MPUs) ecosystem (software,tools support, etc.) is one of themost critical aspects for their MPUdesigns. In this same study, respon-dents indicated that EmbeddedLinux and Android are two operat-ing systems they had planned touse in the next 12 months for theirembedded designs.The expanded list of tools and

software partners enable embed-ded design engineers using theAtmel SAMA5D3 series of ARMCortex-A5 processor-based MPUsthe ability to design on a variety ofsystems including embeddedLinux, Android and other real-timeoperating systems. Each partnercompany offers a tool enablingembedded designers the abilityto get to market faster.• Qt, the leading solution in theembedded market for user inter-face implementation by Timesys,provides the perfect tools for engi-neers to design a rich set of widgetsand reference designs for devicespowered by embedded Linux. • The ARM DS-5™ toolchainoffers a complete developmentstudio for Linux developers thatneed to optimize their code for

performance and power manage-ment, or for designers that wantto debug firmware. • Partnering companies IAR andExpress Logic offer tightly integrat-ed solutions that assist designers indeveloping applications that deliv-er TCP/IP, USB, and OS functiona -lity, along with real-time perform-ance to meet today’s most deman -ding software requirements.

Partner Quotes• ARM Quote • Timesys • Express Logic • IAR More details on all our third-party partners and vendors areavailable here: www.atmel.com/thirdpartysupportATMELwww.atmel.com

Atmel Expands Third-Party Ecosystem of Tools and Software Partners for Recently Launched ARM Cortex-A5Processor-based Family

ULP-COM is now SMARC™



Microchip announces, from the Embedded Worldconference in Germany, its BodyCom™ technology,which provides designers with the world’s first frame-work for using the human body as a secure communi-cation channel. Compared to existing wireless me -thods, BodyCom technology provides lower energyconsumption, whilst further increasing security viabidirectional authentication. Because no RF antennasare required, BodyCom technology simplifies circuit-level design and lowers the bill of materials (BOM).

All of this is enabled by the BodyCom DevelopmentV1.0 Framework, which is supplied through free soft-ware libraries that work on all of Microchip’s 8-, 16-and 32-bit PIC® microcontrollers, of which there aremore than 900.BodyCom technology is activated bycapacitively coupling to the human body.The system then begins communicatingbidirectionally between a centralised con-troller and one or more wireless units.There are many applications where securewireless communication is essential, andthere is no more secure channel than thehuman body. This is especially true whenbidirectional authentication is added tosupport advanced encryption, such asKeeLoq® technology and AES. For example, BodyCom technology helps to preventthe “Relay Attack” problem that is typical in automo-tive passive-keyless-entry security systems.Most secure, short-range communication designs are

battery powered and highly costconstrained. BodyCom technologysignificantly increases battery lifeby eliminating the need for a wire-less transceiver or high-powerinductive fields. It also simplifiesdevelopment and lowers BOMcosts by making antenna designunnecessary and by using a low-fre-quency framework with a commonmicrocontroller and standard AFEfrequencies (125kHz and 8MHz),with no need for external crystals. BodyCom technology also elimi-nates the cost and complexity ofcertification because it complieswith FCC Part 15-B for radiatedemissions.Example applications includeAccess Control such as security sys-tems, home/industrial door locks,pet doors; Personal Safety &Security including equipmentaccess / disable, power tools,firearms and computer systems;Medical equipment for patientmonitoring, hospital-room accessor equipment tracking; andConsumer electronics such as pro-file management for gaming con-soles and exercise equipment.To further enable developmentand speed time to market,Microchip is also announcing theBodyCom Development Kit(DM160213) priced at $149, whichincludes a central controller unitand two wireless mobile units.The free BodyCom DevelopmentV1.0 Framework is also availabletoday and includes a communica-tion library, application code exam-ples and a development GUI foruse on personal computers.

Application note AN1391 offers anintroduction to BodyCom technology.MICROCHIP TECHNOLOGYwww.microchip.com/get/GA5E

PRQA announces support for MISRA C:2012PRQA | Programming Research, a global leaderin static analysis, announces that its tools offersupport for MISRA C:2012 (MISRA C3), withan updated compliance module for QA•CVersion 8.1, in anticipation of the new versionof the coding standard which will be publishedon 18 March. The new standard contains anumber of improvements over previous ver-sions and extends support to the C99 versionof the C language (ISO/IEC 9899:1999).

“This latest version ofMISRA C provides anew benchmark foreffective and safe use ofthe C Language,” saidPaul Burden, co-authorof the MISRA CGuidelines and PRQA’srepresentative on theMISRA C working group

for the past 13 years. “With support for C99, arange of improvements and detailed explana-tion of the rationale behind individual rules,MISRA C3 is an indispensable resource foranyone who cares about software quality andis developing systems in C.”MISRA C:2012 includes support for C99, aversion of the C language that was not widelyimplemented when MISRA C:2004 (MISRAC2) was published, while maintaining guide-lines for C90. Other developments includesome additional rules, and a wide range ofimprovements designed to make it easier forusers to understand and use the guidelines.These are based on the wide experiencegained from users of MISRA C2 around theworld, and include:• Elimination of areas of ambiguity to ensureconsistent interpretation of the rules. • Classification of rules as either “mandatory”,“required” or “advisory”. • A more explicit recognition of “decidability”,the extent to which enforcement of a rule canbe guaranteed.• Improved rule definition providing a closerfocus on dangerous constructs while avoidingunnecessary and irksome restrictions.• Expanded examples and more extensiveexplanation of the rationale behind particularrules.• Better explanation of why a user might want todeviate from (not comply with) a particular rule.MISRA C is the most widely used set of codingguidelines for development in the C language.Although originally developed for automotiveapplications it is now broadly adopted acrossmany different industry sectors.

PRQA | PROGRAMMING RESEARCHwww.programmingresearch.com

Microchip’s BodyCom™ technology is world’s first to use thehuman body as a secure, low-power communication channel

Watch a short video:www.microchip.com/get/9TMM



As part of its 'High-SpeedCompactPCI® Initiative', Kontron isshowcasing its first 3UCompactPCI® Serial (CPCI-S.0)processor board at theEmbedded World 2013 tradeshow in Nuremberg. The KontronCPS3003-SA comes equippedwith latest 3rd generation Intel®Core™ i7 proces-sors and, for thefirst time, offersPCI Express Gen3.0, USB 3.0, SATA6G and GigabitEthernet overbackplane. Thispaves the way forcompletely newapplication sce-narios and cangreatly boost performance incomparison to classicCompactPCI® installations.Significant interest inCompactPCI® Serial has beenidentified in numerous marketsegments including industrialautomation, test and measure-ment equipment as well as trans-port and traffic systems and alsoinclude medical and defensetechnology. At the same time, thenew Kontron CompactPCI® Serialclass processor board is very flexi-

ble. For OEMs, who still rely onclassic CompactPCI® boards,Kontron offers the CPS3003-SA asan option in combination with aCPCI extension module. Thismeans that both CompactPCI®Serial boards as well as classicCompactPCI® boards can operatetogether in a hybrid system. This

enables OEMs to easily continueusing existing boards without anyre-design or requirement toreplace their existing CompactPCI® boards. Not only are OEMsprovided with a clear migrationpath to a high speed point-to-point connection of CompactPCI®Serial, but they can also optimizetheir applications' time-to-marketand product lifecycle to minimizetheir total cost of ownership. KONTRON www.kontron.com

Kontron introduces first 3U CompactPCI® Serialprocessor board for modular, high-speed applications

Wireless infrastructure technologyleaders Freescale Semiconductorand AirHop Communications areestablishing solutions that inte-grate Freescale’s high-performanceQorIQ Qonverge processors withAirHop's eSON™ advanced self-organizing network (SON) soft-ware to speed deployment of het-erogeneous networks (HetNets)around the world. While adding small cell base sta-tions to the world’s networksincreases capacity and comple-ments traditional macrocell basestations, it challenges serviceproviders to coordinate and man-age interaction between growingnumbers of network cells. AirHopeSON™ software support forQorIQ Qonverge processors isdesigned to facilitate the coordi-nation of macrocell base stationsand small cells, optimizing mobiletraffic via dynamic load balancingand inter-cell interference man-agement. This allows QorIQQonverge customers to accelerate4G system deployment and opti-mize broadband coverage ofHetNets. The QorIQ Qonvergeportfolio is comprised of integrat-ed base station-on-chip devices

built on advanced heterogeneousmulticore technology. The broadfamily of products addresses multi-standard requirements scalingfrom residential femto to macrocells with a common architecture.AirHop’s advanced RAN Intelli -gence solutions manage interfer-ence and optimize system capacityin multi-cell and multi-layer net-work configurations. AirHop’seSON™ software suite providesreal-time optimization of availableradio resources via multi-cell coor-dination to effectively providecoverage, capacity and quality ofservice (QoS) for each cell andindividual user. This is especiallycritical in HetNet deploymentswhere a large number of supple-mental small cells can exist within amacrocell coverage area. AirHop’s eSON™ software enablescarriers to deploy small cells on anas-needed basis, leading to signifi-cant opex savings. It also improvesnetwork capacity by utilizing real-time, multi-cell coordination andoptimization. FREESCALE SEMICONDUCTOR www.freescale.com

Freescale and AirHop collaborate on self-organizing network software support forQorIQ Qonverge base station-on-chip portfolio

Renesas Electronics, a premier provider ofadvanced semiconductor solutions, has devel-oped a new split gate (SG) flash memory cir-cuit technology for on-chip flash memorymicrocontrollers (MCUs) that adopts theindustry's leading-edge 40 nanometer (nm)process technology achieve high reliability,low power consumption and the industry’sfastest random access operation speeds.In addition to higher level control, equipmentthat requires high reliability, e.g., automotive,is also now requiring even higher functionalityand diversity of functions, such as support forfunctional safety, security, and networking.Along with such market demands, since evenlarger capacities are required in the flashmemory that stores the MCU's software anddata, it is seen as desirable to increase theintegration density of both the flash memoryand peripheral functions included on a singlechip by the use of even finer features in thefabrication process.

To support these needs, Renesas continues todevelop leading-edge solutions based onadvanced process technologies, and startssample shipments from 2Q CY2013 of auto-motive flash MCUs that adopt both the indus-try's leading-edge 40nm process for flashMCUs and the SG-MONOS structure flashmemory, which has a proven track record interms of high reliability, high speed, and lowpower consumption.

Key Features of the Newly-Developed SG-MONOS Flash Memory Technology:

• Circuit technologies achieve even fasterreadout

• Circuit technologies achieve high rewrite durability

Using these new technologies, Renesas hasprototyped both 4 MB program storage flashmemory and a 64 KB data storage flash memo-ry fabricated in a 40nm generation process,and has achieved operation at over 160 MHz

and high readout speed of 5.1 GB per second– the industry's highest speed for programstorage flash memory. Previously, Renesas ver-ified operation at up to 120 MHz in its 40nmgeneration process products. Leveraging thesenew technologies, Renesas has now verified a33% characteristics improvement. Also, in datastorage flash memory, this technologyachieved 10 million rewrite cycles, a criticalissue in automotive MCUs, even under thehigh-temperature conditions of Tj = 170°C.This indicates that Renesas 40nm automotiveflash memory has great potential in terms ofrewrite cycle counts.

Renesas is hopeful that using this flash memo-ry circuit technology can contribute signifi-cantly to creating automotive flash memorythat provides even higher performance andreliability.

RENESAS ELECTRONICS EUROPEwww.renesas.eu

Renesas Electronics announces development of Flash Memory Technology that achieves High-Speed ReadOperation and High Rewrite Cycle Counts for 40 nm On-Chip Flash Memory MCUs for Automotive Applications



Kontron launches 'High-SpeedCompactPCI® Initiative'Kontron now supports PICMG®CompactPCI® specifications CPCI-S.0and serial mesh based on PICMG® 2.20At the Embedded World 2013 trade show,Kontron (hall 1, stand 460) has announced thelaunch of its 'High-Speed CompactPCI®Initiative'. As of today, Kontron will extend itssupport to include two CompactPCI® stan-dards, CompactPCI® Serial (CPCI-S.0) for 3Uform factors with PCIe, GbE, USB and SATAover backplane, as well as CompactPCI® SerialMesh (based on PICMG® 2.20) for 6U formfactors with 10GbE over backplane. This initiative gives the CompactPCI® busarchitecture a next generation productenhancement providing CompactPCI® appli-cations with an enormous performance boost,and extending the investment security ofCompactPCI® technology for another decade.

The scope of applications using 3UCompactPCI® Serial (CPCI-S.0) is extremelybroad and varied. It ranges from multi-proces-sor systems for the computation of compleximaging processes to high-performance videoor radar data recording through flexibleSATA/RAID architectures, and from high band-width wireless communication with parallelworking radio modules WLAN, UMTS, LTEright up to powerful multi-monitor systems incontrol rooms. “CompactPCI® Serial isa logical enhancementto the very successfuland worldwide-accept-ed CompactPCI® speci-fications. It enables anew generation of high-performance installa-tions requiring massivebandwidth, and alsoprovides a way to boost the performance ofany existing CompactPCI® deployments in awide range of industries,” explains NorbertHauser, Executive Vice President Marketing atKontron. “CompactPCI® Serial allows datatransmission at a rate of several gigabytes persecond. By way of comparison, the highestpossible data throughput via a parallel 32-bit/66 MHz PCI interface in CompactPCI® is0.264 GB/s. That means that withCompactPCI® Serial, completely new systemconfigurations become possible, and with 6UCompactPCI® Serial Mesh, 10 Gigabit Ethernetmeans a huge performance increase too.”For further information, please visit theCompactPCI® Serial (CPCI-S.0) product page

KONTRONwww.kontron.com

As part of its 'High-Speed CompactPCI®Initiative', at the Embedded World 2013trade show in Nuremberg, Kontron is pre-senting a 3U CompactPCI® Serial (CPCI-S.0)compliant chassis platform as well asperipheral boards for 10 Gigabit andGigabit Ethernet, XMC-based I/O and SATAhard disks. These building blocks arealready aligned to the new KontronCPS3003-SA processorboard featuring theIntel® Core™ i7 proces-sor, so that pre-inte-grated modularCompactPCI® Serial sys-tem configurations withPCIe, GbE, USB andSATA over backplaneare also available toOEMs. Customers willalso benefit from arapid time-to-marketand minimal integrationefforts, which Kontroncan also provide as partof its accompanyingservice program. The Kontron chassis familyCPS-ASM4 is a range of 19” (84HP) or42HP rack/wall-mount or tabletop enclo-sures, pre-integrated and EMI tested includ-ing backplane, power supply and fan trayoptions for 3U CompactPCI® Serial installa-

tions. 5-slot and 9-slot variants are available.A 19” hybrid solution includes 5 CompactPCI® serial slots and the assembly option forup to four additional classic CompactPCI®extension boards. With these platform solu-tions Kontron supports both the all-newhigh-speed system configurations and alsoprovides a simple migration path from theclassic CompactPCI® to CompactPCI® Serial.

Thus, users can transfer their systems intothe serial world comparatively easily with-out having to take existing CompactPCI®peripheral cards out of service.KONTRON www.kontron.com

Kontron's systems expertise expanded with the launch of a family ofpre-integrated CompactPCI® Serial building blocks

Easy-to-use, web-based mbed develop-ment tools are now freely available for theKinetis L series of microcontrollers fromFreescale Semiconductor. The FRDM-KL25Z Freescale Freedomdevelopment platform is now included inthe mbed hardware lineup, supported bythe well-established mbed online develop-ment community. The FRDM-KL25Z Freescale Freedomdevelopment platform is a low-cost hard-ware tool used to evaluate and developwith Kinetis L series MCUs, the industry’sfirst microcontrollers built on the ARM®Cortex™-M0+ core. FRDM-KL25Z is idealfor rapid prototyping of MCU-based appli-cations and has proven to be an effectivesolution for developers, with more than10,000 users adopting the platform withinone month of its release. With the addition of FRDM-KL25Z to thembed hardware lineup, FRDM-KL25Z cus-tomers can now utilize mbed.org at nocharge. Users will have full access to thembed online environment, which includes a

software development kit (SDK), powerfulonline tools, an active community of devel-opers and a large selection of reusablecode. The mbed online compiler providesa quick and easy way to get started andexperiment with the FRDM-KL25Z withouthaving to download, install and license anytools. The mbed.org user forums providean interactive community of professionaldevelopers working with a common mbedSDK and tools, sharing best practices andhelping designers be more productive. FREESCALE SEMICONDUCTOR www.freescale.com

Kinetis microcontrollers from Freescale now included in mbed tool for rapid prototyping



As part of its 'High-SpeedCompactPCI® Initiative', Kontronis launching the world’s first 6UCompactPCI® Serial Mesh 10Gigabit Ethernet processorboard together with a compliantchassis at the Embedded World2013 trade show (hall 1, stand460). For the first time, aCompactPCI® sys-tem supports 10Gigabit Ethernetand PCIe overbackplane inaddition to thecommon inter-faces such asPICMG® 2.16,SATA and HDMI.The KontronCompactPCI® processor boardCP6004X-SA is based on the lat-est generation Quad-Core Intel®Core™ processor and offers highcomputing power and perform-ance-per-watt, which reallycomes into its own in the CP-RAPID3 chassis with full-meshfabric configuration.Some applications for the newmodular CompactPCI® SerialMesh (based on PCIMG® 2.20)systems are in high availabilitytelecommunications and data-communications systems for carri-

ers and also with government andsecurity companies. Applicationsinclude secure wireless systems,radar and sonar applications, andcomplex computational algo-rithms such as imaging processing.Now, even with smaller spacerequirements, existing systeminstallations are provided with a

significant increase in datathroughput when new long-termgeneration systems are beingdeveloped which, in the future,could be scaled up to 40 GbE.Multiprocessor board configura-tions have already been fully test-ed in the chassis so that pre-inte-grated modular CompactPCI® sys-tem configurations with 10Gigabit Ethernet and PCIe overbackplane can be supplied toOEMs immediately. KONTRON www.kontron.com

Kontron is launching the world’s first 6U CompactPCI®Serial Mesh 10 Gigabit Ethernet processor board

Mobility Tech Zone has namedFreescale Semiconductor’s QorIQQonverge B4860 macrocell sys-tem-on-chip as a recipient of the2013 LTE Visionary Award. Theprize recognizes products thathave advanced the state of LTEtechnology during the previousyear. Mobility Tech Zone is theleading website dedicated to themobile broadband industry and issponsored by TMC, a global inte-grated media company, andCrossfire Media, an integratedmarketing company with a corefocus on technology. “Freescale’sachievements in advancing 4Gthrough LTE development areimpressive and, as such, the com-pany was selected as a recipient ofthe Mobility Tech Zone LTEVisionary Award,” said Carl Ford,co-founder and community devel-oper, Crossfire Media. “Freescale isa true innovator and one to watchfor the next evolution of solutions.” Introduced last year, the QorIQQonverge B4860 macrocell sys-tem-on-chip is engineered to helpmobile operators accommodaterapid adoption of 4G data-centricwireless devices, while reducingcosts associated with 3G deploy-ments. The B4860 delivers optimalperformance for macrocell base

station applications, utilizingFreescale’s intelligent integrationcapabilities and advanced 28 nmprocess technology to deliver asignificant leap forward in compu-tational capacity. “The Mobility Tech Zone LTEVisionary Award reinforcesFreescale’s role as a proven andcommitted innovator in the wire-less infrastructure market,” saidTareq Bustami, vice president ofproduct management forFreescale’s Digital Networkingbusiness. “Freescale is honored toreceive this prestigious recognitionand looks forward to even moreLTE milestones in 2013.” The B4860 macrocell SoC inte-grates four dual-thread, 64-bite6500 Power Architecture® coreswith AltiVec SIMD engines run-ning at up to 1.8 GHz. The e6500core is ideal for Layer 2, controland transport processing, andincorporates an enhanced ver-sion of the proven, high-perform-ance and widely adopted AltiVecvector processing unit, whichboosts performance for Layer 2scheduling algorithms. FREESCALE SEMICONDUCTOR www.freescale.com

Freescale’s QorIQ Qonverge macrocell base station-on-chip wins 2013 LTE Visionary Award

Atmel® Corporation announced its recentlylaunched AT86RF233 2.4GHz IEEE 802.15.4radio transceiver supports distance measure-ment between two radios, also known as ranging. This novel technology uses the measurement

of phase differencesbetween radio signals andenables items to betracked within a wirelessnetwork for industrial and

consumer applications. Phase measurement isa less complex technology than current dis-tance measurement technologies available onthe market today, with similar measurementaccuracy and lower overall system cost.Many of today’s applications for wireless net-works come with the inherent need to deter-mine the location of the individual nodes withina network to keep installation and maintenancecosts low. The capability to determine the dis-tance between any two nodes within a wirelesssensor network is the reason for implementing

positioning and localization solutions. WhileGPS is very successful outdoor, it cannot sup-port indoor applications efficiently.“We have been evaluating several differenttechnologies including RSSI, Chirp and UWBmethods for radio-based distance measure-ments over the last years,” said AndreasWerner, Managing Director of Agilion GmbH.“We’ve seen all these technologies only suit-able for niche applications, either compromis-ing performance or significantly complexwhich ultimately drive the cost-per-nodemuch higher. The new Atmel technologyestablishes a unique value proposition bycombining wireless sensor networks with real-time location (RTLS) capabilities even for massmarket applications.”“In this era of the Internet of Things, we areseeing an increase of smart, connected deviceswithin the wireless networks,” said OyvindStrom, Sr. Director of Wireless Solutions, AtmelCorporation. “With this increase, many of our

customers are demanding solutions that effi-ciently support the installation and mainte-nance of their systems. This new phase-differ-ence technology can be used to determine theposition of nodes within a network to helpaddress these demanding requirements.”

To accelerate a designer’s project, softwareevaluation kits utilizing the REB233SMA-EK areavailable now. To purchase a kit, contact yourlocal sales representative today. AtmelAT86RF233 transceivers, introduced in April2012, are available in mass production today. ATMELwww.atmel.com

Atmel Announces Ultra-low Power Transceiver to Support Phase-Difference Technology Complementing IEEE 802.15.4 Standards



Freescale and ARM ExtendRelationship with Cortex-A50Processor License for Futurei.MX and QorIQ ProductsFreescale Semiconductor is licensing the ARM®Cortex™-A50 series of microprocessors(MPUs) for future versions of its i.MX applica-tions processor and QorIQ communicationsprocessor product lines. This agreement ispart of a new multiyear subscription licensewith ARM that demonstrates Freescale’s com-mitment to the ARM architecture and its intentto further expand its ARM Powered® portfolio– one of the industry’s broadest range of solu-tions built on ARM technology.ARM recently announced the Cortex-A50series which is based on the ARMv8 architec-ture and is capable of 64- and 32-bit execu-tion. The series initially includes the Cortex-A57 and Cortex-A53 processors. The Cortex-A57 processor is ARM’s most advanced high-performance processor, while the Cortex-A53processor is the most power-efficient ARMapplications processor, as well as the world’ssmallest 64-bit processor. The Cortex-A53 and Cortex-A57 processorscan operate independently or be combinedinto an ARM big.LITTLE™ processor configura-tion, combining high performance with powerefficiency. In big.LITTLE processing, a system-on-chip (SoC) leverages two different, butcompatible, processing engines within thesame device, allowing the power managementsoftware to seamlessly select the right proces-sor for the task. The Cortex-A53 processor provides an opti-mized balance of power consumption andperformance, making it an excellent fit for fea-ture-rich applications, such as auto infotain-ment systems, medical equipment, eReaders,and the next generation of Internet of Thingsrunning i.MX apps processors. The Cortex-A53 processor also delivers hardware virtual-ization and enhanced security features in halfthe size of previous cores. The Cortex-A57 is designed to furtherextend the capabilities of future mobile andenterprise computing applications, includingcompute-intensive 64-bit applications. TheCortex-A57 processor is ideal for marketsFreescale will target with its forthcomingQorIQ communications processors based onthe core-agnostic Layerscape architecture,including the enterprise, data center andwired/wireless communications equipmentspaces. By leveraging the processing power ofthe Cortex-A57 processor, future generationsof QorIQ products can continue to deliverindustry-leading performance/power ratios.FREESCALE SEMICONDUCTORwww.freescale.com

Kontron presents the new Kontron Passau II,a fanless, intelligent data concentrator forSCADA and control room systems. The 2Urackmount server, which can be extended ona modular basis, is designed to meet thehighest requirements in electromagneticimmunity. It furthermore excels with its scala-ble number of protected GbE LAN, CANand galvanically isolated serial interfaces forsurge protection. Thesefeatures enable itsdeployment in heavycurrent environmentswhile also allowing forthe direct control ofdevices with live cur-rent. The server is alsoequipped to protect allsurrounding IT equipment against potentialvoltage surges on its test and control inter-faces. Typical applications areas are, forexample, in substations and SCADA systemsin the fields of power generation and distri-bution as well as in the process industry andin utility management. Further applicationscenarios are to be found in technical facility

management or in passgenger informationsystems at train stations as well as on trains.With its fanless chassis, the robust and long-term available Kontron Passau II is not onlyprotected against dust and moisture infiltra-tion, it is particulary low on maintenance. Byusing PCI (short) and PCI104 extension com-ponents, it can furthermore be very flexiblyconfigured to meet individual application

needs. Kontron's new fanless intelligent dataconcentrator Passau II delivers reliable 24/7performance in a working temperature rangeof 0 to 50 °C over an operating period of 10years which deems it particularly suitable forapplications with long-term service life. KONTRON www.kontron.com

Kontron's fanless data concentrator for SCADA and control roomsystems impresses with its high electromagnetic immunity

As the Internet of Things (IoT) expands toinclude greater numbers of small, intelligent,battery-operated devices, the MCUs thatenable these devices must deliver perform-ance, energy efficiency and connectivity inprogressively smaller footprints. FreescaleSemiconductor is addressing the miniatur-ization trend with its new Kinetis KL02MCU–the world’s smallest ARM Powered®MCU. The KL02 holds great potential forultra-small-form-factor products in applica-tions such as portable consumer devices,remote sensing nodes, wearable devicesand ingestible healthcare sensing.

Measuring just 1.9 × 2.0 mm,the Kinetis KL02 MCU is 25percent smaller than the indus-try’s next-smallest ARM® MCU.Within this miniscule device,Freescale has included the lat-est 32-bit ARM Cortex™-M0+

processor, cutting-edge low-power func-tionality and a range of analog and commu-nication peripherals. This enables systemdesigners to dramatically reduce the size oftheir boards and products while retainingthe all-important performance, feature inte-

gration and power consumptioncharacteristics of their end devices.In addition, space-constrainedapplications that previously could-n’t incorporate an MCU now canbe upgraded to become smartapplications, adding a new tier ofdevices to the IoT ecosystem. FREESCALE SEMICONDUCTOR www.freescale.com

Freescale introduces Kinetis KL02, world’s smallest ARM Powered® microcontroller


DESIGN FPGA

Accelerating software by moving it to massive-ly parallel hardware continues to develop asan attractive methodology. The merit of hard-ware acceleration is significant, but there arehurdles that should be budgeted for.To start, it’s important to understand that withhardware acceleration there are three primaryforms of parallelism: 1) blocks of code that exe-cute in parallel by splitting the data set into mul-tiple parts; 2) pipelined (bucket brigade) blocksof code that operate on streaming data; and 3)hybrid parallelism that combines both forms.On the merit side, each unrolling of a criticalpath loop into an independently streamingprocess can double wall-clock performance.Modern FPGAs, with several million usable (incontrast to advertised) gates can host a dozenor more independent streams. Accordingly,we are seeing solid 10× acceleration in “para -llelizable” designs in application domains suchas image processing, encryption and networkfiltering. The process is not trivial. While wehave student groups from Rochester Instituteof Technology to the University of Naplesdelivering remarkable speed ups, adoption byindustry is concentrated among a few thou-sand classic early adopters. This article covers the current state of the arton software to hardware compilation, pro-vides some realistic tips, and suggests a visionof how to make it more usable by the nextwave of developers.

FPGAs explained for the non-hardware folkMost software developers write code formicrocontrollers or microprocessors. Cremains the dominant language for designstarts. CPUs and microcontrollers typically fea-ture single or low number multiple-cores.They achieve throughput via increasing clockspeeds but are constrained by having to sharelimited cores and common memory. Think of itas driving ever faster through a single or dualtoll booth on a bridge.

FPGAs run at lower clock speeds than micro-processors. They achieve throughput by hav-ing very flexible input and output, so non-sequential tasks can be designed into parallelprocesses. While a conventional processoronly does one operation at a time, a properlyengineered FPGA design will concurrentlyperform hundreds or thousands of operations.Most designs for FPGAs are developed usingan HDL (hardware description language) likeVHDL or Verilog. They are not particularly dif-ficult languages but are sufficiently arcane thatC programmers do not generally take to them.We have seen more cross-over from HDL engi-neers learning C than the other way around.Gate count can be deceptive. Microprocessorsuse the available silicon with high efficiency.

FPGAs can use a significant fraction of theirgates for routing. And, you may not be certainof resource availability. FPGAs have great“blocks” of special purpose gates such as DSP,but they are limited. When you’re out of spe-cial blocks, you’re out, and the routing soft-ware will use less efficient gates, reducing per-formance and space efficiency.Historically FPGAs are descendants of PALs,GALs and PLDs. These precursor devices weremuch smaller. The earliest ones were smallenough to program in Assembly. The devicesgrew and the early HDLs emerged andimproved. Design shifted to HDLs like DataI/O’s ABEL, MMI’s PALASM and others. Jumpto today and the same shift is occurring. HDLslike VHDL and Verilog are more time consum-ing but offer better control over resources. Cto HDL to RTL (Register-transfer level: the gatelevel machine code that characterizes theFPGA) tools like Impulse C, ROCC, C2H,Vivado and others offer a higher level ofabstraction but may not have the same QoR(Quality of results) initially. More about thislater when we talk about the design flow.

Design flowHigh Level Language, or HLL programmingtypically refers to C. System C, C#, C++ are allgreat languages with ardent users. But for thisarticle we’re sticking with ANSI C, as most IP iscreated in this version of C.Design code is entered into a C developmenttool like GCC or Visual Studio. Ideally the sys-tem architect identifies a portion of the particu-lar design to try in hardware, so one can main-tain comparable C code files to compile both toFPGA and to microprocessor. This maintainsequivalence and lets the designer “break onething at a time”. The microprocessor-oriented Ccan be crudely “wrapped” and imported intothe C to FPGA environment – where it willunderperform until refactoring. Refactoringinto individual streaming processes enables thecompiler to better parallelize the code.

A snapshot on the current state-of-the-art in FPGA software tohardware compilation.

by Brian Durwood, Co-founder, Impulse Accelerated Technologies Nicola Granny, President, MNB Technologies

From Visual Studio toFPGA Hardware

Figure 1: The software to hardware stack generates necessaryhardware interfaces post-optimization.


DESIGN FPGA

Identifying the critical path to focus on startswith the basics. You are hunting for key blocksof code, largely free of serial data dependen-cies, which are heavily used by the system (andthey eat up a lot of clock cycles) and which haveopportunities for parallelism (traditional paral-lelism or pipelining), as shown in Figure 1.While open source profiling tools are not fullyrealized, they can be useful. Commercial profil-ing tools are available with improved visibilityand reporting to make run-time analysis easierand more accurate. It is not rocket science; thepoint is to chase clock cycle reduction. Typicaldesign modules that are amenable to paral-lelism include encryption, image processes, FIR,

FFT and any process that wants to sit on a busand look at data streaming by.The point here is to generally offload themicroprocessor and bypass limitations intro-duced by the von Neumann architecture.Some FPGA-enabled boards make this partic-ularly easy by including PCIe connections tohost so you can rapidly experiment with parti-tioning… moving code between FPGAresources and the system processor with singlelines of instruction. In addition to partitioning,you are refactoring. Again, refactoring heremeans breaking C algorithms into coarse-grained logic: single processes that can bemachine parallelized into multiple streamingprocesses. The C to FPGA compilers will unrollas much as they can but you’ve got to refactorinto logic that makes it easier to do so. All whileretaining behavioral equivalency. A great fea-ture of FPGAs is the ability to simplify algo-rithms by creating a behavioral model thatprovides the functionality of the originalmicroprocessor code, but eliminates the over-head generally associated with generic libraryroutines from BLAS, LAPACK, and so on.Verification occurs at every stage. VisualStudio verifies functional operation and equiv-alency to the microprocessor stack. Later in thetool flow you verify that the design will oper-ate in the target FPGA, and what clock cyclereduction is possible given the available FPGA

resources. Post machine compile, the HDL out-put can be directed to an industry standardHDL simulator to provide cycle accurate verifi-cation (Figure 2). Now comes some leaps that can go wrong. Tocompile all the way to FPGA gates, the opti-mizing compiler hands off synthesizable HDLfiles to a place and route tool. This can be oneprovided by the FPGA manufacturer or oneprovided by another EDA firm. This can entailone heck of a coffee break. Place and routetimes for several million gates, taking intoaccount all the special resources involved, cantake hours. This is probably the biggest con-trast in the experience compiling software to

FPGA hardware vs. compiling it to micro-processor (Figure 3). There are newer FPGAsthat enable partial reconfiguration, so if yourparticular process of focus is constrained to anarea that can be isolated, the iterative timescan be significantly reduced. Quartus, Vivadoand Impulse all support partial reconfigura-tion. However, the practical usability of this

technique remains to be fully field tested.In the software to hardware process, anotheroption is to run software on an embeddedcore inside the FPGA. The cores come in soft-and hard-core versions. Soft cores are pro-grammed into general purpose gates. WhileFPGA-hosted processors are slower than thoseof the host machine, they are typically Harvardarchitecture with physically separate storageand signal pathways for instructions and data.This gives them a memory access advantageand direct communication with the FPGAlogic, bypassing host to hardware overhead.Sometimes this can be a less efficient use of sil-icon, but multiple cores can be added as thedesign may need. Cores can be proprietary,provided by the FPGA vendor and useful ifnot familiar. Increasingly FPGA suppliers arestandardizing on ARM cores. This trend isexpected to continue. The use of cores inFPGAs as SoC (system on chip) solutions is anintrinsic benefit. The on-board core can con-solidate microcontroller or light micro pro-cessing tasks on one chip.Development environments such as ImpulseCoDeveloper also interoperate with full-fea-tured heavily used tools such as Microsoft™Visual Studio. A practical example is providedin Impulse App Note 112 by Michael Kreeger“IATAPP-112”.For example, while installing Impulse C andCoDeveloper the Visual Studio plug in is auto-matically installed and just has to be selectedduring setup. When Visual Studio is thereafter

launched, the top pull down-down menu canbe used to select “File->New Project…” whichcreates a sub directory for the new project. Ifbeginning from existing code, those source filesare copied into the solution directory. Headerand source files are added to the Visual StudioProject and then to the source files folder in thesolution explorer directory.

Figure 2: Desktop simulation output.

Figure 3: Building HDL: complex and not very speedy, compared to compiling for a microprocessor.


DESIGN FPGA

To verify before hardware generation a “debug” softwareproject is built, which makes it possible to test the appli-cation from desktop simulation. Next, select the hardwareenvironment from the configuration manager to build theHDL for the target FPGA. This enables the synthesizableHDL to be exported to the appropriate place and routetool to generate RTL for the target FPGA. The wholeprocess may take an hour or so.As a new methodology this tends to be minimally disrup-tive. The pragmas and additions are pretty commonsense. Glitches may arise from “plumbing” (a whole dif-ferent topic), such as getting things lined up with PCIedrivers, DMA, DDR and all the devilish details. Our twocents is that the growing body of known good examplesand reference designs makes this process less risky. Onthe tool side we’d like to see shorter place and routetimes, better back annotation and more useful profiling.Overall as the FPGAs and FPGA based acceleration cardsmature, and the body of available IP expands, this tech-nique becomes more mainstream. n

This article was originally published in the 2013Engineers’ Guide to FPGA and PLD. To view additional FPGA technical articles, news, opinions,product information, market trends and industry researchvisit www.eecatalog.com/fpga. Subscribe to the FPGA andPLD Quarterly Report at www.eecatalog.com/subscribe

About the authors:Brian Durwood co-founded Impulse Accelerated Technologiesin 2002 with David Pellerin, a co-worker from the ABEL® days atData I/O. Impulse has grown to be the most widely used C toFGPA tool, with customers from NASA to Detroit to Wall Street.Mr. Durwood was previously a VP at Tektronix, a VP at VirtualVision and an Analyst at NBC. Mr. Durwood is a graduate ofBrown and Wharton. Impulse now offers tools, IP anddesign/integration services.

Nick Granny is a scientist, engineer, and entrepreneur who hasbeen supporting the EDA and high-performance computing com-munities for more than 25 years. Currently Nick is co-founder andCEO of MNB Technologies, a small company that develops artificialintelligence-based EDA tools and provides technical services to theImpulse C user community. Prior to MNB, Nick was the lead staffscientist in Mentor Graphics research into FPGA-accelerated com-puting and was also a key member of the development and launchteam for the IKOS/Mentor VirtuaLogic emulation system. Earlier inhis career, Nick was an embedded systems engineering consultantto regional electric power utilities and the top-tier critical caremedical apparatus manufacturers. Nick is a medically retired USNaval Officer and further serves his community as an adjunct com-puter sciences instructor and course development consultant toIndiana’s state-wide community college system.

Microsemi Corporation, a lead-ing provider of semiconductorsolutions differentiated bypower, security, reliability andperformance, today unveiled anew family of U.S. DefenseLogistics Agency (DLA)-qualifiedSchottky diodes for aerospaceand defense applications requir-ing high density power andexcellent heat dissipation (typi-cally 0.2-0.85 degrees C per watt(C/W). The new diodes areoffered in Microsemi's patentedThinkeyTM package, which fea-tures a rugged ceramic andmetal construction with no wirebonds to improve reliability.

"Microsemi has decades of expe-rience designing and deliveringhigh-reliability solutions to theaerospace and defense commu-nities," said Dr. SimonWainwright, vice president andgeneral manager of Microsemi'sHi-Rel group. "Our uniqueThinkey package is one of the

many innovations we have devel-oped specifically to meet thestringent requirements of cus-tomers in these segments. Wewill continue to expand ourThinkey Schottky diode productportfolio to include additional

DLA-qualified devices."The new diodes support high-surge capabilities and enabledouble-side cooling. Since softsolder is not used in constructionof the devices, it prevents solderseal from mixing with mounting

solders and eliminates soldercreep and re-crystallization dur-ing power cycling and high-tem-perature storage.

Product Specifications:• Product series 1N6910UTK2thru 1N6912UTK2 and1N6940UTK3 thru1N6942UTK3, anode to strap(AS), cathode to strap (CS) perMIL-PRF-19500/723/726• Voltage: 15V, 30V and 45V• Current: 25A (for1N6910UTK2 thru1N6912UTK2) and 150A (for1N6940UTK3 thru1N6492UTK3)• Available in JAN, TX and TXVDLA qualification levels• 9X lighter than TO254 package

For more information, visithttp://www.microsemi.com/en/design-support/product-brochures

MICROSEMIwww.microsemi.com

Microsemi Delivers U.S. Defense Logistics Agency Qualified Diodes for High DensityPower Aerospace and Defense Applications

High-reliability Thinkey Schottky Diodes Feature Low Thermal Resistance


The latest generation of microcontrollers(MCUs) offers designers higher integrationand a wider range of peripherals and functionswhich demand a wider range of design skillsfor successful implementation. The good news,however, is that MCU manufacturers, such asMicrochip Technology, are investing in resourceswhich enable engineers to quickly gain thenecessary embedded design skills, or signifi-cantly reduce the learning curve for imple-menting new technologies.Microcontrollers are blurring the boundarieswhich once separated the design disciplines ofsoftware engineering and hardware engineer-ing: traditionally, software engineers designedapplication software for microprocessors;whilst hardware engineers created micro-processor-based systems by combining anMCU and external components. The introduc-tion of the first high-integration MCUs changedthis. They replaced the significant engineeringeffort required to develop a design from anMCU, external memory, peripheral, interfaceand discrete circuits, with the need for hard-ware engineers to write firmware for low-leveldriver-type applications.

Microcontrollers and moreThis means that embedded engineers need toadd proficiency in C and Assembly languageas well as debugging to their existing skills ofhardware circuit design, microcontroller inter-facing, PCB layout, design for manufacture anda detailed knowledge of EMC/EMI and otherregulatory standards and product approvals.It is hardly surprising, therefore, that embed-ded engineers are increasingly looking to gaina competitive edge and get products to mar-ket faster. That is why the availability of soft-ware libraries, sub-systems and advanceddevelopment tools has become an essentialpart of the MCU selection process.This article discusses a number of scenarios andsuggests application-based strategies for choos-ing the right microcontroller and using the

manufacturers’ tools and resources to simplifyimplementation and reduce time to market.

Blurring the boundaries of MCU performanceMicrocontrollers provide an extremely widespectrum of computing power and peripheralintegration and the choice is widened still fur-ther with high-performance cores, such as thePIC32MX1 and MX2 MCUs, being offered inreduced pin-count packages. Developmentssuch as this mean that high-performanceMCUs are moving into applications whichwould once have been the domain of themore established cores.Higher levels of integration also present design-ers with complex trade-offs: performanceenhancements such as faster or improved cores

The effective use of COTSsolutions and manufacturerdevelopment kits reducesthe learning curve forimplementing new technologies and enablesembedded products toget to market faster,explains Martin Hill ofMicrochip Technology

by Martin Hill Microchip Technology Inc.

Bridging the skills gap in embedded consumer design

COVER STORY EMBEDDED SYSTEMS

Figure 1: Code examples: http://www.microchip.com/codeexamples


can deliver plenty of MIPs, but very high levelsof analogue functionality can reduce real-timeperformance. Of course, device fit, form and function is onlypart of the selection process. There are otherfactors and decisions which can have a signifi-cant impact, such as design strategy, time tomarket, vendor support and cost. The ques-tions which must be asked are: “Could COTSsolutions help to shorten time to market?”,and “Which design strategy will provide thefastest design cycle?”.

The foundations of embedded designsTypically, embedded systems need a user inter-face, display and some form of connectivity andthese fundamentals have not changed over theyears. Their complexity, however, has increased,whilst the cost has reduced. This means that acolour graphics LCD, touch interface and wire-less connectivity are now considered essential ifa product is to be competitive. There are, ofcourse, still valid reasons for selecting more tra-ditional segmented LCDs, mechanical typeswitches for the user interface and wired con-nectivity, however, fashion sells and customerscan be fickle which makes these design deci-sions crucial to the financial success of thedesign for a consumer electronic product.

Bridging the skills gapIn a fast-paced market, new and improvedtechnologies can come along just as thedesigner has mastered the skills introduced bythe previous technology breakthrough.Fortunately, commercial off the shelf (COTS)solutions can support rapid product develop-ment by bridging the skills gap. COTS solu-tions can take many forms. These can includesource code and configuration tools for imple-menting a communication protocol stack,through to a full software development frame-work including a RTOS, peripheral driver sup-port libraries, system timing analysis, opti-mised for a specific application, such as auto-motive. It is becoming increasingly impracticalto being a new design by writing all of therequired code and starting hardware develop-ment from scratch. Some industries, such asautomotive, actively discourage this ‘blankcanvas’ design strategy by specifying the soft-ware development tools which must be usedduring development. Silicon vendors and third-party developmenttools are also moving designers away from an‘all-my-own-work’ approach by supplying, forexample, a free GUI to ease debugging andtuning of a specific application such as motorcontrol or a touch interface.Another way to achieve higher integration anda faster time to market is to link new microcon-troller peripherals to other peripherals toform sub-systems. Touch-screen applicationscan, for example, be implemented with aCharge Time Management Unit (CTMU), or

using the Capacitive Voltage Divider (CVD)technique, and both could be linked with ananalogue MUX to form a sampling sub-systemfor multiple touch buttons or linear and rotarysliders. Another example would be to use aParallel Master Port (PMP) to drive a Low CostController-less (LCC) graphics panel or usingan integrated graphics controller. New general-purpose peripherals, such as theConfigurable Logic Cell (CLC) can also be

used to reduce the external component count.Designed to link with a range of other internalperipherals, CLCs are a useful addition to theembedded engineer’s tool-kit. Integratedhigh-speed comparators and op amps are alsoparticularly useful real-time system compo-nents for power-supply and motor-controlapplications.

Resources to reduce time-to-marketApplication notes, as well as other COTS solu-tions and tools, support new applications andperipherals and help to speed time to marketby reducing the need for embedded designersto re-invent the wheel. Microchip, for example,offers one of the widest ranges of supportresources including development boards andapplications libraries focussed on specificapplications and product families.

Similarly, code examples address specific pro -blems ranging from initialising peripherals toimplementing algorithms. All of these resourcescan be found on the Microchip website.

Implementing a touch interfaceThe design strategy, and the tools which willbe most helpful in reducing time to market,will depend on the specific challenges pre-sented by each design.

Take for example, the challenge of implement-ing a new touch-screen interface to replace themechanical push-buttons on an existingkitchen-top appliance design using a PIC16 todrive a motor and a segmented LCD.Clearly, the first goal would be to re-use asmuch of the existing design as possible: if theexisting code is written in Assembly language,the aim should be to find a developed touchsolution in a higher-level language. This woulddeliver a fast design process and overcomeany in-house skills gap.

Typically, the strategy for this design wouldbe as follows:• Select a microcontroller from the same ge -neral product family which provides the basisfor compatibility, or partition the design usingtwo microcontrollers;


Figure 2: Upgrading a kitchen-topappliance by adding atouch interface.

18

• Select a microcontroller with the peripheralsrequired to provide the new functionality andmemory, as well as analogue support toreduce component count;• Look for a COTS solution which uses C codefor the touch interface;• Re-use and mix the existing Assembler codeand C code for the new functionality, but parti-tion the design into separate files. Alternatively,partition the design using separate microcon-trollers with communications support for inter-facing between them; • After porting the code, run the applicationwithout the additional functionality and testthe behaviour;• Design and debug the touch interface usingsupport tools;• Environmentally test the touch interface andthen integrate with the complete applicationand re-test.

A potential real-world solution to thisdesign would be to:• Contact Microchip for advice on specificapplication requirements;• Visit the Touch Design Centre athttp://www.microchip.com/mtouch• Evaluate different touch-screen options usinga touch-screen development board;• Use the free Microchip Advanced PartsSelector (MAPS) to identify suitable parts forthe design migration;• Follow the design guidelines for designing atouch interface using Microchip application noteAN1102 and other online resources. Requestadditional support from Microchip if required;• Use the free Microchip CVD FrameworkCOTS software package which is part of theMicrochip Applications Libraries (MAL) todesign and debug the touch interface. At thisstage, also select a suitable debugging tool;• Use the XC8 Microchip C Compiler andMPLABX IDE to construct and debug the com-bined application firmware;• Environmentally test the design. Microchipmaybe able to assist with some qualitativeEMC testing and further design guidance ifrequired.

Interfacing to smart phones or tabletsDesigning a new accessory for a smartphone ortablet computer presents different challenges,such as the need to connect to Android® orApple® phones or tablets, in addition to charg-ing, control and other functions. Given the fast-paced nature of this market, a rapid designcycle is imperative.

The overall design strategy would be to:• Research support for the Apple and Androidplatforms;• Determine the overall application memoryand resource requirements;• Look for a COTS solution for the phone/tablet interface;

• Consider contracting-out the interface designif the skills are not available in-house;• Select a microcontroller which supports theappropriate COTS software framework;• Develop the accessory and the interface.

Again, using online resources and develop-ment tools can help to take time out of thisdesign cycle:• Visit the pages: www.microchip.com/appleand www.microchip.com/android• Contact Microchip for advice on specificapplication requirements; • Obtain any applicable license(s) directly fromApple®. A license is not required for Android®;• Choose a development board with relevantdemos;

• Decide whether to download the free, stan-dard or professional version of the XC16 orXC32 C Compiler;• For Android, download the free AndroidFramework COTS software package which ispart of the Microchip Applications Libraries(MAL). Use the appropriate COTS softwarepackage, XC C Compiler and MPLAB X IDE todevelop and debug the accessory interface;• Request additional support from Microchipif required.The brief for a new design for a smart meterwould probably include a QVGA colourgraphics display, an over-laid resistive touchdisplay panel, radio link to a remote sensorand WiFi connection to the internet. Minimising software development time wouldbe critical as the design would also need aRTOS to perform task scheduling and resourcemanagement.

The typical design strategy would be to:• Use high-level language;• Make extensive use of COTS RTOS, commu-

nication protocol stacks and support libraries/tools;• Adopt a modular approach to the hardwaredesign;• Provide a clear roadmap for the design tobecome the basis for additional products.

To take reduce the time to market for asmart meter, the approach could be to:• Contact Microchip for advice on specificapplication requirements;• Visit http://www.microchip.com/rtos for guid-ance on RTOS support and selection; • Evaluate the microcontroller and multi-mediaoptions with a PIC32, PIC24 or dsPIC33EStarter Kit with a multi-media expansion devel-opment board;

• Follow the strategy outlined in the applica-tion note AN1264 ‘Integrating MicrochipLibraries with a Real-Time Operating System’;• Use the multiple stacks available in theMicrochip Applications Library or requestadditional support from Microchip;• Develop and debug the combined applica-tion firmware using the XC16 or XC32Microchip C Compiler and MPLAB X IDE.

SummaryEmbedded systems are becoming more com-plex but, for many designs, COTS solutionsfrom Microchip enable a fast time to market,and can even minimise the need for designersto undertake the steep learning curve intro-duced with each new technology. Thesedesign resources, such as online design cen-tres, code examples, application notes, devel-opment boards and more, have become anessential part of the tool-kit for innovativeembedded design and a critical factor inmicrocontroller selection. nwww.microchip.com

EP&Dee | March, 2013 | www.epd-ee.eu


Figure 3: Accessories design has different challenges.


DESIGN MCUs

ST Microelectronics has developed two inte-grated micro-controller solutions, especiallyfor human-machine-interface solutions. TheSTM8T controller family thus supports bothcapacitive approximation and touch key tech-nology. The latest addition to the family is theSTM8T143, which is based on capacitive tech-nology with charge transfer. It uses a channelto measure the capacitance of the humanbody in the vicinity of an electrode locatedbehind a dielectric material such as glass oracrylic. The 8-pin micro-controller has a reallylow power consumption of around 10μA.

Integrated sampling resistors and a voltageregulator optimize the sensitivity of the deviceand increase the noise immunity of the systemin loud environments. Thanks to its electrodeparasitic capacitance compensation (EPCC),the STM8T143 has a detection range of up to25 cm. One interesting feature for use inmobile phones is on-ear proximity detection,which detects when the user is holding thephone to his or her ear whilst making a call. Atthis point, all touch-screen functions are deac-tivated to prevent them from being acciden-tally triggered during the call.

The STM8T143 also reduces radio frequencystrength if the device is in the vicinity of theuser. As a result, it meets FCC SAR (specificabsorption rate) requirements and can also beused in tablets. The ST micro-controller is alsosuitable for use for on/off switches, proximitycontrol for the backlighting of user interfaces,alarm and control functions, and energy-savingperipheral modules for portable devices.For the development of high-end user inter-faces, ST offers a touch sensing softwarelibrary with a completely free source codesolution for capacitive touch key controllers.

Today, touch functionality is absolutely indispensable for a modern human-machine-interface. All leading micro-controller manufacturers offer sophis-ticated solutions here. This means that developers are spoilt for choice.

Microcontrollers forhuman/machine interfaces

Please touch!


DESIGN MCUs

It allows keys, wheels, and controllers tobe linked with traditional peripheralapplications of micro-controllers such ascommunication, LED, and LCD indicatorsand beepers. The library is part of theapplication firmware; evaluation boardsare also available.In the first quarter of 2013, ST Micro -electronics will launch a new STM32Cortex M4 core micro-controller withfloating point unit (FPU) calculation. Thiswill be specially designed for HMI appli-cations. The IC has an integrated TFT LCDcontroller, USB OTG support, Ethernet,and generously sized SRAM and flashmemories. The implemented slide andtouch functionality is based on the capac-itive method and is executed via GPIOswith the help of the ST software library.The new micro-controller builds uponthe second generation of the familiarSTM32F4 series.In order to reduce the development timefor prototypes for various applications,Infineon has developed a modular,extendable application kit - the HexagonKit System - for its XMC4000 micro-con-troller family. Various satellite boards ori-ented towards different applications canbe docked onto the central board withthe XMC4500 micro-controller. The HMI satellite (HMI_OLED-V1) isavailable for HMI applications. It compris-es an OLED display as well as audio,touch, and SD/MMC functions. In con-junction with the COM (communication)satellite, it can be enhanced by Ethernet,CAN, and RS485 functions. The manufacturer recently launched theFM3touch library for applications based onthe FM3 ARM® CortexTM M3 MCU familyby Fujitsu. It enables the realization of amodern, capacitive touch screen withextensive functions and an aesthetic design.

For example, the FM3touch library allowsthe design of buttons, slides, and touchpads with a minimum of additional hard-ware and software. Thanks to flexible sen-sor grouping, there are numerous layoutoptions here. In addition to a simple on/offstatus indication, the signal strength ofeach touch sensor can also be read. It isalso possible for a notification function tobe triggered when a button is pushed or aslide is moved. The 12-bit analogue-to-digital converter(ADC) in all FM3 micro-controllersensures fast data capture, high sensitivity,and a low system load. One or two moreADC converters ensure the required levelof flexibility and resource availability. Theconfigurable drift calibration guaranteesstable operations.

Touch-sensitive metalMicrochip offers a whole range of touchsolutions with the name mTouch. A tech-nology which even enables the realizationof metal surfaces with touch functions isparticularly noteworthy: The Metal OverCap touch system. This involves placing aconductive surface which acts as a secondcapacitor plate over the capacitive sensor.If this surface is touched, the distancebetween the surface and the sensordecreases. The capacitance changes andthe micro-controller computes the touch.The conductive surface can be a thinpiece of metal or a piece of plastic with ametal coating. This allows the power-sav-ing properties of the capacitive touchtechnology to be exploited for touchapplications with metal or plastic surfaces.The user interface can be flexiblydesigned to carry Braille writing, forexample. It works if the operator is wear-ing gloves and even under water. nwww.rutronik.com

Different touch principles

• Capacitive: When you touch the surface with oneof your fingers, the capacitance of the capacitorchanges. The micro-controller uses this change todetermine the place which was touched. Capacitivetouch screens are now extremely popular, since theyare easy and inexpensive to realize and enable aneasy-to-clean, enclosed user interface. Capacitivetouch sensors allow controls such as knobs, slides, andwheels to be simulated. You operate them with yourfingers or with a special stylus. One disadvantage isthe relatively large effect of temperature, air humidity,and water on the efficiency of the detection function.Appropriate hardware and tailored software canreduce the consequences of these influences. Whendesigning capacitive touch sensors, it is particularlyimportant to pay attention to the EMC conformity ofthe circuit.

• Resistive: The pressure resulting from the surfacebeing touched can be determined by means of themeasurement of two resistances of the touch point.The measurement takes place using an analogue-to-digital converter and is thus supported by a largenumber of micro-controllers. The resistive principlealso reacts regardless of what or who triggered theactuation. Resistive touch screens enable a higher res-olution at low cost in comparison with capacitive vari-ants. They are therefore particularly widespread incost-sensitive applications. Unlike other touch tech-nologies, this technology scores particularly well withregard to its high EMC conformity. Resistive touch sen-sors frequently take the form of screen overlays, butbuttons, slides, and wheels are also possible. A disad-vantage of this technology is that the film wears eachtime it is touched; however, this disadvantage hasbecome much less significant during recent years.

• Inductive: In the case of inductive sensors, a mini-mal mechanical movement of inductive componentsacts upon a resonant circuit. The sensitivity can beextremely accurate - in the single- or double-digitmicrometre range. The effects of environmental influ-ences on inductive touch sensors are significantly lessthan for capacitive touch sensors. They react to beingtouched by a finger, gloved finger, or stylus. They areprimarily used in industrial applications because oftheir robustness. Here, the fact that the reversiblemechanical deformation of the touch surface allowstactile feedback (a click) is also useful, since it makes itpractically impossible to trigger something withoutbeing aware of it. On the negative side, the inductiveprinciple has a higher power consumption and doesnot enable slides or wheels.

• The multi-touch procedure enables the detec-tion and evaluation of multiple touch points at thesame time, thus opening up further scope for controls.This procedure is reali zed with different overlays:Resistive types are used as single-touch variants and asmulti-touch variants with multi-finger control. Opticaloverlays are available for total reflection.


DESIGN POWER

Just as it is important to properly size andspecify your AC/DC supply, it's also importantfor designers not to over-specify this vitalcomponent. It may seem counterintuitive, but"too much" of a good thing can have negativeconsequences in efficiency, cooling, overallproduct size, and even available vendors,besides the obvious downside of higher cost.The first and largest factor to consider is match-ing the supply output capability to the load itmust support. For example, if the maximumload (DC voltage × current) is 500W, then a1000W supply provides much more design-margin insurance than you actually need.

What are the consequences of a supply thathas so much headroom? The good news isthat, obviously, you'll have plenty of amps atthe nominal voltage-rail values you need.Simple enough, end of story - but not quite.There are significant drawbacks to having allthis extra, unused power available.

The biggest one has to do with inefficiencyand its many consequences. Every supply hasan efficiency vs. load graph, such as the one inFigure 1. For a well-designed switching supply,this efficiency is usually at its highest in therange of 80-95% of maximum rated load. [Thisgeneral guideline does not apply to linear reg-ulators and supplies, but those are unusualabove fairly low power levels of a few watts.]When operating at low loads, which may bemost of the time in an application like a datacenter, the power supply can generate a lot of

extra heat, and this is where the engineer'snightmare of both obvious and unintendedconsequences starts. The obvious effect is thatyou are wasting more AC-mains power, soyour system costs more to operate, and thatcost is straightforward to quantify. The largersupply is also more expensive to buy, and it’seasy to put a cost number on that, as well.

But beyond those easily assessable factors areones that are much harder to grasp. As a consequence of the additional heat, whichyou must get rid of, you are now dealing with

more complex design and budget issues relatedto convection cooling (which may no longer bepossible), fans, airflow layout, and heat sinks.

These alternatives add direct cost, materials,unreliability, and constraints on packaging andlayout to the design, and even limit yourdegrees of freedom as you need to squeeze

more into the product box, or make the boxbigger. In addition, the larger-capacity supplyhas a larger footprint, with clear negative con-sequences.

You can buy more supply than you need, yet less is actually often better,says Don Knowles, VP Engineering at N2Power.

by Don KnowlesVP Engineering N2Power

It's easy to over-specifyyour AC/DC supply

but you shouldn't

Figure 1: The efficiency of a supply varies with load, and most peak in the zone of 80-95% of their maximum rated capacity; this chart shows the N2Power XL280-48 curve.


Further, as you select larger supply sizes, you'lllikely find fewer vendors to choose among,and fewer direct alternatives or secondsources to your primary or preferred source.This may not bother you, but your purchasingdepartment or contract assembly source maybe uncomfortable and even push back.

For these reasons, most AC/DC supply ven-dors offer a broad family with many similarunits, except for capacity, so you can match thesupply size to the load with little excess capa -city. For example, members of the XL series ofAC/DC supplies from N2Power are availablewith closely spaced 125, 160, 275, and 375Wratings.

Note that adjacent-rated units from some ven-dors often differ only in their power rating,but have the same physical size and connector,so you can “interchange” painlessly if it turnsout your actual power needs are different thanyou anticipated, as the N2Power XL125 andXL160 photos show (Figures 2a and 2b); bothhave the identical 3" × 5" inch (7.5cm ×12.5cm) footprint.

Of course, it's easy to say “just design to useless overall power, and then size the supply tothe maximum load.” The problem is that formany designs, the ratio between the maximum(peak) load and the typical load is large; 2:1 oreven 3:1 ratios are common. So you must sizeyour supply for the peak load, but most of thetime it is running at far less, and is in the ineffi-cient zone.

There are ways to circumvent this problem,such as by using an auxiliary booster for peakloads, a supercapacitor, or other techniques.However, each of these brings new designproblems of switching them to the load, andthe overall response to load transients.Therefore, to avoid over-specifying, try to getthe maximum load of the system down to asclose as possible to the typical load value.

Beyond efficiency, what else?Other factors to consider are operating tem-perature range, operation voltage range,line/load regulation, various types of protec-tion, redundancy, and I/O.

Given your ambient operating environmentand the cooling scheme you'll be using, whatoperating temperature do you need on thesupply? Certainly, a supply which is specified tooperate at higher temperatures costs more -but perhaps that allows you to “get away” withreduced cooling requirements, so that's a trade-off to consider. Don’t forget low-temperatureoperation, as well, if your application is one ofthose where your supply has to survive or evenjust start up below freezing.

DESIGN POWER

Figure 2: a) The XL125 125W AC/DC supply and b) the XL160 160W supply from N2Power differ primarily in their power rating; their footprint, physical size, connector, and many other specifications are otherwise identical.

Figure 2a

Figure 2b

page 23


DESIGN POWER

Good representative of this group is a familyof E364xA power supplies by AGILENTTECHNOLOGIES. Excellent technical parame-ters and good functional features characterisethe devices. What is most important is thatthey have a feature, which is crucial for manyusers, i.e. communication with PCs throughdedicated interfaces RS232 and GPIB.On the front panel of power supplies, besidesreadable display, necessary regulation compo-nents are located, including buttons to set thevoltage values and protection thresholds, or callup to 5 earlier pre-set and memorised va lues ofpower supply parameters from non-volatilememory. Maximum resolution for setting upvoltage and current is 10mv/1mA respectively.Additionally, comfortable “analogous” knoballows rough and quick setup of voltages andcurrents.When it comes to electric parameters, thepower supplies offer output power from 30W

to 100W at voltages up to 60V and currentsup to 8A. Excellent electric parameters werereached thanks to the application of specificregulation solutions. (see Figure 1)Above all, it is not only linear transistor that hasbeen applied for regulating voltage and current,but also so called phase-controlled pre-regu-lator. Thus a wide range of output voltage reg-

ulation has been achieved at significantlyreduced power dissipation at series regulationelement (transistor). On the other hand connection of series linearregulation, characterised by high precision andspeed with monitoring actual output parame-ters (voltage feedback control) ensures veryhigh stability of output voltage regardless of

One of the product groups offered byTME consists of so-called programmableDC power supplies. Besides fulfilling anobvious function for a power supply,that is delivering defined, pre-set andexpected voltage and current, thesedevices also allow, by means of special

commands and buttons, programming and readout of many parametersthrough front panel or dedicated interface.

by Marcin Zając

Programmable DC Power Supplies

Figure 1

Agilent E363xA Series


changes in power supply load. Such solutionalso has a very favourable impact on timecharac teristics of the power supply response.Short-term pulse changes of load that canaffect voltage stability are compensated veryquickly. Declared response of the circuit isbelow 50us. Also, voltage protections have been appliedthat prevent damage to connected devicesand protect against excessive voltage rise.Please note that protection is carried outthrough short-circuiting the power supply out-put. Should there be another power source insupplied device, it may lead to significant riseof current that flows through the power supply.Control by means of GPIB or RS-232 connec-tor substantially expands power supply func-tionality transforming it into multifunctionalworkshop device.Communication through the interface runs intwo directions: readout of voltage and currentand setting of these parameters is possible.This can be done by means of easy to use SCPIstandard (Standard Commands forProgrammable Instruments), allowing quickand effective control.

Possible applications of the interface:• quick, earlier predefined, setting of values ofpower supply parameters depending onrequirements of connected devices;• measuring current and voltage characteristics(without additional measuring devices) ofsemi-conductor components (by means ofadditional software);• studying reactions of circuits to changes inpower supply with pre-programmed scenarioof voltage changes when designing and start-ing up circuit prototypes.

These are only a few examples of possibleapplications. In the case of using additionalVXI controllers there is an option of integra-tion of the power supplies with other, morecomplex, software like LabVIEW orLabWindows by National Instruments.It opens up the possibility to create alreadyvery complex systems dealing with control,measurements or testing newly designedequipment.Power supplies of E364xA family are solid,state-of-the-art power supplies with excellentelectric parameters, both static and dynamic,offering stable output voltages at low values ofripple and noise. Connected devices are pro-tected against excessive voltage rise, while GPIBand RS232 interfaces allow reading and settingup power supply parameters through PC. Described components are offered by TME atattractive prices. We kindly invite you to have a look at our offeravailable on the website http://www.tme.euwhere you can make purchases. nwww.tme.eu

DESIGN POWER

What is the nominal value of your AC line(mains)? Do you need a supply for only115VAC, only 230VAC, or a wide-rangesupply that handles both ranges?As usual, there's a tradeoff: in general, asupply for both AC values is slightly moreexpensive, but the extra cost may beworthwhile because you'll be able to buymore of the single unit, and your stocking,inventory, and support costs will belower.

More complicated is the tolerance youneed around the nominal AC line. Doesyour supply have to work with a fairlymodest ±5% swing, a mid-range ±10%span, or a wider variation of ±20% aroundnominal? Supplies which can work withmore-poorly behaved AC mains, yet stillmaintain regulation within specification,are more costly, and you'll have fewersuitable vendors. If you need to tolerate awide mains tolerance, it may be less costlyto get a separate pre-regulator to keepthe AC line in a tighter range and then useless-costly supplies.

What level of output absolute accuracy,stability, and regulation does your systemrequire? Most supplies have a factoryadjustment for nominal output value, sothe supply should be fairly close to speci-fied output. But keep in mind that whilestability and regulation vary from vendorto vendor, and tighter specs cost more,you may not need that performance.

The reason is that many AC-to-final DC-rail supply paths now consist of multiplestages, where the first-stage AC/DC con-verter feeds an intermediate bus conver -ter (IBC) or point of load (POL) convert-er, not the final rail itself. These DC/DCconverters provide the actual voltages thesystem uses, and they may be able to tole -rate modest variations coming from theAC/DC supply to their DC inputs.

Nearly all credible vendors offer featuressuch as overvoltage protection, short-cir-cuit protection, and output crowbar.Some offer extra protection againstextreme line transients, including light-ning-induced spikes and surges. If youdon’t expect these upset events, or preferto protect the supply with external, dis-crete components, you can use a supplythat meets basic industry-wide transientspecifications, rather than one withgreater protection.

Some supplies offer N+1 capability,where you can set up an array of supplieswith automatic switchover in case onefails. If you don’t need this level of reliabi -lity, or prefer to use a single AC/DC supply,this feature is unneeded.

There's also a trend, especially in largersystems, to have the supply report manyof its operating conditions (especiallyvarious internal temperatures) back to asystem monitor, and even change opera -ting parameters under direction of a sys-tem controller. For applications that don’trequire this level of supply/system inter-action, don't spend for the I/O port (I2C,PMBus, SPI) and related circuitry withinthe supply.

Whether you over-specify your supplydue to lack of understanding of your sys-tem needs, supply parameters, or just tosleep better at night, there's really noneed to do so. As with most engineeringdecisions, you'll be able to specify whatyou need and not more, once you under-stand the priorities of your project and itsmarket, as well as the tradeoffs yourchoices bring to the design. n

www.n2power.com

About the author:Don Knowles joined N2Power as VP Engineering 12years ago after more than two decades’ experience inpower electronics design and manufacturing spanningindustrial, ICT and medical electronics sectors. Prior to joining N2Power, he ran his own power elec-tronics business for 20 years, designing power sup-plies and high-power AC and DC loads, and workingwith contract manufacturers. Don holds a degree inElectronics from American River College, Sacramento,California, USA.

It's easy to over-specify your AC/DC supply -but you shouldn't

page 21


In recent years, the automotive industry hasmade remarkable advances in driver assistance(DA) systems that truly enrich the drivingexperience and provide drivers with newforms of information about the roadwayaround them. This article looks at how FPGAscan be leveraged to quickly bring new driverassistance innovations to market. [Editor’snote: driver assistance systems are sometimesreferred to as ADAS: Advanced Driver AlertSystems.]

Driver Assistance IntroductionSince the early 1990s, developers of advancedDA systems have envisioned a safer, more con-venient driving experience. Over the past twodecades, DA features such as ultrasonic parkassist, adaptive cruise control and lane-depar-ture warning systems in high-end vehicles havebeen deployed. Recently, automotive manu-facturers have added rear-view cameras,blind-spot detection and surround-vision sys-tems as options. Except for ultrasonic parkassist, deployment volumes for DA systemshave been limited. However, the research firmStrategy Analytics forecasts that DA systemdeployment will rise dramatically over the nextdecade, including growth from $170 billion in2011 to $266 billion by 2016 – a compoundaverage annual growth rate of 9.3%.

In addition to government legislation andstrong consumer interest in safety features,innovations in remote sensors and associatedprocessing algorithms that extract and inter-pret critical information are fueling an increase

in DA system deployment. Over time, theseDA systems will become more sophisticatedand move from high-end to mainstream vehi-cles, with FPGA-based processing playing amajor role.

Driver Assistance Sensing Technology TrendsSensor research and development activitieshave leveraged adjacent markets, such as cellphone cameras, to produce devices that notonly perform in the automotive environment,but also meet strict cost targets. Similarly,developers have refined complex processing

algorithms using PC-based tools and are tran-sitioning them to embedded platforms.While ultrasonic sensing technology has led themarket, IMS Research (Figure 1) shows camerasensors dominating in the coming years.

A unique attribute of camera sensors is thevalue of both the raw and processed outputs.Raw video from a camera can be directly dis-played for a driver to identify and assess haz-ardous conditions, something not possible withother types of remote sensors (for example,radar). Alternatively (or even simultaneously),the video output can be processed using

DESIGN AUTOMOTIVE

Driver Assistance Systems

with the Powerof FPGAs by Paul Zoratti

Xilinx

FPGAs can be leveraged to quickly bring new driver assistanceinnovations to market.

Figure 1: Driver Assistance Sensors Market.


image analytics to extract key information, suchas the location and motion of pedestrians.Developers can further expand this “dual-use”concept of camera sensor data by bundling

multiple consumer features based on a singleset of cameras, as illustrated in Figure 2.From such applications, it is possible to draw anumber of conclusions regarding the require-ments of suitable processing platforms forcamera-based DA systems:

•They must support both video processing andimage processing. In this case, video processingrefers to proper handling of raw camera datafor display to the driver, and image processingrefers to the application of analytics to extractinformation (for example, motion) from a videostream.•They must provide parallel data paths foralgorithms associated with features that willrun concurrently.•Given that many new features requiremegapixel image resolution, connectivity andmemory bandwidth are just as critical as rawprocessing power.

Meeting DA Processing Platform RequirementsFPGAs are well suited to meet DA processingplatform requirements. For example, in a wide-field-of-view, single-camera system that incor-porates a rear cross-path warning feature, thesystem’s intent is to provide a distortion-cor-rected image of the area behind the vehicle. In

addition, object-detection and motion-esti-mation algorithms generate an audible warn-ing if an object is entering the projected vehi-cle path from the side.

Figure 3 illustrates how the camera signal issplit between the video- and image-process-ing functions. The raw processing power need-ed to perform these functions can quicklyexceed what is available in a serial digital signalprocessor (DSP). Parallel processing along with hardware accel-eration is a viable solution.

FPGAs offer highly flexible architectures toaddress various processing strategies. Withinthe FPGA logic, it is a simple matter to split thecamera signal to feed independent video- andimage-processing intellectual property (IP)blocks. Unlike serial processor implementa-tions, which must time-multiplex resourcesacross functions, the FPGA can execute andclock processing blocks independently.

Additionally, if it becomes necessary to make achange in the processing architecture, the abil-ity of the FPGA to reprogram hardware blockssurpasses solutions based on specializedapplication-specific standard products(ASSPs) and application-specific integratedcircuits (ASICs), giving FPGA implementationsa significant advantage when anticipating thefuture evolution of advanced algorithms.

Another benefit of FPGA implementation isdevice scalability. As OEMs look to bundle morefeatures, processing needs will rise. For example,the rear-view camera might need to host amonocular ranging algorithm to provide driverswith information on object distance.

DESIGN AUTOMOTIVE

Figure 2: Bundling Multiple Automotive Features.

Figure 3: Video and Image Processing Functions.


The added functionality requires yet anotherparallel-processing path. Implementing this in aspecialized ASIC or ASSP could be problematic,if not impossible, unless the designers madeprovisions for such expansion ahead of time.Attempting to add this functionality to a serialDSP could require a complete re-architectureof the software design, even after moving to amore powerful device in the family (if it isplausible at all). By contrast, an FPGA-basedimplementation allows the new functionalblock to be added, utilizing previously unusedFPGA logic and leaving existing blocks virtual-ly intact. Even if the new function requiresmore resources than are available in the origi-nal device, part/package combinations fre-quently support moving to a denser device(that is, one with more processing resources)without the need to redesign the circuit boardor existing IP blocks.

Finally, the reprogrammable nature of FPGAsoffers “silicon reuse” for mutually exclusive DAfunctions. In the rear-looking camera example,the features described are useful while a vehi-cle is backing up, but an FPGA-based systemcould leverage the same sensor and process-ing electronics while the vehicle is moving for-ward, with a feature such as blind-spot detec-tion. In this application, the system analyzesthe camera image to determine the locationand relative motion of detected objects. Sincethis feature and its associated processing func-tions are not required at the same time as thebackup feature, the system can reconfigure theFPGA logic within several hundred millisec-onds based on the vehicle state. This allowsthe complete reuse of the FPGA to providetotally different functionality at very little cost.

Meeting DA External Memory Bandwidth RequirementsIn addition to raw processing performance,camera-based DA applications require signifi-cant external memory access bandwidth. Themost stringent requirements come from multi-camera systems with centralized processing,for example, a four-camera surround-view sys-tem. Assuming 4 megapixel imagers (1,280 ×960), 24-bit color processing, and perform-ance of 30 frames per second (FPS), just stor-ing the imagers in external buffers requires 3.6Gb/s of memory access. If the images need tobe simultaneously read and written, therequirement doubles to 7.2 Gb/s. With an 80percent read/write burst efficiency, therequirement increases to 8.5 Gb/s. This esti-mate does not include other interim imagestorage or code access needs. With theserequirements, it is clear that camera-based DAapplications are memory bandwidth-intensive.

These systems also commonly require memorycontrollers; however, adding one in a cost-effective manner requires efficient system-level

design. Again, developers can leverage theflexibility of the FPGA to meet this need. Tosummarize, FPGA memory controllers providecustomized external memory interface designoptions to meet DA bandwidth needs andoptimize all aspects of the cost equation (mem-ory device type, number of PCB layers, etc.).

DA Image Processing Need for On-ChipMemory ResourcesIn addition to external memory needs, cam-era-based DA processing can benefit from on-chip memory that serves as line buffers forprocessing streaming video or analyzingblocks of image data. Bayer transform, lens dis-tortion correction and optical-flow motion-analysis are examples of functions that requirevideo line buffering. For a brief quantitativeanalysis, a Bayer transform function using 12-bit-pixel Bayer pattern intensity information toproduce 24-bit color data is examined.Implemented as a raw streaming process, abicubic interpolation process requires buffer-ing four lines of image data. Packing the 12-bit-intensity data into 16-bit locations requiresapproximately 20.5 kb of storage per line, or82 kb for four lines of data.

As part of their suite of on-chip resources,today’s FPGAs offer localized memory calledBlock RAM. The BRAM supports line bufferstorage of image data in close proximity tofabric-based image processing cores. As FPGAs now target vision applications, therelative amount of BRAM resources hasincreased with each product family.

A Single All-Programmable PlatformIn addition to external memory bandwidthrequirements and image processing needs,having a single, all-programmable system on achip (SoC)-based platform for DA applicationsoffers automotive manufacturers the uniqueability to address both the technical challengesand business goals in their DA designs. Thistype of all-programmable platform offersdesigners an integrated, flexible, power opti-mized solution with high computational per-formance that automotive manufacturers and

their electronics suppliers can combine withtheir own hardware and software, available IPand design frameworks to reduce develop-ment time, bill of material (BOM) costs and riskfor next-generation DA solutions.

Currently, this type of platform has only beenoffered as a multi-chip solution, which canrequire additional processing that keeps BOMcosts high, and reduces flexibility options toscale between vehicle platforms. Yet automo-tive designers can now take advantage of theindustry’s first SoC family that incorporates anARM dual-core Cortex-A9 MPCore processingsystem with tightly coupled programmablelogic on a single die. This combination dramat-ically increases performance, which is criticalfor processing-intensive real-time DA applica-tions, and enables greater system integration,allowing the bundling of multiple DA applica-tion, while simultaneously reducing BOM costsby minimizing device cost and the cost ofadditional hardware platforms.

Automakers are eager to offer car buyersincreasingly advanced DA applications, whichhave already proven to be quite popular inmanufacturers’ high-end vehicles. By presentingnew DA applications and being able to offermultiple DA applications per vehicle using an allprogrammable, customized solution, automak-ers are now given the opportunity to differenti-ate their vehicles from those of their competi-tors in a hotly contested market.

ConclusionSystem designers working on DA processingplatforms must consider architectural flexibili-ty, platform scalability, external memory band-width, on-chip memory resources, high-speedserial interfaces, and parallel/serial processpartitioning. The challenge is to strike anappropriate balance between meeting theseneeds and maintaining a competitive productcost structure. In this quest, FPGA technologyis a viable alternative to standard ASSP andASIC approaches. n

www.xilinx.com

DESIGN AUTOMOTIVE

About the author:Paul Zoratti is a member of the Xilinx Automotive Team. As a senior systemarchitect and manager of driver assistance platforms, his primary responsi-bility is the global application of Xilinx technology to automotive driverassistance systems. Zoratti holds master’s degrees in both electrical engi-neering and business administration, both from the University of Michigan.He also has a specialized graduate certification in intelligent transportationsystems, also from the University of Michigan. Zoratti has been awarded 16United States patents associated with vehicle safety technology.

This article was originally published in the 2013 Engineers’ Guide to FPGA and PLD. To view additional FPGA technical articles, news, opinions, product information, market trends andindustry research visit www.eecatalog.com/fpga. Subscribe to the FPGA and PLD Quarterly Report at www.eecatalog.com/subscribe


As high-speed rail networks expand and jour-ney times shrink, international rail travel isonce again in vogue. For the traveler, at least within the EuropeanUnion, national borders have all but ceased toexist, but the same is not entirely true for thetrains themselves. However, the traction unitpulling the train still makes a transition as itcrosses from one country to another. Thepower drawn from the overhead line will stillbe supplied, in most cases, by a national rail

service, and will be discontinuous at the bor-der: it may – or may not – be of similar voltageand frequency each side of the transition.

This presents the designers of rail traction sys-tems with a number of problems. Firstly, andmost obvious, the traction set must be “multi-standard” – it must be able to operate fromany power supply it will encounter across thenetwork, which with today’s schedules canmean multiple countries in a single journey.

On-board power measurementA somewhat less obvious requirement is theneed for accurate monitoring of the powerdrawn by the traction motors, at all times. Eachauthority supplying power on the networkneeds to be paid for the energy that an inter-national train uses while it is within its nationalboundaries. Monitoring power supplied toeach train from the supply side is impractical, sothe only feasible way of producing accuratebilling information is to measure and log power

New standards impose accuracy requirements for energy metering thatrepresent a massive step-change relative to prior industry practice.

DESIGN AUTOMOTIVE

by Hartmut Gräffert

Measuring energy flowsin cross-border railtraction operations

Figure 1: Concept of prEN 50463.


at frequent intervals throughout the journey.Each entry in that record must also carry infor-mation on where the train was at each measure-ment point; the ubiquitous GPS can readilyprovide this part of the data. A further compli-cation is that modern traction systems employregenerative braking, running their motors asgenerators and returning power to the net-work as they slow. For full fiscal accuracy, there-fore, the energy measurement system has to bebidirectional.To harmonise operation of international servic-es, a European standard has been drawn up thatspecifies exactly how the energy measurementfunction (EMF) is to be carried out (Figure 1);still at the provisional stage, and therefore car-rying the prefix “pr”, the document isprEN50463. It specifies a record that includesparameters such as data and time, train identifi-cation, network frequency – which can be16.7Hz, 50Hz, 60Hz or DC - locations, and, cen-tral to the core function, the load profile. The record must include absolute energy va -lues for both “real” and reactive power,embodying information on power factor thetrain presents to the line at each instant.

New levels of measurement accuracyWhere EN50463 sets new challenges is in theaccuracy it demands, specified in terms of“class R” (rail) figures. The overall requirementis for measurement of energy to an accuracy of±1.5% for AC and ±2% for DC. There are three

main elements in the measurement andrecording of energy; transducers to measureinstantaneous current and voltage, and ameter that takes those readings, computes andstores the energy values. For AC readings, itmust take account of the phase angle differ-ence between voltage and current waveformsto derive both real and reactive power figures.Each of those elements contributes someuncertainty to the process, and those errorsadd in a root-mean-square manner (that is, thetotal error is the square root of the sum of thesquares of the individual error figures).Therefore, transducers and the energy metermust individually perform to considerablytighter tolerances than the overall error figuressuggest. To enable traction system designersto meet the requirements of EN50463, LEMhas assembled a matched offering of transduc-ers with enhanced accuracy, together with thenew EM4T II energy meter.

High-accuracy current, voltage transducersMeasuring current to the levels of accuracyspecified in the new standard is challenging; itis even more difficult in the rail traction envi-ronment, where transducers may be exposedto large variations in ambient temperature, tolarge external magnetic fields, and high levelsof electrical noise. The current to be measuredalso contains very high peak values and largetransients: all of which constrains the choice of

an appropriate measurement technology.Direct measurement by a resistive shunt is pos-sible, and LEM can meet the required specifi-cations with transducers from its DI serieswhen this method is specified. However, an in-line resistance of sufficiently low value to keeplosses acceptable at the upper end of the cur-rent range is challenged to yield the requiredaccuracy at low current values, and self-heatingeffects make achieving linearity across the cur-rent range very difficult. Also, that approachdoes not provide galvanic isolation.

Therefore, in many cases an indirect currentmeasurement is preferred. Several technolo-gies exist to make indirect measurementsbased on the magnetic field generated by thecurrent in the conductor feeding the tractionsset. In their physical form, they appear superfi-cially similar, a toroidal sensor with the primaryconductor passing through the centre of thering. Measurement of the magnetic fluxinduced in the toroid by the primary currentrelates directly to the value of that current.One form of this sensor uses the Hall effect totranslate magnetic flux to a measurement value.However, Hall Effect sensors have difficulty inmaintaining linearity over a large dynamicrange, such as that specified in EN50463: andmay also exhibit limited accuracy at high andlow extremes of their measurement range, dueto saturation and residual magnetisation effectsin the magnetic material.

DESIGN AUTOMOTIVE

Figure 2: Block diagram of EM4T II.


LEM has chosen to address the problem with aFluxgate sensor, that will carry certification toClass 0.5R (±0.5% accuracy); the company hasmany years’ experience in sensors usingFluxgate technology. The Fluxgate sensor usesa nulling principle; it generates a magnetic fluxin the toroidal core to exactly balance thatinduced by the primary current, and from thedrive level needed to do so, infers the value ofthe current. In operation, the sensor drives themagnetic core with a high-frequency AC signalthat constantly reverses the magnetisation ofthe core, driving it around the characteristicknown as its B-H curve. The additional mag-netic field that arises due to the primary cur-rent modulates this behaviour; using thisapproach the sensor can detect the null condi-tion with extreme sensitivity. At all values ofmeasured current the magnetic material isdriven to saturation in both directions, remov-ing any direct dependency between measure-ment linearity and the magnetic core’s charac-teristics. The result is a sensor that can offervery high accuracy and linearity over a widedynamic range.Nominally rated at 4000A, the ITC 4000 willmeasure ±6000A, consuming less than 80mA (atzero primary current) to under ±340mA (at4000A primary current) from a supply voltageof ±24V to its measurement (secondary) circuit.As noted previously, Fluxgate technology is

capable of extremely high levels of both accu-racy and linearity; the ITC 4000’s linearityerror is under 0.05%. The device’s offset cur-rent is less than ±10μA and it also exhibitsextremely very low temperature drift. The ITC4000 operates over –40 to +85°C, and meetsor exceeds all relevant standards for safety andoperating environment. Voltage measurement is provided by a trans-ducer from LEM’s DV series, that will be avail-able with 1% or 0.75% full-range accuracy(Class 1R or Class 0.75R certified accuracy, asexpressed in the rail traction standards). Theseries meets or exceeds all of the performanceand safety demands of rail traction systems,both those in use today and planned for thefuture. It offers insulated voltage measure-ments from 1200 to 4200V, in a package signif-icantly smaller than any other product in themarket. (Figure 2)

Metering to fiscal standardsCompleting the offering is a new, enhanced,version of LEM’s EM4T energy meter, EM4T II,also rated and certified to Class 0.5R accuracy.The compact unit is a single-phase energymeter that meets all current and proposedstandards for on-board rail-traction energymonitoring and, specifically, complies with allof the requirements of the new EN 50463draft. The EM4T offers four input channels to

accept measurements from any existing AC orDC traction supply network. From voltage andcurrent measurements, it calculates active andreactive power, compiles a load profile, andstores the values in internal flash memory; datapoints are recorded at selectable intervalsranging from 1 to 60 minutes. Data points inthe record are stamped with information suchas time and date, train identification, and theprecise location of the train at every interval:location is derived from a dedicated GPS inputto the EM4T II. Recording at 15-minute inter-vals, the EM4T II has sufficient internal memo-ry for over 300 days’ data. Real-time datainterfaces also support exchange of data withother train systems, including a driver display.The EM4T II features excellent immunity tothe high levels of electrical noise that are typi-cal of the traction environment. It supports bi-directional monitoring of energy flows and cancorrectly record energy returned to the sup-ply network during regenerative braking.

While the EM4T II offers complete interfacingflexibility, and can connect to any suitably-rated transducer, the combination of the ener-gy meter with LEM’s optimised current andvoltage transducers offers the industry’s onlyroute to full compliance with prEN 50463, withcertified class accuracy. nwww.lem.com

DESIGN AUTOMOTIVE

Murata announced the availabili-ty of the RBQ series of isolated12Vout, 400 Watt DC-DC con-verters from Murata PowerSolutions. The RBQ-12/33-D48model provides what arebelieved to be the industry’shighest efficienciesof up to 96% froma standard DOSA-compliant quarter-brick package. TheRBQ-12/33-D48 isdesigned to oper-ate in most appli-cations with con-vection cooling.The RBQ-12/33-D48 can operatefrom a standard telephone net-work voltage (TNV) networksupply of 36 to 75 VDC arounda nominal 48 VDC. The RBQseries is ideal for use in a host oftelecommunications, enterpriseIT and wireless networking infra-structures that use a distributedpower or regulated intermediatebus architecture. The RBQ-

12/33-D48 will reliably deliver33A/400W @ 60°C with 200LFMairflow.Optional features include a loadsharing option for higher poweror N+1 applications, conformalcoating for harsh environments,

Positive & Negative on/off con-trol logic and optional pinlengths to accommodate variousapplication board thicknesses.Protection features include out-put over voltage, over currentprotection, input under voltageand over temperature shutdown. MURATA EUROPEwww.murata.eu

400 W quarter brick DC-DC converter from Muratadesigned for telecom applications

Murata announced availability ofthe Murata Power Solutions PAQseries of 150 Watt isolated DC-DC converters. Packaged as anindustry standard open-framequarter brick module withthrough-hole mount-ing, the converterhas a 2:1 input rangeof 36 – 75 VDCaround a nominal 48VDC input. The sin-gle 29.8 VDC nomi-nal output can betrimmed from 23.83(-20%) to 32.78(+10%)VDC in orderto accommodate anapplication’s specific voltagerequirements.This highly efficient DC-DC con-verter, typically 92.5%, can deliv-er up to 5 Amps output currentfrom the fully regulated output.An optional base plate is avail-able for use in conduction-cooleddesigns such as sealed box RFpower amplifiers that are oftenmounted on radio masts. For

these applications the converterand base plate are fixed directlyto the host equipment chassis. Inthis way the operation of end-equipment is efficiently main-tained despite being situated in a

potentially demanding physicalenvironment. The high efficiencyof the PAQ converter greatlyassists in reducing operating costsdue to less waste power beinggenerated and helps maintain theunit’s reliability. Input / Outputisolation is provided by 2,250VDC Basic insulation.MURATA EUROPEwww.murata.eu

150 W DC-DC converter suits micro cell transmitterand power amplifier applications


MAZeT GmbH, a leadingprovider of Electronic Design andManufacturing Services of opto-electronic and embedded com-puting solutions, present theirnew Android based solutions forembedded systems. Modernrequirements for embedded sys-tems are that defined tasks needto be solved at maximum efficien-cy. This is at limited usage ofoperating systems and at minimalhard- and software effort. Priceperformance-radio, reliability andflexible data rates are high indemand. Embedded systemsoften use modified versions ofoperating system, such asEmbedded Windows, Linux orAndroid. For systems with high

requirements on performanceand reliability it is also recom-mended to use customizedFPGAs or ASICs. In relation to theEmbedded World 2013 inNuremberg the newest systemsolutions and applications for themobile sector will be presented.MAZeT for example, demon-strates the wireless communica-tion via Bluetooth with a com-pact pre-engineered Androidsystem. The unit addresses alighting system with feedbackcontrol loop, which uses special-ly developed color sensor tech-nology to control and optimizethe color value output and long-term stability of LEDs.Android implementation and

IAR Systems® announces furtherenhancements to its innovativePower Debugging technology.At Embedded World inGermany, the company demon-strates new functionality in itsworld-leading development toolsuite IAR EmbeddedWorkbench® forARM®. This func-tionality includesdetailed in-circuitpower measure-ments made possi-ble by IARSystems’ new I-scope™ probe. Power Debuggingprovides develop-ers with informa-tion about how the softwareimplementation of an embed-ded system affects the system’spower consumption. This tech-nology is available using the C-SPY® Debugger included inIAR Embedded Workbenchtogether with IAR Systems’ in-circuit debugging probe I-jet™.By adding I-scope, developerscan gain knowledge of thepower consumed by individualmodules, detect if design flawsin the code are causing unneces-sary power consumption andpossibly extend battery lifetime.At any designated point on the

target board, I-scope can meas-ure current and voltage with 12-bit resolution at a sampling rateof up to 200 kHz. The informa-tion is transferred to I-jet, whichsynchronizes the data with therunning program counter,enabling it to be graphed in real

time and analyzed using the C-SPY Debugger. The analysis canbe used to reveal the powerconsumption of individual func-tions and peripherals, which I/Oactivity causes current spikes,and how much power is con-sumed in various sleep modes.I-scope is designed to be usedas an extension to I-jet and willbe available for delivery in Q22013. IAR EmbeddedWorkbench is a complete devel-opment tool suite with supportfor all ARM cores.IAR SYSTEMS www.iar.com

IAR Systems further strengthens product portfoliofor power analysis

AndroidTM Meets Embedded Systems

Mouser Electronics, Inc. is nowstocking the new Cortex™-M0+microcontrollers from NXPSemiconductors. LPC800 MCUsoffer easy-to-use peripherals,32-bit capabilities, and excep-tional power efficiency.Based on an ultra-low-power 30-MHz ARM Cortex-M0+ proces-sor, LPC800 is fully compatiblewith the Cortex-M architectureand instruction set and offerssuperior code density to 8- / 16-bit architectures. The Cortex-M0+features a two-stage pipeline thatreduces power consumptionwhile improving performance.LPC800 MCUs also take advan-tage of the Cortex-M0+ peripher-al bus, allowing single-cycleaccess to the GPIOs. These fea-tures enable NXP LPC800 MCUsto offer deterministic, real-timeperformance – a key requirementfor 8-bit developers. LPC800includes game-changing features,such as a switch matrix thatenables designers to assign on-chip peripherals to any pin with asingle line of code or a single clickin the configuration tool. With its

broad product line and unsur-passed customer service, Mousercaters to design engineers andbuyers by delivering What’s Nextin advanced technologies.Mouser offers customers 19global support locations andstocks the world’s widest selec-tion of the latest semiconductors

and electronic components forthe newest design projects.Mouser Electronics’ website isupdated daily and searchesmore than 10 million products tolocate over 3 million orderablepart numbers available for easyonline purchase. MOUSER ELECTRONICSwww.mouser.comNXP SEMICONDUCTORSwww.nxp.com

Symtavision has launchedSymTA/S 3.3, a major new versionof its award-winning system-leveltool suite for planning, optimizingand verifying embedded real-time systems. SymTA/S 3.3 fea-tures significant new timing analy-ses including supportfor FlexRay SystemDistribution, the newCAN-FD standard,Buffer Fill Level analy-sis for COM andGateways, as well asenhanced Gantt chartcustomization. Majorimprovements to thedesign features ofSymTA/S 3.3 includea new Customizable WizardFramework and improved drag-and-drop functionality providingresolution of dependencies. Atthe same time, Symtavision hasannounced TraceAnalyzer 3.3, anew version of its powerful solu-

tion for visualizing and analyzingtiming data from both measure-ments and simulations whichseamlessly integrates withSymTA/S. With the launch ofSymTA/S 3.3, the SymTA/S SystemDistribution analysis framework is

extended with support forFlexRay covering both Static andDynamic Segments, with the lat-ter supporting cycle multiplexingfor FlexRay 2.x frames. SYMTAVISIONwww.symtavision.com

Launch of SymTA/S 3.3 features significant new timing analysis and design capabilities

Mouser First to Stock NXP LPC800 Series Cortex™-M0+ MCUs

PRODUCT NEWS EMBEDDED SYSTEMS


Highest Accuracy BatteryFuel Gauge from MaximIntegrated MaximizesBattery Capacity andBoosts User ConfidenceMaxim Integrated Products, Inc. announced that it issampling the MAX17050, the industry’s highest accu-racy battery fuel gauge. It uses the company’s patent-ed ModelGauge™ m3 algorithm, which improves uponthe traditional coulomb counting approach by com-bining it with the patented voltage-basedModelGauge approach. The MAX17050 provides thebest possible accuracy, uses 4x less power, and shrinkssolutions by 3× compared to competitive devices. Thedevice is ideal for portable applications where batterycapacity and state-of-charge accuracy is critical, includ-ing smartphones, tablets, mobile radios, battery back-up, and portable medical/fitness equipment.

Battery fuel gauge designs often depend on coulomb-counting technology. Although coulomb counting ishighly accurate for a short period of time, it tends todrift over time. The MAX17050 uniquely integrates theshort-term accuracy and linearity of a coulomb counterwith the excellent long-term stability of a ModelGaugefuel gauge. The ModelGauge m3 technology makescontinuous microcorrections and eliminates suddenjumps in a battery’s state of charge (SOC). Users cantrust the resulting battery capacity and state informa-tion, including the health and maximum battery usage.The MAX17050 functionality is also integrated into thecompany's TINI® Power System-on-Chip (SoC) to makeit easier to upgrade designs to more complex systems.Key Advantages• Best possible accuracy: integrated coulomb count-ing and ModelGauge technology automatically com-pensate for aging, temperature, and discharge rate;provides accurate SOC and time-to-empty over awide range of operating conditions, as well as batteryhealth information.• Extends battery capacity: sophisticated SOC, voltage,and temperature alerts allow the system microproces-sor to remain in sleep mode for a longer period of time.• Upgrade path to full power management: TheMAX17050 functionality is available in a Power SoCwith identical IP to facilitate a complete power-man-agement solution in a shortened time to market.MAXIM INTEGRATEDwww.maximintegrated.com

Mouser Electronics, Inc. announces theavailability of STMicroelectronics STM32F4 Expansion Boards that help discoverthe features of the STM32 F4 High-Performance Cortex™-M4 microcon-trollers. The STMicroelectronicsExpansion Boards aim to expand thefunctionality of the STM32 F4Discovery board, built aroundthe STM32F4 processor and fea-turing 32-bit ARM Cortex-M4architecture. The newly availableaccessories include an LCD mod-ule (a 3.5 inch LCD and driverboard) and a camera module(contains an OV9655, which is a1.3 megapixel CMOS SXGAimage sensor), both of whichconnect to the third accessory, ahardware extension base boardthat provides Ethernet connec-tivity by plugging directly intothe Discovery Board. The baseboard helps to extend out andconveniently offer all the inter-faces on the STM32F4 Discoveryboard. The new hardware acces-sories are custom designed for theDiscovery Board STM32 F4 and comewith the necessary software drivers,making them easy for engineers to useto quickly extend the functionality oftheir STM32F4 design. With its broad product line and unsur-passed customer service, Mouser catersto design engineers and buyers by deliv-ering What’s Next in advanced tech-

nologies. Mouser offers customers 19global support locations and stocks theworld’s widest selection of the latestsemiconductors and electronic compo-nents for the newest design projects.Mouser Electronics’ website is updateddaily and searches more than 10 million

products to locate over 3 million order-able part numbers available for easyonline purchase.Mouser.com also houses an industry-firstinteractive catalog, data sheets, supplier-specific reference designs, applicationnotes, technical design information, andengineering tools. MOUSER ELECTRONICSwww.mouser.com

STMicroelectronics STM32 F4 Discovery ExpansionBoards Now Available at Mouser

system customization can be a challeng-ing task - especially with non-standard-ized hardware. Embedded Android sys-tems are capable of modular upgradingso that own applications can be integrat-

ed via common interfaces. All knownARM processors from Freescale to TexasInstruments are supported throughoutthe implementation process. Energymanagement, safety protocols and per-formance are regarded as key aspects.As a system and design company andwith more than 20 years of experience inthe field of embedded computing solu-tions - MAZeT GmbH offers design andmanufacturing of system solutions from asingle source. MAZeT is specialized inporting operating systems to specificprocessor platforms, developing driversfor customer-specific interfaces, as well asdesigning application software anddeveloping firmware for various hard-ware solutions. The company developsspecific drivers and application softwarefor Windows, Linux and Android.MAZeTwww.mazet.de


Android is a trademark of Google Inc.


Maxim Integrated Products, Inc.and Freescale Semiconductor®have collaborated to showcase acomprehensive LTE/3G picocellbase station at the 2013 MobileWorld Congress. This innovativesmall cell base station platformdesign is field deployable bymobile operators and also servesas a production-ready referencedesign to accelerate time to mar-ket for equipment manufacturers.Ideal for outdoor public accessor enterprise small cell applica-tions, the LTE/3G picocell bringstogether a Freescale BSC9132base station SoC card, a MaximMAX2580* RF transceiver card,RF power amplifiers, and net-work management cards - all in a

passively cooled enclosure andweighing in at 6.8 kg. TheBSC9132 features two PowerArchitecture e500 cores and twoStarCore® DSP cores, along withFreescale’s high-performanceMAPLE-B2P baseband accelera-tor platform. The MAX2580 is anRF to Bits®, high-dynamic-per-formance LTE/3G radio transceiv-er. The BSC9132 and MAX2580work seamlessly to deliver a com-plete multi-band, multi-standardsmall cell solution. Key Advantages• MAX2580: Operates in all LTEbands from Band 1 to Band 41with selectable channel band-widths from 1.4 MHz to 20 MHz;supports FDD-LTE, TD-LTE, and

Ericsson has launched the indus-try’s first digitally controlledAdvanced Bus Converter specifi-cally aimed at poweringRedundant Array ofIndependent Disks (RAID) andNetwork Attached Storage(NAS) hard-disk applicationsthat operate with high capacitiveloads up to 15 millifarads (mF).Based on the Ericsson 3E* FRIDAII platform, the BMR456 ‘High-Cap’ series (BMR4560004/018)guarantees stable andaccurate 12V busvoltage over the fullinput-voltage range.RAID and NAS appli-cations developedfor datacenters aredesigned for highperformance and reli-ability and requirevery stable andsmooth bus voltageswithout disturbances, such asvoltage glitches or delays thatresult from power sources oper-ating with current-mode limita-tions. Designers will commonlyuse large arrays of capacitors,often resulting in an averagevalue of 12 to 15 mF, to securethat the bus voltages deliveredto hard disks and other sub-assemblies are exempt fromnoise. Developing these types of

products can be a significantchallenge, but the BMR456‘High-Cap’ converter makes fulluse of the flexibility offered bydigital control in addition to itsEricsson Energy andPerformance Optimizerfirmware, and embeds a voltage-mode-control algorithm thatmaintains a tight output voltagebetween 11.88V and 12.12Vand accurately controls the 20millisecond ramp-up time.

Designed for 12V bus applica-tions that might only require 6Vin standby mode, controlled viaPMBus commands, the outputvoltage of the BMR4560004/018can also be adjusted to as little as4V, even powering 5V board-level subsystems in idle mode,delivering significant headroom toboard designers.ERICSSON POWER MODULES www.ericsson.com/powermodules

New Ericsson Advanced Bus Converter Ideal forNetwork-Attached-Storage Applications

Maxim Integrated Collaborates with Freescale to Showca

With IPElog, the IPETRONIK-business unit IPEmeasure offers adata logger for particularly highmeasuring requirements in themobile data acquisition area.IPElog adds to the successful M-/S-LOG- and FLEETlog-family forconsistent further develop-ment based on the more than15 years of experience ofIPETRONIK in the field ofstand-alone data logger equip-ment. The new data logger isdistinguished by enhancedfeatures and higher performancefor data logging. These include,twelve ISO 11898-2-compliantCAN-Bus-measuring inputs,Quick start with No MessageLost-Function, as well as, a remov-able 1.8-inch-Solid-State-Disk forthe storage of large amounts ofdata. GPRS/UMTS/ 3G/WiFi-Hardware and a GPS-receiver areintegrated for the wireless data-transmission and global position-ing. For CAN-protocols there areoptional CCP, XCPonCAN,XCPonETH, KWPonCAN andUDS available.

IPElog based on a 1.33 GHz fastAtom-processor, has 1 GByteRAM and uses the real-timeoperating system RTOS-32 withthe field proven data acquisitionsoftware TESTdrive. Other fea-tures are two optional 100Base-

TX-Ethernet-measuring inputs, aUSB-2.0 port, four digital inputsand four digital outputs as well asintelligent power managementfeatures including Wake-On-CANfor all CAN-inputs and multipletrigger-functions. The transfer ofonline-measurement data toIPEmotion, ETAS INCA, andVector CANape is done by CANor XCPonETH. All settings of themeasuring configuration can bedone easily using the Windows-Software IPEmotion. IPETRONIKwww.ipetronik.com

DSM Computer presented at theembedded world for the firsttime the NanoServer® NN-QM67embedded system that offersmaximum power in the tightestspace. The 58 mm flat industrialcomputer is based on Intel®Core™ processors ofthe third generationand on the energy-saving Intel® QM67mobile chipset. Withthe compact, slot-lessNN-QM67, DSMaugments itsNanoServer® familywhose models differin the number ofslots and in the performanceclass. A sophisticated ventilationconcept ensures a low internaltemperature of the extremelycompact housing and so increas-es significantly the service life ofthe system despite its high per-formance. DSM Computer has

specified the maximum full-loadtemperature of the high-per-formance NanoServer® NN-QM67 not just 45 °C, but 50 °C.As standard, the NanoServer®NN-QM67 is equipped with theIntel® Core™ processor of the

third generation i5-3610ME withtwo cores (2.7 GHz, 3 MB). For high-end applications, amodel with quad-core Intel®Core™ i7-3610QE (2.3 GHz) canbe delivered as an option. DSM COMPUTER www.dsm-computer.com

DSM Computer at the embedded world: Flat embedded system of the third generation

IPETRONIK offers high-performance data loggerswith twelve CAN-Bus-measuring inputs for automobile tests



Maxim Integrated toDemonstrated a MoreIntegrated World at MobileWorld Congress At the 2013 Mobile World Congress in Barcelona,Spain, Maxim Integrated Products, Inc. presented itsunique and highly integrated portable equipment andcommunications solutions for mobile platforms. Maximwill host nine interactive demo kiosks with solutions forconsumer and RF/communications applications. AllMaxim mobility products feature high integration forbetter performance and substantial savings in space,cost, and design time. Some of the products demon-strated at Hall 6, Stand 6G100 include:

Consumer Applications• The MAXQ614, a low-power MAXQ® 16-bit micro-controller with infrared (IR) module, is designed forvery small applications such as smartphones andtablets. Available in a tiny 2mm × 2mm × 0.65mm, 25-ball wafer-level package (WLP), it shrinks solution sizeby 70% compared to previous solutions. TheMAXQ614 includes 80KB of flash memory and 2KB ofdata SRAM for applications flexibility. It provides theindustry’s best low-power battery-operated perform-ance with an ultra-low-power stop mode (0.2 μA typ). • The MAX98096 is a highly integrated, high-per-formance, smart audio hub. It integrates Maxim’sFlexSound® processor, which supports a plug-in-capa-ble audio framework. With low power consumptionand advanced audio signal processing, this single-chipaudio solution supports wideband voice, includingnoise cancellation and echo suppression, in a widerange of portable applications. • Maxim addresses another key piece of the mobilesystem with its new MEMS motion sensor, theMAX21000. The device is an ultra-accurate, 3-axis,digital-output gyroscope that delivers unprecedentedsensitivity over temperature and time. It offers lownoise, high bandwidth, and a high-speed interface andis suitable for both user interface (UI) and opticalimage stabilization (OIS) applications.RF/Communications Applications• The MAX2580* is a highly integrated LTE/3G smallcell RF transceiver that supports all major cellularbands. (*Future product - contact factory for availability.)MAXIM INTEGRATEDwww.maximintegrated.com

Atollic® announced the next release oftheir award-winning embedded inte-grated development environmentTrueSTUDIO® v4.0. Providing profes-sional embedded developers with acomprehensive integrated suite of tools,this latest version of TrueSTUDIO addssupport for many new ARM®-baseddevices in addition to tool chainimprovements.Suiting high volume, deeplyembedded automotive and con-sumer applications, the Cortex™-R4, Cortex-R5 and Cortex-R7real-time processors are nowsupported within TrueSTUDIOv4.0. The Cortex-R series ofdevice cores offer highly per-formance and real-time responsecharacteristics as required bydemanding system-on-chip(SoC) applications.Also in this release, the supportfor generic ARM Cortex-A-basedseries of application processors hasbeen extended from the Cortex-A5 coreto include the Cortex-A7, Cortex-A8,Cortex-A9 and Cortex-A15 processorcores. The Cortex-A series of cores areaimed at very high volume and low costapplications that are likely to be using afeature-rich real-time operating system.Atollic TrueSTUDIO v4.0 now supportsa total of over 1300 ARM devices fromleading semiconductor manufacturers.This support now covers devices usingARM7™, ARM9™, Cortex-

M0/M0+/M1/M3/M4, Cortex-R4/R5/R7and Cortex-A5/A7/A8/A9/A15 cores.Complementing the addition of targetcore support, a number of specificdevices and evaluation boards have alsobeen added to the list of support tar-gets. These include the InfineonXMC1000 family of Cortex-M0 devicesand the NXP LPC4xxx family of Cortex-

M0/M4 dual core microcontrollers. Withthe LPC4xxx targets TrueSTUDIO v4.0can debug both cores simultaneouslyusing only one GUI debugger instance,vastly simplifying dual-core develop-ment. Additional IDE improvementsinclude the upgrading of the Eclipseplatform and the GNU compiler toolchain. The compiler now has bettercode size optimizations in addition to asmaller runtime library. ATOLLICwww.atollic.com

Atollic TrueSTUDIO C/C++ IDE adds Cortex-R support andextends coverage of Cortex-A application processors

WCDMA modes; JESD207 data interfaceprovides a seamless radio interconnectbetween the MAX2580 and BSC9132.

• BSC9132: Supports multimode opera-tion that enables it to process LTE-FDD/TDD and HSPA+ users simultane-

ously. It is engineered toaddress the perform-ance and cost require-ments of up to 20MHzsingle sector LTE-FDD/TDD. The BSC9132can process 64 simulta-neous users and handles150 Mbps downlink with75 Mbps uplink rates in20 MHz LTE-FDD/TDDmode, and 42 Mbpsdownlink with 11.5Mbps uplink rates in 5MHz HSPA+ mode.MAXIM INTEGRATEDwww.maximintegrated.com

ase a Comprehensive Outdoor LTE Picocell at Mobile World Congress



Silicon Labs. unveiledPrecision32™ software and hard-ware development tools thatenable engineers to maximizethe power efficiency of their 32-bit embedded designs. A leading supplier of mixed-sig-nal microcontrollers (MCUs),Silicon Labs offers a diverseportfolio of Precision32 MCUsbased on the ARM® Cortex™-M3 processor. This portfolio issupported by a rich set ofdevelopment tools that enablesdesigners to optimize theirdesigns for the lowest powerconsumption without compro-mising performance. SiliconLabs’ new low-power tools ondisplay at Embedded World

include the following:Power-Aware software tools:Silicon Labs’ complimentaryEclipse-based IDE andAppBuilder software forPrecision32 MCUs include first-of-a-kind tools that enable devel-opers to estimate power con-sumption and receive configura-tion guidance to minimize systempower. The Power Estimator toolgives developers a top-levelgraphical view of how aPrecision32 MCU uses power inactive and sleep mode. Low-Power 32-bit humiditysensing demonstration: Thisdemonstration highlights thefunctionality of Silicon Labs’ newSi7005 relative humidity (RH)

Power Integrations announcedRDR-347, an LED-downlight ref-erence design. The new circuitdemonstrates the capabilities ofthe recently launchedLYTSwitch™ IC family, whichoffers the industry’s best high-end dimming performance froma single-stage LED driver–together with allthe associatedefficiency, spaceand cost benefitsthat the single-stage approachbrings.The RDR-347 12W referencedesign, based onthe LYT4313E,delivers a powerfactor greaterthan 0.95 and reduces total har-monic distortion (THD) to lessthan 10% - easily meetingEN61000-3-2C requirements.Efficiency is greater than 86% at120 VAC – industry-leading per-formance for an isolated solutioncapable of operating with awide range of dimmers. Thisperformance is possible becauseLYTSwitch ICs use one com-bined PFC and CC power con-version stage which minimizeslosses and cuts componentcount, which in turn increases

reliability and decreases cost.TRIAC dimming is challenging,especially deep-dimming whereTRIAC asymmetry between half-cycles can have a significanteffect. RDR-347 shows thatPower Integrations’ LYTSwitchdriver IC works excellently, evenat very low output currents,

without any shimmer or flicker.The design also demonstratessystem start-up without notice-able hysteresis (so-called “pop-on effect”), even in deep-dim-ming. The IC features a very faststart-up time (under 500 ms)even when dimmed to 10% out-put current, achieving the‘instant-on” condition that manycustomers want, but that manydesigns in the market cannotprovide.POWER INTEGRATIONS www.powerint.com

RDR-347 combines industry-leading efficiency,low component count and great dimming

Silicon Labs Unveils Latest Low-Power Development Tools

Murata’s innovative RFID tag -MAGICSTRAP® - integrated withCogiscan’s TTC Middlewaresolution has won the 2013 NewProduct Introduction (NPI)Award from Circuits Assembly inthe labelling equipmentcategory. Murata’s MAG-ICSTRAP® series is thelatest in UHF RFID tagtechnology designed tobe placed directly onthe printed circuitboard. Seamlessly com-bined with Cogiscan’smiddleware technology,it offers a complete RFIDtrack, trace and controlsolution for the PCB market.Designing MAGICSTRAP® on thePCB is made easy by selectingone of the recommended refer-ence antenna designs based onthe users’ read range require-ments. This eliminates the needfor UHF antenna design knowl-edge, greatly simplifying andaccelerating integration.The Murata-Cogiscan solutionenhances PCB tracking function-ality by using UHF RFID technol-ogy to provide a read-writabledata repository on the PCB andenables faster, more reliableidentification and communica-

tion while overcoming the chal-lenges posed by traditional opti-cal track and trace systems. Italso enables tracking of the fin-ished goods throughout the sup-ply chain, for customer service

and field support.With what is believed to be thesmallest UHF RFID module in theindustry, MAGICSTRAP® turnsthe standard PCB into a “Smart”PCB that truly provides a com-plete “cradle to grave” solution.For more information on MAGIC-STRAP® visithttp://www.murata.com/prod-ucts/rfid/index.html, for “cradle tograve” RFID solutions visitwww.rfid-valuecreators.comMURATAwww.murata.euCOGISCANwww.cogiscan.com

CUI Inc has introduced a newfamily of board mount dc-dc con-verters developed for railwayapplications. The modules aredesigned to comply with theEN50155 standard, which speci-fies input, EMC, mechanical, andenvironmental requirements. Theinternally potted and encapsu-lated design provides increasedsystem reliability throughadded protection from envi-ronmental factors such as dust,moisture, and shock and vibra-tion. Offered in industry stan-dard quarter brick and half brickfootprints respectively, the 100 WVQB100R series and the 150 WVHB150R series are highly effi-cient, reaching levels up to 92.5%.Though the modules aredesigned primarily for railway

applications, the VQB100R andVHB100R series also targetdesigns that may experience hightransients, including telecom sys-tems and battery-poweredequipment. The dc-dc convertersprovide a 3:1 input range of66~160 Vdc and output voltage

options of 5, 12, or 24 Vdc. Theyare designed to deliver a caseoperating temperature range of -40°C to 100°C and provide 2250Vdc I/O isolation.CUIwww.cui.com

CUI Launch EN50155 Compliant Dc-Dc Convertersfor Railway Applications

PCB Mounted RFID TAG Solution from Murata andCogiscan wins NPI award

PRODUCT NEWS ACTIVE COMPONENTS


High-current, high-voltage andhigh-performance applicationsbenefit from Microchip’s new LDOvoltage-regulator familyKey Facts:• LDOs provide industry’s best combination of per-formance, size and price• High output current of 300 mA, high input voltageof 16V and high noise rejection of 70 dB• Ideal for designs with 9V and 12V power-supplyrails which require very clean supply voltages• Suited to a wide range of applications includingmedical, mobile devices and LED lighting controllers

Microchip announces the MCP1755 and MCP1755Shigh-performance, high-input-voltage and high-out-put-current family of CMOS Low-Dropout (LDO) volt-age regulators. These LDOs can accept a wide inputvoltage range of 3.6-16V and deliver an output currentof 300mA at output voltages of 1.8-5.5V, while con-suming only 68μA of quiescent current (typical).Additionally, the MCP1755/1755S LDOs have a shut-down pin, enabling them to consume less than 0.1μA(typical) during shutdown mode, which extends bat-tery lifetimes. To conserve space, the LDOs are offeredin SOT and 2 × 3mm DFN packages. These featuresprovide designers with a high degree of flexibility foran extremely broad range of applications, such as med-ical, mobile devices and LED lighting controllers.Many of today’s electronic designs need to operate inenvironments that contain high levels of electrical andmagnetic noise. Noise is generated when switchingtransients and RF signals get coupled onto power-supply lines. Applications such as ground- and arc-fault circuit interrupter (GFCI/ AFCI) circuit breakers,wireless devices, automotive-aftermarket electronics,and medical devices are sensitive to this power-sup-ply noise. The MCP1755 and MCP1755S LDOs havea high Power Supply Rejection Ratio (PSRR) of 70dB@ 1kHz, which filters input noise and provides signif-icantly cleaner output voltages, making them well suit-ed to noise-sensitive applications.Both of the new LDOs are available today for samplingand volume production. The MCP1755S is offered in3-pin SOT223 and 8-pin, 2x3 mm DFN packages;whilst the MCP1755 is available in 5-pin SOT223 andSOT23 packages, as well as the 8-pin, 2 × 3 mm DFN.

MICROCHIP TECHNOLOGYwww.microchip.com/get/LMUT

CUI Inc announces a family of compactac-dc power supplies featuring base-plate-cooling. The VBM-360 seriesoffers exceptionally high power densityof 14.8W/in3, providing 355W in anindustry standard footprint.The space-saving power supplies arehoused in two optional case styles: acompact open frame version measuring5 × 3 × 1.6 inches (127.00 × 76.00 ×40.60mm) and an enclosed version.Featuring typical efficiency above 93%,the power supplies are designed forapplications where forced-air cooling isnot preferred due to audible noise orspace limitations.The VBM series isalso ideally suitedfor sealed-boxapplications whereheat must be con-ducted throughthe chassis, a com-mon requirementin commercial andindustrial equip-ment that’sdesigned for out-door and harshenvironments.CUI’s VBM-360 power supplies have auniversal 90~264Vac input for globaloperation and are available with singleoutputs of 12, 24 and 48Vdc. Utilizingconduction-cooling via an integratedbaseplate, open frame models are ratedfor operation at 80% from -20°C to 40°C

ambient, derating to 40% load at 70°C,while enclosed versions are rated foroperation at 100% load from -20°C to40°C, also derating to 40% load at 70°C.For applications that can incorporate airflow of 10 CFM, the open frame unitswill operate at 100% load from -20°C to50°C, while enclosed versions will oper-ate at 100% load from -20°C to 60°C.The VBM-360 series features includeactive power factor correction (PFC) andprotections for over voltage, over cur-rent, short circuit, and over temperature.Connections for remote on/off control,remote sense, 5V standby, and a 12V fan

output are also included. All modelscarry UL/cUL and TUV 60950-1 safetycertifications for ITE, commercial andindustrial equipment, and are compliantwith the EN 55022 Class B EMI standard.CUIwww.cui.com

CUI’s Baseplate-Cooled Ac-Dc Power Supplies Deliver 355 Wat 93.5% Efficiency

and temperature sensor in a SiM3L1xxMCU evaluation board environment.The standalone, battery-powereddemonstration board includes an LCDpanel that shows changes in temperatureand humidity readings from the Si7005sensor. The demo includes examplecode that can be used to speed devel-

opment of applications requiring bothan ultra-low-power 32-bit MCU andhumidity and temperature sensors suchas smart thermostats and in-home energymonitoring systems. Low-Power SiM3L1xx developmentboard: Roughly the same size as an IDbadge, this compact developmentboard showcases the power efficiencyof SiM3L1xx MCUs. The board containsan ultra-low-power SiM3L1xx MCU,segmented LCD, supercapacitor, LEDand photodiode sensor, debug inter-face and USB port. The board can dis-play information continuously on theLCD for up to three days after a quick90-second charge of the supercapacitorthrough a USB cable. SILICON LABSwww.silabs.com

s at Embedded World 2013

PRODUCT NEWS ACTIVE COMPONENTS

With its rich feature set, the EAeDIPTFT57-A intelligent displayfrom Electronic Assembly is theideal choice for implementationof interactive control in mechani-cal engineering and industrialelectronics applications. The high-contrast colorscreen, which measures5.7" in the diagonal, hasLED background illumi-nation and a resolution of640 x 480 pixels. Its mostimportant feature, how-ever, is the built-in intelli-gence which greatly facili-tates integration of thedisplay modules into the applica-tion. The Evaluation Kit, which istailored specifically to this display,makes development work eveneasier. Electronic Assembly sup-plies the kit to speed up the com-missioning process and facilitatefamiliarization with the world ofintelligent displays. It includeseverything needed to get startedincluding an evaluation/program-mer board, EA eDIPTFT57-ATP

color display, USB cable, touchpanel, plug-top power supply(with international adapters) and amini DVD containing software,documentation and samplemacros. The evaluation board has

a USB port for communicationwith the PC. Interface boards withRS-232, RS-485, I²C and SPI inter-faces for communication with thehost are optional features butthey are nevertheless includedwith the kit. An array of LEDs pro-vides information about data traf-fic and the status of the digitaloutputs of the display. ELECTRONIC ASSEMBLY www.mediaberatung.de

Gleichmann Electronics present-ed the innovative multi-touchprojected capacitive touch(PCT) technology from Zytronicon its Booth 2-219 in Hall 2 atembedded world2013, February26 28, 2013 inNuremberg,Germany. Thistechnology hasthe capability tosupport up to 10simultaneoustouch points ondisplays with up to 55-inch(139.7 cm) screen diagonal. The multi-touch solution, whichwas developed by Zytronic forlarge display formats used ininformation and self-service ter-minals as well as industrial applica-tions, is based on its ZXY200touch controller and a ruggedi-zed, scalable touch sensor. Inaddition to offering 15 inch (38.1cm), 22-inch (55.9 cm), 32-inch

(81.3 cm), 46-inch (116.8 cm) and55-inch (139.7 cm) sensors, cus-tomer-specific sizes in low vol-umes are also available. Like all of Zytronic’s PCT touch

sensors, the multi-touch sensorsolution uses a matrix of 10micron diameter copper elec-trodes embedded within a 4 mmthick, durable glass laminate whichis both impact and scratch resist-ant. The sensor also works withgloved hands, thus considerablyincreasing the application areasfor this solution. GLEICHMANN & CO. ELECTRONICSwww.msc-ge.com

Multi-touch projected capacitive touch (PCT)technology for displays with up to 55-inch(139.7 cm) screen diagonal

Evaluation Kit Provides Simple Pathway toDisplay Development

Gleichmann Electronics presentsnew 5.7-inch (14 cm) industrialdisplays, G057AGE-T01 withQVGA and G057VGE-T01 withVGA resolution, from Innolux.Innolux guarantees a minimum offive yearsavailability forboth displays. The displays,which areequippedwith a TTLinterface, fea-ture unusuallywide viewingangles for TNdisplays of160° horizon-tal and 140° vertical, and a lowpower consumption of just 3.35W. The lifetime of the LED back-light is specified at 50000 hoursminimum. 500 cd/m2 brightnesswith QVGA and 450 cd/m2brightness with VGA resolution, acontrast ratio of 800:1 and a wide

operating temperature rangefrom -30°C to +85°C allows foruse of the displays outdoors.Both displays have identical formfactors and are 100% pin-com-patible. Furthermore, as with all

Innolux’s "Industrial Line" of dis-plays, a reverse scan function isprovided. Detailed information can berequested by sending an emailto [email protected] & CO. ELECTRONICSwww.msc-ge.com

DSM Computer has augmentedits new ultra-slim panel PC familywith the PN18-A2 widescreenmodel that provides a 47 cm(18.5 inch) TFT display with PCTmultitouch screen and LED back-light. The PCT technology offersthe familiarease-of-useknown withsmartphones,also for indus-trial displaysystems, whilealso satisfyingthe highrequirementsthat industryplaces onrobustnessand longevity.The PN18-A2 panel PC ideal forthe harsh industrial environmentis characterized with its wear-freeand scratch-resistant dual-touchscreen (two-finger operation)that has a flat surface without anydirt-accumulating edges. Onrequest, a multitouch for opera-tion with more than two fingers

can be deployed. The maximumresolution is 1366 x 768 pixelswith luminance of 300 cd/m². Thecontrast is specified as 1000:1.As an alternative to thewidescreen panel PC, DSMComputer also offers the D18

with integrated A/D converter aspure multitouch display. Thefront display part is identical withthe panel PC. The monitor can becontrolled with DVI, VGA and S-Video; the multitouch can beconnected via USB to an externalcomputer.DSM COMPUTERwww.dsm-computer.com

Flat widescreen panel PC and industrial monitorwith multitouch

Low power consumption 5.7-inch displays with QVGAor VGA resolution and at least five years availability


PRODUCT NEWS DISPLAY


n Optical sensorsn Sensors for logistic applicationsn Safety at work

n Optical Sensorsn Inductive Sensors

n Color Sensorsn True Color Sensors, Spectrometersn Gloss Sensors

Leuze

Contrinex

Sensor Instruments

n Linear Sensorsn Angle Sensorsn Tilt Sensors

n PLCsn Temperature Controllern Timer

n Flowmetersn Level Indicators and Switchesn Pressure Sensors and Switches

ASM

Selec

Koboldn Linear Solenoidsn Permanent Electromagnets

Westec

HTP

Intertec

Visit our online shopwww.oboyle.ro

n Circular connectors M8; M12; M23n Cable and Connectors for Sensorsn Valve Connectorsn Distribution Blocks

n Heavy Duty Industrial Connectorsn Power and Data Transmission Connectorsn Aluminium Junction Boxes

AUTOMATION

Honeywell announced it has expandedits portfolio of Relialign™ door interlockswitches with a version designed specif-ically for commercial applications, theCommercial Door Interlock Switch, CDISeries. This new device is Honeywell’salternative to large, installation-inten-sive interlocks and OEM custom-madeproducts. Potential applications for theCDI Series include commercial swing-door elevators, dumbwaiters, and plat-form/vertical lifts. Honeywell’s Relialign™ CommercialDoor Interlock Switches are designedto hold the swing door in place andprevent it from being opened whennot desired, such as when the platform

lift is not present at the door. A numberof design features, including a door-closure retention cam, contribute to theCDI Series’ enhanced safety, reductionof nuisance stoppages, simplifiedwiring, and ease of installation. Reliableand smooth operation is enhanced bythe use of a metal key that is less sus-ceptible than plastic to bending andbreakage. The lack of open or exposedcontacts minimizes the possibility ofowners making manual adjustments.Featuring a custom internal solenoidcontrol, Relialign™ Commercial DoorInterlock Switches can reduce complex-ity of the host controller, trim powerconsumption, extend solenoid life, and

reduce solenoid “time outs,” reduc-ing customer aggravation. The CDISeries includes a manual override foreasy actuation without user hazardthat reduces the potential for call-backs. To meet customer and indus-try requirements, the Relialign™Series is compliant with ASME A17.1& 18.1, and cULus. Honeywell’s Relialign™ CommercialDoor Interlock Switches join an estab-lished product line that features tworesidential versions: a metal housingversion (RDI Series) and a plastic-molded housing version (RDI2 Series). HONEYWELLwww.honeywell.com

Honeywell Expands its Relialign™ Door Interlock Switch Portfolio with a Version for Commercial Applications

PRODUCT NEWS SENSORS

EP&Dee no 3 - March 2013

Documents

f1 evaluation platform

e c t r o n

s t e r n e u r o p

amicrochip f1 evaluation

f1 evaluationplatform

eurostandard press

euro standard press

f1evaluation platform