Elektor Industry Magazine February 2020 [PDF]

Zitiervorschau

www.elektormagazine.com • Edition 1/2020 • € 9.75

innovate

> launch > trade

Edition 1/2020

Special Edition

11 Ultra-fast real-time data exchange 3rd Generation Computer-on-Modules

191219-Cover-EN.indd 1

18 Open source ISA RISC-V What’s All the Hype About?

22 Technology Solution Provider Guidance through platform selection

03-02-20 15:39

EI-20.01-Adv-EN.indd 1

03-02-20 16:08

Preface

The market for embedded systems continues to show strong growth — the Internet of Things, autonomous vehicles and AI are just three developing applications fuelling this trend (see page 32). A visit to the Embedded World trade fair in Nuremberg really should be mandatory for all electronics professionals in 2020. As usual Elektor will be in amongst the fray; why not drop in to see us at our stand (Hall 4A-646)? You will be able to get a closer look at a range of popular controller projects published in our sister magazine Elektor and also checkout some of the latest boards and tools available from our online shop. With so many representatives from all areas of the industry in one place it’s an ideal opportunity to make contacts and network. My lab colleagues Clemens Valens and Mathias Claussen will also be on hand so while I’m there (by the way, from now on, I shall be acting as editor-in-chief of Elektor Industry) I plan to be out on the prowl to pick up the latest industry gossip and breaking news at this important trade fair. For sure I will be dropping in at the RISC-V stand where mem-

bers from the foundation will be touting the advantages of its open-source hardware instruction set architecture (see articles on pages 18 and 38 of this issue). Of course we can’t miss out Arm Holdings; it will be interesting to see how they respond to challenges in the 32 to 128-bit range. On top of that there will also be representatives from all the major electronics distri­ butors and semiconductor manufacturers. It’s a good chance to talk with engineers developing solutions in AI and to IoT security issues; you can always learn a lot chatting with people who are actually involved in solving problems. There are also lots of innovative solutions in so many other areas that really can’t afford to miss out. If you want to find out more about this now, checkout our cover story on page 11 which looks into the emerging COM-HPC specification designed to meet the increasing requirements of edge-computing applications. See you in Nuremberg! Jens Nickel 191220-01

Advertisement

USB Type-C . 24 reasons to connect with us. TM

USB Type-CTM Würth Elektronik offers the complete product range from connectors, components for filtering of EMI noise, ESD surge protection as well as AC/DC to DC/DC power conversion for USB 3.1. The specially designed products are featured in the Würth Elektronik USB Type-CTM EMC Dongle to easily and quickly validate the EMC and Surge compliance of your application. Our products are referenced in the USB Type-CTM reference designs of leading IC manufacturers.

GND RX1+ RX1- +V SBU2 D- D+ CC2 +V TX2- TX2+ GND

#USBTYPEC

© eiCan / eiSos

GND TX1+ TX1- +V CC1 D+ D- SBU1 +V RX2- RX2+ GND

For further information, please visit: www.we-online.com/typec App Note ANP007 – Effective USB 3.1 filtering and protection: www.we-online.com/anp007

n USB licensed and TID listed Type-CTM Connectors WR-COM n 60 W and 100 W Vbus filter n High efficiency molded power inductors WE-MAPI for Vbus filter

n Low parasitic capacitance WE-TVS diode n High speed, low noise data transmission filter embedded world Hall 3 Booth 247

with WE-CNSW HF

n Pulse stable SMT Chip Bead Ferrites WE-MPSB for hot-pluggin

Embedded & Tools

191220-Foreword-EN.indd 3

3

03-02-20 15:40

Contents Edition 1/2020

Focus:

Embedded Technology, Microcontrollers & Tools

Regulars 32 Infographics

Hot electronics technologies and trends to consider before and after the Embedded World 2020 trade show.

66 Index of Advertisers | Our Contributors | Next Edition

11 Whispers

about the Computer-on-Module standard COM-HPC The third performance cycle for COMs is coming

Articles

22

6 “Zero Energy“ Indicators 8 SIGLENT SDS2000X Plus Series Super Phosphor Oscilloscope

Technology Solution Provider

9 German Start-Up Introduces 3D Fabrication to Electronics 10 Incooling Collaborates with GIGABYTE to Develop a New Class of Two-Phase Liquid Cooled Servers

Embedded Platforms

11 Whispers about the Computeron-Module standard COM-HPC

The third performance cycle for COMs is coming

16 EMI Debugging with an Oscilloscope 18 Open Source ISA RISC-V What’s All the Hype About?

21 Honeywell launches New Media Isolated Pressure (MIP) Sensor Platform

28 Low-Voltage Motor Control System Design Considerations For Stepper Motors

34 Microchip Simplifies HardwareBased IoT Security

22 Technology Solution Provider

36 Miniaturized Wireless Medical Wearables

25 Review: LimeSDR Mini

38 RISC-V: Questions, Answers, ... and Embedded World 2020

– Embedded Platforms

4

44 BeagleBoard.org introduces the PocketBeagle® Grove Kit 45 Expert Paper: Tips and Tools for PCB Designers 46 Interference Sources in Automotive Applications 50 Ultra-Compact AutomotiveGrade MOSFETs

Provide Superior Mounting Reliability

Elektor Industry 1/2020

191221-Contents-EN.indd 4

03-02-20 15:41

Colophon Elektor Industry Edition 1/2020 embedded world Special Edition

Printers Pijper Media, Stettinweg 15, 9723 HD Groningen

© 2020

18

Open Source ISA RISC-V What’s All the Hype About?

54 Long-range Wireless STM32WL Microcontrollers 55 PCAN-MicroMod FD Series Robust I/O Devices with CAN FD

Reader Notices Elektor Industry contains www.elektormagazine.com contributed/sponsored content. The Publishers acknowledge Elektor Industry, English edition, all trademarks that may exist is published four times a year by in connection with products, Elektor International Media services, materials and company PO Box 11 names that appear in this NL-6114-ZG Susteren publication. The Netherlands The views expressed in Elektor Phone: +31 46 4389444 Industry are not necessarily Fax: +31 46 4370161 those of the Publisher or the Editor. Editor-in-Chief Jens Nickel Copyright Notice Email: [email protected] The content of this magazine is for educational use only. All Deputy Editors drawings, photographs, printed Eric Bogers, Jan Buiting, circuit board layouts, and Raoul Morreau, Tanja Pohlen, article texts published in this Robert van der Zwan magazine (other than third-party advertisements) are copyright Advertising Elektor International Media b.v. Margriet Debeij and may not be reproduced Tel. +49 241 955 09-174 or transmitted in any form Email: margriet.debeij@ or by any means, including elektor.com photocopying, scanning and recording, in whole or in part Layout without prior written permission Jack Jamar | Graphic Design, from the Publisher. Such written Maastricht permission must also be obtained before any part of this publication Publisher is stored in a retrieval system of Don Akkermans any nature. Mission Elektor Industry offers to electronics engineers, innovators, and start-up companies, essential information and insights into the latest products, research and intelligence from the industry. Distribution and Supply Elektor Industry appears synchronously in English and German and is supplied free to selected Elektor Magazine Gold members in print, and to Elektor Green members as a pdf download. The magazine is also on free distribution at selected trade fairs including Productronica (Munich), electronica (Munich), and embedded world (Nuremberg).

Patent protection may exist in respect of circuits, devices, components etc. described in this magazine. The Publisher does not accept responsibility for failing to identify such patent(s) or other protection. The Publisher disclaims any responsibility for the safe and proper function of reader-assembled projects based upon or from schematics, descriptions or information published in or in relation with Elektor Industry.

© Elektor International Media b.v. 2020 www.elektormagazine.com Printed in the Netherlands

56 Research Project “KI-FLEX” 58 Has the Hydrogen Car Lost Out on the Electric Car or Will There Be a Comeback?

Embedded & Tools

191221-Contents-EN.indd 5

5

03-02-20 15:41

“Zero Energy“ Indicators Provided by SCHURTER AG

Indicators for the operating status of electrical

3D Polymer Liquid (iStockphoto)

equipment generate heat during operation. Wasted energy, which also promotes the ageing of surrounding components. But that does not have to be the case. A new technology reduces the power loss by up to a factor of 20. Status displays, often called indicators, for electronic components are helpful wherever they assume safety-relevant functions. In an electronic circuit, for example, it may be of interest to know whether a fuse is still functioning. A rapid visual status display facilitates testing. In the past, glow lamps were used as an indicator for this. However, due to stricter requirements to reduce the standby current, such indicators have not been allowed to be used for some time. Up to now, indicators based on LED technology have taken their place.

Applications for indicators There are plenty of possible areas of application for such indicators. In principle, any application can be considered that requires minimal downtime: safety-critical applications as well as particularly cost intensive ones. Example medical technology: Permanently installed medical devices have their own fuses. And these fuses can be equipped with an indication. Fault detection is possible immediately. In the event of a fault in a device, this prevents the circuit breaker from responding and disconnecting other - possibly life-supporting - devices

6

from the power supply. Another example is laboratory equipment of various kinds in industry, research and development. In limit load operation, the fuse protects the device. If the fuse interrupts the circuit in such a case, the cause can be identified immediately by the indicator.

LED solution The LED solution has proven itself over many years. The technology is under control. A real disadvantage of the LED solution is its basic nature. If the diode is to light up, it needs electrical power. It emits light, but more than half of this power is wasted as heat. Heat loss is generally unwanted in electronic circuits. It also causes components to age faster than necessary, which shortens their service life.

Intelligent glass For some years now, developers have been able to use „intelligent glass“ (smart glass). What all available variants have in common is that their transparency or opacity varies depending on an electrical 2 White Paper Circuit Protection schurter.com/

Elektor Industry 1/2020

191041-Schurter-Zero Energy-EN.indd 6

03-02-20 15:42

data/download/WP_ZPI Circuit Protection voltage, changing lighting conditions or temperature fluctuations. The process is always reversible. This effect is best known for spectacles that become darker with increasing brightness and thus protect against sunlight - called thermochromism.

About SCHURTER SCHURTER continues to be a progressive innovator and manufacturer of electronic and electrical components worldwide. Our products ensure safe and clean supply of power, while making equipment easy to use.

Solution with PDLC glass However, a different principle is used for new indicators. PDLC glass (Polymer Dispersed Liquid Crystal) becomes transparent by applying an electrical voltage. PDLC glasses are based on a polymer liquid crystal film embedded between two flat glass panes. This is connected to a power source. The randomly oriented liquid crystal molecules are located inside the solid polymer. Incident light is scattered by these molecules and the pane is opaque, so it does not appear transparent to the human eye, but rather like frosted glass. When an electric voltage is applied, the liquid crystal molecules in the electric field rearrange themselves and the glass appears transparent. If the voltage is removed again, the liquid crystal molecules are disordered again and the pane becomes opaque again.

The use of PDLC glass instead of LEDs as an indicator offers a wealth of advantages. On the one hand, the power dissipation caused by the component can be reduced by a factor of 10 to 20. Electrical insulation (capacitive) is another advantage. Thanks to this, the technology is suitable for applications with increased safety requirements. The necessary space requirement for the indicator is also reduced, which is very convenient for use, for example, for fuseholders (Fuseholder [1], Fuseholder Blocks & Clips [2] ) on densely populated printed circuit boards. PDLCs are just like LEDs significantly less susceptible to vibrations or temperature fluctuations, which opens up new areas of application (e.g. electro-mobility or IoT). Such PDLC status displays for visual operational inspections can be integrated into all conceivable safety-relevant equipment such as measuring and laboratory equipment, fire detectors, air conditioning systems and much more.

Advertisement

Advantages of the new technology

Competence and cooperation SCHURTER, with its more than 80 years of experience in the field of circuit protection, is known as a competent and innovative supplier. This new indicator technology was developed in close cooperation with the CSEM in Alpnach (Homepage CSEM [3] ). CSEM is a private, non-profit Swiss research and technology organization focused on creating value for a sustainable world.

191041-01

References / Document Downloads [1]: www.schurter.com/pg02 [2]: www.schurter.com/pg02b

Complete Electronic Solutions

Visit us! Hall 1 / Booth 239

- Complete solutions from idea to the delivery - Focus on safe power supply and easy operation - Worldwide presence with competence centres - Service over the entire product life cycle [email protected] | +49 7642 6820 schurter.com

[3]: www.csem.ch

Embedded & Tools

191041-Schurter-Zero Energy-EN.indd 7

7

03-02-20 15:42

SIGLENT SDS2000X Plus Series Super Phosphor Oscilloscope

Siglent has released a new 2000-class digital oscilloscope. The new SDS2000X Plus series includes four models: one 2-channel model with 100 MHz bandwidth (software upgradeable to 350 MHz) and three 4-channel models, with bandwidths of 100, 200 and 350 MHz. In addition, the 350-MHz models can be upgraded to 500 MHz (Max Bandwidth available on two independent channels).

The SDS2000X Plus inherited the superb user interface with the higher-performance series SDS5000X. This includes a 10-inch touch screen, external mouse and keyboard control, and the built-in webserver makes this series easy to use and reduce the learning curve to a minimum. Furthermore, the oscilloscope offers a 10-bit acquisition mode that uses oversampling to achieve higher resolution. Combined with the lowest vertical setting of 500 µV/div, the X Plus can root out the smallest signal details. The 4 channel models come with two 2-GS/s ADCs and 2x 200-Mpts memory modules. This enhanced memory depth ensures a high sample rate at larger time/div settings. This is very useful when analyzing high frequency content on slow changing signals by providing the detail you need, regardless of timebase settings. The sequence mode boosts the waveform capture rate up to 500,000 wfm/s. This helps to maximize waveform capture rate and avoid missing critical events. This reduces oscilloscope “dead time” by a factor of >4 times and enables engineers to find rare signal anomalies faster. The special mode also enables optimized use of the memory. The built-in 50-MHz function generator option together with the free Bode plot function delivers convenient and low cost frequency analysis without investing in any other instrumentation. If you are developing switch mode power supplies (SMPS), Bode plots are a convenient way to measure Phase and Gain margin of feedback loop systems and help to determine the stability of the design.

8

The X Plus series also features a Power Analysis Option that delivers automatic on-screen performance analysis of common power supply characteristics. Together with Siglent’s current and differential voltage probes, this option is a must-have for perfecting your designs. Standard functions also included: Serial bus trigger and decoding for I2C, UART, SPI, LIN, CAN. Options include I2S, CAN-FD, FlexRay and MIL-1553B decoding as well as maximum bandwidths to help the SDS2000X Plus grow with your test needs.

Elektor Industry 1/2020

200015-SDS2000X Plus-EN.indd 8

03-02-20 15:43

All in all, the SDS2000X Plus is a powerful oscilloscope that can be used flexibly in many applications. Optional extensions (bandwidth, functions) allow it to keep pace with increasing requirements and making it a secure investment. The starting price for the 2-channel version is €899 (net), the 4-channel models start at €1199 (net). 200015-01

SIGLENT TECHNOLOGIES started in 2002 with the development of their first oscilloscope. Now, the portfolio has rapidly expanded to cover many areas of general purpose test instrumentation, including oscilloscopes, signal and function generators, digital multimeters, lab power supplies, spectrum analyzers and RF-signal generators. Today SIGLENT TECHNOLOGIES is a global leader producing electronic test and measurement equipment that combines innovative features and functionality with a strong commitment to quality and performance. SIGLENT is ISO 9001:2000 and ISO 14001:2004 certified for its product quality and environmental management programs.

German Start-Up Introduces 3D Fabrication to Electronics So far electronic packaging is in the Big Player domain. The production of customized packages is only feasible at high output and demands high changeover costs. As a result, customers interested in prototyping and small to medium size orders are left with standard versions of devices, and customization remains unaffordable. The newly developed KONEKT technology offers a solution allowing even small numbers of packages to be produced in a cost efficient manner. Soon, medium and small sized companies can gain access to products such as highly adaptable and

Materials can be customized as well, offering adaptable properties to ensure safety from heat, physical stress and explosion protection amongst others.

high performance sensors and connected devices. KONEKT uses smart manufacturing for electronic packaging so the package can be manufactured as a service in ranges from prototype to medium-volume and scalable to high volume. The manufacturing of whole System-inPackages is also possible because KONEKT also offers a completely new toolbox for electrical engineers and designers. A higher degree of design freedom can be reached by 3D packages offering great potential for miniaturization.

High frequency interconnections for a higher data transmission as well as the possibility to put your electronics closer to the actual place of measurement in the case of sensor applications, may give you the decisive advantage for the next generation of products. The KONEKT Project is funded through the EXIST Program by the German Federal Ministry for Economic Affairs and Energy and the European Social Fund. Visit us at: www.micropack3d.de or meet us at the Embedded World Nuremberg 2020. 200006-01

Embedded & Tools

200015-SDS2000X Plus-EN.indd 9

9

03-02-20 15:43

Incooling Collaborates with

GIGABYTE to Develop a New Class of Two-Phase Liquid Cooled Servers

As the requirement for greater computing power in applications such as high performance computing, artificial intelligence, scientific simulation and financial trading is growing, air cooling of these systems is becoming ever more problematic, consuming more energy and causing greater damage to the environment. Since liquid is a much more efficient method of cooling than air, GIGABYTE is investing more research and development into liquid cooling technology, and co-operating with a variety of industry partners to offer servers equipped with different types of liquid cooling solutions. GIGABYTE is now pleased to announce another new partnership with cooling technology provider Incooling, who has adopted GIGABYTE’s overclockable R161 Series server platform as the test-bed and prototype model for a new class of two-phase liquid cooled overclockable servers designed for the high frequency trading market.

True innovations only come together with truly innovative partners “We are very excited to be able to collaborate with GIGABYTE because we know that true innovations only come together with truly innovative partners. GIGABYTE not only makes best in class products but are also able to push the boundaries on multiple technology frontiers. These capabilities come together in their R161 series server, sporting both a highly modular and flexible system design for easy customization as well as excellent power delivery and custom overclocking BIOS capabilities that allow us to fully exploit the potential of phase change cooling.” said Rudie Verweij, CEO at Incooling. “We are always looking into new cooling technologies that help our customers push the boundaries of performance and efficiency in their data centres. Incooling offers a unique

10

solution, different from others in the marketplace, and we are looking forward to working with them to see what new possibilities we can achieve using our servers together with their cutting edge two-phase cooling technology” said Mike Chang, Thermal R&D Manager at GIGABYTE.

Turbocharged cooling Incooling’s technology is capable of pushing temperatures far below the traditional data center air temperatures, unlocking a new class of turbocharged servers. It does this by leveraging specialized refrigerants using phase-change cooling, inside a pressure-controlled loop. This is a highly efficient method that allows exchange of the heat from the chip with the datacentre air with far less thermal resistance.  Coupled with the R161 overclockable server from GIGABYTE, Incooling’s solutions are able to push performance further than ever before. First system tests showed up to 20°C lower core temperatures contributing up to 10% increase in boost clock-speed whilst lowering total power draw by 200 watts. Continuous R&D efforts will focus on further pushing these boundaries but it’s not all just about raw performance increase. In the near future, Incooling is planning to adopt their cooling technology for other types of GIGABYTE server systems such as its H-Series multi-node servers for high performance computing and G-Series GPU servers for artificial intelligence. This will further contribute towards reducing the currently excessively high energy consumption in servers that are using the highest powered CPUs or GPUs. 200026-01

Elektor Industry 1/2020

200026-News-Incooling-EN.indd 10

03-02-20 15:44

Whispers about the Computer-on-Module standard COM-HPC

The third performance cycle for COMs is coming By Zelkjo Loncaric, Marketing Engineer at congatec

Figure 1: ETX and the COM Express standard that has evolved from it, are two Computer-on-Module specifications that have been standardized by independent bodies in line with technological progress. With COM-HPC, a third standard is now being launched to meet the high performance requirements of broadband and 5G connected devices, machines and systems.

The Computer-on-Module industry is about to launch a new product lifecycle that will create a new performance class after ETX/XTX and COM Express. Called COM-HPC, the new form factor standard is currently primed for PICMG certification by companies such as congatec. Ultra-fast real-time data exchange will be a major application field due to 5G network rollout. But users of preceding COM standards needn’t worry about the availability of their solutions, because every transition takes time and products based on existing standards will be on the market for many years to come.

Embedded & Tools

200021-Whispers COM-HPC-EN.indd 11

11

03-02-20 15:44

multiple performance variants. Another important argument was to reduce the complexity of I/O board design. As a rule, I/Os require significantly fewer layers, which lowers the cost of PCB design. How to reduce power consumption and the amount of heat waste with each new module already played a role back in those days as well. And ultimately, customers always want the latest CPU — that was true then just as much as it is today. This advantage can also be easily ensured with modules.

Modules also solved cable clutter

Figure 2: The first ModulAT modules from JUMPtec were based on the then common AT/ISA96 bus and equipped with a 9.54-MHz Intel CPU 80C88 and 640 kByte DRAM memory. Type 6/7 today. From the outset, two footprints are specified for each class which means that the larger high-end COM-HPC modules will probably be able to host up to eight DIMM sockets. The graphic also clearly shows the identical placement of the inter-board connectors for the different sizes.

Computer-on-Modules (COMs) have established themselves as the most important design principle for embedded computer systems over the many years that the COM concept has existed. Studies such as those by IHS Markit predict that COMs will account for around 38% of the total sales of embedded computing boards, modules, and systems by 2020 [1]. The first modules appeared in the early 1990s, when Hans Mühlbauer — then owner of German company JUMPtec and still involved in the activities of congatec AG today — introduced the first ModulAT modules based on the then common AT/ISA96 bus. They were equipped with 9.54 MHz Intel CPU 80C88 and offered 640 kByte DRAM memory (Figure 2). The aim was to make office PC technology suitable for industrial use. That was something the embedded computer world hadn’t seen before. At the time, industrial PCs predominantly came as 19-inch rack systems. An industrial-grade computer on a board measuring only 100 x 160 mm was unheard of. The module had 120 pins on the same side as the CPU and the components; processors obviously didn’t require complex thermal management back then.   The goal of these first modules was to avoid putting all functions on one card in order to cushion the fast CPU innovation cycles. Back then, Intel and AMD launched new CPUs every six months. As it wasn’t certain how long old CPUs would be available, it was necessary to ensure the required long-term availability via modules. Of course, this scalability also helped to create

12

However, compared to the related AT/ISA96 bus, the ModulAT modules came rather late to the market since embedded computing for harsh industrial applications was still in its infancy. For example, x86 and Windows had yet to be established in the industry, and the fight against the blue screen was still ongoing at the time. In this respect, the modules were more like early ‘pirate’ products of a new start-up industry than the distinct product lifecycle of a whole generation of an established module standard. Nonetheless, JUMPtec was the company that opened the specification and pioneered the global module business. And it proved to be a success, as developments until today show. The PC/104 SBC form factor, which offered too little space for connectors if they had to be mounted on the same side as the CPU and chipset, provided another sound argument for modules in the mid-1990s. With customers starting to demand more connectivity, connectors were custom mounted on the other side of the PCB to connect even more peripherals. The PC/104 design principle also meant that cables had to be used to route the I/Os to the housing. This led to bigger and bigger cable tangles, which in turn increased system error susceptibility. At that time, good system design was characterized by clean and tidy cabling. Containing cable clutter by connecting the external I/Os to the housing without cables via an application-specific carrier board was therefore an important argument in favour of the module concept. The launch of the first ETX modules developed by JUMPtec marked the breakthrough for the Computer-on-Module market.

Hard contest for best module concept However, to start with, the ETX form factor for these ISA/ PCI-based modules with 400-pin connectors to the carrier board did not have an easy ride – despite the fact that JUMPtec had opened the specification. Many companies and competing module concepts, which are remembered only by die-hard insiders today, courted OEMs with comparable solutions. There was fierce competition between embedded computing vendors, who were not as big back then as they are today. JUMPtec and Advantech joined forces in November 2001 to found the ETX Industrial Group (ETX-IG), which introduced the first open, manufacturer-independent module standard and maintained a version that is still valid today. “Advantech, I-Base, IBR and PCISystems, for example, developed alternative ETX boards and rapidly brought them to market maturity. To ensure the uniform worldwide development of the ETX standard, it was therefore necessary to establish an open ETX consortium,” explained Mühlbauer at the time (Figure 3). Within a few months, the most important other ETX proponents also joined the consortium, as they all felt the great benefits of an open standard. Later on, the module forest of alternative form factors significantly thinned out due to the growing importance of ETX-IG as well as mergers & acquisitions. Ultimately, it was almost

Elektor Industry 1/2020

200021-Whispers COM-HPC-EN.indd 12

03-02-20 15:44

completely cleared, so that the next technology cycle was able to establish itself as the new standard of the embedded Computer-on-Module industry with significantly less sabre rattling.

COM Express becomes official PICMG standard in 2005 The widespread introduction of the new PCI Express bus and the elimination of ISA support in the new processors and chipsets required a new concept in 2004: COM Express, which was comparatively easier, although not entirely painless to establish (Figure 4). Some trench warfare and delay tactics had to be fought within the PICMG, which was to host this standard. Ultimately, the embedded community managed to agree on the COM Express standard within the PICMG, but not before July 2005. So from the first presentation of the concept in collaboration with Intel in autumn 2003, it took a good 18 months until standardization. From Rev 2.0 in 2010 up to the current Rev 3.0 from 2017, the specification was developed under the continuous leadership of draft editor Christian Eder, who initially worked under Mühlbauer at JUMPtec, later at Kontron and today is working at congatec.   Today, a full 14 years after the PICMG launched COM Express, the Computer-on-Module market is, as previously mentioned, the largest and most dominant embedded computing submarket, with all major embedded computer manufacturers offering a wide variety of COM Express modules. Having said that, ETX/XTX modules are still offered today, which means that the first COM cycle has not yet ended. And it took several years, until 2012 to be precise, for COM Express to outperform the ETX/XTX form factor in terms of quantities. ‘Never change a winning team’ is a motto that’s not only popular in the embedded computing market.

Figure 3: The foundation of ETX-IG at the SPS/IPC/Drives in Nuremberg by Advantech and JUMPtec.

COM Express modules are unrivalled Today, COM Express modules are also the undisputed standard for new embedded carrier board designs in the mid to high-end performance class. And there’s not a competitor in sight. The standard, incidentally, went through various rather silent revisions up to the current version 3.0, which was published in May 2017; this certainly also counts as a merit of standards. In its relatively new Type 7 specification (Figure 5), COM Express is predestined for embedded edge servers, and for extreme applications it is even used as a basis for VITA specifications. Alternative module standards such as Qseven and SMARC 2.0, which both also support ARM-based application processors, have established themselves only for the low-power/ small form factor segment. Lessons from the past were learned for the new COM-HPC module standard: right from the outset, this standard for high-performance embedded computing was developed within the PICMG in order to avoid module battles as far as possible. There is no better fitting ecosystem for this third generation of module standards than this manufacturer-independent consortium.   The third generation of high-performance COMs is coming Since October 2018, the PICMG working group has been working under the chairmanship of congatec’s Christian Eder on the new COM-HPC module specification, which is urgently needed because the COM Express inter-board connector can no longer support the upcoming high-bandwidth, high-frequency commu-

Figure 4: The first COM Express modules were launched into series production in 2005 with the Intel Pentium M processor, which was a real milestone processor for the embedded computing market.

nication buses required for all types of new IoT/5G-connected devices. This means the global Computer-on-Module specifications have been conceived and promoted for decades in the same German COM forges. Currently, the driving force is congatec AG, the company that started in 2005 as a pure module company to avoid competing against its own customers with system solutions. It is also the idea generator behind Qseven and SMARC 2.0.

Broadband Internet needs broadband computers As with the switch from ETX to COM Express, the introduction of new bus technologies is leading to a new standard. The aim of the COM-HPC specification is to create a new Computer-on-Module standard for broadband computing over broadband Internet that is suitable for the new high-frequency signals

Embedded & Tools

200021-Whispers COM-HPC-EN.indd 13

13

03-02-20 15:45

Figure 5: The most powerful COM Express Type-7 Server-on-Module to date is equipped with the AMD EPYC processor.

needed and must be thrown overboard. That’s because the new standard targets applications with requirements well above the current top COM Express performance class. However, the overall aim is to offer OEMs the benefits of the large ecosystem and reputation of the PICMG and its standards, which is why there is an emphasis on making migration as easy as possible for customers. Ample experience from migrations from ETX to COM Express should be available.   For these reasons, there will be two more or less entirely new next-gen performance classes above the COM Express specifications Type 7 and Type 6. One is aimed at edge server technology, which requires more communication interfaces instead of integrated, powerful graphics, and will bring an extremely large number of cores for workload consolidation. The other expands the existing high-end embedded computing with new performance options that COM Express can no longer cover, which also comprises graphics. The list includes USB 3.2 with 20 Gbit/s, USB 4.0 with 40 Gbit/s, PCIe Gen4/5 with x2/x4 port configuration and re-timer, 100/200 Gbit/s Ethernet, NVMe, and much more.

Twice as many pins and up to eight DIMM sockets from PCI Express Gen 3 to 5. However, COM-HPC should not be seen as a replacement for COM Express, just as COM Express was not the replacement of ETX. As mentioned earlier, ETX/ XTX modules are still available today, allowing customers to continue to work on the basis of the same design principles even after 20 years. In this respect, the upcoming new COM-HPC module standard is also proof that the basic concepts that were valid then are still valid today. In addition, the tasks associated with the design-in of new processors have become much more complex, which is why it makes even more sense today to decouple the I/Os from the processor module via an application-specific carrier board.   But why not simply develop COM Express further? The new COM-HPC standard doesn’t just require a new connector; there are more legacy COM Express features that are no longer

An essential part of the new specification is the connector. COM Express, for instance, is limited to PCIe Gen 3.0 with 5.0-GHz clock rate and 8 Gbit/s. The new connector supports transfer rates of more than 32 Gbit/s, which is adequate up to PCIe Gen 5.0. In addition, there are now up to 64 PCIe lanes to the carrier board – enough to connect many powerful GPGPUs for machine learning, for example. COM Express, by comparison, supports a maximum of 32 lanes. The COM Express performance, currently limited to 10 Gb Ethernet per signal pair, will also increase to at least 25 Gb Ethernet per signal pair, enabling support for up to 100 Gb Ethernet. The new processor generations for edge computing also require more interconnects than before, as well as more space for DIMM sockets than previously possible. Current plans allow for up to eight DIMM sockets and 800 pins to the carrier board; COM Express has only 440 pins (Figure 6).  

Figure 6: COM-HPC modules from congatec will be available in two performance classes: a client and a server class, as with COM Express Type 6/7 today. From the outset, two footprints are specified for each class which means that the larger high-end COM-HPC modules will probably be able to host up to eight DIMM sockets. The graphic also clearly shows the identical placement of the inter-board connectors for the different sizes.

14

Elektor Industry 1/2020

200021-Whispers COM-HPC-EN.indd 14

03-02-20 15:45

Developing a new standard is no easy matter, even if some might think so. The complexities arising just from increasing the signal frequency are already enormous. To give an example: congatec and Samtec have been working together for a full two years on COM-HPC connector requirements and tests to be able to support modules up to 300 W. The PICMG working group, however, was not set up until October 2018. This shows how much ground work was already done, speeding up the decision making process in the working group itself.

When will COM-HPC modules be available? The development departments are already working at full speed on first design studies based on the latest processor technology, which semiconductor manufacturers such as Intel are sharing via early access programs. The first production-ready COM-HPC modules from congatec will probably become available end 2020, when the next major embedded computing generations are expected to launch. This leaves about as much time as there was for COM Express. So it should be achievable.

PICMG — a strong team The PCI Industrial Computer Manufacturers Group, or PICMG for short, is a consortium of over 140 companies that jointly develop patent-free specifications for high-performance telecommunications and industrial applications. The members of the consortium are for the most part technology pioneers with many years of development experience in their industries. Currently, the group is developing the COM-HPC specification as an open architecture for the next generation of Computer-on-Modules. Companies involved in the development of the COM-HPC specification include Adlink, congatec and Kontron as sponsors of the working group, as well as Acromag, Advantech, AMI, Amphenol, Elma Electronic, Emerson Machine Automation Solutions, Ept, Fastwel, HEITEC, Intel, MEN, MSC Technologies, N.A.T., nVent, Samtec, Seco, TE Connectivity, Trenz Electronic, University Bielefeld, VersaLogic Corp. . The chairman is Christian Eder of congatec.   200021-01

Reference [1] http://embeddedtechtrends.com/2017/PDF_Presentations/ M06%20-%20IHS%20Markit.pdf

  Advertisement

Visit us at the EMBEDDED WORLD! We are pleased to present our new products.

NEW 2020!

Nuremberg Germany 25.-27. 2. 2020 (hall 4, booth 216)

Our worth is measurable.

Embedded & Tools

200021-Whispers COM-HPC-EN.indd 15

15

03-02-20 15:45

EMI Debugging with an Oscilloscope Shortens Power Electronics Development

By Marcus Sonst, Application Development Engineer at Rohde & Schwarz

Oscilloscopes are the workhorses for power electronics engineers. With powerful and easy- to-use fast Fourier transformation (FFT) analysis capabilities, their application fields extend to electromagnetic influence (EMI) debugging, saving a lot of time and money. Figure 1: EMI measurements during the design process can be done with an oscilloscope such as the R&S RTM3000 and without the need of a test chamber. Image: Rohde & Schwarz.

Conducted emission testing is a mandatory measurement at the end of each design process of a switching mode power supply (SMPS). The developer has to verify that the product is compliant with the applicable standard before it can be released on the

A typical task is verifying the effectiveness of an EMI input filter of a switching mode power supply – early in the development phase.

market. Full compliance testing requires a test chamber and a suitable EMI receiver. If the product falls outside the limits of the standard, it may be necessary to modify the power supply and that can greatly affect many parts of the SMPS, e.g. EMI input filter, PCB design, concept decisions like selection of a suitable switching frequency. This can significantly impact the time to market. It is often necessary to partially redesign the product. This risk can be drastically reduced by performing the conducted emission tests in an earlier stage of the development cycle. For precompliance testing, you do not necessarily need a test chamber, but you do need an instrument that is able to measure the spectrum of the power supply input and output lines in a comparable manner. This can be a spectrum analyzer but also an oscilloscope (Figure 1).

Interference spectrum analysis with an oscilloscope

Figure 2: It is sensible to do conducted emission measurements early in the development phase of a SMPS. This avoids unpleasant surprises during the final compliance testing. Image: Rohde & Schwarz.

16

Rohde & Schwarz oscilloscopes offer a powerful, easy-to- use FFT analysis functionality to measure the magnitude of the frequency component. Users are able to see the time domain related signals at the same time and can therefore correlate unwanted spectral emissions with time domain events. This

Elektor Industry 1/2020

200018-EMV-Fehlersuche-EN.indd 16

03-02-20 15:46

makes these oscilloscopes powerful stand-alone instruments for performing early conducted emission tests on power electronics designs. This is particularly helpful when no dedicated equipment such as an EMI receiver is available in the R&D lab to support the precompliance measurement during the design phase. The earlier EMI compliance is taken into account, the more likely it will not become an issue at the end of the development process. Uncovering an EMI problem early on is less costly and easier to fix. Since oscilloscopes are typically the main instrument used during hardware development and system testing, they are a valuable tool for EMI related tests in R&D (Figure 2).

ated by a factor of 10 dB, which has to be taken into account when comparing measured values with emission limits. The bottom window shows the spectrum in dBµV measured on the input terminal of the power supply. Without an EMI filter, the noise spectrum generated at the input of the DC/DC converter is clearly visible. In contrast, the measurement with an EMI filter (Figure 4) shows that the conducted emissions on the input line are effectively attenuated. For some frequencies, up to 30 dB attenuation is visible. To verify emissions in the lower frequency region, the user has to repeat the measurement with a focus on the lower frequencies.

Figure 3: EMI spectrum of the SMPS without EMI input filter. Image: Rohde & Schwarz.

Figure 4: EMI spectrum of the SMPS with EMI input filter: the conducted emissions on the input line are effectively attenuated especially at lower frequencies. The user has to repeat the measurement with a focus on the lower frequencies. Image: Rohde & Schwarz.

Test setup

Summary

To measure conducted emissions of a power supply, a line impedance stabilization network (LISN) is required to decouple the device under test from the external power supply. The coaxial output of the LISN has to be connected to the oscilloscope with a coaxial cable with 50-ohms input impedance activated at the oscilloscope to ensure proper matching. To measure the spectrum the user has to activate the FFT functionality of the oscilloscope. Then he configures the minimum and maximum frequency and the resolution bandwidth. Afterwards, he adjusts the vertical sensitivity in the time domain window. This shall avoid that the input channel is overdriven when the device under test (DUT) is powered. For a reference measurement the user has to switch off the power of the DUT. With this measurement he knows the noise floor of the setup, which has nothing to do with the DUT. Then he can switch on the DUT power again and make his measurements. He can verify the measured spectrum against known conducted emission limits for the DUT. But he should take into account any additional attenuation due to the LISN he uses.

The FFT functionality of Rohde & Schwarz oscilloscopes is a powerful feature that enables designers to debug conducted emissions of power supplies. Since the oscilloscope is a standard measurement instrument for power electronics design, it can save a lot of time and money by also using it to evaluate EMI early in the development phase. This makes it more likely that the product achieves EMI compliance at once without an essential redesign after failing a test.

Case study — effectiveness of the EMI filter For verifying the effectiveness of the EMI input filter of a SMPS a R&S RTM3000 oscilloscope was used to measure conducted emissions without and with an EMI filter. In Figure 3 and Figure 4, channel 1 displays the measured time domain signal connected to the LISN. Due to the LISN, this signal is attenu-

200018-01

The Author Marcus Sonst is an Application Development Engineer at Rohde & Schwarz in Munich and responsible for power management and power electronics tests. Marcus holds a Master degree in Electrical Engineering with focus on power and energy from the University of Applied Sciences in Dortmund (Germany). He has more than 20 years experience with development, design, and test of power supplies and electronic ballasts.

Embedded & Tools

200018-EMV-Fehlersuche-EN.indd 17

17

03-02-20 15:46

RISC-V:

What’s All the Hype About? By Mark Patrick, Mouser Electronics

New processor instruction set architectures (ISAs) really don’t come along very often. The open-source RISC-V ISA developed within the labs of the University of California at Berkeley has created a significant buzz in the embedded industry, however. The much talked-about RISC-V ISA was developed on the premise that any designer could use it to create processor cores and software compilers. The project is now run by the RISC-V Foundation, with members that include a wide range of universities and multinational technology companies (Google, IBM, Microsoft, NVIDIA and Oracle, most notably), as well as chip makers and startups.

The aim of RISC-V was to learn from the mistakes of other processor ISAs. The key is stability — for the instruction set and the core, plus chip designers, compiler makers, operating system architects and development tool vendors. This is essential for encouraging as many engineers as possible to use such open-source technology across the ecosystem, making powerful processor cores much more accessible and usable. Application developers can optimize their code for a frozen ISA with minimal memory footprint and power consumption, but still be scalable and compatible with future devices. This allows processor core developers to work on all kinds of different

implementations of the instruction set – from a simple pipeline to ones with multiple states and out-of-order execution. These will have different latencies, sizes and power consumption, but will all have an underlying compatibility with each other, and with the tools encompassed within the ecosystem.   Providing this stability across the whole ecosystem was a critical part of the new instruction set. This was designed with 32-bit, 64-bit and 128-bit address spaces in mind, so compatibility across them can be maintained. The architecture is also designed with extensions, to provide the customization that

Figure 1: The U500 64-bit multicore open-source processor from SiFive.

18

Elektor Industry 1/2020

200011-RISC-V Hype-EN.indd 18

03-02-20 15:46

chip makers need for differentiation and for future application scenarios, but the ISA’s foundations remain deliberately untouched.   The 128-bit ISA is still intentionally undefined, because as yet there is little practical experience with such large memory capacities in embedded systems. However, the fact that the architecture will support this larger address space highlights the forward-thinking approach that has been taken here. All this means software written for, or ported to, RISC-V will run on all similar RISC-V cores forever, giving software managers a solid foundation that preserves their software investments. As the ISA is open, multiple hardware implementations can be developed, so software architects can become more influential in the final hardware implementation.   The input to hardware designers makes the RISC-V core more software-centric. This has led to a wide range for processor cores implementing the ISA, and a number of system-on-chip (SoC) devices based around those cores. Cores have been developed by Codasip, Syntacore, Hex Five and T-Head, while SiFive has pushed forward with a range of 32-bit and 64-bit SoCs.   SiFive, which was co-founded by Yunsup Lee (one of the original creators of RISC-V), launched its first RISC-V core in 2017 as a family of SoC platforms, adding support around the cores and chips. The devices are being built on a 28-nm process for a 64-bit multicore Linux implementation, or on 180nm for the 32-bit low-cost IoT market with various peripherals (Figure 1).   The company’s Freedom platforms comprise a complete software specification, board support packages (BSPs) to bring up an operating system, development boards and base silicon, allowing customers to create their own silicon enhancements and customizations. The Freedom U500 series is a fully Linux-capable embedded application processor with multicore RISC-V CPUs, running at a speed of 1.6 GHz or higher with support for accelerators and cache coherency for machine learning, storage and networking. This supports standard high-speed peripherals including PCIe 3.0, USB 3.0, Gigabit Ethernet, and DDR3/DDR4.   The Freedom E300 series (Figure 2) is designed as an embedded microcontroller for the IoT and wearables markets. Based on the Freedom E310, the HiFive1 Arduino-compatible RISC-V development kit incorporates SiFive’s E31 CPU Coreplex — a high-performance, 32-bit RV32IMAC core that is capable of running at over 320MHz.   SiFive has also used the RISC-V instruction set (Figure 3) for what it says is the world’s smallest embedded processor core. The S2 core IP series is a configurable core that can be as small as 13,500 gates (in the case of the RV32E 32-bit version). The S21 64-bit embedded core has separate instruction and data buses, along with two banks of tightly integrated memory (TIM). This enables SoCs to have an always-on low-power 32-bit CPU that can be combined with a high-end 64-bit CPU that switches on when applications demand elevated performance (such as in voice-activated smart devices). Developments of this kind help address the growing need for connected devices with machine learning and IoT, where real-time workloads have

Figure 2: The E300 series of open-source 32-bit microcontrollers using the RISC-V ISA.

generated a massive demand for greatly enhanced embedded intelligence at the edge.   The open-source nature of RISC-V has opened up SoC design to startups such as Kendryte, efabless and low RISC, but more mainstream chip vendors are also using the technology. Microsemi (now part of Microchip) has produced some development boards for SiFive, while NXP has its own RISC-V chip. Andes Technology and Greenwave have also developed ICs around the ISA. Faraday Technology has used the ISA for an ASIC platform for the design and mass production of next-generation edge AI and IoT SoCs. It brings together the RISC-V core IP integration and SoC design verification, as well as a full-featured reference design kit consisting of real-time operating system (RTOS) and peripherals drivers, all on a 55nm process for battery-powered edge devices. This highlights how the hardware manufacturers can differentiate around a standard ISA. Faraday has included dynamic voltage and frequency scaling (DVFS), power mode switching and fast system wake-up in the platform, but can safely include the software libraries and drivers to ensure the chips work seamlessly for specific interfacing, sensing and power-management functions.

Tools Another strength of the RISC-V ISA is its ability to be used with a wide range of tools. Microsemi has used the ISA across its FPGAs with a range of embedded operating systems such as Express Logic’s ThreadX, Huawei LiteOS and Micrium µC/ OS-II. Boards include the RTG4 development kit, the PolarFire

Embedded & Tools

200011-RISC-V Hype-EN.indd 19

19

03-02-20 15:46

The Author Mark joined Mouser Electronics in July 2014 having previously held senior marketing roles at RS Components. Prior to RS, Mark spent 8 years at Texas Instruments in Applications Support and Technical Sales roles and holds a first class Honours Degree in Electronic Engineering from Coventry University.

evaluation kit, etc. Debug dongles from Microsemi and Olimex, first-stage bootloaders and multiple soft peripherals are also included. Examples of drivers, firmware and projects are available on GitHub. Another tool company benefiting from the stability is UltraSoC, which develops hardware that can be embedded in a SoC to monitor the activity. This can be used for debugging the chip more effectively and even used in the field for monitoring. It has been working with Andes on integrating the monitoring hardware into the high-end AndesCore processor IP, and into an AI “supercomputer-on-a-chip” from Esperanto Technologies that uses thousands of RISC-V cores. The Esperanto engineering team is led by RISC pioneer Dave Ditzel – who helped develop the SPARC RISC processor. Esperanto completed the RTL for a 10-stage high-performance core called ET-Maxion last September using the RV64GC 64-bit ISA, with samples expected in the second half of 2019 and production around mid-2020, as well as a smaller, more power-efficient 64-bit core called ET-Minion. This uses an in-order pipeline with multi-threaded instructions, and adds a vector floating point unit that supports extensions for graphics and Google’s Tensor AI instructions, all intended to be built on a leading 7nm process. These have been built with a single toolchain around the RISC-V ISA that includes the test and verification tools. The cores are combined in the supercomputer-on-a-chip, which has 16 Maxions and 4,096 Minions, running Linux and other high-level software on the Maxion cores while delegating AI-intensive workloads to the Minions. All this is made much easier by using the same ISA across all the devices, thereby highlighting the value of the RISC-V approach.  

ers can focus on the particular processor implementations, whether for research projects, IoT nodes or a supercomputer-on-a-chip. All these use the same compilers, the same development tools and the same debug tools, minimizing fragmentation and allowing companies to keep pushing performance benchmarks rather than worrying about maintenance of multiple software products for multiple cores — and all, of course, with an open-source ethos that allows improvements to be fed back into the industry. Extensions allow for differentiation and optimization, particularly with regard to security, without compromising the stability of the tool ecosystem. In this way, RISC-V allows hardware developers to focus on innovation, driven by the software needs to meet the cost, power, security and performance requirements of end users. 200011-01

RISC-V Challenges Though considerable headway has already been made in relation to RISC-V’s development and proliferation, there are obstacles ahead. Researchers at Princeton University have uncovered a number of flaws in the RISC-V open-source processor core that they believe to be significant. They found more than 100 errors involving incorrect ordering of the storage and retrieval of information from memory in variations of the RISC-V processor architecture that, if uncorrected, could cause issues in software running on RISC-V chips. The RISC-V Foundation said the errors would not affect most versions of RISC-V, but might have been more problematic for higher-performance systems.   With a common, frozen ISA across 32-bit, 64-bit and even 128-bit (when they emerge) address spaces, core develop-

20

Figure 3: The HiFive1 RISC-V development kit.

Elektor Industry 1/2020

200011-RISC-V Hype-EN.indd 20

03-02-20 15:47

sement

Honeywell launches New Media Isolated

Pressure (MIP) Sensor Platform For HVAC and Industrial Market

New multi-purpose sensors offer performance, durability for harsh applications

Honeywell announced the launch of a new, high-performance, value-priced line of Media Isolated stainless steel pressure sensors (MIP) for a wide range of media, including the industrial and HVAC market. The new MIP series will give customers a single, customizable solution for use in a wider range of harsh media. Media isolated pressure sensors play a key role in making process control systems more efficient by maintaining pressure throughout the system for effective air or fluid distribution. Designed to withstand extreme temperatures and resist corrosion, the sensors are deployed by engineers for designing high-performance HVAC systems, medical and biomedical processes, marine systems, aircraft systems, environmental engineering, laboratory, wastewater and other industrial applications. “Industrial and HVAC customers can now choose a single product series for the bulk of their industrial and HVACR sensing needs—without the premium price tag for similar products available in the market,” said Joseph Coakley, senior product manager. The MIP Series eliminates the leakage paths associated with O-rings and adhesives. The hermetic design combined with stainless steel construction makes MIP sensors compatible with a wider range of media including aggressive fluids and water. In addition the MIP series is NSF certified allowing it to be used in food, water, and commercial foodservice applications. The MIP series is a platform product based upon industry standard configurations. The new sensor has improved performance (Total Error Band, Accuracy, and Response time) and burst ratings over previous offerings and the platform concept allows for customizations for almost any application. The MIP series is a platform product based upon industry standard configurations. The new sensor has improved per-

formance (Total Error Band, Accuracy, and Response time) and burst ratings over previous offerings and the platform concept allows for customizations for almost any application. Within the highly price sensitive industrial sensing market customers will typically need to balance price vs. performance. Many portfolios today offer value-priced products with less robust material such as brass and will rely on O-rings and adhesives to seal the media from the sensors integrated circuitry. For customers requiring a more robust product like the MIP Series they will often pay a premium for that added value. Key Features and Value Lower system operating costs and improved system efficiency due to increased TEB Accuracy from ±1% to ±.75% • Elimination of leakage due to elastomeric O-rings and adhesives due to laser-welded hermetic joints • Increased safety factory with higher burst rating • Improvement in sensor and system reliability due to internal dampening mechanism • NSF certification reducing implementation costs for water and food service applications • Improved troubleshooting of electrical failure with built-in diagnostics • Additional reduction in overall operating costs through supply chain and SKU consolidation through the use of one product for all applications within the portfolio regardless of media type.

https://sensing.honeywell.com/ 200024-01

Embedded & Tools 21

200024-News-Honeywell-EN.indd 21

03-02-20 15:48

Technology Solution Provider – Embedded Platforms By By Amir Sherman, Director of Engineering Solutions & Embedded Technology EMEA, Arrow Electronics

Figure 1: The Thor96 Boards from Arrow Electronics.

Engineering teams are increasingly using specialist platforms to form the basis of their new product designs. This enables them to concentrate on perfecting their own application rather than spending time on the more fundamental elements of the technology. A Technology Solution Provider, such as Arrow Electronics, is able to provide support and guidance through the platform selection process, helping to ensure that the chosen solution meets the needs of the customer’s application and specification. To see an example of this, let’s look at the popular i.MX 8 Family from NXP. Built with advanced media processing, secure domain partitioning and innovative vision processing, the i.MX 8 applications processor family is revolutionizing multiple display automotive applications, industrial systems, vision, HMI and singleboard computers. Many users want to develop a solution based on this family but it is not a simple

22

process to get to production due to the technical competence needed to develop with multi-core , Linux based and DDR3L or DDR4 high end technology . To support the wide variety of i.MX 8 users, Arrow have developed a platform based on a Thor96 community board. The Thor96 is a single-board computer powered by the NXP i.MX 8M SoC, incorporating a quad-core 64-bit Arm-A53, dedicated GPU and VPU, 4K support,

Wi-Fi, Bluetooth and a wide range of I/O. 96Boards is a 32-bit and 64-bit Arm® Open Platform hosted by Linaro™ with the intension to serve the software/ maker and embedded OEM communities. Some customers will be able to develop their own board using this community board schematic and progress this design by using Arrow FAE’s for technical support. They can also customize their own design by using Arrow’s eIn-

Elektor Industry 1/2020

191228-Embedded Platforms-EN.indd 22

03-02-20 15:47

fochips engineering services. This is the first option with the platform. More information can be found at www.96boards. org and on arrow.com. The second option, for customers who don’t have the time to develop their own board but still want to use the i.MX 8, is to use a SoM – System On Module. The System on Module (or CoM – Computer On Module) is designed to plug into a carrier, or base board, and is generally a small processor module with a CPU and standard I/O capability. The complex effort associated with designing a CPU subsystem is avoided by using SoM and a custom base board. Arrow work with many SoM suppliers to support a wide variety of form factors and solutions. This includes companies like Variscite and iWave Systems, which are part of Arrow’s Embedded World 2020 Supplier showcase. The Variscite SoM has a proprietary form factor while iWave supports the SMARC community form factor. Using a proprietary form factor is popular because users can find the best fit for their applications dependent on size, connectors, layout and many other features. However, some customers want to use a standard form factor and the SMARC form factor is popular among Arm based SoMs. SMARC (Smart Mobility ARChitecture) is a versatile small form factor computer module definition targeting applications that require low power, low costs, and high performance. The modules typically use Arm SOCs similar or the same as those used in many familiar devices such as tablet computers and smart phones. Alternative low power SOCs and CPUs, such as tablet oriented X86 devices and other RISC CPUs, may be used as well. The module power envelope is typically under 6W. Two SMARC module sizes are defined: 82 x 50 mm and 82 x 80 mm. The module PCBs have 314 edge fingers that mate with a low profile 314 pin 0.5-mm pitch right-angle connector (the connector is sometimes identified as a 321-pin connector, but 7 pins are lost to the key). The modules are used as building blocks for portable and static embedded systems. The core CPU and support circuits, including DRAM, boot flash, power sequencing, CPU power supplies, GBE and a single channel LVDS display transmitter are concentrated on the module.

The modules are used with application specific carrier boards that implement other features such as audio CODECs, touch controllers, and wireless devices. The modular approach allows scalability, fast time to market and upgradability while still maintaining low costs, low power and small physical size. The new global standard under the brand name ‘SMARC’ is based on ULP-COM, the term that up to now was used for Ultra Low Power Computer-on-Modules.

the i.MX 8 family. With DART-MX8-MINI and VAR-SOM-MX8M-MINI based on the i.MX 8 Mini family that have 1.5 GHz Dual/Quad Cortex™-A53 and 266 MHz Arm Cortex™-M4 to DART-MX8M and VAR-SOM-MX8 based on i.MX 8 with 2 x 1.8 GHz Cortex™-A72 + 4 x 1.2 GHz Cortex™-A53 & 2x Arm Cortex™-M4F up to the latest i.MX 8X family SoM called VAR-SOM-MX8X with 4 x 1.2 GHz Cortex™-A35 and 264 MHz Arm Cortex™-M4F and the last SPEAR-MX8

Figure 2: Variscite i.MX 8 Mini System On Module.

Variscite Variscite has developed, produced and manufactured a powerful range of System on Modules, consistently setting market benchmarks in terms of speed and innovation. All Variscite production is performed at fully ISO 13485, 9001 and 14001 compliant facilities, satisfying international customer and regulatory requirements for a broad range of industries including medical devices and related services. The company’s production facilities are equipped with the most advanced SMT machines that ensure punctual deliveries and highquality products. It’s no surprise that in less than a decade Variscite has taken a leading position in the design and manufacture of System on Modules. Variscite serves more than 1500 customers in over 50 countries worldwide, delivering a cost-sensitive high-performance portfolio that combines interface flexibility with advanced power management. Variscite has six SoM Families that can be customized on demand based on

based on the i.MX 8 QuadMax / Quad Plus with 2 x 1.8GHz Cortex™-A72 + 4 x 1.2 GHz Cortex™-A53 and 2 x 266 MHz Cortex™-M4F.

i-Wave iWave Systems Technologies, an ISO 9001:2015 certified company, established in 1999, focuses on standard and customized System on Module/ SBC product development in industrial, medical, automotive and embedded computing application domains. iWave Systems also provides comprehensive engineering design services involving embedded hardware, FPGA and software development. The i.MX 8M Quad, Quad Lite, Dual based SMARC System On Module integrates Quad/Dual Cortex A53 at up to1.5 GHz, H.265 4K60 decode, GC7000 Lite GPU, MIPI CSI/DSI, HDMI2.0 TX, USB3.0, PCIe2.0 with on SOM 10/100/1000 Mbps Ethernet PHY and IEEE 802.11 a/b/g/n/ ac Wi-Fi and Bluetooth 5.0 module. The i.MX 8M SMARC SoM is aimed to

Embedded & Tools

191228-Embedded Platforms-EN.indd 23

23

03-02-20 15:47

resulted in Arrow maintaining its position as a true Technology Solution Provider as technology itself continues to evolve. In 2020 Arrow will continue to assist customers who want to develop with technology at the leading edge. This will include more support for services, cloud solutions and security, and will ensure that time-to-market is always minimized, whichever solution platform is chosen. 191228-02

Figure 3: iWave SMARC based i.MX 8 System On Module.

support applications such as digital media adaptors, HD digital signage, industrial HMI, building automation, imaging & scanning, audio/video streaming devices, and machine vision. The third possible platform option is for users to purchase a complete working solution. Many require a ready-to-use Proof of Concept (PoC) but do not have the time and the money to develop this complex solution. A number of Arrow’s suppliers can offer a ready-made solution based on the latest technologies including the NXP i.MX 8. One example is SolidRun, a global developer of embedded systems and network solutions focused on a wide range of energy-efficient, powerful and flexible products. Their innovative compact embedded solutions are based on Arm and x86 architectures, and offer a variety of platforms including SoMs, SBCs and industrial mini PCs. SolidRun offer a one-stop-shop for developers and OEMs, providing a complete service from hardware customization to software support and even product branding and enclosure design. Their CuBox Pulse solution is claimed to be the smallest computer in the world. The CuBox Pulse is silent and tiny — at just 2″ x 2″ x 2″ — based on NXP i.MX 8M Dual/Quad core Arm Cortex A53 up to 1.5 GHz (with Arm M4 GPP). CuBox Pulse is the perfect home entertainment device with the highest standard of visual resolution support, including 4 K UltraHD

24

at 60 Hz and full HDR. The CuBox Pulse makes it easy to turn any screen into a smart TV or online streamer. This tiny multimedia powerhouse is also the perfect device for digital signage and a wide variety of audio and visually rich Internet of Things solutions.

Figure 4: SolidRun CuBoX Pulse, i.MX 8M ready-made solution.

True technology solution provider The nature of product development and the requirements of companies have changed dramatically in recent years and Arrow Electronics has adapted to ensure that it is always ready to deliver what its customers need next. This has

Elektor Industry 1/2020

191228-Embedded Platforms-EN.indd 24

03-02-20 15:48

Review:

LimeSDR Mini

LimeSDR Mini – Functions Maxi! A pocket-sized SDR for transmitting and receiving By Mathias Claussen (Elektor Labs)

A first test run with the LimeSDR Mini During experiments and developments you might wish to occasionally see the spectrum produced by your own circuit, or whether the built-in receiver could receive a signal at the location. It is also nice to have the right tool at hand when it comes to eliminating interference or analysing it. For this purpose many people use an RTL-SDR, which is versatile and cheap. This solution works in the frequency range between 24 MHz and 1700 MHz, depending on the tuner used, and is a flexible but somewhat “deaf” solution for your own experiments. Also the bandwidth, or here the sampling rate, is usually 2.4 MS/s and allows viewing a maximum of 2.4 MHz in the spectrum. Beside the “deafness” we can only receive signals; generating test signals is not possible although that’s sometimes desirable to test your own developments in a defined way.

Out of the box

The hardware

further accessories. There is, however, an optional version with case, which was also made available to us for the first test. The case not only protects the electronics, but also dissipates the heat of the individual ICs better. For actual use, it is advisable to install the stick in a housing, and do not forget to provide sufficient ventilation. Proper cooling of the LimeSDR Mini is recommended, considering that oscillators may drift from their set frequency. The heart of the SDR is the LMS7002M transceiver [1], which is equipped with 12-bit DA/AD converters for signal processing. The chip itself should cover a frequency range from 100 kHz to 3.8 GHz, which is however limited to 10 MHz to 3.5 GHz, as well as in bandwidth from 61.44 MHz to an actual 30.72 MHz. Being a transceiver, the LimeSDR Mini can transmit as well as receive signals in “broadband” mode. Up to 10 dBm of transmit power is available at the output.

The LimeSDR Mini is designed in stick form for the USB 3.0 port and is delivered bare (i.e. without housing) and without

To be able to connect your own antennas, there is one SMA connector each, clearly

Compared to the RTL-SDR, the LimeSDR Mini is a different category of SDR and definitely not a toy. On the one hand the price of the LimeSDR Mini is higher than that of a cheap RTL-SDR, on the other hand it’s a true transceiver for SDR, so besides receiving you can also use it to transmit. When operating, the regulations of the country in which the SDR is used must be considered. In case of doubt it can be helpful to turn to a friendly radio amateur from the neighbourhood.

marked ’RX’ or ’TX’. The possibility to connect multiple antennas for transmitting and receiving as with the big brother, the LimeSDR, was not implemented here to reduce hardware costs. Beside the transceiver and the USB bridge there is a MAX10 FPGA in the centre, which controls the communication between the transceiver and the USB bridge. The bitstream required for the FPGA can be generated with the free version of Intel Quartus and loaded into the FPGA with the appropriate tool. Using the data sheets and the circuit diagrams the FPGA can be supplied with a new bit stream to adapt the hardware to your own needs, as you can see from the work of Gaspar Karm [2]. But when we now generate a signal, we depend on the quality of the reference clock for frequency stability. On the board there is a Rakon 40 MHz VCTCXO with +/-1 ppm (RTX5032A) installed. If you don’t want, or can’t use it for your applications, you may opt to connect your own

Embedded & Tools

200012-LimeSDR Mini-EN.indd 25

25

03-02-20 15:49

SMA connections. SMA connections.

reference for the system. On the board, suitable UL.F connectors are provided to measure the internal reference or to feed a different clock signal from an external source, the latter requiring the removal of a pair of resistors. Not only the technical data for the SDR is important, but also the software support.

Drivers and software The Windows 10 drivers of the FTDI FT601 USB3.0 Bridge are automatically

MAX10 FPGA and FTDI USB Bridge.

available, but you may have to manually start the search for the drivers after the initial plug in. For Windows 7 it may be necessary to download the drivers from the FTDI side. With the “Pothos SDR dev dev enviroment” we have a good basic package allowing us to install GNUradio/GRC/OsmoSDR, GQRX and CubicSDR with the necessary plugins for the LimeSDR Mini. This provides a well-equipped toolbox under Windows, including the LimeSDR Suite for hardware

diagnosis and configuration. The GNUradio support makes the signal generation and the signal reception very versatile. The generation of a DVB-T signal, the transmission and reception of FM as well as FM + RDS are provided as examples. SDR# can also be used with the LimeSDR Mini if a suitable plugin is used [3]. Looking at the software, you will notice that many of the tools have been ported from the Linux world onto Windows. The tools mentioned, as well as others, are

Radio reception using GNU Radio.

26

Elektor Industry 1/2020

200012-LimeSDR Mini-EN.indd 26

03-02-20 15:49

DVB-T using GNU Radio.

available for the common Linux distributions and can usually be installed there directly using the package management. These are also not limited to the X86/AMD64 architecture, running the LimeSDR Mini on the Raspberry Pi 3B+ is possible, which has already been proven with the setup of GSM base stations [4]. And those of us on the road with a Mac and don’t want to or can’t do without the MacOS, have a suitable catalog of software to choose from, even if it means that on occasions you can’t avoid compiling code yourself.

The DVB-T example was used and an RTL-SDR was used as receiver. The signal was successfully generated, and the data stream was reproduced on another computer, a signal generation for different tests is possible and can be realized, for example, by GNUradio under Windows. In addition to DVB-T, experiments with other transmissions can also be carried out in the company’s own RF-tight laboratory. DAB+ or DAB are only one possibility. Even FM with RDS 2.0 or classic AM or SSB are no obstacle with the right tools. Due to the more than 30 MHz bandwidth there are a lot of possibilities.

Test run The Elektor laboratory in Aachen does not offer good reception conditions, which we already noticed during the development of the piRadio [5]. In addition to the reception test, a short transmission test was carried out under laboratory conditions.

The compact size is pleasant when it comes to stowing the SDR in a sheet metal housing, but has the disadvantage of high component density. Due to the proximity of the components to each other, it is not easy to shield unwanted interference from them to the receive

path. Some compromises have been necessary in terms of price, size and performance. Some side effects are discussed in the myriadrf forum [6].

Conclusion Before use, it is advisable to take a closer look at the peculiarities of the miniaturised version. Maybe one or the other will modify the LimeSDR Mini to avoid the side effects of miniaturization. Nevertheless, the LimeSDR Mini, at a price of 180 euros, is an interesting hardware base that stands out due to its wide frequency range and transmission and reception bandwidth. For those who are looking for a way to generate test signals in their laboratory that are broadband and beyond 900 MHz, the LimeSDR is an inexpensive way to achieve this. 200012-02

Weblinks [1]: http://limemicro.com/technology/lms7002m/ [2]: http://limemicro.com/community/limesdr-mini-fpga-accelerated-real-time-spectrogram/ [3]: http://discourse.myriadrf.org/t/limesdr-mini-with-sdrsharp/2772/31 [4]: http://limemicro.com/news/lime-provides-limesdr-mini-and-raspberry-pi-base-stations-for-emf-camp/ [5]: http://www.elektormagazine.com/labs/piradio-for-fm-radio-receiver-with-rds-for-raspberry-pi-160520-1 [6]: http://discourse.myriadrf.org/t/limesdr-mini-and-spurious-signal-around-100-mhz/2801

Embedded & Tools

200012-LimeSDR Mini-EN.indd 27

27

03-02-20 15:49

Low-Voltage Motor Control System Design Considerations For Stepper Motors By Bernhard Dwersteg, Trinamic

Figure 1: Open-source TMC2300-MOTOR-EVAL board for evaluation of the TMC2300 stepper motor driver for battery-powered applications.

Despite having a limited energy supply, more and more hand-held devices and IoT applications rely on actuators. And while a bigger battery has some benefits, it’s the ones with a low cell count that are preferred by device manufacturers for cost, reliability, size and weight, safety, and other reasons. Whether it’s for home automation, security, medical, or handheld POS devices, batteries with a voltage range of 2.4-4.3 V seem the way to go.

These batteries, however, have certain characteristics that pose new challenges not only for the motors themselves, but also for designs and technologies related to motor control and motion control. Characteristics including a restricted peak power and long-term energy consumption, which even forms an issue for applications with a low operating duty cycle. This poses the question for motor and motion control: How do you efficiently control a stepper motor from a low voltage power source?

28

Driving a battery-powered stepper motor As battery-powered applications normally require a compact solution, standard hybrid steppers like the NEMA 17 types are too large. Even if NEMA 17 offers the best value-for-money due to high volume usage, e.g. in 3D printers. Smaller hybrid stepper motors like NEMA 11 and NEMA 8 or even smaller come at an increased price. As such, inexpensive permanent magnet stepper motors are often preferred for mobile solutions and widely available in different sizes with manufacturer-specific

Elektor Industry 1/2020

191224-LowVoltage-EN.indd 28

03-02-20 15:50

mounting schemes. Standard types often come with a motor coil wound for 5 V or for 12 V. Both voltages are neither intended nor good for battery operation. Instead, they are intended for mains operated constant voltage driving scheme with quite limited motor velocity as there is no headroom for back EMF, which builds up at increased velocity.

Parameter name

Parameter value

Nominal (RMS) coil current

ICOILNOM [A]

Nominal coil resistance

RCOIL [Ω]

Rated coil voltage

UN = RCOIL * ICOILNOM [V] (sometimes specified instead of ICOILNOM)

Holding torque at ICOILNOM

Holding Torque [Nm]

Selecting the right motor To understand which stepper motors are suited for battery-powered designs, the motor supply voltage requirements for a stepper motor need to be taken into account. Motor torque is directly proportional to the coil current multiplied by the number of windings, since each Ampere of current flow contributes with a certain amount to the magnetic field and thus to motor torque. The specified motor torque is reached with the rms ICOIL current in both motor coils to build up the required magnetic field strength. A lower current will basically generate a proportionally lower torque, e.g. 70% of torque at 70% current. Even a reduction to 70% saves a lot of energy since power dissipation goes with the square of the current. Thus, a motor with more reserves can offer better efficiency. With this, the required supply voltage UBAT for motor standstill can be calculated, taking into account the driver’s power stage resistance plus a few 100 mV loss in the sense resistor (170 mΩ per bridge MOSFET for the TMC2300 low voltage stepper driver and 0.3 V peak at the sense resistor):

Table 1: Main relevant parameters for a stepper motor.

Measured item

Standard coil Modified coil

Unit

RCOIL

5.2

1.5

Ω

ICOIL

400

760

mA

PCOIL = 2*R*I²

1.66

1.62

W

Required supply (motion)

5.0

3.2

V

Current draw

487

850

mA

Power

2.24

2.76

W

Difference

-

+23

%

Table 2: Measurement results from stepper motor with standard coil vs. adapted coil.

ICOIL is the RMS motor current which gives the desired torque at standstill. At slow motion with negligible back-EMF, there not much difference. For higher velocity operation (more than a few electrical rotations per second), the motor’s specific back EMF constant CBEMF should additionally be taken into account (see below explanation). With this, the lowest feasible supply voltage for a given motor and a maximum velocity [RPM] calculates to:

This formula uses the quotient of holding torque and assigned coil current (taken from motor datasheet) to calculate the motor’s back EMF-constant. With most motor suppliers, the coil winding has to be adapted for battery-powered operation. This allows trading in a lower motor voltage for battery operation versus higher motor current. For example, a motor with a short, thick coil wire can work at a lower voltage than the same motor with a long, thin coil wire, but it needs a higher current for the same torque. Coil power dissipation and motor efficiency stay identical for both motor windings, see Table 1.

Advertisement

Embedded & Tools

191224-LowVoltage-EN.indd 29

29

03-02-20 15:50

Figure 2: ’Scope shot showing coil current of a 1.5-Ω motor (adapted winding) at 3.3 V (left) vs. 5.0-Ω motor at 5 V (right). Supply current (purple) and supply (green).

Test setup: linear actuator for smart thermostatic radiator valve To prove the importance of dedicated low voltage motor driver ICs, a test setup is made using a linear actuator for a heating valve. The original stepper motor used in this setup has a 5 Ω coil resistance, compared to a modified stepper motor with 1.5 Ω coil resistance (Table 2). Driving both motors is the TMC2300 low voltage driver IC, driving the modified motor at 3.3 V and the standard one at 5 V, with the assumption that the motor power dissipation is identical due to the same degree of copper filling in the coils. The motor is operated at 320 Hz full-step (beyond resonance). The scope shots in Figure 2 show the different ’height’ of supply current and motor coil current (increased current for lower resistive motor) and the supply voltage. Based on the measurements we can conclude that using the same driver IC, the lower voltage motor can deliver the same torque with a lower supply voltage. Due to increased coil current, power dissipation within the driver stage is higher and adds the power demand. On the other hand, when a single Li-Ion cell is used as power supply, a step-up converter (which also has limited efficiency of maybe 90-95%) can be eliminated. Further, this example shows that the power stage’s resistance in a low voltage motor driver IC is a key feature to efficiency. A higher resistance not only wastes power in the power stage, it also reduces voltage headroom to drive the actuator, which means it has to be designed for an even lower voltage and thus higher current. Conventional driver ICs, even for low voltage, have a problem with low-voltage. This is shown using a standard MOSFET as an example, which is similar to MOSFETs integrated into ICs: The final on-resistance (RDSon) of the FET is reached between 4 V and 6 V gate voltage (Figure 3). The area below 4 V shows drastically increased resistance. To yield a low RDSon, the battery supply driver IC needs an internal step-up converter, at the minimum to drive the power MOSFETs. Figure 4 illustrates RDSon vs. supply voltage

30

Figure 3: Low-voltage operation of a MOSFET (BSS138) shows similar characteristics to a power stage of a standard motor driver IC.

for an IC which is specially designed for low voltage operation, integrating a voltage multiplier for control of the power stage instead of a standard transistor. A comparison is made of the shape of the curves, not the actual value. The area of operation with 2 AA cells is circled in blue, showing differences for the MOSFET, resp. a conventional driver IC in RDSon of a factor of 2, while the single Li-Ion cell operation voltage of 3-4 V still shows significant benefits for the optimized IC. As the test setup shows, dedicated low voltage motor driver ICs such as the TMC2300 optimize system designs so batteries can be used till the last drop – whether it’s for home automation, lab automation, or medical devices to be used in the field. With a low overall power-draw, the solution allows Engineers to either reduce the size of the energy storage needed or to increase lifetime and service interval for improved user experience. However, it’s important to optimize the entire system

Elektor Industry 1/2020

191224-LowVoltage-EN.indd 30

03-02-20 15:50

Figure 4: Low-voltage operation of a dedicated low-voltage stepper driver IC, which uses internal circuitry to enhance MOSFET conductivity.

for best result, meaning the power source and motor – and their inherent characteristics – should be carefully considered as well to drive innovation and transform digital information into perfect physical motion for battery-operated devices.

Advertisement

191224-02

With Trinamic’s plug and play solutions for motion control, any engineer can easily optimize embedded system designs and be the market leader The Author Bernhard Dwersteg is one of the founding members of TRINAMIC Motion Control and joined their team more than 20 years ago, being the mastermind behind Trinamic’s stepper motor drivers. By continuously pushing the boundaries of current control, he is one of the driving factors why stepper motors are what they are today. Bernhard holds a degree in information technology from the University of Hamburg.

Visit us at Embedded World Hall 3 - Booth 253

Embedded & Tools

191224-LowVoltage-EN.indd 31

31

03-02-20 15:50

What

Applications Are We Talking About?

When we are talking embedded systems using a MCU, what kind of applications are we talking about? Not surprisingly, we are mostly talking automotive. One could think of Advance Driver Assistance Systems (ADAS) such as lane departure warning, parking assistance and collision avoidance. However, during the coming years consumer electronics will become ever more important, claiming a larger market size than the 2018 pie chart shows. The coming years, cameras and robots will no longer be associated with large premises only, but literally will become household names. (Source: Grand View Research)

Global Microcontroller Market Share, by Application, 2018 (%) Military & Defence

Automotive Medical Devices

Consumer Electronics

Industrial

By Robert van der Zwan

What Does the Market for MCUs Look Like? How will the market for MCUs evolve during the coming, say, five years? According to research company Grand View Research, a solid growth rate of a little more than 9% is highly probable. The compound annual growth rate (CAGR) of 9.2% is founded on three sub markets, namely the 8 bit, 16 bit and 32 bit microcontrollers. Research within North America (US, Mexico and Canada) shows that these three markets will be growing more of less equally in the near future, thus keeping their respective shares intact: a quarter for 8 bit MCUs, a half for 16 bit MCUs and the remaining quarter for 32 bit MCUs.

(Source: Grand View Research) Global Microcontroller Market Size, by Product, 2018-2025 (USD Billion)

≈34

Security? It is Still An Issue

How do vendors face the very real threat of attacking and hacking their embedded systems? Well, the awareness to make security a core value is definitely on the rise. As a result, the global market for embedded security items (both software and hardware based) is growing 6% a year, according to research company MarketsandMarkets. However, a word of caution is called for. Since the MCU market as a whole is roughly growing at 9% these coming years, there still seems to be a quite large group of engineers that does not take security into account at all. (Sources: Barr Group, MarketsandMarkets)

Embedded Security Market, by Product, 2023 (USD Billion) CAGR (2017-2023)

Hardware Security Module

CAGR 9.2%

Secure Element and Embedded SIM

18.6 Hardware Token

Trusted Platform Module

Market Size in 2023 2018

32

2019

2020

2021

2022

2023

2024

2025

Elektor Industry 1/2020

191222-Infographics-EN.indd 32

03-02-20 15:51

Which

Players Are There (and On the Rise)?

Now that machine learning (ML) and artificial intelligence (AI) become an integral part of embedded systems, there is a new market looming: that of AI software platforms. These are platforms providing tools to analyse all kinds of unstructured information, leading to, for example, facial recognition or speech recognition. Although there are a lot of of rather small parties active in this emerging market, some big names are making quite substantial inroads, growing significantly faster than the total market for AI software platforms. Take Amazon, Google and Microsoft.

(Source: IDC)

Worldwide AI Software Platforms 2018 Share Snapshot IBM $240.6; +26.6% SAS $89.0; +104.6%

9.3% Microsoft $88.5; +139.4% Palantir $75.2; +5.2% Google $58.5; +61.7% Amazon Web Services $51.3; +106.3%

Total Market

$2.6B

Digital Reasoning $49.8; +30.9%

+26.6%

Rest of Market $1,946.6; +20.7%

74.9%

Market Share, see Pie Chart in % Revenue company in USD Million Growth company in %, 2017 2018

ML and AI are the Way to Go

The organisers of embedded world tell us that machine learning (ML) and artificial intelligence (AI) are “hot topics” during its 2020 exhibition. This is because ML and AI can now be done on “the embedded node” rather than on a central server somewhere in the cloud. This creates extra demand for new embedded systems that can, for example, determine its own maintenance intervals locally. So… are ML and AI indeed hot topics? They undoubtedly are, according to research carried out by Gartner. Gartner notices a massive tendency from online non embedded systems to online embedded systems, or ‘edge computing’.

(Sources: Gartner Inc., Politecnico di Torino) Expected Growth of IoT Devices: from Non Embedded to Embedded

> 100 Billion 30 Billion

10 Billion

500 Million 5,000

1950

2.5 Billion

10,000

1975

2003

2009

2014

2020

2050

Embedded & Tools

191222-Infographics-EN.indd 33

33

03-02-20 15:51

Microchip Simplifies

Hardware-Based IoT Security with the Industry’s First Pre-Provisioned Solutions for Deployments of Any Size

With a minimum orderable quantity of 10 units, Microchip’s Trust Platform provides hardware-based secure key storage for low-, mid- and high-volume deployments. As the number and types of connected devices proliferates, market fragmentation and security vulnerabilities in the Internet of Things (IoT) have created significant challenges for developers. Hardware-based security is the only way to protect secret keys from physical attacks and remote extraction, but extensive security expertise, development time and costs are

34

required to configure and provision each device. With companies producing anywhere from hundreds to millions of connected devices per year across the globe, scalability of architecture can be a major barrier to deployments. Manufacturers typically have only been able to support configuring and provisioning for high-volume orders, leaving companies with low- to mid-sized deployments with low performing options. To address this need in the mass market, Microchip Technology Inc. today introduced the industry’s first pre-provisioned solution that provides secure key storage for low-, mid- and high-volume device deployments using the ATECC608A secure element. Microchip’s Trust Platform for its CryptoAut-

Elektor Industry 1/2020

200042-IND2001 Microchip-EN.indd 34

03-02-20 15:52

hentication™ family enables companies of all sizes to easily implement secure authentication.   Microchip’s Trust Platform consists of a three-tier offering, providing out-of-the-box pre-provisioned, pre-configured or fully customizable secure elements, allowing developers to choose the platform best suited for their individual design. As the first solution to provide ready-to-go secure authentication for the mass market, the first tier – Trust&GO – provides zero-touch pre-provisioned secure elements with a Minimum Orderable Quantity (MOQ) as low as 10 units. Device credentials are pre-programmed, shipped and locked inside the ATECC608A for automated cloud or LoRaWAN™ authentication onboarding. In parallel, corresponding certificates and public keys are delivered in a “manifest” file, which is downloadable via Microchip’s purchasing e-commerce store and select distribution partners.   In addition to saving up to several months of development time, the solution significantly simplifies provisioning logistics, making it easy for mass market customers to secure and manage edge devices without the overhead cost of third-party provisioning services or certificate authorities.   With the ability to authenticate to any public or private cloud infrastructure, Microchip’s Trust Platform is also flexible and customizable. For customers who want more customization, the program includes the TrustFLEX and TrustCUSTOM platforms.   The second tier in the program, TrustFLEX, offers the flexibility to use the customer’s certificate authority of choice while still benefiting from pre-configured use cases. These use cases include baseline security measures such as Transport Layer Security (TLS) hardened authentication for connecting to any IP-based network using any certificate chain, LoRaWAN authentication, secure boot, Over-the-Air (OTA) updates, IP protection, user data protection and key rotation. This reduces the time and complexity involved in customizing the device without requiring customized part numbers. For customers who would like to entirely customize their designs, the third tier in the program – TrustCUSTOM – provides customer-specific configuration capabilities and custom credential provisioning.   “The uptick in successful attacks on software-based security solutions underscores the need for companies to adopt industry best practices, including isolating private keys in secure elements,” said Nuri Dagdeviren, vice president of Microchip’s secure products business unit. “Microchip’s Trust Platform makes hardware-based security simple and cost-effective for companies of all sizes to implement, removing the barriers traditionally associated with configuring and provisioning devices.”   Microchip worked with Amazon Web Services (AWS) to enable a straightforward and simplified onboarding process into AWS IoT services for products designed with all variants of the Microchip Trust Platform.    The ATECC608A provides Common Criteria Joint Interpretation Library (JIL) “high”-rated secure key storage, giving customers confidence that devices implement industry-proven security practices and the highest level of secure key storage. With

hardware-based root of trust storage and cryptographic countermeasures, the device protects against the widest classes of known physical attacks. Microchip’s secure manufacturing facilities safely provision keys, ensuring that keys are never exposed to any party during provisioning or the lifetime of the device.   Development Tools The ATECC608A can be paired with any microcontroller and microprocessor. For rapid prototyping of secure solutions, designers can use the Trust Platform Design Suite, which includes: • A guided “use case tool” • Executable Python tutorials running on Jupyter notebooks • C code examples for each use case • A “secret exchange” utility • The Trust Platform hardware development kits   Pricing and Availability Devices in Microchip’s Trust Platform are available in volume production today with the following minimum order quantities (MOQ): • Trust&GO for TLS (ATECC608A-TNGTLSx-B): $1.20 with a MOQ of 10 units* • Trust&GO for TLS (ATECC608A-TNGTLSx-G): $0.77 with a MOQ of 2000 units* • Trust&GO for LoRaWAN (The Things Industries ATECC608A-TNGLORAx-B and Actility ATECC608ATNGACTU-B): $1.40 with a MOQ of 10 units* • TrustFLEX for LoRaWAN any join servers (ATECC608ATFLXLORAx): $0.938 with a MOQ of 2000 units* • TrustFLEX (ATECC608A-TFLXTLSx): $0.845 with a MOQ of 2,000 units* • TrustCUSTOM (ATECC608A-TCSTMx): $0.883 with a MOQ of 4,000 units* *uDFN (x = U) or SO8 (x = S)   Development tools in Microchip’s Trust Platform are available at: • CryptoAuth Trust Platform kit: $13 • ATECC608a Trust Platform kit: $14 For additional information and to purchase products mentioned here, visit www.microchipdirect.com or contact a Microchip authorized distributor. 200042-01

Embedded & Tools

200042-IND2001 Microchip-EN.indd 35

35

03-02-20 15:52

Miniaturized Wireless Medical Wearables New 2 x 5 mm embedded chip antennas By Manuel Carmona, Business Development Manager, Johanson Technology

Tiny RF chip antennas are no longer relegated to PCB “keep out” area, allowing product designers to further miniaturize wireless biosensors and medical wearables. Traditionally, small chip antennas used in RF-enabled medical devices have required a designated ground “keep out” area to minimize interference from other components and ensure the ideal radiation pattern for wireless signals.  In some cases, this reserved space can eat up as much as 15 x 20 mm of the printed circuit board. However, with the drive to further miniaturize next generation medical biosensors and wearables, new alternatives are entering the market that allow the chip antenna to be mounted directly above metal surfaces.  By doing so, as much as 10-20% of the space traditionally reserved for the keep out area is no longer required, allowing designers to reduce the overall size of the product. This has major implications for wireless medical devices in which even miniaturized PCBs along with coin-cell batteries often utilized are limiting factors in the minimum form factor. Products that could be positively affected by this development include an array of “smart” devices such as watches, clothing, eye glasses, patches, pills and even adhesive bandages.

cal parameters than can be tracked include vital signs, sleep, emotions, stress, breathing, movement, efforts, posture, gait, body shape, lesions, mental acuity, toxins, blood glucose, ECGs, and drug adherence. The information collected is then wirelessly transmitted to nearby cell phones, remote monitoring stations or through Wi-Fi over the Internet to back-end servers for further analysis, assessment and decision making. Collecting data in this manner is expected to facilitate the development of disease models and an understanding of the complex behaviour of biological networks. Mobile health data can also be a valuable tool for drug discovery and clinical research. Among the products already incorporating this type of technology are adhesive bandages that contain built-in sensors that measure heart rhythm, respiratory rate and temperature.  These readings can be used, for example, to determine the precise amount of insulin that should be dispensed from wirelessly controlled insulin pumps worn by diabetics.

Embedded chip antennas Mobile health biosensors and wearables For the medical industry, the “future is now” when it comes to miniature battery powered sensor devices that can be located near, attached to, or implanted in the body to monitor physiological signs such as temperature, blood pressure and pulse rate. These smart devices will soon monitor everything from fitness to health, environment, lifestyle and behaviour. Biologi-

36

To transmit and receive RF wireless signals in the appropriate frequency range, smart devices must contain small RF chip antennas embedded on the PCB or behind the scenes underneath the encasement of the product. These chip antennas radiate and receive electromagnetic waves as other types of antennas, but the most notable difference is its small size. In fact, today’s mobile phones incorporate a min-

Elektor Industry 1/2020

180748-Johanson Technology-EN.indd 36

03-02-20 15:53

imum of four antennas and up to 13 in some models. Smaller wearable devices may only contain one or two antennas. To work properly, chip antennas have typically been ground plane dependent, meaning they require an appropriately sized and positioned ground plane to form a complete, resonant circuit. While the PCB can serve as the ground plane, the antenna itself must typically be placed on the edge of the board in an isolated section which is free from ground and metal components that would distort its radiation. Without the isolation distance, the performance of the antenna is significantly affected.

just 2 x 5 mm, designed for IoT, 2.4 BLE, wearables, ISM, ZigBee, and 802.11-standard applications

“The ‘keep out area’ is fundamental to ensure the chip antenna can electromagnetically radiate to antenna applications, because everything affects the radiation pattern including the package size, where the antenna is mounted and its proximity to the human body,” says Manuel Carmona of Johanson Technology. According to Carmona, Johanson Technology has been able to eliminate the requirement for a designated ground keep out area through optimization of materials (ceramics and inks), manufacturing processes and RF circuit design. The new 2.4 GHz antenna can now be mounted directly onto the metal ground plane. The product measures 2 x 5 mm and is designed for small coin-cell battery operated IoT, 2.4 BLE, wearable, ISM, ZigBee, and 802.11-standard applications where metal or a battery/display covers the entire length or side of the PCB. “With PCB real estate at a prime, the size and placement of the chip antenna is critical because as everything gets smaller it becomes increasingly difficult to place more components on the board,” explains Carmona. “Therefore, design engineers are looking to component manufacturers to deliver miniaturised solutions that occupy next to no real board space.” The design of the antenna itself is also critical to its range and performance. With medical devices, radio interference or some other glitch could result in interrupted connectivity. There can be legal ramifications as well. As with any wireless device, products that utilise RF technology, including Bluetooth, to collect or transmit information are subject to regulation by the Federal Communications Commission (FCC). Therefore it is critical the device perform at the designated frequency and the design and placement of the antenna is critical to proper tuning. Despite the critical nature of the antenna, Carmona says it is often overlooked until late in the design process, at which point optimal antenna performance may not be achievable within the space provided.

The Author Manuel Carmona, MSEE, is a Business Development Manager at Johanson Technology. He is Johanson’s Senior RF expert on Integrated Passive Components and Chip Antennas, with extensive Process and Material Engineering expertise, as well as broad knowledge on Electromagnetic Simulation.

Johanson Technology Phone: (805) 389-1166 e-mail: [email protected] web: www.johansontechnology.com/ant Address: 4001 Calle Tecate, Camarillo, CA 93012, USA.

To assist with chip antenna design and selection, Johanson Technology offers a program where design engineers can send in a miniaturised device and the company will tune the antenna for optimum functionality. “A chip antenna that can be mounted over a ground plane opens up many applications for products that want to incorporate wireless,” says Carmona. “To date, we have received everything from smart shirt buttons to jewelry and other wearables in various shapes and sizes.”

Embedded & Tools

180748-Johanson Technology-EN.indd 37

37

03-02-20 15:53

RISC-V:

Questions, Answers, ... and Embedded World 2020 The RISC-V Foundation What is RISC-V and what is the Foundation’s role in promoting RISC-V?

Calista Redmond is CEO of the RISC-V foundation.

Curious to hear both the basics and the latest about the RISC-V ecosystem, Elektor Industry sent out a questionnaire to RISC-V cofounders and members like Imperas, CloudBEAR, and Syntacore, not forgetting the RISC-V Foundation themselves. Their answers should interest anyone looking for methods and tools to implement a licence-free ISA in their product employing embedded technology.

38

RISC-V is a free and open instruction set architecture (ISA) enabling a new era of processor innovation through open standard collaboration. Born in academia and research, the RISC-V ISA delivers a new level of free, extensible software and hardware freedom on architecture, paving the way for the next 50 years of computing design and innovation. The RISC-V Foundation’s role is to build an open, collaborative ecosystem of software and hardware innovators while directing the future development and adoption of the RISC-V ISA. Our top priority is driving the technical collaboration to enable a broad and deep ecosystem of RISC-V extensions, tools and resources to bolster adoption. The Foundation provides several programs to support the success of members and the adoption of RISC-V. These programs include education, developer advocacy, visibility of progress, compliance frameworks and an online forum to showcase offerings. To gather the community together, each year the RISC-V Foundation hosts multiple global events to discuss current and prospective RISC-V projects and implementations, commercial and open-source implementations, software and silicon, vectors and security, applications and accelerators, simulation infrastructure and much more. We also actively promote independently hosted Meetups and events centered around RISC-V. Our goal is to encourage organizations, individuals and enthusiasts around the globe to join our ecosystem and together enable a new era of processor innovation through open standard collaboration.

Why is RISC-V so interesting for so many different companies? The RISC-V ISA offers a variety of advantages including its openness, simplicity, clean-slate design, modularity, extensibility and stability. Below are the benefits of RISC-V and these key points are driving companies to join our ecosystem. • Unlocks architecture and enables innovation. RISC-V is a layered and extensible ISA so companies can easily implement the minimal instruction set, well defined extensions and custom extensions to create custom processors for cutting-edge workloads. • Reduces risk and investment by enabling companies to leverage established and common IP building blocks with

Elektor Industry 1/2020

200043-RISC-V Questions-EN.indd 38

03-02-20 15:53

the development community’s growing set of shared tools and development resources. • Provides the flexibility to create thousands of possible custom processors. Since implementation is not defined at the ISA level, but rather by the composition of the SoC and other design attributes, engineers can choose to go big, small, powerful or lightweight with their designs. • Accelerates time to market through collaboration and open source IP reuse. RISC-V not only reduces development expenses, but also enables companies to get their designs to market faster.

RISC-V allows compatibility and reuse across the base configurations and extension while at the same time supporting the addition of custom instructions and extensions. The ecosystem of tools and software are essential to the adoption of any ISA, in the case of RISC-V the RISC-V Foundation has paved the way with a large number of members coming together and supporting the inherent flexibility that RISC-V offers. The technology and specifications are clearly important, but its the ecosystem that will really support the adopters of RISC-V.

What can we expect from RISC-V in the future?

Imperas code morphing simulation technology, virtual platforms and tools are used by lead customers for early software development and high-level architectural exploration. We support over 12 ISA’s, and we have worked with all the top architectures and our virtual platforms help the complexities of software development and hardware verification in complex multi-core design. Many projects are heterogeneous and our first interest was with RISC-V starting to be used by customers in a companion role, which quickly developed in to arrays of minion cores. However, RISC-V is unlike previous ISA’s with the flexibility and configurability built into the base structure. So, to be successful and easy to use it’s vital that the ecosystem leads with adaptable solutions that support the flexibility of RISC-V. The Imperas business model is to provide the fastest and highest quality simulator that can be used for software development and also hardware tape-out verification. Most of our models and platforms are provided in source form, originally this flexibility was used to customize processor peripherals, but with RISC-V it’s an ideal approach to support the flexible configuration, custom instruction and extensions.

The RISC-V ecosystem is poised to significantly grow over the next five years. Semico Research predicted that the market will consume a total of 62.4 billion RISC-V central processing unit (CPU) cores by 2025, with the industrial sector forecasted to be the largest segment with 16.7 billion cores. By that time we look forward to seeing many new types of RISC-V implementations including innovative consumer devices, industrial applications, high performance computing applications and much more.

Imperas What are you showcasing at Embedded World 2020? Imperas are presenting two technical papers at the Embedded World 2020 conference on RISC-V verification methods and virtual platforms for mixed safety critical systems. At our demonstrations and presentation on the RISC-V booth (Hall 3a-536) will feature RISC-V processor and extension verification methods, custom instruction analysis and SoC architectural analysis of many core designs using virtual platforms.

Are there any new developments compared to the event last year? Imperas are supporting the RISC-V Foundation booth again in 2020, together with 14 other members, which is an increase from the 7 supporters in 2019. This growth not only reflects the additional members over the past 12 months, but the recognition in the potential for RISC-V to the industry and the importance of the exhibits and conference at Embedded World in Nuremberg. New for Imperas this year will be our support for the RISC-V draft specifications for Bit Manipulation and Vectors extensions. We will also be showing the latest developments on RISC-V verification and out tools in the design flows for custom instructions.

We are aware that the RISC-V ISA set can be used license-free. Are there other benefits like technical ones, when compared to other ISAs? The RISC-V ISA (Instruction Set Architecture) is an open standard, this openness allows implementations great flexibility and configurability together with the optimization of custom extensions and instructions. But, as a standard, the real advantage is the community activities within the RISC-V foundation to develop and enhance the future functionality. This has allowed members to actively participate in the next generation of extensions for Bit Manipulation and Vector instructions. From an ecosystem perspective the built-in support for the flexibility of

What was your motivation to join the RISC-V ecosystem?

Impact of RISC-V Adaptability on SoC Verification Methods

Do you also make contributions on technical aspects? Imperas is actively contributing to the RISC-V Foundation working group on Compliance and in November 2019 announced the contribution of the highest quality compliance suite for RISC-V RV32I. This builds on our release of a free single core reference model, riscvOVPsim available on GitHub. We are also actively supporting the working groups on Bit Manipulation and Vectors and announced the first delivery of a vectors model to lead customers in 2019.

Critical issues and bugs have been reported for other processor architectures. Is RISC-V is safer, and if so, on what grounds? Processors are probably one of the most complex machines ever developed, however, the verification of processors has, until recently, been an in-house skill developed around the well-known and closed ISA’s. Now with RISC-V, implementors are exploring configuration flexibilities and optimizations with

Embedded & Tools

200043-RISC-V Questions-EN.indd 39

39

03-02-20 15:54

custom instructions. In the traditional one-size-fits-all approach, a design technique for performance may have unintended consequences in a high reliability or mixed safety critical application. Now with RISC-V developers can configure the processor for the ideal balance of requirements. At Embedded World 2020, Imperas will be presenting a technical paper on using virtual platforms for mixed critical safety systems based on our work within the Horizon 2020 SafePower project.

Vendor-specific instructions are an important feature of RISC-V. Does RISC-V’s flexibility constitute any liabilities? While the traditional or closed ISA’s of the past had fixed configurations and implementations, the RISC-V open ISA allows implementations with both microarchitectural features and custom extensions and instructions. This offers increased flexibility for optimizations that address the needs and requirements of domain specific custom devices. As with any complex design, the verification requirements are of paramount importance and the Compliance working group of the RISC-V Foundation is a great starting point. However, the open and flexibility of RISC-V presents some new issues for the DV teams and at Embedded World 2020 we will be presenting a technical paper with an update on the verification methods based on the Google random instruction stream generator and our reference model to cover some unique and hard to cover corner case situations.

Virtual Platform Based Development Environments for Low Power, Mixed Level Safety Critical Systems

over an array of data items which enables acceleration of key computational workloads. Using Imperas models and tools allows system designers to evaluate advanced SoC architectural analysis of many core designs using virtual platforms and full software application workloads with real-world datasets, instead of limited benchmarks or small sample test cases.

What direction would you like RISC-V to take? The open ISA of RISC-V is a great starting point for innovation and future development, but its more than a specification. The RISC-V Foundation represents the community of members from all aspects of the ecosystem from tools, OS and software providers, IP cores and silicon providers, to academic and commercial adopters. Collaboration will be the hallmark of RISC-V in bring together experts from across the industry in the spirit of constructive support and mutual efficiency. With healthy competition on solutions, RISC-V adopters will benefit tremendously from the innovations and flexibility of RISC-V. The future for RISC-V is being shaped within the community driven ecosystem.

CloudBEAR What are you showcasing at Embedded World 2020? CloudBEAR is showcasing application processor SoC prototype based on our BI-671 cores.

Are there any new developments compared to the event last year? We plan to introduce new BR series cores on EW2020 which target high-performance embedded applications.

We are aware that the RISC-V ISA set can be used license-free. Are there other benefits like technical ones, when compared to other ISAs? Clean-slate design. Opportunity to do standardization right from the beginning.

What kind of products do you offer in support of RISC-V? Imperas has developed a reference model of RISC-V which covers all the optional configurations and the latest draft specifications for Bit Manipulation and Vector extensions. Together with our verification and analyst tools our customers are designing RISC-V implementations, developing custom extensions, and SoC designs with multiple clusters of processor cores. We released the reference model, riscvOVPsim, on GitHub with free use for commercial and academic users, more details are available at www.imperas.com/imperas-riscv-solutions

Are there plans to expand these in the future? The draft specification for Vector instructions is expected to be ratified in 2020, this represents a great opportunity for processor developers in addressing the requirements for advanced ML (machine learning) and AI (artificial intelligence) applications. The RISC-V Vector extensions are designed to support complex arithmetic operations required for applications involving linear algebra, such as supercomputers, cryptography, AI, ML, and deep learning (DL). A traditional or scalar ISA is based around operations on single data items, a Vector processor operates

40

What was your motivation to join the RISC-V ecosystem? Royalty-free ISA and great software ecosystem.

Do you also make contributions on technical aspects? Mostly patches to open source software for RISC-V.

Critical issues and bugs have been reported for other processor architectures. Is RISC-V is safer, and if so, on what grounds? RISC-V addressing these issues since standardization process on going and many thinks could be designed considering these threats.

Vendor-specific instructions are an important feature of RISC-V. Does RISC-V’s flexibility constitute any liabilities? It could lead to fragmentation but most vendor-specific instructions are for targeted workload in specific domains.

Elektor Industry 1/2020

200043-RISC-V Questions-EN.indd 40

03-02-20 15:54

What kind of products do you offer in support of RISC-V?

What was your motivation to join the RISC-V ecosystem?

Processor IP cores based on RISC-V from microcontrollers to high performance out-of-order.

Syntacore is a founding member of the RISC-V Foundation, which means we were there right from the beginning. Being a customizable CPU IP company, such an open ecosystem provides a perfect environment for our technology.

Are there plans to expand these in the future? Adding new ISA extension and more features to existing core series

What direction would you like RISC-V to take? Improve new extensions for particular emerging markets like AI.

Do you also make contributions on technical aspects? Yes, we do — we are on a Technical committee and contributing in several Working Groups.

Imperas delivers first RISC-V Simulator for new Vector and Bit Manipulation specifications to Lead Customers

Syntacore What are you showcasing at Embedded World 2020? Syntacore, as a leading provider of the RISC-V compatible processor IP and associated services, showcases several live demos, based on our IP. These include production silicon samples, Linux-capable Octa-core RV64GC system running full Debian and examples of SDKs — presented jointly with our commercial tools partners.

Are there any new developments compared to the event last year?

Critical issues and bugs have been reported for other processor architectures. Do is RISC-V is safer, and if so, on what grounds? We believe yes, on multiple levels. First and foremost, as Foundation iteratively stresses, RISC-V as technology has a unique opportunity to do security right. There is no legacy burden and it’s possible to specify clean-slate security mechanisms which would factor in all the recent discoveries and developments in this area. This has been continued focus of the Foundation, which is backed, for example, in a dedicated Security Technical Committee with several Working Groups, focusing on different aspects of security. Next, the open-source nature of the

Yes, we demonstrate customers silicon, based on our highly popular open-source SCR1 core, which is now in a full wafer production. Another new item is full Debian, hosted on the Linux-capable Octa-core SCR7-based setup. We believe this may be the first live public RISC-V Octa-core SMP Linux system demonstration.

We are aware that the RISC-V ISA set can be used license-free. Are there other benefits like technical ones, when compared to other ISAs? In our opinion, at the moment, majority of the benefits are still organizational. In few words, this ISA is extensible, and industry driven. Indeed, starting from 2019, RISC-V ecosystem can already enjoy solid base in a stable and ratified by the RISC-V Foundation standard ISA specifications. That establishes solid grounds and baseline insurances for a wide commercial adoption and further ISA development. On the technical side, as multiple studies have shown, base RISC-V ISA is high-quality and provides good basis for highly competitive implementations. There are multiple ongoing developments of further standard extensions, some of which are quite interesting (and differentiating), - and all this information is open, while in development!

The new reality and tremendous opportunity of open source processing

Embedded & Tools

200043-RISC-V Questions-EN.indd 41

41

03-02-20 15:54

Another new item is full Debian, hosted on the Linux-capable Octa-core SCR7-based setup.

ecosystem means that RISC-V itself is a product of worldwide collaboration, ISA is fully observable and is widely tested by multiple companies prior to be finally frozen in the specifications, and this reduces chances to miss critical issues. Then, it’s likely that over time RISC-V ecosystem will develop considerable amount of the fully open-source IP (like our SCR1), which would create opportunity to have comparable levels of visibility in the semiconductor industry which open-source SW enjoys. Overall, in our opinion, RISC-V ecosystem creates a perfect environment for a value-add technology, focusing on security. This is confirmed by a high number of companies, focusing on a different aspects of security, among RISC-V foundation members.

Vendor-specific instructions are an important feature of RISC-V. Does RISC-V’s flexibility constitute any liabilities? People talk about fragmentation, but properly handled, we don’t see it as an issue.

What kind of products do you offer in support of RISC-V? Being RISC-V IP Specialist, Syntacore focuses on two things. First, we design and license SCRx family of the state-of-theart RISC-V compatible processor IP, varying from minimalistic open source and SHL-licensed 32bit MCU core SCR1, to SCR7, our high-perf 64bit out-of-order APU with up to 8 cores SMP support, which we have at the exhibit here, at the EW2020. Second, for all our baseline IP we provide a one-stop customization services (including compiler and tools support and SW porting), which enables customers to create truly differentiated products.

Are there plans to expand these in the future? Yes, we have full roadmap with several additions planned later this year, both new IPs and interesting new features for the existing product line.

What direction would you like RISC-V to take? Just keep going. 200043-01

Advertisement

Post your ideas and electronics projects all sizes/all levels/allsorts, at www.elektor-labs.com and become famous! Elektor Labs: www.elektor-labs.com Design, Build, Share... Electronics!

42

Elektor Industry 1/2020

200043-RISC-V Questions-EN.indd 42

03-02-20 15:54

Next generation phase change cooling

www.incooling.com

EI-20.01-Adv-EN.indd 5

03-02-20 16:09

BeagleBoard.org introduces the

PocketBeagle® Grove Kit

PocketBeagle® Grove Kit is an infinitely scalable tool for working with sensors and actuators aimed at helping beginners and professionals alike. “With BeagleBoard.org® PocketBeagle® Grove Kit, one could go from learning programming, Linux and physical computing to developing projects and products on PocketBeagle® quickly and conveniently.” says Jason Kridner, Co-founder of the BeagleBoard.org® Foundation. “I love that I can just turn it on and play with it as a music synthesizer.” The kit comes with PocketBeagle® and PocketBeagle® Grove Cape to connect Grove modules without soldering or stray jumper wires. There are 10 different Grove modules that are pre-installed in an acrylic case, including sensors, a display, LEDs, and buttons. They are all ready to use with drivers pre-installed. In addition, it contains all the essentials that you need to get started with PocketBeagle®, such as a USB cable, SD card, and WiFi dongle. Provided tutorials take you from novice to building your own musical instrument, performing voice recognition and connecting to Internet services.

Features • • • •

Complete set of accessories 10 Grove modules ready to plug in and use Pre-installed software to get started quickly 8 well written tutorials

Accessories List • • • • • • • • • • • • • • • • • • •

BeagleBoard.org® PocketBeagle® x 1 Grove - Analogue Microphone x 1 Grove - Chainable RGB LED x 2 Grove - Ultrasonic Distance Sensor x 1 Grove - Rotary Angle Sensor x 1 Grove - Slide Potentiometer x 1 Grove - Button x 2 Grove - 12-key Capacitive I2C Touch Sensor V2 x 1 Grove - 3-axis Digital Accelerometer x 1 Grove - Speaker Plus x 1 Grove - 16x2 LCD (white on blue) x 1 BeagleBoard.org® PocketBeagle® Grove Cape x 1 Alligator Cable x 10 SD+TF Card Reader x 1 16GB SD Card x 1 WiFi Dongle x 1 Acrylic shell x 1 Grove Cable x 10 USB Cable x 1

“We believe the BeagleBoard.org® PocketBeagle® Grove Kit will give engineers and educators practical everyday examples of sensing the physical world through touch, sound, force and distance” stated Christine Long, Executive Director of the BeagleBoard.org® Foundation. “Our goal is to give all audiences the opportunity to experience the creation of music using the PocketBeagle® Grove Kit and its included accessories.” https://beagleboard.org/grove

44

Elektor Industry 1/2020

200028-News-PocketBeagle-EN.indd 44

03-02-20 15:59

Exp

PAPE t R er

Tips and Tools for PCB Designers

Reducing error rates and costs by using Eurocircuits • What is this Elektor Expert Paper about? Developers of hardware need prototypes for testing. Eurocircuits offer a one stop shop solution, not only offering fast turn around bare board services (2 days delivery) but also offering a superfast assembly service (3 days). All backed by their own free to use online intelligent tools to ensure the designs are “Right First Time for Manufacture”.

• How much time does it take to read this Elektor Expert Paper? To read this paper online you need approximately 7 minutes.

• Who can you contact for further information? Eurocircuits contact: Dirk Stans +32 (0)15 28 16 30 / [email protected]

Explanatory videos and tools help hardware developers to design PCBs in a productionready manner and to work out error-free assembly data. This saves costs in PCB production and avoids faults and delays when populating the PCB.

Summary Faultless production data and the planning of the manufacturing effort early in the design process have a substantial impact on the product quality and the fabrication costs of the product. The term DFM (Design For Manufacturability) is important. Tolerances are decisive. The smaller the tolerances specified by the designer, the higher the manufacturing costs. The concept of tolerances, measures for uniform copper distribution and many other things are explained by PCB service provider Eurocircuits in freely available design tips and videos. This helps the developer to cut costs and minimize errors that

may occur during production. In addition, Eurocircuits has developed special software to allow detailed inspection of PCBs before production. This has reduced error rates from up to 25% to under 3%. With a calculation tool it is possible to calculate prices in advance and select the optimal configuration. Eurocircuits takes the logical step towards the assembled circuit board and includes the later assembly production, because even one mistake means problems and delays the assembly. The PCBA Visualizer Tool pre-checks all components and compares their dimensions with the design of the PCB to be assembled. The BOM and CPL are also examined. 200020-03

Do you need fully assembled prototype circuit boards? Then use the free PCBA Visualizer tool to create a validated PCB design with a validated bill of materials, CPL and 3D data. Reduce errors before the actual production of the prototype. Tools, tips and videos help you reduce costs and speed time to market. You can launch an online ordering process and have your PCB manufactured and assembled within six days. Smart Tools can be used to validate layout parameters or find more cost-effective options.

You can find the complete article at www.elektormagazine.com/news/ tips-and-tools-for-pcb-designers-reducing-error-rates-and-costs

Elektor Expert Papers are produced jointly by Elektor and selected partners active in educational and industrial ranges and are partly based on Expert Meetings. All Elektor Expert Papers are accessible online at : www.elektormagazine.com

200020-ExpertPaper-Tipps-EN.indd 45

03-02-20 15:59

Interference Sources in Automotive Applications And how to get rid of them By Tam Hanna (Slovakia)

It is easy for anyone whose background is in the world of industrial control to fall into one of the traps for the unwary when turning their hand to designing a circuit for use in a vehicle. This article will look at the ways in which things can go pear-shaped, and at the risks in store for designers and their hardware.

Just a couple more words before we get down to the nitty-gritty: the author knows of a large number of low-volume designs that in some cases have been in working use for years despite implementing few if any of the protection measures we will be discussing. One of the reasons for this can be understood by thinking of an automotive system as comprising a network of filtering elements with an exponential decay characteristic as shown in Figure 1, with some circuits being in effect closer to the sources of trouble others: these circuits take the brunt of the interference, while others are left relatively unscathed. However, providing a design with the appropriate protection measures generally does not cost the earth and thus in any case makes economic sense.

Figure 1. The circuits have different levels of exposure to the sources of interference.

The savage alternator Let us start with the most pernicious of all types of interference, known as ‘load dump’ in the trade. Imagine a quad bike or similar vehicle being driven hell-for-leather along a dirt track. Unfortunately the integrity of the connection between the alternator and the battery is not all that could be wished for: perhaps the manufacturer has been trying to cut costs, the mechanic was asleep or the dreaded tin worm has taken its toll on the terminals. And so, as you might expect, when the tyres hit an especially rocky patch what happens is that the connection between alternator and battery is briefly interrupted. Now the alternator wants, and indeed has to, unload its excess output energy somewhere. Fortunately this is not that common an occurrence in practice. At this point we encounter for the first time a problem that we will frequently face in automotive applications: competing

46

manufacturers specify more or less whatever they want to, following the approaches used by different standardization organizations. It’s a jungle out there! Back to load dump. There are two relevant ISO specifications: the first is ISO standard 7637-2, which can be downloaded at [1]. That was in force until 2010, but continued to be used subsequently by some car makers, including Ford. The second is the version currently in force, ISO 16750-2, which is not available to view free of charge. The semiconductor manufacturer Vishay has prepared a brief overview at [2], summarizing the requirements of the current standard. At [3] Texas Instruments reveals the form of the test impulse, as shown in Figure 2. The bottom line is the observation that in a 12 V system we must be able to withstand a pulse of up to 101 V with a duration of 400 ms, with a source impedance of 0.5 Ω minimum and

Elektor Industry 1/2020

180345-Störquellen-EN.indd 46

03-02-20 16:00

4 Ω maximum. For 24 V systems the maximum duration is 350 ms, and the voltage is in the range 151 V to 202 V with a source impedance from 1 Ω to 8 Ω. From these we can determine, with some not-too-taxing arithmetic, that a 12 V system has to be able to withstand an instantaneous power of 20.4 kW, and a 24 V system has to withstand 40.8 kW. Particularly tricky to deal with in this context is the very short rise time of just 10 ms or so. Any circuits we design need to react very quickly, which rules out from the start the use of components such as relays. Just as tricky is the requirement laid down in the new standard that the circuit must be able to withstand a series of ten such impulses within the space of ten minutes. The older standard only required tolerance to a single pulse. However, the requirement of the new standard does make sense, as a vehicle with corroded battery contacts can easily produce a series of load dump impulses. A classical approach to providing protection against such pulses, beyond just adding a huge smoothing capacitor with a series resistor, is to design in a protection switch that disconnects the circuit from the vehicle supply when emergency strikes (Figure 3). Fortunately the semiconductor industry has fully realized how onerous complying with the standards can be, and offers a range of components designed to deal with transients. The company Littelfuse has been particularly successful in this department. Their TVS (transient voltage suppression) diodes, described briefly at [4], are components that act as a kind of crowbar, limiting the voltage when an exceptional transient occurs. STMicroelectronics have also been busy: their range of suppression diodes, going under the name ‘Transil’, is grouped according to the ISO test pulses, making it easier for designers to choose the right component (Figure 4). Vishay’s ESD protection diodes go under the name ‘TransZorb’. Be aware that TVS diodes are almost always available in unidirectional and bidirectional variants. Unidirectional TVS diodes behave much like conventional Zener diodes, with a pulse of the ‘wrong’ polarity being clamped immediately. Bidirectional TVS diodes have a minimum threshold voltage in each direction before they start to conduct. The author is a fan of the LM2576HV switching voltage regulator from Texas Instruments’ ‘simple switcher’ range, which is somewhat tolerant of high voltages (although its cold-start voltage performance leaves something to be desired). Using this reduces the burden on the diodes as they can be arranged to start acting at 35 V: the power they are required to dissipate is then considerably lower. TVS diodes are of course not the only weapon that can be used in the battle against voltage spikes. For example, varistors (voltage-dependent resistors, or VDRs) are bidirectional devices that in principle exhibit characteristics similar to diodes. One way they differ from TVSs is in their considerably higher energy absorption capacity; on the other hand they switch more slowly and degrade over the course of time (even though Vishay claim at [6] a ‘negligible wearout within safe operating area’). Another potential problem in this regard is the tendency of metal-oxide varistors (MOVs) to suffer ‘punch-though’ failure: the author has seen several systems where aging MOVs have shorted out one or more power rails. In practice one often finds a MOV paired with a TVS diode with a rather higher rated breakdown voltage. The idea behind this

Figure 2. It’s no fun dealing with a spike like this (source: Texas Instruments [3]).

Figure 3. Sometimes it pays to be cowardly!

Figure 4. STMicroelectronics categorizes its Transil family of devices according to the level of ESD protection offered (source: STMicroelectronics [5]).

Embedded & Tools

180345-Störquellen-EN.indd 47

47

03-02-20 16:00

Unfortunately in practice things are not so simple, as the wetware behind the wheel cannot always be relied upon. One example is when a vehicle’s battery is left to run completely flat (whether through negligence or stupidity) and then an attempt is made to jump start it from another vehicle. It is surprising how often people will try to jump start a 12 V car from a truck with a system voltage of 24 V. Also, in the English-speaking world at least, it is not uncommon to wire the two batteries in series when jump starting. At any rate, it is worth just making sure that the protection circuits in your design will not see a continuous 24 V supply as a voltage spike that should be clamped or shorted out. For a switching regulator an input voltage of 24 V for a brief period should not present a grave problem, as long as the circuit is reasonably carefully designed.

Don’t trust the voltage! Figure 5. Unoriginal, but simple: a voltage limiter using a Zener diode.

Figure 6. The unpleasant effect of cold cranking (source: Texas Instruments).

is to combine the benefit of the fast switching characteristic of the TVS diode with the higher energy capacity of the MOV. But note that the IEC 60950-1 standard prescribes the use of MOVs for over-voltage protection and forbids the use of TVS diodes!

Vicious voltages As well as protecting a circuit from its main power supply, it is important also to look at adequately protecting sensitive components, and in particular their inputs, using voltage limiters. A standard and highly effective, although not exactly elegant, configuration is the Zener diode circuit shown in Figure 5. When using this kind of protection circuit it is necessary to ensure that the components are correctly dimensioned: the resistor and the Zener diode must be able to absorb the energy in any applied voltage spike without overheating: the resistor must be able to limit the current through the Zener diode to a safe level without itself being destroyed. In theory in a 12 V system we might reckon with a maximum continuous supply voltage of at most 14 V or perhaps 15 V.

48

Getting voltage spikes out of the way is only half the battle: it is not just impulses that cause interference. A further source of aggravation is voltage drop-outs, which can arise for a variety of reasons. One is the sudden application of a heavy load to the system. Starting an internal combustion engine places extreme strain on the battery, especially at low temperature (‘cold cranking’) when the performance of the battery is in any case degraded. The behavior of the system voltage can be as extreme as that shown in Figure 6. At first sight the designer might be inclined to ignore the cold crank phenomenon. A voltage regulator (such as the LM2576 mentioned above) might easily be able to operate from voltages as low as 8 V. According to the ISO standard, however, systems must be able to cope with voltage dips down to 3 V, depending on their criticality. There are several ways to attack this problem. The first, of course, is to fit as large a capacitor as possible across the input, which will also have the effect of helping to smooth out rapid rises in voltage. The second possibility is to arrange things so that the system shuts down when the voltage falls below a certain level during cold cranking, and the restarts when the voltage becomes high enough once more. This approach is not always recommended, in particular in the case of controllers running complex operating systems that can take a long time to start up, as this can lead to further significant problems. Another, rather more interesting, solution is to use buck-boost technology. Here we use a switching regulator which is capable of either stepping down or stepping up its input voltage. Application Note 728 from Texas Instruments [7] gives more information. If the regulator can operate with a 3 V input voltage, then the problem of cold cranking is eliminated entirely.

Fighting negative voltages As a designer you might think it amusing to imagine a vehicle owner or mechanic accidentally connecting the battery the wrong way around, but don’t laugh: it really does happen! Fortunately protecting against this kind of stupidity is not particularly onerous, and it does not take a genius to design the protection circuit shown in Figure 7 using a single diode. In practice however, the demands on this protection diode are far from light. This is because a vehicle is not just a collection of

Elektor Industry 1/2020

180345-Störquellen-EN.indd 48

03-02-20 16:00

exponential-decay elements connected in series, but instead is a complex agglomeration of different inductive loads that generate all sorts of spikes when switched on and off. As well as positive-going spikes, against which we install TVS diodes, we can also expect negative-going spikes at the inputs to our circuits, sometimes going as low as –200 V! This means that our choice of diode is rather critical. For the main supply the author has had good experiences with the Diodes Incorporated S5BC-13-F, while the BAS20 comes in very handy for protecting voltage attenuators and similar circuits. Standard diodes have long ago lost their monopoly in this kind of application scenario. Their relatively high forward voltage drop means that at high currents it is very desirable to replace them with more efficient components. Here designers have found a new field of applications for both P-channel and N-channel field effect transistors. Figure 8 shows schematically how one kind of protection circuit can be constructed.

Figure 7. Sometimes the simplest solutions are the best.

What more is there to say? This article has confined itself to 12 V and 24 V systems. As vehicle electronics becomes more and more complicated it is becoming increasingly common to see vehicles with 48 V system supplies. The specifications for these systems are not yet completely finalized, and so if in doubt consult the manufacturer of the vehicle. Automotive is far from being a niche market, and so it will come as no surprise to anyone that there is a huge amount of literature on the subject of automotive electronics. The ISO documents mentioned in the introduction to this article are certainly the standard references, although they are rather difficult to get hold of. In some countries it is possible to inspect standards for free in libraries or colleges, although this will often involve a long journey. Figure 8. A somewhat unusual application for a MOSFET.

As we said in the introduction: there are thousands of systems that crassly flout some or even all of the rules that we have discussed, but which nevertheless function without problems. Protect your design with a TVS diode and a couple of other components, and the likelihood is that everything will work fine. 180345-02

Weblinks [1] ISO 7637-2: http://www.smd.ru/upload/medialibrary/a3d/iso_7637_2.pdf [2] Load dump protection: http://www.vishay.com/docs/49748/49748.pdf [3] Load dump and cranking protection: http://www.ti.com/lit/an/snva681a/snva681a.pdf [4] Littelfuse automotive TVS diodes: https://www.littelfuse.com/~/media/electronics/application_notes/littelfuse_tvs_diode_ automotive_circuit_protection_using_automotive_tvs_diodes_application_note.pdf.pdf [5] Automotive protection devices: http://www.st.com/content/st_com/en/products/protection-devices/automotiveprotection-devices.html [6] Transient suppressors, a competitive look: http://www.vishay.com/docs/88444/tvscomp.pdf [7] Output voltage regulation during automotive cold crank: http://www.ti.com/lit/an/snva728/snva728.pdf

Embedded & Tools

180345-Störquellen-EN.indd 49

49

03-02-20 16:00

Ultra-Compact AutomotiveGrade MOSFETs Provide Superior Mounting Reliability

By Jochen Hüskens, Senior Product Marketing Manager ROHM Semiconductor Europe GmbH

With the RV4xxx series, ROHM offers ultra-compact 1.6x1.6mm size MOSFETs. The AEC Q101qualified components are characterized by high reliability during assembly and guarantee high reliability and performance in automotive applications even under extreme conditions. In addition, ROHM’s Wettable-Flank technology guarantees the electrode height of 130 μm required for automotive applications.

Many of the latest vehicles are equipped with a host of features, but new functions are continually being added using a variety of technologies. Examples include large instrument panels, animated LED head/rear lamps, high end audio systems, and ADAS (Advanced Driver Assistance Systems).   50

ADAS in particular is being developed for practical realization in many countries around the world, with onboard cameras playing an important role in the configuration of these systems. Cameras mounted at the front, rear, left, and right of the vehicle allow drivers to view the surrounding environment on the

instrument panel, significantly improving safety.   The market for vehicle cameras, which totaled around 100 million in 2018, is expected to increase three-fold by 2025. The reason for this rapid expansion can be attributed to not only the growing

Elektor Industry 1/2020

191225-MOSFETs-EN.indd 50

03-02-20 16:01

Figure 1: Onboard camera locations.

Figure 2: RVxx5 lineup.

number of camera-equipped vehicles, but an increase in the number of cameras used per vehicle as well. A number of today’s vehicles incorporate 4 cameras, but this is sure to increase as functions such as ADAS become standard and more sophisticated (Figure 1).

The demand for greater miniaturization Recent years have seen a demand for

more compact vehicle cameras. This is largely due to the increasingly smaller available installation space following the trend towards electrification coupled with the growing number of cameras mounted per vehicle to further improve safety. Decreasing the size of automotive cameras allows for greater flexibility regarding the mounting position/location, minimally impacting vehicle designs.  

Current cameras are mounted on a circuit board with outer dimensions measu­ring 20 x 20 mm. In the coming years even smaller iterations are sure to be developed. At the same time the size of PCBs is shrinking, which in turn demands smaller components.   However, increasing functionality and higher resolution require larger currents. In addition, components to pro-

Figure 3: New ultra-compact MOSFETs for reverse-connection protection.

Embedded & Tools

191225-MOSFETs-EN.indd 51

51

03-02-20 16:02

tect against reverse connection are being mounted to prevent damage to the camera system in the event a reverse voltage is applied to the power supply line. In the past Schottky barrier diodes (SBDs) were used, but with high-VF SBDs heat generation becomes problematic at large currents. As a result, many designers are replacing SBDs with MOSFETs that generate much less heat.

Adopting a bottom-electrode package ROHM commands a large share of the discrete semiconductor market which includes transistors and diodes. Smallsignal transistors and diodes in particular enjoy a high market share and have been adopted in a variety of applications worldwide.   In response, ROHM is focused on smallsignal MOSFETs featuring a bottom-terminal package. This increases the electrode surface area during solder bonding, minimizing heat generation even if high power is supplied. The result is significantly higher power handling capability over conventional leaded packages.   ROHM’s new RV4xxx series of MOSFETs (Figure 2) optimized for reverse connection protection is offered in the compact bottom electrode DFN1616 package (1.6 x 1.6 mm) that meets the demands of automotive cameras for high current handling capability in a compact size. The result is 78% smaller mounting area over conventional SBDs and 68% even when compared to standard leaded MOSFETs with the same specifications. ROHM engineers were faced with a not insignificant challenge during development. However, with the conventional DFN1616 package, it is difficult to verify solder condition via visual inspection after mounting. To solve this problem, designers in recent years have turned to wettable-flank technology. Wettable-flank technology is a technique for plating the sides of the lead frame. Products using this technology exhibit superior toe fillet formation after mounting. The greater the height of the toe fillet, the easier it is to verify solder conditions during visual inspection, improving mounting reliability.

52

Figure 4: Wettable flank technology.

Figure 5: Wettable flank technology comparison.

Figure 6: Mounting visibility comparison.

Elektor Industry 1/2020

191225-MOSFETs-EN.indd 52

03-02-20 16:02

Minimizing burrs Normally, forming a wettable flank involves cutting a notch into the lead frame using a blade before carrying out the plating process. The depth of this notch can significantly affect the height of the toe fillet after mounting. To achieve a consistent toe fillet height this notch must be sufficiently deep, but unfortunately the amount of burrs increases in proportion with this depth.   And when mounting, these burrs become part of the mounting surface, possibly causing package tilt or even connection failure. For larger packages this poses no problems, but the smaller the package, the greater the susceptibility (Figure 5). After taking these burrs into account ROHM was able to achieve a guaranteed toe fillet height of 100 μm after mounting.   As a result, ROHM’s new RV4xxx series has been completely redesigned utilizing wettable flank technology. Perhaps the biggest difference from conventional products is that a material (barrier layer) with a hardness different from the lead frame is applied to the surface of the lead frame. This makes it possible to reduce burrs by approximately 50% over conventional methods.

ROHM’s RV4xxx series significantly improves mounting visibility Conducting a thorough review of construction methods allowed ROHM to develop the industry’s first package (DFN1616) featuring a guaranteed electrode height on the side of the package of 130 μm by suppressing burrs. Currently, a toe fillet of 100 μm may result in NG during visual inspection, but the RV4xxx series is able to sufficiently clear the fillet height required by the automotive market, ensuring superior mounting visibility (Figure 6).  

Conclusion ROHM has a proven track record of developing and launching products ahead of the market, including ultra-compact MOSFETs. And going forward, ROHM will continue to develop compact products that leverage this technology, including bipolar transistors and diodes, allowing us to expand our extensive product portfolio and achieve greater miniaturization along with higher reliability.

Advertisement

Especially in the automotive market which demands high mounting reliability, a minimum solder height is required. But engineers encountered another obstacle in stabilizing this height.

You CAN get it... Hardware & software for CAN bus applications

Hall 1, Booth 483

191225-01

PCAN-MicroMod FD Universal plug-in module with I/O functionality and CAN FD interface. Available with an evaluation board for developing custom applications.

The Author Jochen Hüskens has been working in the semiconductor market for more than 20 years. Responsible in ROHM`s product marketing division as Senior Product Manager for discrete components (diodes, transistor) and optodevices (LED, sensors, LD), including distribution DACH responsibility for power products (SiC MOSFETs & SBDs), IGBT. In recent years worked also as a business development manager for automotive accounts.

PCAN-MiniDiag FD Compact handheld for basic diagnosis of CAN and CAN FD buses. Measurement of bit rate, termination, bus load, and voltages at the D-Sub connector.

PCAN-M.2 CAN FD interface for M.2 slots. Available with up to four channels incl. software, APIs, and drivers for Windows® and Linux.

Otto-Roehm-Str. 69 64293 Darmstadt/Germany Phone: +49 6151 8173-20 Fax: +49 6151 8173-29 [email protected]

Embedded & Tools

191225-MOSFETs-EN.indd 53

53

03-02-20 16:02

Long-range Wireless STM32WL Microcontrollers Complementing the STM32 RF connectivity portfolio, the STM32WL System-On-Chip integrates both an ultra-low-power general-purpose microcontroller and a LoRa sub-GHz radio on the same chip.

Built on an Arm® Cortex®-M4 core architecture, STM32WL microcontrollers also support multiple modulations– LoRa®, (G)FSK, (G)MSK, BPSK — to ensure flexibility in wireless applications with LoRaWAN® or any other suitable protocol in a fully open way. STM32WL microcontrollers feature a LoRa-compliant radio (with multiple ST patents pending for, radio power management Architecture, among other things) to meet the requirements of a wide range of Low-Power Wide Area Network (LPWAN) wireless applications in industrial and consumer Internet-ofThings (IoT). The embedded sub-GHz transceiver comes with a LoRaWAN® radio stack available on demand. Thanks to a deep integration, the innovative and open architecture is optimized for LoRaWAN® legacy / proprietary protocols, flexible resource use, power management and helps lower BOM cost while offering a better user experience.

use and reliability, while being perfectly tailored for a wide range of industrial and consumer applications.

Wireless connectivity Developed using the same technology as the one implemented in ST’s ultralow-power STM32L4 microcontrollers, the STM32WL series provides similar digital and analog peripherals for basic or complex application use cases requiring an extended battery life and a long RF range through its sub-GHz transceiver. To ensure worldwide compatibility, the STM32WL MCUs feature a dual-power output and a wide linear frequency range fitting any unlicensed RF spectrum need. Overall, the STM32WL series is the STM32 family’s pioneer in sub-GHz wireless connectivity, offering ease-of-

54

STM32WLE5 microcontrollers comply with the physical layer requirements of the LoRaWAN® specification released by the LoRa Alliance®. Available LoRa®, (G)FSK, (G)MSK and BPSK modulations can also be used in legacy and proprietary protocols. The radio is suitable for systems targeting compliance with radio regulations including, but not limited to, ETSI EN 300 220, FCC CFR 47 Part 15, China regulatory requirements and the Japanese ARIB T-108. Continuous frequency coverage from 150 to 960 MHz enables the support of all major sub-GHz ISM bands around the world.

System peripherals The STM32WLE5 line includes a wide variety of communication features including up to 43 GPIOs, an integrated SMPS for power consumption optimization, and multiple low-power modes to maximize battery life. A dual-power output and a wide linear frequency range ensure worldwide compatibility.

Security & ID In addition to its wireless and ultralow-power features, STM32WL microcontrollers include embedded security hardware functions such as 128- / 256-bit AES hardware encryption, PCROP (Proprietary Code Read-Out Protection), and public-key cryptography with an elliptic curve encryption engine. 200017-01

Elektor Industry 1/2020

200017-STM32WL-EN.indd 54

03-02-20 16:02

PCAN-MicroMod FD Series It All Starts with the Processor and Evaluation Board

With their PCAN-MicroMod FD, the German company PEAK-System Technik has released a universal solution for the integration of a CAN FD interface and I/O functionality into custom hardware. Developers can plug the CPU module into their motherboard and configure it with the PCAN-MicroMod FD Configuration Windows software included in the scope of supply. The configurations are transferred to the PCAN-MicroMod FD via the CAN bus. Several modules can be configured independently of each other in a CAN bus. In addition to mapping the signals to CAN messages, the software enables a variety of operations and processing functions for analog and digital I/O signals. Once the configuration has been transferred, the PCAN-MicroMod FD forms an autonomous CAN node with the motherboard that can forward the information from sensors, actuators, and switches via the CAN bus. Together with the PCAN-MicroMod FD, PEAK-System is launching an evaluation board onto the market that is intended to facilitate the development of an individual motherboard. With the so-called PCAN-MicroMod FD Evaluation Board, users can access all resources of the attached PCAN-MicroMod FD via taps, screw terminals, switches, and potentiometers and test configurations or circuits. The PCAN-MicroMod FD can now be purchased individually or together with the PCAN-MicroMod FD Evaluation Board. Furthermore, a kit is available that contains a CAN FD interface for the USB port and a CAN cable in addition to the processor and evaluation board.

Robust I/O Devices with CAN FD

In addition to its use as plug-in module in in-house development, the new PCAN-MicroMod FD I/O module from PEAK-System is now also available with ready-to-use motherboards in black aluminum profile casings. The devices offer peripherals in variants for specific requirements. The data exchange is handled through CAN FD, which is downward compatible to classic CAN 2.0. The PCAN-MicroMod FD Analog 1 motherboard emphasizes analogue inputs and outputs, where 8 inputs, among others, have a resolution of 16 bits. On the Digital versions of the motherboards, 8 outputs are provided either as Lowside switches (Digital 1) or High-side switches (Digital 2). All motherboards have one analogue input in common for voltage monitoring up to 30 volts and two additional frequency outputs up to 20 kHz. The configuration of the motherboards with PCAN-MicroMod FD is done comfortably with the supplied Windows software and is transmitted via CAN. The motherboards then run as independent CAN nodes. The cabling of the devices is done using spring terminal connectors. Operation is possible in the extended temperature range from -40 to +85 °C (-40 to +185 °F). 200024-01

www.peak-system.com/PCAN-MicroMod-FD.500.0.html?&L=1 www.peak-system.com/PCAN-MicroMod-FDEvaluation.501.0.html?&L=1

Embedded & Tools 55

200030-News-Peak System-EN.indd 55

03-02-20 16:03

Research Project “KI-FLEX” How new AI-based electronic solutions can make autonomous driving safer Contributed by Fraunhofer IIS

Fully automated and autonomous vehicles should be able to respond appropriately in every situation. Together with partners in the “KI-FLEX” project, the Fraunhofer Institute for Integrated Circuits IIS is developing a platform that uses artificial intelligence methods to help measure vehicle position and determine vehicle environment in the future.

As part of the “KI-FLEX” project funded by the German Federal Ministry of Education and Research (BMBF), Fraunhofer IIS is leading the development of a software programmable and reconfigurable hardware platform that processes sensor data with AI-based methods for autonomous driving. The project

56

is a key step in the development of technology components that are urgently required to make autonomous driving safe and reliable.   Autonomous driving is dependent on the fast and reliable processing and merging of data from laser, camera and radar sensors in cars. As a result, the vehicle always has an accurate picture of the actual traffic conditions, can locate its own position in this environment, and, on the basis of this information, make the right decision in every driving situation. The data that the vehicle must process to determine its environment is so complex that artificial intelligence methods are required to ensure a high level of safety on the roads.   To this end, Fraunhofer IIS and partners are working on the “KI-FLEX” project to develop a powerful hardware platform and the associated software framework. The algorithms used for sensor signal processing and sensor data fusion are largely

Elektor Industry 1/2020

191257-KI-FLEX-EN.indd 56

03-02-20 16:04

based on neural networks and enable the vehicle’s exact position and environment to be determined.

Reconfigurable, secure and efficient

Future-proof, neuromorphic technology component

Project consortium comprising research and industry partners

The planned platform is a new development in the field of neuromorphic hardware, the functionality of which is inspired by the human brain and specially designed and optimized for the efficient use of neural networks. A key consideration in the project is that while product cycles in the automotive sector are very long, AI algorithms are advancing very rapidly. The project partners are therefore working towards a hardware platform that can be quickly and easily adapted to new software and hardware requirements in the field of machine learning. To achieve this, they are focusing on using a flexibly programmable multi-core deep learning accelerator in the form of a specially developed chip (ASIC). Such ASICs help reduce costs and power consumption compared to conventional multi-purpose processors (CPUs) or graphics processing units (GPUs). On that basis, the project is playing an important role in driving forward both science and the automotive industry in the field of autonomous driving.

The joint “KI-FLEX” project which runs through August 2022, is funded by the German Federal Ministry of Education and Research (BMBF) within the framework of the guidelines on promoting research initiatives in the field of “AI-based electronic solutions for safe autonomous driving (AI element: autonomous driving).” Fraunhofer IIS is leading the project consortium, which comprises the following research and industry partners: Ibeo Automotive Systems GmbH, Infineon Technologies AG, videantis GmbH, Technical University of Munich (Chair of Robotics, Artificial Intelligence and Real-time Systems), Fraunhofer Institute for Open Communication Systems FOKUS, Daimler Center for Automotive IT Innovations (DCAITI, Technical University of Berlin) and FAU Erlangen-Nürnberg (Chair of Computer Science 3: Computer Architecture). 191257-01

Advertisement

Embedded & Tools

191257-KI-FLEX-EN.indd 57

57

03-02-20 16:04

Advertisement

The relevance and usability of individual sensors varies depending on the traffic situation and on the weather and light conditions. To account for this, the platform is being designed as software programmable and reconfigurable hardware, which means that the algorithms used for sensor evaluation can be switched in line with changing driving conditions. This enables the vehicle to respond flexibly if individual sensors are compromised or if they fail. In addition, the project team will develop suitable methods and tools for ensuring the functional safety of the AI algorithms used and their interactions, even if the algorithms are reconfigured while the vehicle is on the road. To enable all algorithms and reconfigurations to be executed efficiently, the hardware platform’s computing resources are allocated dynamically according to load.

Has the Hydrogen

Car Lost Out on the Electric Car or Will There Be a Comeback?

By Richard de Jong (The Netherlands)

It seemed to be an exciting race, the car running on hydrogen versus the one running entirely on electricity. Until the sheer tension took a disastrous turn. In Sandvika, near Oslo, a high-pressure tank at a hydrogen fuel station exploded in mid-2019. Two motorists were lightly injured by the pressure wave, but this could have ended completely differently with the hydrogen stored under high pressure in the tanks. What does this mean for the race between the hydrogen car and the EV?

Recap: what happened in Sandvika?

Is hydrogen so incredibly dangerous?

Bck in June 2019, a high-pressure tank in Sandvika started leaking, causing an explosion. How the gas eventually came into contact with a spark or other form of ignition, is unknown. Because the storage tanks are not located near the filling points themselves but are stored in a separate place with a strong wall to prevent pressure waves and large explosions, the damage caused by the explosion was limited.

Hydrogen is dangerous because hydrogen gas is highly flammable. It needs a simple spark to ignite immediately. Hydrogen gas is already highly flammable, but it is not stored in the tanks in the ’normal’ state. It is stored under high pressure, which increases the risk of an explosion considerably.

The damage to the hydrogen car’s image is another story. A video clearly shows how the emergency services remain at a safe distance. There was great fear among the local residents, who had no idea what the risks were. The fact that an expert from Denmark had to be flown in to determine the risks for other stations is of course a sign on the wall too.

58

Does this make hydrogen gas an immediate major risk? Technically not, as the properties and risks are known. But when specially designed filling stations explode despite all the precautions, you can only imagine why people don’t want high-pressure hydrogen in their residential area.

What progress with those hydrogen cars? The hydrogen car has not yet conquered the world. And with the current supply of hydrogen-powered cars and the number

Elektor Industry 1/2020

191256-Hydrogen Car-EN.indd 58

03-02-20 16:05

of hydrogen-powered cars on the road, we can safely say that this will not happen in the short term either. The hydrogen-powered car has completely lost out on the electric car now with more than 20,000 electric cars on Dutch roads in 2018 compared to 13 hydrogen-powered cars. These figures say enough. Especially when we put the number of hydrogen powered models on a (very small) list: the Nexo and the Mirai. In addition, there are many developments, but few changes for the Dutch market.

charging stations alone is more than 18 times as high as the number of hydrogen stations. And that’s not all. Do you have an electric car and would you like to charge the car when you sleep or work? Nowadays, a simple type-2 EV charge cable is usually enough. Whether you are investing in a charging station at home or at work, you will be ready immediately and be able to charge significantly faster without having to create a separate electrical group to cope with peak demand while charging the EV.

Of course, this is not only due to the explosion in Norway. The moderate (or non-existent) success of the hydrogen car has to do with ambiguities, thresholds, an insufficient number of hydrogen filling stations and the enormous costs associated with the hydrogen car. Let’s get these factors straight.

Uncertainties: EV versus hydrogen car When it comes to ambiguities, the hydrogen car takes a back seat to the combustion engine cars and the EV. Ask an average person about the operation or nature of electricity or fossil fuels and an answer will come within seconds. Then ask the same person about the operation of a hydrogen-powered car and chances are that the answer is not there or at most there is an incoherent story. Hydrogen is relatively unknown. There are even people who do not know whether hydrogen itself is harmful, whether it is flammable or how it can be extracted. So how likely is it that these people choose to put their child and partner in a hydrogen-powered car?

Thresholds: EV versus hydrogen car When we talk about the thresholds, we mean the technological challenges that remain before a certain technology can conquer the world. Here, too, hydrogen is losing ground. At the moment, for example, it is even the case that a bus on green hydrogen produces 3.8 times as much CO2 emissions per kilometre than a bus on grey hydrogen. Right, the grey variant is more environmentally friendly than the green variant. This has everything to do with the way in which the hydrogen is extracted, transported and stored in order to be offered to people who want to drive ‘hydrogen’. The technology is far from where it should be to make hydrogen green and profitable, which makes driving on hydrogen far from attractive at the moment. That does not apply to the electric car. On the contrary, thanks to many tests, developments and innovations, battery life increases, the range of the average electric car grows and the vehicle can now even be charged ‘green’.

Petrol stations: EV versus hydrogen car Speaking of refuelling: the hydrogen car loses points again. We have four filling points in the Netherlands and there should be 11 more as a minimum. That means there will soon be fifteen filling points on a good 18 million people. Do you see yourself standing in line when everyone chooses a hydrogen car? Electric charging? That’s not really a challenge any longer. The Netherlands boast more than 48,000 charging bays. More than 280 stations can charge EVs rapidly. So the number of fast

Cost: EV versus the hydrogen car The Toyota Mirai comes with a price tag of around €80,000 (without options). With the Hyundai IX35 we can see the price difference even better. It had to cost €150,000, but the price eventually dropped to €67,000 because production was increased. Fine, but if you buy an IX35 with a fuel engine, the price is €25.000. That’s a difference of €42,000 and for more than forty thousand euros you’ll have a car that you can’t ‘refill’ anywhere because there are too few charging points. In addition, the hydrogen car loses out even further when we look at the second-hand options. The limited number of hydrogen-powered cars in the Netherlands means that the supply on the second-hand market is small. Also, because the business fleet never made the step to hydrogen. The business driver opted for electric and often on a lease basis. All these models are now available on the second-hand market. Dealer maintain electric cars at competitive prices, making the EV available to everyone. In contrast to the hydrogen-powered car that requires a second mortgage in an average household.

The offer: EV versus the hydrogen car When we’re looking for a new car, we’re not just looking at the fuel. We also look at the price, the model, the space and in rare cases even the colour. This is where the hydrogen car gets its ass kicked. There is a handful of models on the market, which for most motorists do not have all the features and are not within the price range within which they are looking for. If we look at the electric car, we’ll find a wide range of new

Embedded & Tools

191256-Hydrogen Car-EN.indd 59

59

03-02-20 16:05

city cars under €30,000, but we don’t have to go far up in price to find the bigger models for anyone who doesn’t live in the middle of Central and West Holland (Randstad). In fact, with the Ford Mustang Mach-E (probably around €50,000 at introduction) you can make the most of your time and do your shopping while driving from Groningen to Middelburg without having to recharge anything. And that’s not all. The list of electric cars already on the market is long. But since all car manufacturers are now really using, innovating and developing electric powertrains, from 2020 onwards we can really get our act together. What about the Mini Electric, the BMW IX3, or the Tesla Cybertruck that doesn’t look like a car at all anymore? Well, if Tesla can stick to the schedule this time, because it has to come on the market in 2021. No price announced, of course.

The future: is the hydrogen car going to make a comeback? We should not forget that developments in the field of hydrogen cars continue and that major steps are being taken. Explosions like those in Sandvika bring more knowledge with which to work. The hydrogen car is not a stillborn child, but at the same time we have to admit that a runaway race is hard to win. After all, the electric car is already firmly in the saddle. And then making a comeback becomes very difficult. After all, car manufacturers see that EVs sell well and that the average motorist knows the advantages of the electric car and wants to use them to benefit as well. A mist hangs over the hydrogen-powered car, partly thanks to its mediocre image. There are therefore few car manufacturers who are still fully working on this, although Audi is going to focus on hydrogen with the Audi h-tron to become a leader in this market. Well, if you can’t beat Tesla in their own market, you just change the focus, right? At the moment everything the electric car has the upper hand. The infrastructure is ready to charge anywhere in the Netherlands. With the many rapid chargers that can now also be installed at home and at work — with tax advantages for the employer — and the charging cables that can simply be used in any socket, range is becoming less and less important. Unless of course you have to drive more than 200 kilometres to get to work, but that chance is very small in the Netherlands.

How green is the future of our hydrogen and electric mobility? The hydrogen-powered car is not yet included when it comes to green mobility. Grey hydrogen is associated with a lot of CO2 emissions, green hydrogen needs so many operations that it only becomes an alternative when fossil fuels have been used up completely. Then we end up at the EV and there are different opinions. To help you in the next discussion where you end up on a birthday party: the direct CO2 emission of an EV is 0, because no CO2 is emitted by the engine while driving the car itself. The indirect emissions of the EV are higher. After all, electricity has to be generated, and most of the electricity in the Netherlands still comes from coal and gas. If you charge your car with green electricity, you can cross out this factor. If you’re driving on ‘grey’ power, the CO2 emissions are still more than

60

30% lower than if you were driving on fossil fuels. In addition, electric cars are heavier than fossil fuel cars, which may result in extra particulate matter being produced by the car tyres while driving. However, we are working hard on new types of tyres for electric cars, which should reduce this. Not perfect, but considerably better, especially if you fuel-up a little more ‘green-style’ in the future.

Ha, but then the battery comes and it’s not green at all! There are a lot of falsities online about the battery of the electric car. This would make the EV grayer than the petrol or diesel car and would cost a lot more energy than the production of a ’normal’ battery. Fortunately, Dutch TNO has also carried out research into this and we can quote concrete figures: upwards of 39,000 kilometres, the EV is more environmentally friendly than a fossil-fuelled car (depending on numerous variables, including type, driving style and size of the car). With an average service life of over 200,000 kilometres, there is therefore a clear advantage, despite the fact that extra batteries have to be produced. Especially if you include the increase in lifespan in the calculations.

Conclusion: the hydrogen car doesn’t make it to our driveways yet The hydrogen-powered car is an exciting journey with numerous obstacles that have yet to be overcome. When it comes to storage, transport, price and safety, there is still a lot to do before the EV can feel the hot breath of the hydrogen car in its neck. If that moment is ever going to happen at all, because the EV has already built up a huge lead. 191256-01

Elektor Industry 1/2020

191256-Hydrogen Car-EN.indd 60

03-02-20 16:05

Win up to a

€75,000 marketing package from Elektor and exposure at the electronica trade fair

COMPETE TO

LAUNCH YOUR STARTUP AT ELECTRONICA 2020 Is your electronics-related startup ready to disrupt? It is time to introduce your company’s solution to the global market! ENTER NOW! IT’S FAST AND SIMPLE

Visit elektormagazine.com/fastforward to read the Rules, Terms, and Conditions and to start an application. Good luck!

FINALIST BENEFITS First Prize = €75,000 marketing package from Elektor, plus a booth at electronica 2022. Second Prize = €50,000 marketing package from Elektor. Third Prize = €25,000 marketing package from Elektor. Get exclusive, specially priced (only €1,905!) booth space (5 m²) at electronica 2020. Get great exposure before, during, and after electronica 2020. (10.-13.11.2020)

>Sign up here electronica Fast Forward is brought to you by

elektormagazine.com/fastforward

ADV-FAST FORWARD 3START-UP AWARD-2020-A4 vs04 (1).indd 1 EI-20.01-Adv-EN.indd

24-01-2020 13:05 03-02-20 16:09

WALL FEBRUARY

Elektor International Media is proud to present its 2020 Wall of Fame! We partner with top electronics companies – from global parts suppliers to device manufacturers – to promote exciting products and services that power high-tech innovation. The companies listed on this month‘s Wall of Fame offer electrical engineers, makers, and students a wide range of indispensable tools and solutions for developing next-generation electronics projects and products.

62

Elektor Industry 1/2020

200000-Wall of Fame-EN.indd 62

03-02-20 16:06

WALL OF FAME 2020

Your logo could be here. Please call +49 (0)241 955 09 186

Embedded & Tools

200000-Wall of Fame-EN.indd 63

63

03-02-20 16:06

welcome in your

EDITOR’S CHOICE

ONLINE STORE

Elektor Bestsellers 1. Learning Python with Raspberry Pi www.elektor.com/19106

Andonstar AD407 Digital Microscope with 7’’ Display With its large 7” colour LCD display, the new Andonstar AD407 model is a stand-alone microscope. In the Elektor laboratory we used to add a large screen to the predecessor called ADSM302.

2. Raspberry Pi 4 B www.elektor.com/rpi4

The

AD407 eliminates the need for an external

3. The Ultimate Compendium of Sensor Projects www.elektor.com/19103

screen, freeing up space on the workbench. This can be sufficient reason to give the AD407 a clear preference if and when a new device is going to be purchased. Our

4. Andonstar AD407 Microscope www.elektor.com/19079

budget allowing, we’d replace the existing ADSM302 USB microscope with the new AD407.

5. Raspberry Pi Zero WH www.elektor.com/18567

Luc Lemmens (Elektor Labs)

6. Mendocino Motor AR O-8 www.elektor.com/19129

www.elektor.com/ad407-hdmi-digital-microscope-with-7-lcd-screen

7. Elektor SDR Shield 2.0 www.elektor.com/18515

The Ultimate Compendium of Sensor Projects

KiCad Like a Pro

SDR Hands-on Book

This book is about developing projects using the sensor modules with Arduino Uno, Raspberry Pi and ESP32 microcontroller development systems. More than 40 different sensor types are used in various projects described in the book. The book explains in simple terms how to implement the sensors in your project, using tested and fully working example projects.

This book will teach you to use KiCad. Whether you are a hobbyist or an electronics engineer, this book will help you become productive quickly, and start designing your own boards. This book takes a practical approach to learning. It consists of four projects of incremental difficulty and recipes.

Elektor’s SDR-Shield (SKU 18515) is a versatile shortwave receiver up to 30 MHz. Using an Arduino and the appropriate software, radio stations, morse signals, SSB stations, and digital signals can be received. In this book, successful author and enthusiastic radio amateur, Burkhard Kainka describes the modern practice of software defined radio using the Elektor SDR Shield. He not only imparts a theoretical background but also explains numerous open source software tools.

Member Price: €31.46

www.elektor.com/19103

64

Member Price: €35.96

www.elektor.com/18822

Member Price: €26.96

www.elektor.com/18914

Elektor Industry 1/2020

Ei-20.01-Store-EN.indd 64

03-02-20 16:06

SHOPPING

BOOKS

CD/DVD

DIY PROJECTS

DEVELOPMENT TOOLS

SALE

Learning Python with Raspberry Pi This book teaches the Python programming language using the Raspberry Pi 4 computer. The book introduces the Raspberry Pi 4 hardware and then teaches Python using the topics: variables, strings, arrays, matrices, lists, dictionaries, user functions, flow of control, printing, keyboard input, graphics, GUI, object-oriented programming, and many more.

NEW

The book is aimed at beginners, students, practicing engineers, hobbyists, and anyone else wanting to learn Python programming. The book includes many example programs and case studies. All example programs and case studies have been fully tested by the author and can be proven to work.

Member Price: €31.46

www.elektor.com/19106 JOY-iT 3-in-1 Device (Oscilloscope + Signal Generator + Multimeter)

Elektor ESP32 Smart Kit Bundle

Camera Projects Book

This practical and flexible 3-in-1 device combines the functions of an oscilloscope, a function generator and a multimeter. Two batteries type 18650 allow operation for about one day. The batteries are charged via a USB-C port which can also be used to operate the device during charging.

Experimenting with the ESP32 was never easier! With the Elektor ESP32 Smart Kit Bundle, you’ll absorb programming with the ESP32 IoT Microcontroller using the Arduino IDE and MicroPython programming languages. The kit consists of the highly popular ESP32 DevKitC development board, breadboard, sensors, LEDs, LCD, etc. for use in your experiments.

The book explains in simple terms and with tested and working example projects, how to configure and use a Raspberry Pi camera and USB based webcam in camera-based projects using a Raspberry Pi.

Member Price: €215.10

www.elektor.com/19157

Member Price: €62.96

www.elektor.com/19033

Member Price: €26.96

www.elektor.com/18943

Embedded & Tools

Ei-20.01-Store-EN.indd 65

65

03-02-20 16:06

Our Contributors SCHURTER AG . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . “Zero Energy” Indicators | Bruno Zemp

Advertisers congatec ......................................................... 1, 67 Arrow ................................................................... 2

SIGLENT TECHNOLOGIES . . . . . . . . . . . . . . . . . . . . . . . . . . SIGLENT SDS2000X Plus Series Super Phosphor Oscilloscope

congatec . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Whispers about the Computer-on-Module standard COM-HPC | Zelkjo Loncaric

Würth Elektronik .................................................... 3 SCHURTER AG ........................................................ 7 PeakTech ............................................................. 15 SIGLENT .............................................................. 29 Trinamic ............................................................. 31

Rohde & Schwarz . . . . . . . . . . . . . . . . . . . . . . . . . . . . . EMI Debugging with an Oscilloscope | Marcus Sonst

Mouser Electronics . . . . . . . . . . . . . . . . . . . . . . . . . . . . RISC-V: What’s All the Hype About? | Mark Patrick

Incooling ............................................................. 43 PEAK-System ....................................................... 53 STMicroelectronics ............................................... 57 electronica Fast Forward ...................................... 61 multi-cb .............................................................. 68

Arrow Electronics . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Technology Solution Provider – Embedded Platforms | Amir Sherman

Trinamic . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Low-Voltage Motor Control | Bernhard Dwersteg

Microchip . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Microchip Simplifies Hardware-Based IoT Security with the Industry’s First Pre-Provisioned Solutions for Deployments of Any Size

Johanson Technology . . . . . . . . . . . . . . . . . . . . . . . . . . . Miniaturized Wireless Medical Wearables | Manuel Carmona

RISC-V . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Questions, Answers, … and Embedded World 2020

Eurocircuits . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Expert Paper: Tips and Tools for PCB Designers

ROHM Semiconductor Europe GmbH . . . . . . . . . . . . . . . . . . . . Ultra-Compact Automotive-Grade MOSFETs Provide Superior Mounting Reliability | Jochen Hüskens

STMicroelectronics . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Long-range Wireless STM32WL Microcontrollers

Fraunhofer IIS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Research Project “KI FLEX”

Infographics . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Infographics | Robert van der Zwan

Elektor . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Review: LimeSDR Mini | Mathias Claussen

Elektor . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

Next Edition Edition 2/2020 of Elektor Industry magazine will cover sensors as well as equipment and methods for test and measurement, with contributions from companies, industry specialists, Elektor editors, and free-lance authors. Edition 2/2020 has a special focus on the Sensor + Test trade show in Nuremberg, Germany.

Interference Sources in Automotive Applications | Tam Hanna

Elektor . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Has the Hydrogen Car Lost Out on the Electric Car or Will There Be a Comeback? | Richard de Jong

66

The issue will be available to Elektor GOLD Members in print, and Elektor GREEN and GOLD Members as a PDF download at www.elektormagazine.com. Publication date: 18 June 2020.

Elektor Industry 1/2020

200000-Contributors-EN.indd 66

03-02-20 16:07

EI-20.01-Adv-EN.indd 4

03-02-20 16:10

PRINTED CIRCUIT BOARDS

NEW:

1-CLICK STENCIL ORDER Simply add to the PCB

Hightech PCB

Precision SMD-Stencil

Inclusive 0.1mm tracks 0.2mm drills

or

or

2 layers e.g. 1pc. 80mm x 100mm

€ 35,00 net € 41,65 gross

8 WD

€ 29,80 net € 35,46 gross

+ SMD-Stencil

UNIT PRICE

TOTAL PRICE

€ 9, 1 WD €11,

90 net

78 gross

incl. German VAT 19%

ALREADY INCLUDED

ALREADY INCLUDED

0.1mm tracks & annular ring, 0.2mm drills, FR4 1.55mm 35µm Cu, surface HAL lead-free, 2x solder-stop green, 1x legend print white, E-Test, Design Rule Check

Unlimited pads, thickness 100120µm, PCB name half-lasered on squeegee side, optional pad reduction, finishing treatment deburring on both sides, axial tolerance only ±2µm

in 4 WD

€ 44,90 net € 53,43 gross

in 8 WD

€ 39,70 net € 47,24 gross

YOUR ADVANTAGE

SMD-Stencil for €9,90 - Hightech makes it possible:

Our high-performance laser systems for SMD stencils meet the highest quality standards with optimum cutting quality (axial deviation only ± 2µm). All common quick clamping frames (for example ZelFlex, Paggen, VectorGuard, etc.) and aluminum frames are available. All SMD-Stencils from Multi-CB are deburred on both sides. Additional finishing treatments include electropolishing and nano coating. Your advantage - Deburring avoids time-consuming reworking.

2 Layer PCB together with

Only 1x shipping

FR4 1.55mm

net

1 pc. 80 x 100mm HAL lead-free

gross

5 WD

incl. German VAT 19%

6 Layers €98,62 €116,

net

1 pc. 80 x 100mm

chem. Gold (ENIG)

gross

6 WD

€49,80 net €59,

26

gross

€93,10 net

€110,

9 WD

79

9 WD

gross

DEBURRING:

Special material e.g. • Rogers • FR4 HTg • Alu core (1.0 - 8.0 W/mK)

DISTRIBUTORS wanted PCB experts only

ALL-INCLUSIVE PRICING: - FR4 1.55mm 35µm Cu - Surface finish - 2 x solder-stop green - 1 x legend print white - E-Test, Design Rule Check - A.O.I. & X-Ray - Milling (inner, outer) - PTH slits

Calculate price and production time comfortably online: Circuit Boards HTg, HDI, HF, etc.

SMD-Stencils also clamping frames

• Custom layer-buildup • Thin laminates • PCB thickness up to 7mm • Backplanes, Backdrill • PCI cards, Goldfingers • Thick copper • Panel production

Surface finish e.g.: • HAL lead-free • chem. Gold (ENIG) • ENEPIG, Bondgold. ...

Comparing pays off, why pay more for less performance?

4 Layers €59,21 €70,

• Plated-through slits • Countersink holes • Z-axis milling • Plated halfholes • Sideplating

• UL certified (E198312) • IPC-A-600

SMD-Stencil

PCBs including 0.1mm Tracks, 0.2mm Drills, 4WD Standard! High-tech makes it possible without surcharge!

• Ultra finest-line • Filmless laser-direct-imaging (LDI) • Impedance control, from 4WD • HDI boards • Blind & Buried Vias • Microvias, also lasered • Via Filling / Via in Pad overplated • Stacked & Staggered Vias

2 layer PCB

e.g. 1pc. 100mm x 120mm

UNIT PRICE

4 WD

All from one source

1-48 layers | from 1WD Express

Flex & Rigid-Flex up to 12 layers

Contact England: Multi Circuit Boards Ltd. Business Centres, HC GB-Poole BH17 7FJ

Contact Germany: Multi Leiterplatten GmbH Brunnthaler Str. 2 DE-85649 Brunnthal

[email protected] Hotline DE / EN: +49 (0) 81 04/628 0 +44 (0)1425/489 111 Prices plus delivery costs from €8,03 gross/ €6,95 net. Sales only to business customers and public institutions.

www.multi-cb.com Alu core (IMS) 1 & 2 layers PTH

Zelflex, Quattroflex, Paggen, Essemtec, Metz, VectorGuard etc.

Payment options:

EI-20.01-Adv-EN.indd 2

Invoice

Prepayment

Directly from your data e.g.:

03-02-20 16:10