0 3 5 ) . & 3 - Power Systems Design · operating conditions, and then order free evaluation...

November 2018

Special Report: Iot + Wearables (pg 21)

䤀渀昀漀爀洀愀琀椀漀渀琀漀倀漀眀攀爀夀漀甀爀䐀攀猀椀最渀猀一伀刀吀䠀䄀䴀䔀刀䤀䌀䄀

VIEWpoint

Connecting the World Through a Network of Connected DevicesBy: Jason Lomberg, North American

Editor, PSD

POWERline

Precision is Key to Today’s Power Measurements

DESIGNtips

Voltage Dividers in Power SuppliesBy: Frederik Dostal, Analog Devices, Inc.

MARKETwatch

The Transportation Connected SensorBy: Kevin Parmenter, FAE Director North America, Taiwan Semiconductor

COVER STORY

Breaking Boundaries with

Infineon’s New GaN Solution

By: Eric Persson, GaN Applications, Infineon Technologies AG

TECHNICAL FEATURES

Wireless Charging

High Performance Design for Wireless Power TransferBy: Mark Patrick, Mouser Electronics

GAN Series Part 2 of 6

The Power and Evolution of GaN: By: Alex Lidow, CEO and Co-founder of EPC

SPECIAL REPORT:IoT + WEARABLES

Is that a Battery, is that a

Capacitor? – No, it’s Supercap

By: Takahide Morikane, Manager, Product Management Department, Capacitor Division, KEMET

Selecting Connector Solutions

for Electric Vehicles

By: Albert Culetto, Technical Support

Connectors & Cables

Driving the Gates of SiC

Cascodes is a Breeze

By: Zhongda Li, NPI Manager, UnitedSiC

IoT Systems Need High Integration

& Small Form Factor

By: Tony Armstrong Director of Product Marketing Power Product Group Analog Devices, Inc.

5 Steps to Secure Electrical Devices

in a Connected World

By: Max Wandera – CISSP, GLSC, Director of the Cybersecurity Center of Excellence at Eaton

A Challenge to the Potential of

Industrial IoT

By: Harry Sim, CEO, Cypress Envirosystems

Opening Up New Opportunities in

Long Life Battery Powered IoT

By: Ken Bednasz, VP Application Engineering, Telit

2

Wearable Devices Need to Change

their Power Consumption

By: Florian Bousquet, Market Development

Manager, Product Center Positioning, u blox

FINALthought

Steering Wheels in Self-Driving Cars?By: Jason Lomberg, North American Editor, PSD

Dilbert

48

6

34

4

Highlighted Products News, Industry News and

more web-only content, to:

www.powersystemsdesign.com

37

POWER SYSTEMS DESIGN 2018NOVEMBER

1WWW.POWERSYSTEMSDESIGN.COM

42

4513

8

48

COVER STORY

Breaking Boundaries with Infineon’s New GaN Solution (pg 9)

29

22

25

27

9

14

17


Nanocrystalline WE-CMBNC

MnZnWE-CMB

NiZnWE-CMB NiZn

MnZn/NiZnWE-ExB

HorizontalWE-CMBH

High VoltageWE-CMB HV

#CRYSTALCLEAR

The WE-CMBNC is a VDE certifi ed series of common mode chokes with a highly permeable

nanocrystalline core material. Despite the small size, it delivers outstanding broadband attenuation performance, high rated currents and low DC resistance values. Low profi le and high

voltage ratings can also be realized by the common mode chokes of the WE-CMB family.

For further information, please visit:

www.we-online.com/we-cmb

crystal clearfrequencies.

© e

iSos

High permeability nanocrystalline core material

High IR & low R

DC in a small size

Broadband suppression

Stable inductance values at high

temperatures

Improved isolation through plastic case and

patented winding spacer

electronica Hall B6 Booth 404

High permeability nanocrystalline core material

2

VIEWpoint

WWW.POWERSYSTEMSDESIGN.COM

POWER SYSTEMS DESIGN

The Internet of Things (IoT) will change the world. While the internet put the sum of humanity’s knowledge at a keystroke, the IoT will unite a plethora of systems and devices in a connected ecosystem. And the possibilities are endless.

The IoT requires small-form-factor power conversion devices – “the proliferation of wireless sensors supporting the numerous IoT devices has increased the demand for small, compact and efficient power converters tailored to space and thermal constrained device form factors,” notes Tony Armstrong from Analog Devices.

Armstrong discusses that and the fact that industrial and medical products “typically have much higher standards for reliability, form factor and robustness.”

Cypress Envirosystems focuses on the industrial IoT (or IIoT), which “promises to revolutionize our industrial infrastructure by improving efficiency at existing power plants, refineries, off-shore oil platforms, pharmaceutical plants, hospitals etc.”

And while the IIoT stands to unlock $6.2 trillion in potential economic value by 2025, “lack of data visibility in existing plants hinders Industrial IoT deployment.” To renovate legacy industrial systems for full IoT compatibility, plants must look towards non-invasive instrumentation upgrades.

“The only way to proceed is to look for alternative ways to retrofit manual instrumentation without the associated high cost and disruption,” claims Harry Sim from Cypress.

Meanwhile, Max Wandera with Eaton delves into “5 Critical Steps for Securing Electrical Devices in an Increasingly Connected World.”

Businesses spent $964 billion last year on IoT devices, and by 2020, 31 billion devices will be part of the fabled connected ecosystem.

Through their participation in UL’s Data Acceptance Program, Eaton is “helping establish measurable cybersecurity criteria for network-connected power management products and systems.”

Lastly, I’d like to highlight S3 Semiconductor’s piece, “The New Challenges of Industry 4.0” – aka, the fourth industrial revolution and a flagbearer for the IoT.

In short, “Industry 4.0 adds an extra layer of data exchange and automation to computerised manufacturing, drawing on Internet of Things (IoT) technologies to create smarter factories,” says S3’s Edel Griffith.

“The IIoT and Industry 4.0 are a natural fit. IIoT technologies are increasingly providing the capability to deploy high numbers of relatively low-cost devices, each including one or more sensors or actuators, and the ability to send that data to other devices or a central node.”

Best Regards,

Jason LombergNorth American Editor, [email protected]

Connecting the World

Through a Network

of Connected Devices

Power Systems Corporation 146 Charles Street Annapolis, MD 21401 USA Tel: +410.295.0177Fax: +510.217.3608 www.powersystemsdesign.com Editorial Director Jim Graham [email protected]

Editor - EuropeAlly [email protected]

Editor - North AmericaJason [email protected]

Editor - ChinaLiu [email protected]

Contributing Editors Kevin Parmenter, [email protected]

Publishing DirectorJulia [email protected]

Creative Director Chris [email protected]

Circulation Management Sarah [email protected]

Sales Team Marcus Plantenberg, [email protected]

Ruben Gomez, North America [email protected]

Registration of copyright: January 2004ISSN number: 1613-6365

Power Systems Corporation and Power Systems Design Magazine assume and hereby disclaim any liability to any person for any loss or damage by errors or ommissions in the material contained herein regardless of whether such errors result from negligence, accident or any other cause whatsoever.

Free Magazine Subscriptions, go to: www.powersystemsdesign.com

Volume 14, Issue 9

The Smart Choicefor Small Spaces

WWW.COILCRAFT.COM

Boost the performance and battery life of your wearable products with these tiny RF and power inductors from Coilcraft:• Wirewound chip inductors as small as

0201 size for wireless communications• Shielded power inductors as thin as

0.71 mm for power management• Coupled inductors as small as 2.2 x

1.45 mm for LED display drivers

Get started by using our suite of web tools to quickly locate the perfect parts for your design. Compare and analyze multiple inductors based on your actual operating conditions, and then order free evaluation samples with just a click.

Learn why we’re the biggest name in ultra-miniature inductors. Visit us at www.coilcraft.com.

Coilcraft is the brand engineers trust most when specifying tiny inductors for wearable technology

®


4

POWERline


Power design is more complex than it has even been, and that trend is set to accelerate due

to many factors. At board level, shrinking IC geometries demand higher currents in smaller spaces, as well as very precise slew rates and boot up sequences. At the higher power end of the scale, alternative energy sources and electric vehicles are designed to extract every available joule of en-ergy from the source. There is also a rise in the demand for complex electromechanical products, such as robotics, which demand precise voltages applied with strict timing limitations. As if those demands weren’t enough, designers also must tread a narrow path between getting the best design possible and complying with a slew of regu-lations on efficiency, interoperabil-ity and safety.

To meet all of these demands requires a comprehensive testing methodology, and reliable and precise equipment. Precision espe-cially is in demand as a few mil-livolts or milliseconds either way could reduce battery life signifi-cantly over time, fail to meet strict limits laid down in legislation or cause a nasty intermittent fault as a vital component fails to start up correctly. The number and variety of different tests that need to be

Precision is Key to Today’s Power Measurements

performed can often require a full benchtop of instru-mentation. Yokogawa has looked at the prob-lems facing power designers and de-veloped a solution that offers both high precision measurement, and a flexible modular approach which allows developers to create a bespoke instrument for each test. The Yokogawa WT5000 features an accuracy of ± 0.03% at 50/60 Hz along with high levels of stability and noise immunity, allowing the easy and accurate measurement of power consumption, loss, and efficiency for the device under test.

At the heart of the instrument is an 18-bit analog to digital con-verter with a sampling rate of 10MS/s, which allows the device to accurately capture waveforms and provide stable measurements from the seven input channels. In the past, capturing seven sepa-rate inputs would require several instruments to be linked and syn-chronized.

The seven inputs are designed to make multi-system measurements easier to perform. For example, the WT5000 can carry out two harmon-ic measurement functions simulta-

neously, each at up the 500th order and up to 300kHz fundamental waveform. This makes it possible to measure the carrier frequency from the rotational speed of a motor in the inverter drive and also to check the influence of the carrier frequen-cy on the motor drive.

The modular design of the WT5000 allows elements to be switched out depending on the requirements of the desired test. For example, the 30 A and 5 A elements can be switched to allow developers to evaluate a number of different mo-tors.

For larger current measurements, an external current sensor input function is fitted as standard in the input element of both the 30 A and 5 A input elements of the WT5000. For even larger currents of up to 2,000 A rms, a measurements can be taken using a dedicated high-current sensors can be used.

Yokogawahttps://www.yokogawa.com/

You can do more with THA Thinpack on board.

Using the most advanced aluminum electrolytic capacitor technology, type THA and THAS offer the highest energy density in a low profile. Design-in one component versus banks of alternative SMT capacitors for incredible size and cost savings. Improve reliability with one component versus many.

THA: 3,000 hr. life @ 85 ºCTHAS: 3,000 hr. life @ 105 ºC

Replaces banks ofSMT capacitors

Highest energy density inlow-profile: 1.1 J/CC

Uses about 70% less boardspace than alternatives!

For technical information and samples visit cde.com/THA

One Capacitor does the job of many in a fraction of the space.

6

DESIGN t ips


Voltage Dividers in Power SuppliesBy: Frederik Dostal, Analog Devices, Inc.

In the design of power sup-plies, the desired output voltages can be set manually. This is accomplished in most

integrated power supply circuits, as well as switching and linear regulator ICs, with the help of volt-age dividers. The ratio of the two resistance values must be suit-able to enable the desired output voltage to be set. Figure 1 shows a voltage divider. The internal refer-ence voltage (VREF) and the desired output voltage determine the ratio of the resistance values, as seen in Equation 1:

Especially in switching regulators, this behavior is critical, because noise arises due to the rapid switching of currents and can

couple to the feed-back node.

Useful resistance values for R1 + R2 lie between 50 kΩ and 500 kΩ depending on the noise expected from other circuit segments, the value of the output voltage, and the need to re-duce the power loss.

The reference voltage VREF is defined by the switching regulator or linear regular IC, and is usually 1.2 V, 0.8 V, or even 0.6 V. This voltage represents the lowest voltage that the output volt-age (VOUT) can be set to. With the reference voltage and output voltage known, there are still two unknowns in the equation: R1 and R2. One of

the two resistance values can now be selected relatively freely, as usual values lie below 100 kΩ.

If the resistance values are too low, the power loss due to the constantly flowing current VOUT/(R1 + R2) during operation is extremely high. If R1 and R2 each

had a value of 1 kΩ, then a continu-ous leakage current of 1.2 mA would flow at an output voltage of 2.4 V. This corresponds to a power loss of 2.88 mW generated by the voltage divider alone.

Depending on how precisely the out-put voltage should be set and how high the current in the power supply error amplifier at the FB pin is, Equa-tion 1 can be specified more precisely through consideration of this current.

However, the resistance values should not be too high. If the resis-tance values were each 1 MΩ, we would only have a power loss of 2.88 µW.

A major disadvantage of this resistor dimensioning with very high values is the fact that it results in a very high feedback node impedance. The

Figure 1: Voltage divider in a voltage regulator for adjustment of output voltage

Figure 2: Example of a well-placed voltage divider in a power supply

Figure 3: Adjustment of the output voltage without a continuous power loss in the voltage divider.

current flowing into the feedback node can be very low depending on the voltage regulator. As a result, noise can couple to the feedback node and directly affect the control loop of the power supply. This can

stop the regulation of the output voltage and lead to control loop instability. Another important aspect is the placement of the voltage divider on the board layout. The feedback node should be designed to be as small as possible so that very little noise can couple to this high impedance node.

Resistors R1 and R2 should also lie very close to the feedback pin of the power supply IC. The connection between R1 and the load is usually not a high impedance node and can hence be designed to have a longer

trace. Figure 2 shows an example of resistors placed close to the feed-back node.

To reduce the power loss of a voltage divider, especially in ultra low power applications such as energy harvesting, some ICs, like the ADP5301 step-down regulator, feature an output voltage setting function in which the value of a variable resistor at the VID pin is only checked once during startup. This value is then stored for on-going operation without current constantly flowing through a volt-age divider. A very sensible solution for highly efficient applications.

Analog Deviceswww.analog.com


Your Complete Power Solutions

www.aosmd.com

See us at Booth C4-102Powering a Greener Future™

Mobile/Consumer • Automotive • Industrial/Renewable • Cloud Computing

ALPHA & OMEGAS EM ICONDUCTOR

AOS_PDS_CompletePower_Nov18_Ad2.indd 1 10/3/2018 12:32:09 PM

8

MARKETwatch


By: Kevin Parmenter, PSD Contributor

With more electronics in the vehicles than ever before,

we’re seeing the industry move from a basic set of electronics applications, with some automation for safety and road assistance, to high and full automation and complete comfort driving. To make this work, the transportation industry is capitalizing on developments in other areas, including 3D video, virtual reality, AI and gaming (which helps innovation in self driving vehicles), smart cities and the Internet of things.

Navigant research says that 4.8 million autonomous vehicles will be in service by 2025. But to make this possible, innovation and rapid development of GPUs and video technology, sensors/ Lidar and more are taking place. Autonomous vehicles will have to be connected with anywhere from five to no more than 50 mS of latency to enable cooperative maneuvering and trajectory planning, predication of situation and intention, plus cooperative perception. This means connected cars will be communicating with the infrastructure: traffic

The Transportation Connected Sensor

signals, other vehicles, other sensors (smart city sensors), IoT devices and a backend server. And as the drive train and body electronics become ever more sophisticated, soon our cars will be communicating with the “cloud.” Meanwhile, the consumer electronics market is merging with transportation electronics more than ever, with hopes of making traffic accidents and vehicle theft a thing of the past. However, there are – and will be – many problems to be solved in the many interdisciplinary areas. The good news is that the technologies that are needed to drive all the associated parts and pieces are already evolving quickly. The leap from driver-assisted vehicles to autonomous vehicles is a progression; however, the growth in the electronics content along the way will provide lots of content growth in the transportation market.

At the top of the list for design engineers to consider will be safety and security. For instance, what if a bad actor or nefarious entity decides to get rid of someone or several ‘someones’

at once by taking control of a vehicle and driving it off a cliff? Or what happens if hackers are intent on wreaking mischief? Or, more mundanely, what if advertisers want to tap into the big data on the back end to access our personal information – with the vehicle becoming a connected mapping sensor who owns the data? Other problems to be solved involve reliability. With critical safety systems you need high reliability and ASIL-D functionality. Your cell phone can’t kill, but a vehicle can, so the entire system must be thought through very carefully.

The additional good news for those of us in power electronics is that all of these interrelated systems will need innovation in power to work efficiently, reliably and effectively with 5, 9’s uptime or more. From energy harvesting in IoT sensors, to communication systems, vehicular-based mobile power and propulsion to the 5G systems, power electronics will be a part of the innovation. That means we are integral the explosive growth in automotive infotainment and the trajectory electrified, connected and autonomous transportation.

Breaking Boundaries with Infineon’s New GaN Solution

By: Eric Persson, GaN Applications, Infineon Technologies AG

Introducing the 600 V CoolGaN™ with matching EiceDRIVER™

Today’s demand for high-performance, low-cost power conversion products is driven by

consumer expectation for longer battery life, faster charging for their phones, electric vehicles (EVs) or power tools. And consumers want faster data communication along with powerful artificial intelligence (AI) capabilities, delivered at lower cost from the vast network of hyper-scale datacenters, telecom server farms, and upcoming 5G com-munication towers integrated into our everyday living environment. At the heart of achieving high perfor-mance as well as lowering the up-front and operating costs for these applications, is advanced power electronics that process electric util-ity voltage through multiple stages to ultimately power our devices efficiently and cost-effectively.

For many years this progression has been enabled by continuous innovation resulting in families of high and low-voltage silicon power transistors that never fail to sur-prise us with further improvements generation after generation. Now Infineon continues this innovation by introducing a family of 600 V GaN power transistors and related

driver ICs at electronica 2018, the world’s leading trade fair for elec-tronic components, in Munich. This article addresses what GaN is, what it can do for us, and provides application examples of where it is used to reliably provide ultra-high efficiency at the lowest system cost.

What is GaN and how does it work?Gallium nitride (GaN) is a wide bandgap semiconductor material in the same category as silicon carbide (SiC). If it were possible to grow large-diameter single-crystal GaN to make wafers for processing, ver-tical transistors could be fabricated in a similar way as SiC MOSFETs are made today. But growing GaN substrates has proven to be pro-hibitively difficult. Instead it is far more effective to use readily-avail-able, low-cost silicon wafers as a

substrate, and grow GaN epitaxially so it can be fabricated into lateral transistors known as high electron mobility transistors (HEMTs).

Figure 1 shows the cross section of a CoolGaN™ ™ transistor based on this structure. Since GaN and Si have different crystal lattice con-stants and coefficient of thermal expansion, proprietary transition-layers are first grown to provide a suitable base on which to grow the GaN layers. At the interface be-tween the GaN and AlGaN layers, a sheet of free electrons, known as a 2-dimensional electron gas (2DEG) is formed. The 2DEG is highly conductive due to the exceptional electron mobility of this layer. Drain and source contacts can be added, and a gate structure that provides a local electric field is used to deplete or enhance the 2DEG, enabling the

Figure 1: Cross section of a 600 V CoolGaN™ ™ power transistor

COVER STORY


1110

COVER STORY

WWW.POWERSYSTEMSDESIGN.COM WWW.POWERSYSTEMSDESIGN.COM


to prevent spurious gate turn-on due to fast positive dv/dt on the drain (a C dv/dt induced turn-on due to the current injected by the gate-drain capacitance).

The negative bias can be supplied by a fixed negative supply to the gate driver, and/or negative bias supplied by the capacitor in the RC network.

But constant negative gate bias voltage has a disadvantage: when the GaN transistor conducts in the reverse direction during dead time between high and low-side conduc-tion, the effective voltage drop is approximately 2 V plus the negative gate bias. Therefore it is desirable to use only the smallest negative bias necessary to keep the gate fully turned-off, in order to minimize power dissipation when in third-quadrant conduction (diode mode) during dead time. Ideally, VGS would return to zero after the dead time.

This is where the new Infineon

1EDx5673 EiceDRIVER™ isolated gate driver comes-in. The 1EDx5673 and its 1EDS package variants all feature a unique three-level gate drive solution shown in figure 4, with the waveforms in figure 5. The new driver is among the fastest isolated drivers offered by Infineon, with a 37 ns propagation delay and a -6/+7 ns delay uncertainty for highly accurate driver timing necessary for adaptive dead time management at high frequencies. Moreover, the driver has a common mode transient immunity (CMTI) of >200 V/ns to handle the high switching speed that GaN transis-tors are capable of. This driver operates from a single supply volt-age, but provide negative gate drive

temporarily from the RC capacitor, or by inverting its output during startup or burst-mode. Otherwise the gate voltage is returned to zero to optimize the GaN third quadrant conduction at all other times. The negative drive duration is program-mable with a resistor on the TNEG input pin. These operational modes provide the best combination of speed, performance, and safety from spurious turn-on in any GaN gate driver available today. To ac-commodate smaller PCB layouts requiring only functional isolation, the driver is offered in both a 5x5 mm LGA and a 16-pin narrow-body DSO – as well as a 16-pin wide DSO fully rated for UL/VDE safety isolation.

Figure 3: Simple gate drive method using IGBT gate driver

Figure 4: The new high-performance EiceDRIVER™ isolated GaN gate driver

2DEG to essentially be turned on and off. Using p-doped GaN for the gate makes this an enhancement, rather than depletion-mode device – in other words, it is normally off, with a gate threshold of about +1.4 V.

Note the diodes shown on the gate structure. This combination of Titanium gate metal and p-GaN forms an ohmic contact in series with a GaN diode between the gate and source. The diode has a knee voltage of about 3.5 V in the forward direction. In addition, 3 series GaN protection diodes are monolithically integrated onto the GaN transis-tor to protect the gate from electro static discharge (ESD). These ESD diodes begin clamping when VGS goes more negative than about -10 V. In this way, the gate voltage of CoolGaN™ is self-clamping by in-tegrated diodes in both the forward and reverse direction, making it extremely rugged and immune to overvoltage spikes. Competitor’s enhancement-mode GaN struc-tures use a Schottky gate contact, which blocks the GaN diode clamp-ing function. As a result, the com-petitive devices require external circuits to manage and limit gate voltage transients which would oth-erwise damage the Schottky gate and lead to failures. Figure 1 also shows the p-GaN “hybrid drain” structure that effectively manages trapped charge, thus mitigating dynamic RDS(on) variations that can plague competing GaN transistor design approaches.

Another key competitive advantage of CoolGaN™ is its higher satura-

tion current capa-bility, especially at maximum tempera-ture (150°C) where it matters most. Figure 2 compares a 70 mΩ-rated Cool-GaN™ transistor versus a competi-tor 650 V, 50 mΩ-rated transistor. Even when driven to the maximum recommended gate voltage of 6 V, the competitor’s GaN is rated to only 25 A maximum. In contrast, the CoolGaN™ device is not only lower RDS(on) over the entire current range, but it is rated at 35 A maximum drain current, 40% greater than other devices in the market. This peak current handling capability is critical because many applications require the device to operate/survive in transient high-current conditions (startup, line-cycle dropout, lightning-strike transient, short-circuit faults, etc.). This combination of a rugged gate structure and higher peak current handling capability helps make Infineon’s CoolGaN™ the most rugged and reliable GaN power transistor on the market today.

Driving GaN – what is different about driving the gate?As described in the previous sec-tion, the CoolGaN™ ™ gate is self-clamping with a forward voltage of approximately 3.5 V. So instead of driving the gate-source with a fixed voltage, the gate is ideally driven

with a current source, and the resulting voltage on the gate simply follows the diode characteristic. For the 70 mΩ device described above, the typical steady-state gate current to fully enhance the gate is about 10 mA. But in order to quickly charge the gate capacitance, a much higher pulse current is necessary for about 20 ns. A 2-stage current-source driver could accomplish this, but a much simpler solution is to use a conventional gate driver with an R-C network to the gate (instead of the conventional R), as shown in figure 3. With this simple circuit, a conventional 12 V gate driver can be used to effectively drive the GaN gate. Not only does the RC network provide the pulse of peak current necessary to achieve fast turn-on, but it also then charges-up to VCC – VGS, and that charge results in a negative gate bias at turnoff.

The negative bias is actually benefi-cial as it provides additional margin

Figure 2: Comparing saturation current of CoolGaN™ ™ versus competitor

12

COVER STORY


GaN PFC reference design.

When operated in continuous conduction mode (CCM) transis-tors Q1 and Q2 operate alternately as boost switch and synchronous rectifier under hard-commutated turn-on conditions. If conventional silicon high voltage power transis-tors were used for Q1 and Q2, the reverse recovery characteristic of their parasitic body-diodes would incur huge switching losses lead-ing to their destruction. But GaN transistors have zero reverse recovery, as there are no minority carriers involved in channel con-duction. Thus the switching losses are quite low, especially since the output capacitance, and related energy is smaller than any compet-ing transistor technology.

The losses are so low, that the PFC circuit shown in Figure 6 can deliver efficiencies exceeding 99% over most of their output power range, as shown in the measured efficien-cy plot in figure 7. The 99.3% peak efficiency is truly extraordinary, rep-resenting only 7 Watts of loss when operating at 1 kW output. This PFC circuit is a real-world reference design from Infineon that passes all of the normal PFC requirements including EN5022 conducted EMI, line-cycle dropout, and lightning-strike surge tests.

In summary, Infineon’s new Cool-GaN™ offering represents the most rugged 600 V high-performance GaN transistors available today. When combined with Infineon’s coreless-transformer isolated Eice-

DRIVER™ gate driver family, power electronic engineers can now cost-effectively deliver higher perfor-mance power converters than ever before possible using conventional silicon transistors.

The new Infineon GaN transistors introduced today are the 600 V, 70 mΩ IGx70R060D1 available in 4 different surface-mount pack-

Applications – where GaN delivers highest performance:The first instinct for any power electronics design engineer is to try substituting a new GaN device for their existing CoolMOS™ – to see how much better it performs. But they are often disappointed that GaN only provides a modest im-provement in switching loss. Why is that – why doesn’t GaN necessar-ily provide a big improvement when dropped-into an existing circuit design? It is because these circuits where CoolMOS™ is commonly-used are often unipolar topologies like traditional boost PFC, single-ended flyback or two-transistor forward converter. In these topolo-gies, the common-theme is that the transistor only conducts in the forward direction. Thus, the body-diode performance is irrelevant because the transistor never con-ducts in the third quadrant. So the only parameters that really mat-ter besides RDS(on) are the turn-on energy loss EOSS, and the switching speed, and the small advantage of GaN here would not be worth the

benefit – CoolMOS™ is already a proven technology and does a great job in these applications.

So, if not unipolar topologies, where does GaN work best and provide the most benefit? It is commonly believed that GaN is at its best at high-frequencies. Yet it can offer extraordinary efficiency in half-bridge based topologies even at modest operating frequen-cies like 65 kHz. Shown in figure 6 is the full-bridge totem pole PFC topology used in Infineon’s 2.5 kW

Figure 5: The three-level gate drive waveforms provided by the new GaN EiceDRIVER™

Figure 6: The full-bridge totem pole PFC topology

INTRODUCING THE NEW XOETM DIODETECHNOLOGY

www.deantechnology.com

TO THEPERFORMANCEEXTREME

Contact DTI today to get a quote or place your order!

Dean Technology is excited to announce its new eXtreme Optimized E�ciency diode technology, XOETM. By tightly controlling all possible variables in the manufacturing process, and remaining focused on getting the maximum overall e�ciency from the silicon, diodes made with XOETM

technology have dramatically better performance over similar sized parts using the same materials.

XOETM diodes can be used in ALL applications where standard high voltage diodes are already used, and will help you devel-op new solutions with higher perfomance in smaller sizes.

Figure 7: Measured efficiency of Infineon 2500 W GaN totem pole PFC demonstration board

ages including top- or bottom-side cooled DSO, TOLL, and 8x8 DFN. In addition a 190 mΩ version in the TOLL package is offered. The product line will soon grow to cover the range of 35 mΩ, up to 340 mΩ in the same range of four packages.

Infineon Technologieswww.infineon.com/ganwww.infineon.com/eicedriver


1514

WIRELESS CHARGING



High Performance Design for Wireless Power Transfer

By: Mark Patrick, Mouser Electronics

Wireless charging must be efficient and safe

In the more compact portable and wearable devices now being introduced onto the market,

it is extremely difficult, or even impossible, to incorporate a charge connector. For example, some may simply have too little available space in which to house an appropriate connector. Other devices may require sealed enclosures to protect sensitive electronic circuits from adverse environments and allow convenient cleaning or sterilisation, while for medical wearables replacing a battery could require invasive surgery with risks to patient health thereby being entailed.

Many state-of-the-art battery-powered devices provide for the possibility of charging without the physical connection to a power supply through the use of wireless power transfer (WPT). The extraction of electro-magnetic energy by an antenna from the impact of external radiation is the basis of wireless communications and WPT. Maximum energy transfer is ensured when the antenna has adaptive conjugate properties,

i.e. when it is resonant and has a coupling with the free space equal to that of its load. This condition, however, can be easily influenced by changes in the environment, thereby preventing optimal functioning of a WPT system. How WPT works?The principle behind wireless charging is well known and now available in all transformers. The Faraday-Lenz law explains that the variation of the magnetic field generates a variation of potential at the ends of a loop. By using the inverse reasoning, a variation of potential at the ends of a loop can generate a variation in the magnetic field. Using mutual induction, it is possible to couple

two turns - the first has the task of generating the magnetic field, while the second generates an induced current directly proportional to the intensity of the magnetic field (as described in Figure 1).

The BD57015GWL and BD57020MWV from ROHM are wireless transmitter and receiver devices respectively. Made of ferrite cores, they use inductive magnetic coupling and require precise alignment between primary and secondary coils to achieve a secure connection. Movement restriction limits the use of this technology in portable electronic devices due to difficulties when it comes to maintaining high levels of

efficiency and the weight of the cumbersome magnetic materials required. Starting from the basic principles expressed in inductive coupling, it is possible to increase the transmission distances through the technique of resonant magnetic coupling. Highly effective wireless power transfer is achieved by magnetic resonance, which uses oscillating magnetic fields to exchange energy. A sizeable inductive spiral excited by a radio frequency (RF) source can exploit its resonance to induce a resonant mode in another similar structure, placed at a certain distance. In this way, a power transfer is obtained, without using a radiative field, on a distance that can even be 4 times the size of the spiral. The absence of ferrite cores make them lighter, and therefore easier to integrate. The efficiency is less than the inductive coupling, but still sufficient to guarantee its use with acceptable quality standards.

The Qi protocolInteroperability between various devices is made possible thanks to the presence of open standards, developed by the cooperation of various companies. The Wireless Power Consortium (WPC) developed the Qi system for the inductive transfer of electricity. It operates at frequencies between 100kHz and 200kHz, and consists of two primary modules - namely the base station and the mobile device. Its architecture is

represented in Figure 2.

The Qi standard defines the detailed requirements for the coils in the transmitter, the covering materials, the geometries and the electrical characteristics. This means that the products of different manufacturers are fully interoperable.

The base station can include one or more power transmitters, while the mobile device contains a single power receiver. The power transfer is obtained by inductive coupling between the primary and secondary coils. These coils realise a resonant transformer in the air. This is able to transfer electric power with an efficiency that is comparable to those of a metal conductor.

The transmitters and receivers from the main microelectronic

companies perform all the key control functions, ensuring compliance with the most critical standards on the market and providing customers with certified turnkey Qi solutions. The STWLC family of receivers from STMicroelectronics can efficiently charge any battery, such as single-cell Li-ion or Li-polymer, and support direct charging. The STWLC04 is an integrated wireless power receiver suitable for use in wearable applications. It is designed for 1W power transfer based on the Qi protocol, with digital control and I2C interface for a high degree of personalisation. Efficient power transfer is enhanced by keeping power consumption low while waiting for a receiver, plus the foreign object detection (FOD) function maximises safety. This is complemented by the STWBC-MC 15W wireless battery charger transmitter - which can

Figure 1: Block Diagram for a Typical WPT System.

Figure 2: Block Diagram Describing a Qi System

16

WIRELESS CHARGING


control multiple charging coils. Compliant with the latest Qi 1.2.4 specifications, it is specifically designed to support the Qi MP-A15 extended power profile (EPP) topology.

NXP's NXQ1TXH5 has a dedicated analogue ping circuit to detect devices available according to the Qi standard. It offers a fully integrated transmitter solution that includes a 5V full-bridge power stage to drive a Qi transmitter coil directly, as defined in the Qi A5, A11, A12 and A16 5V standards.

Semtech offers the revolutionary

LinkCharge technology to address the need for wireless charging hardware that supports multiple portable devices at the same time. The LinkCharge LP platform simultaneously charges multiple wireless charging

receivers with spatial freedom for devices. The TS51223 is a fully integrated wireless power receiver for low power wearable applications requiring a low cost, space-saving solution (Figure 3).

Development boardsTexas Instruments’ TIDA-00881 is designed to add the Qi-compliant wireless power functionality to other low power boards available

Figure 3: Typical Application Circuit for TS51223/TS80002/TS51231 Transmitters.

Figure 4: Block Diagram for the TIDA-00881.

from the company. It employs a 19mm loop, with a wireless receiver that is compatible with the Qi protocol. For charging, it is sufficient to position the coil on any wireless charger, and thus to recharge a 3.6V battery. This battery will manage the Launchpad with 3.3V and provide 5V for any additional modules (see Figure 4). Texas Instruments offers in its design the use of a LIR2032 Li-ion cell. This can supply an ongoing maximum current of 50mA (with peaks up to 60mA). Another design solution for initial prototyping purposes is the IDT WP3W-RK. This is suitable for wireless

applications ranging from 0.5W to 3W. The reference kit includes both the transmitter (P9235A-R-EVK) and the receiver (P9027LP-R-EVK) elements and offers up to 80% system efficiency at 3W.

ConclusionWPT clearly has all the features needed to revolutionise the electronics market -

enabling efficient power transfer in a far more convenient manner. Future developments of this technology are mainly destined to focus on extending its scope to serve medium and high-power applications. This should allow it to prove its long-term value to modern society.

Mouser Electronicswww.mouser.com

The Power and Evolution of GaN:

By: Alex Lidow, CEO and Co-founder of EPC

Part 2 of a 6 part series

In the first article in this series, how gallium nitride(GaN)-on-silicon low voltage power devices have enabled many

new applications, such as light detection and ranging (LiDAR), envelope tracking, and wireless power was discussed. These new applications have helped develop a strong supply chain, low produc-tion costs, and an enviable reliabil-ity record for GaN devices. All of this fundamental work with GaN provides adequate incentive for the more conservative design engi-neers in more mainstream applica-tions, such as DC–DC converters, AC/DC converters, and automotive to start their evaluation process. In this edition, the first of these traditional applications, the 48 V DC/DC converter will be examined.

With the power architecture transition from a 12 V to a 48 V bus power distribution in modern data centers, there is an increased demand to improve 48 V power conversion efficiency and power density. In this context, DC-DC converters designed using eGaN® FETs and ICs provide a high efficiency and high power density solution. Additionally, with the advent of 48 V power systems in mild-hybrid, hybrid and plug-in hybrid electric vehicles, GaN transistors can provide a

reduction in size, weight, and Bill of Materials (BOM) cost.

Why 48 V for Cars? Automotive technology has entered a renaissance with the emergence of autonomous cars and electric propulsion as the driving forces. IHS Markit estimates that 12 million cars will be autonomous by 2035 and, by the same year, 32 million cars will have electric propulsion according to Bloomberg New Energy Finance, Marklines. Both trends translate into a large growth in demand for power semiconductors.

Within the car, the need for 48 V bus power distribution has become evident with all the new power hungry electronically-driven functions and features appearing on the latest cars. These features include electric start-stop, elec-tric steering, electric suspension, electric turbo-supercharging, and

variable speed air conditioning, to name just a few examples. And now, with the emergence of autonomous vehicles, additional demands from systems such as LiDAR, radar, camera, and power-ful graphics processors, are placed upon the power distribution sys-tem. These graphic processors in particular are very power hungry and put a large additional burden on traditional automotive 12 V electrical distribution buses.

The solution to providing the high-power levels to increased electri-cal loads turns out to be a similar solution to the one being applied to high performance gaming sys-tems, high performance servers, artificial intelligence systems, and even cryptocurrency mining – a 48 V distribution bus.

Meeting the Demands of the Modern Data CenterIn addition to the automotive

GAN SERIES PART 2 OF 6


1918

GAN SERIES PART 2 OF 6



as function of load current at 5 − 12 V output, when operating at 500 kHz.For benchmarking purposes, a silicon MOSFET-based design is

compared with the EPC9205 in figure 6. The Si MOSFET design used a lower switch-ing frequency of 300 kHz, and the largest volume 5.6 µH inductor from series IHLP-5050FD-01 (Vishay) was selected. The electrical perfor-mance com-parison with Si design is shown figure 6.

GaN technology – More Efficient, Smaller Size, Lower CostMigrating an intermediate 48 V to 5 − 12 V bus converter

design from silicon MOSFETs to eGaN FETs offers reduction in both size and cost, while maintain-ing or exceeding efficiency targets. The bill of materials of an eGaN-

Figure 4: EPC9205 efficiency vs. output for 48Vin to 12 Vout when operating at 700 kHz

Table 1: Bill of materials, eGaN-based 48 V to 12 V buck converter

Figure 5: EPC9205 efficiency vs. output for 48Vin to 12 Vout when operating at 500 kHz

Figure 6: 48 VIN –to–12 VOUT electrical performance comparison between EPC9205 and Silicon design

based 48 V to 12 V buck converter yields a cost per watt of less than $0.05. This same bill of materials can be used for output voltages as low as 5 V.

Suitable controllers for the EPC9205 include the TPS53632G from Texas Instruments. When the EPC9205 is configured in a multi-phase system for higher output current the dsPIC33EP128GS704 from Mircochip can be used.

ConclusionThe eGaN FET based 48 V to 5 − 12 V, 10 A load converter was demonstrated to yield 5−12 V out-put with a peak efficiency of 96%, a power density of 1400 W/in3, and a cost that can go below $0.05 per watt when operated with 12 V output. This establishes eGaN FETs as a faster, more efficient and low-cost semiconductor solu-tion for a fast-growing portfolio of very high performance, very high power density 48 V applications, most notably data center power delivery and automotive applica-tions.

Efficient Power Conversion (EPC)http://epc-co.com/epc

requirements mentioned, emerg-ing advanced computing appli-cations demand more power in much smaller form factors. On top of the expanding needs of the server market, some of the most challenging applications are multi-user gaming systems, arti-ficial intelligence, and cryptocur-rency mining. Using eGaN FETs and ICs to build the smallest and most efficient 48 V – 12 V DC/DC converter suitable for these high-performance computing appli-cations increases the efficiency, shrinks the size, and reduces the overall system cost. The EPC9205 GaN-based power module, configured as a synchro-

nous buck converter, yields a power density of 1400 W/in3 when operating at 48 V input, 12 V output and a 10 A load. This design can produce an output voltage ranging from 5 V to 12 V and deliver 14 A per

phase of output current. The EPC2045, shown in figure 1, the latest generation eGaN, FET, rated at 100 V with 7 mΩ on-resistance, is capable of car-rying a continuous current of

16 A. Moreover, the EPC2045 is nearly one fifth the footprint of a comparable Si MOSFET and was chosen because it has lower parasitic capacitances and can switch much faster than equivalent silicon devices, yielding lowest switching loss even at higher switching frequency.

EPC9205 DrGaNPLUS Power Module – 1400 W/in3The EPC9205 (figure 2) power module, with the block diagram schematic shown in figure 3, is configured in a synchronous buck topology that uses two EPC2045 eGaN FETs. The EPC9205 power module, shown in figure 2, also features the uP1966A half-bridge gate driver IC from uPI Semicon-

ductor Corp., input and output filters, as well as current and temperature sensing. The high frequency capability of eGaN FETs greatly reduces the filtering re-quirements, allowing for an opti-mized output filter inductor with much smaller size and lower loss.

EPC9205 Performance ValidationWhen stepping down 48 V to 12 V at 700 kHz, the EPC9205 achieves a peak efficiency of 96% at 10 A load, with a maximum FET tem-perature of 100ºC under 400 LFM airflow. Figure 4 shows the power efficiency curve for 12 V load up to 15 A output current. The same EPC9205 is also capable of pro-ducing output voltages as low as 5 V. Figure 5 shows the efficiency

Figure 1: 100N eGaN with 7 m ohm on-resistance

Figure 3: EPC9205 DrGaNPLUS Power Module block diagram schematic

Figure 2: Development board boost a power density of 1400 W/in3

Reliable. Available. Now.

TRACO POWER North America, Inc.http://us.tracopower.com

[email protected]+1 (408) 916-4570

ALL MODELS

Single or Dual Outputs: 3.3 | 5 | 9 | 12 | 15 | 24 | ±5 | ±12 | ±15 VOUT

2:1 INPUT MODELS 4:1 INPUT MODELS

Input Ranges:4.5–13.2, 9–18, 18–36, 36–75 VIN

Input Ranges: 4.5–18, 9–36, 18–75 VIN

TEC 2 TEC 3 TEC 2-WI TEC 3-WI

2 WATT 3 WATT 2 WATT 3 WATT

NEXT-GEN SIP-8 PACKAGE DC/DC CONVERTERS

TEC Family2 & 3 WATT MODELS

DESIGN WITH CONFIDENCE• Cost Optimized Design• Compact SIP-8 package• I/O-isolation voltage 1’600 VDC• Tight regulation 0–100% load (no minimum required)• Operating temperature –40°C to +90°C• Continuous short circuit protection• Remote On/Off Control• Designed to meet UL 62368-1 (UL 60950-1)• > 5 mil hours MTBF (MIL-HDBK-217F)• 3-year product warranty

TRC ELECTRONICS, INC.

“

“

“

Contact TRACO POWER NORTH AMERICA for evaluation units and technical supportDatasheet, Cert’s, EMC Reports, Installation Manual & 3D Drawings available at

tracopower.us/tecAvailable stock from our North American distribution network

USA_en_TecFamily_PSD_Full_203x273.indd 1 26.09.18 14:48

Special Report:Iot + Wearables

Inside:

Is that a Battery, is that a Capacitor? – No, it’s Supercap...

Selecting Connector Solutions for Electric Vehicles...

Driving the Gates of SiC Cascodes is a Breeze...

IoT Systems Need High Integration & Small Form Factor...

5 Steps to Secure Electrical Devices in a Connected World...

A Challenge to the Potential of Industrial IoT...

Opening Up New Opportunities in Long Life Battery Powered IoT...

Wearable Devices Need to Change their Power Consumption...

22

25

27

37

29

42

34

45

22

SPECIAL REPORT : IoT + WEARABLES


Is that a Battery, is that a Capacitor? – No, it’s Supercap

By: Takahide Morikane, Manager, Product Management, Capacitor Division, KEMET

Supercaps provide long-lasting, reliable, compact backup power for many apps, including IoT devices.

With capabilities including high cycle life and fast charge and dis-

charge times, small-cell superca-pacitors can oust coin-type bat-teries from back-up power duties in equipment ranging from IoT devices, smart meters, or medical devices, to automotive electronics and industrial computing. Typical applications include maintaining the system’s real-time clock or vola-tile memory when the main system power is removed, such as during a power outage or when the main system battery has been removed for replacement.

Using a supercapacitor in these situations enables product manu-facturers to set users free from the restrictions imposed by finite battery lifetimes. They can also eliminate the expensive and bulky battery holder from the PCB bill of materials, in favor of a small, sol-dered-down device, and eliminate manufacturing challenges such as managing battery shelf lives and in-serting batteries prior to shipping. The supercapacitor’s benign open-circuit failure mode contrasts with typical short-circuit battery failures

that may result in outgassing or ignition.

Supercapacitors in capacitance val-ues up to 5 Farads present the most cost-effective alternative to small backup batteries, and can store enough energy to provide backup

for durations ranging from a few seconds to several days, depend-ing on the type of load and current demand.

A Peek Under the CapeThe supercapacitor, also known as the Electric Double-Layer Capacitor

(EDLC), comprises two electrodes coated in a porous material, which is usually carbon-based, sepa-rated by an electrolyte that is itself divided by a membrane.

Unlike a battery, the superca-pacitor stores and releases energy quickly through physical adsorp-tion and desorption of ions in the electrolyte contained between its electrodes. These processes are much faster than the chemical re-actions involved in battery charg-ing. Given the supercapacitor’s low internal resistance, the device can be fully charged within a few seconds whereas a secondary cell can take from ten minutes to several hours to be fully charged. Moreover, there is no theoretical limit to cycle life, whereas a lithi-um-ion secondary cell has a finite lifetime of about 500 cycles.

Modern advances in carbon-based materials enable porous electrodes to have a large surface area resulting in a high capacitance value and small external dimensions.

The electrolyte also has an important influence over device properties, and typically is either an organic compound or an aqueous solution. Aqueous electrolytes are highly conductive, have low environmental impact and are non-flammable, yielding strong performance and safety credentials. They also typically have greater resistance to moisture absorption than organic compounds, resulting in longer

Figure 1: Small-cell supercapacitor with aqueous electrolyte

Figure 2: example of aqueous-electrolyte small-cell supercapacitor in rugged resin-molded case.

info@paci�cpower.com

DC PowerAC Power AC+DC Power

www.paci�cpower.com

“The Power of Expertise”© 2018 Paci�c Power Source, Inc., 17692 Fitch, Irvine CA, 92614

1.800.854.2433

More Power, Less Space...The Enhanced AFX Series “A” version

Precision ProgrammableAC and DC Power Sources

60kVA AFX Cabinet

• New “A” version with higher Voltage & Frequency Ranges• Get up to 15kVA/15kW of power in 4U rack space• Start at 6kVA, add Power later as needed• Parallel for Higher Power Systems up to 150kVA• Test for compliance to Mil-Aero or IEC Standards using available test sequences• LXI - Connect with LAN and GO! No drivers, no code to write

Call or Email to learn more...

THE BEST JUST GOT BETTER....DIGITAL POWER CONVERSION - UNPRECEDENTED PERFORMANCE

C

M

Y

CM

MY

CY

CMY

K

AFX-A_version_Ad-PSD_Magazine_v1.pdf 1 5/7/2018 1:13:11 PM

lifetime with better stability.

For all device types, the electrolyte properties determine the overall supercapacitor terminal voltage. The voltage, when fully charged, is usually less than 3V.

A conventional approach to con-struction of small-cell superca-pacitors is comparable to that of

a coin cell, comprising lower and upper metal cases that are joined by swaging and enclose the carbon electrodes and organic electrolyte. Although an internal gasket aids sealing, the electrolyte can dry within a relatively short lifetime, and thermal shock can compromise structural integrity.

No Ordinary Supercap


24



KEMET’s small-cell supercapaci-tors feature a high-strength vul-canized rubber bond that ensures high safety against liquid leakage. The cross-section of figure 1 shows how these supercapacitors are constructed, including the aque-ous electrolyte, rubberized elec-trodes, and separator membrane.

To achieve a desired output volt-age higher than the base voltage of a single cell, several cells are con-nected in series. Coin-type super-capacitors, for example, are often stacked inside an external can, or within a tube of heat-shrink mate-rial, and electrodes attached to the upper and lower surfaces.

Supercapacitors containing aqueous-electrolyte can be stacked efficiently to achieve higher volt-age ratings within smaller case sizes, with the added advantage of a strongly bonded seal to protect against thermal and mechanical shock. Robust and reliable devices can cover a wide range of operat-ing voltages, from 3.5V to 12V. Figure 2 shows how such a multi-cell structure is encased in a resin-molded package. Alternatively, the

outer packaging can be a sealed metal can, and terminations can be either through-hole or surface-mount.

In the Real WorldIn addition to having special pow-ers and an aka – like any favorite superhero – supercapacitors also possess vulnerabilities. They require current above a minimum threshold, called the absorption current, while charging, and just like ordinary capacitors are prone to leakage and parametric change due to environmental effects and aging.

The absorption current that oc-curs during charging results from the redistribution of ions within the porous electrode material. Ions that are initially absorbed on the surface tend to diffuse into the electrode structure over time, consuming a proportion of the current flowing into the device. For this reason, a high initial current is needed to continue charging the supercapacitor. It can take many hours for the absorption current to decay to a stable value of leakage current typically in the order of a

Figure 3: Comparison of ESR stability with increasing temperature

few microamps (µA).

Parametric changes can occur as the result of environmental conditions such as high ambient temperature or humidity, as well as aging. Devices with aqueous electrolyte are inherently better able to withstand high temperatures and humidity without drying or absorbing moisture, compared to alternatives with organic electrolyte and hence typically exhibit greater stability. Figure 3 compares the shift in equivalent series resistance (ESR) for aqueous and organic types when exposed to increasing temperature, showing how the aqueous type has significantly greater temperature stability.

ConclusionSupercapacitors offer a high-performing alternative to batteries in a wide variety of backup-power applications, delivering far greater cycle life and avoiding any need for designers to worry about battery replacement or recharging.

Improvements in electrode materi-als and electrolyte formulations enable these powerful devices to store more energy and stack more easily to provide the required out-put voltage. The latest supercapac-itors featuring aqueous electrolyte are ready to come to the rescue in RTC hold-up and NVM backup applications throughout energy, security, automotive, and medical market segments.

Kemetwww.kemet.com

Selecting Connector Solutions for Electric Vehicles

By: Albert Culetto, Technical Support Connectors & Cables, Rutronik

Designing the powertrain for EVs, choose the correct connector for each individual application

An electric vehicle contains many sub-systems that need to inter-operate so the

vehicle can function as intended. These disparate systems range from the battery string, comprising Lithium-Ion modules to various drive systems, converters, auxiliary modules, and charging and moni-toring systems. As a result, choos-ing the right connector for each in-vehicle application is essential to a reliable vehicle.

Given the safety implications as-sociated with vehicles, both for the driver / passengers and other road users, it is not surprising that many standards and automotive-specific requirements exist. Electri-cal safety is one key point, such as the use of a High-Voltage Inter-connect Loop (HVIL) to shut the system down as a hazard devel-ops. Other more basic protections such as ensuring high voltage connectors cannot be touched are important, especially for mechan-ics working on the vehicle. Other automotive requirements include contact position assurance (CPA) / terminal position assurance (TPA) and secondary locking of

the connector assembly.

The trade-off between safety and costImproving safety is es-sential and a constant quest for designers, but this often leads to longer develop-ment cycles, with higher costs – as well as the potential for more ex-pensive components (such as con-nectors) being required. Correctly assessing the needs at an early stage can ensure that costs are kept to a minimum without safety being compromised. Key consider-ations for designers include:

• What are the basic electrical requirements (current and voltage)?

• Will a locking system be re-quired for safety or vibration?

• Does the connection form part of the service process?

• Is keying required to avoid potentially hazardous mis-connection?

Establishing answers to these

questions will go a long way to defining the connector type that will meet the needs of the specific application. However, apart from the basic op-erational and safety requirements, there are more considerations that will shape the decision and selection process. Many modern electric vehicles are very densely packed, so the available space may be small. The demand for ever higher efficiency in vehicles means that the weight of every compo-nent (including connectors) is under scrutiny.

As the market for electric vehicles grows, so does the number of available options, giving design-ers a huge choice. However, as the market remains in a relatively early stage, relevant standards are

Figure 1: Electric Vehicle Charging



26 WWW.POWERSYSTEMSDESIGN.COM


few and far between, making the choice challenging.

A broad range of connectors with significant optionsOne supplier with an extensive range of connectors for electric vehicles is Amphenol. Most of its products are based on the com-pany’s RADSOK technology, which guarantees coverage up to 65% of the contact surface area. Its com-prehensive range offers multiple options allowing the designer to select the right product for their own individual application:

• Electromagnetic shielding can be selected to cover individual contacts or the entire connector.

• Sealing / ingress protection is available up to IP6K9K when mated which keeps fingers out as well as dirt and mois-ture that could be a source of failure.

• Additional interlocks and / or secondary locking ensure con-nectors remain mated.

• Optional keying and colour variations are available to en-sure connectors are not mis-plugged.

• Products such as the Power-Lok series offer choices of up to four poles as well as metal and plastic housings, coding configurations and colour options.

• For high current applications such as electrical distribution, drive systems or motor control connectors are available from 2.8 mm to 14.0 mm diameter,

giving the ability to safely handle currents up to 650 A. The Manual Service Discon-nect (MSD) series provides an additional level of safety, especially for the distribution box.

Shock and vibration resistant solutionsOne of the most rugged products on the market is the Imperium High Voltage / High Current (HVHC) connector system from Molex that is able to withstand more heavy shocks and vibrations than any other available product. Currently available in diameters of 8.00 mm and 11.00 mm, new configurations are being devel-oped to broaden the range as well as to demonstrate how customer-specific requirements can be met by repackaging. StandardisationAs things are evolving so rapidly in the electric vehicle space, there are few, if any, standards for connec-tors. In Europe, the most common charging connector is the Men-nekes (Type 2) plug that is covered by IEC 62196-2 & IEC 62196-3.

This plug is used to charge the Tesla electric vehicles and is avail-able as part of Amphenol’s HVCO series.

The majority of today’s electric ve-hicles are in Asia. Here a bidirec-tional charging standard known as CHAdeMO is the most popular and products from manufacturers such as JAE are compliant with this standard.

JAE’s KW01 and KW02 series are rugged and highly reliable solu-tions with rust resistant contacts, and a flame-retardant and weath-er-resistant resin housing and cable. Both series can be supplied as complete cable assemblies, giv-ing OEMs a turnkey solution. The KW02 also allows energy from the vehicle battery to supply power to the home when plugged in – the so-called ‘Vehicle-to-Home’ (V2H) standard that allows users to reduce energy costs by using energy stored in their vehicle at peak times and then recharging when demand is lower.

Rutronikwww.rutronik.com

Figure 2: Imperium High Voltage / High Current (HVHC) connector system from Molex

Driving the Gates of SiC Cascodes is a Breeze

By: Zhongda Li, NPI Manager, UnitedSiC

To get the best performance from a power switch, the gate drive circuit has to be carefully designed

In the days when bipolar transistors were the only option for semiconductor power switches, their base

drive circuit was critical for fast and efficient operation. With power device gains often just in double digits, base currents could be huge. Complex arrangements had to be used to drive current in and out at different levels with significant associated power dissipation. The big bugbear was ‘base storage charge’ which slowed switch-off and had to be overcome with negative base voltages and proportional drive circuits or ‘Baker Clamps’. A typical complete base drive circuit might have had 20 passive and active components.

MOSFETs promised easy gate driveAs high voltage, low RDS(ON) Si-MOSFETs became available, gate drive seemed to become easy – just a positive voltage for ON and 0 V for OFF. Power drawn was close to zero at low switching frequencies and all you needed was a series resistor to stop parametric oscillations. Textbooks and application notes made the drive circuits look absurdly simple.

Engineers are a skeptical bunch, though, and they soon realized that if you want to use Si-MOSFETs at the high frequencies they are capable of, some effects creep in that make the gate drive more complex to keep efficiency high and prevent failures. The problem is that Si-MOSFETs are far from the ideal switch – they have real input and output capacitance and a ‘Miller’ effect which looks like a variable gate-drain capacitance depending on switching state. They also have non-zero package inductances. Cyclically driving the capacitances around the MOSFET implies power consumption and the Miller effect can allow dangerous spikes of current into the gate ON turn-off, preventing clean and loss-free switching. The

source package inductance has a similar effect with high di/dt levels – a voltage is generated opposing the gate drive.

The solution was to choose a positive gate drive voltage for the ON-state for good saturation and a negative voltage for the OFF-state, ensuring that the device is fully OFF when required. Gate drive power is significant, though, proportional to the total gate voltage swing and total gate charge and frequency, with larger devices typically needing several watts. A buffer between control electronics and gate is normally needed to handle the power and peak currents.

Still, Si-MOSFETs are basically

Figure 1: Gate drive voltage comparisons, recommended and absolute maximum





voltage-operated devices and the gate drive circuit remains quite simple with good margin achievable between operating and absolute maximum voltage ratings.

Wide band-gap – nice but…Now, wide band-gap switches in Silicon Carbide (SiC) and Gallium Nitride (GaN) technology are available. The combination of extreme switching speed, low conduction and switching losses, high temperature operation and high voltage has designers reaching for their calculators. How small can the magnetics and other passives be if frequency is increased and can that bulky heat sink be reduced or even eliminated? Spectacular gains in efficiency and size were forecast and demonstrations show real, practical results.

The headline specifications of the new devices are great but the gate drive has now become more critical. Figure 1 shows the recommended drive voltage levels for some SiC MOSFETs and the absolute maximums, which can be frighteningly close; look at the ‘Generation 3’ SiC MOSFET from one supplier, with a recommended negative drive of -3 V and an absolute maximum of -4 V. There is a similar tight margin for the positive recommended and absolute maximum voltage. Designers are rightly worried about

reliability with the smallest of transient gate overvoltages. GaN is little better with the latest products needing typically +6 V positive with a +10 V breakdown level.

Wide band-gap devices do have very low total gate charge but the Miller capacitance is still there and non-zero. With potential dV/dt rates upwards of 100 kV/µs, even a few picofarads produce current close to an amp, injected back into the gate drive circuit. Actually, slowing down SiC and GaN devices is usually needed in practical circuits to keep within EMI levels and mitigate the effects of Miller capacitance and package inductance.

There is another solution though; a cascode arrangement of a Si-MOSFET and a normally-ON co-packaged SiC JFET. This combination gives the ease of a MOSFET gate drive with typical absolute maximums of +/-25 V and the speed and low loss advantages of SiC. In a vertical construction, the Miller capacitance is almost immeasurable, at around 1 pF, meaning that a simple 0 V/+12

V gate drive can be used. This makes the gate drive compatible with older Si-MOSFET and IGBT designs, making a drop-in upgrade to SiC in existing systems realistic. You can see the way it compares in Figure 1.

Switching is dramatically fast but can be slowed if needed simply by using a series gate resistor (RG). Figure 2 shows how different RG values control the gate energy EON and dV/dt and di/dt levels for a UnitedSiC UJC1206K device. For fine-tuning of switching speed and EMI a separate resistor can be used for the OFF-drive through an isolating diode.

Get that calculator out againSiC cascodes really do give the best of all worlds with their easy gate drive and robust construction. Do the math with your magnetics and passives using cascodes and see how you can turn miniaturized high-frequency power converters into reality.

UnitedSiChttps://unitedsic.com/

Figure 2: RG effectively controls switching speed in SiC cascodes

IoT Systems Need High Integration & Small Form Factor

By: Tony Armstrong Director of Product Marketing Power Product Group, Analog Devices Inc.

It is not unusual for an industrial IoT system designer to use linear regulators in a system

At the medium-to-low end of the power spectrum there

are modest power conversion requirements such as those commonly found in “Internet of Things” (IoT) equipment, which necessitate the use of power conversion ICs that deal with modest levels of current. These are usually in the range of 100’s milliamps of current, but can be higher if there are peak power demands that are needed by an onboard power amplifier for the transmission of data or video. Accordingly, the proliferation of wireless sensors supporting the numerous IoT devices has increased the demand for small, compact and efficient power converters tailored to space and thermal constrained device form factors.

However, unlike many other applications, many industrial and medical products typically have much higher standards for reliability, form factor and robustness. As you would expect, much of the design burden falls on the power system and its

associated support components. Industrial, and even medical IoT products, must operate properly and switch seamlessly between a couple of power sources such as the AC mains outlet and a battery backup. Furthermore, great lengths must be taken to protect against faults, while also maximizing operating time when it is powered from batteries to ensure that normal system operation is reliable whichever power source is present. Accordingly, the internal power conversion architecture used within these systems need to be robust, compact and require minimal heat sinking.

Power Supply Design ConsiderationsIt is not unusual for an industrial

IoT system designer to use linear regulators in a system that incorporates wireless transmission capability. The primary reason being that it minimizes EMI and noise emissions. Nevertheless, although switching regulators generate more noise than linear regulators, their efficiency is far superior. Noise and EMI levels have proven to be manageable in many sensitive applications if the switcher behaves predictably. If a switching regulator switches at a constant frequency in normal mode, and the switching edges are clean and predictable with no overshoot or high frequency ringing, then EMI is minimized. Moreover, a small package size and high operating frequency can provide a small tight layout,

Figure 1: 3.3V to 5V Input, Delivering 12V at up to 800mA with an External Clock



3130 WWW.POWERSYSTEMSDESIGN.COM WWW.POWERSYSTEMSDESIGN.COM

SPECIAL REPORT : IoT + WEARABLES POWER SYSTEMS DESIGN 2018NOVEMBER

in the development cycle. With little time and perhaps limited specialist power design resource, there is pressure to come up with a high efficiency solution with the smallest possible footprint; while potentially utilizing the underside of the PCB as well for maximum space utilization.

This is where the µModule regulator provides an ideal answer; the concept is complex on the inside, simple on the outside – the efficiency of a switching regulator and the design simplicity of a linear regulator. Careful design, PCB layout and component selection are very important in the design of a switching regulator and many experienced designers have smelt the distinctive aroma of burning circuit board in the earlier days of their career. When time is short or power supply design experience is

limited, the readymade µModule regulator saves time and reduces risk to the program.

A recent example of ADI µModules, is the LTM4661 synchronous step-up uModule regulator in a 6.25mm x 6.25mm x 2.42mm BGA package. Included in the package are the switching controller, power FETs, inductor and all support components. Operating over an input range of 1.8V to 5.5V, it can regulate and output voltage of 2.5V to 15V, set with a single external resistor. Only bulk input and output capacitor are needed.

The LTM4661 is efficient and can deliver efficiencies of greater than 87% when stepping up from a 3.3V input to a 12V output. See Figure 2 efficiency curve below.

Also, Figure 3 shows the

measured thermal picture of the LTM4661 running form a 3.3V input to 12V at 800mA DC current with 200LFM airflow and no heat sink.

ConclusionThe deployment of IoT equipment has exploded in recent years and includes a wide variety of products for the military and industrial application spaces. A new wave of products, including sensor-filled medical and scientific instrumentation have been key market drivers in recent years and are only now starting to see signs of significant growth. At the same time, the space and thermal design constraints of these systems has necessitated a new class of power converters that can deliver the necessary performance metrics of small, compact and thermally efficient footprints to power the internal circuits, such as the power amplifier. Fortunately, devices such as the recently released LTM4661 step-up uModule regulator elevates the power supply designers task.

Finally, using µModule regulators make sense in these types of environments since they can significantly reduce the debug time and allow for greater board area usage. This reduces infrastructure costs, as well as the total cost of ownership over the life of the product.

Analog Deviceswww.analog.com

Figure 3: Thermal Image of LTM4661; 3.3V Input to 12V Output at 0.8A, 200LFM Air Flow & No Heat Sink

which minimizes EMI radiation. Furthermore, if the regulator can be used with low ESR ceramic capacitors, both input and output voltage ripple can be minimized, which are additional sources of noise in the system.

It is common for the main input power to today’s industrial and medical IoT devices to be a 24V or 12V DC source from an external AC/DC adapter and /or battery bank. This voltage it then further reduced to either 5V and/or 3.xV rails using synchronous buck converters. Nevertheless, the number of internal post-regulated power rails in these medical IoT devices has increased while operating voltages have continued to decrease. Thus, many of these systems still require 3. Xu, 2.xV or 1.xV rails for powering low power sensors, memory, microcontroller cores, I/O and logic circuitry. Nevertheless, the internal power amplifier used for data transmission can require a 12V rail with up to 0.8A of current capability to transmit any recorded data to a remote centralized hub.

Traditionally, this 12V rail has been supplied by step-up switching regulators, requiring specialized switch-mode power supply design know how, and needs a large solution footprint on the printed circuit board (PCB).

A New Compact Boost ConverterAnalog Devices’ µModule® (micromodule) products are complete System in a Package

(SiP) solutions that minimize design time and solve the common problems of board space and density issues commonly found in industrial and medical systems. These µModule products are complete power management solutions with integrated DC/DC controller, power transistors, input and output capacitors, compensation components and inductor within a compact, surface mount BGA or LGA package. Designing with ADI’s µModule products can significantly reduce the amount of time needed to complete the design process by up to 50%, depending on the complexity of the design. The µModule family transfers the design burden of component selection, optimization and layout from designer to the device, shortening overall design time, system troubleshooting and ultimately improving time to market.

Furthermore, ADI’s µModule solutions integrate key components commonly used in discrete power, signal chain and isolated designs within a compact, IC-like form factor. Supported by ADI’s rigorous testing and high reliability processes, the µModule product portfolio simplifies the design and layout of power conversion designs.

The µModule family of products embraces a wide range of applications including point of load regulators, battery chargers, LED drivers, power system management (PMBus digitally-managed power supplies), isolated converters, battery chargers and LED driver. As highly integrated solutions with PCB Gerber files available for every device, µModule power products address time and space constraints while delivering a high efficiency, reliable and with select products a low EMI solution compliant with EN55022 class B standards.

As design resources become stretched by increased system complexity and shortened design cycles the focus falls on development of the key intellectual property of the system. This often means the power supply gets put to one side until late

Figure 2: Efficiency vs. Output Current for the LTM4661 from a 3.3V Input to Outputs Ranging from 5V to 15V

VPP PC MOUNT TRANSFORMERS

VPL CHASSIS MOUNT TRANSFORMER

VPS CHASSIS MOUNT QUICK CONNECTQUICK CONNECT

VPT TOROIDAL MOUNT

VPM MEDICAL TOROIDAL MOUNT

FEATURES• Dual secondary windings for series or parallel connections• Low leakage current and low stray fi elds.• Low temp rise: 25 to 55°C

SPECIFICATIONS• Frequency: 50/60 Hz• Primary: 100, 120, 208 or 240 Vac• Sec voltage: 5 to 36V

APPLICATIONS• Hospital equipment• Biomedical equipment• Test equipment• Audio equipment

Triad’s VPS Chassis Mount Transformers are chassis mount devices requiring higher power—up to 175 VA. They meet major U.S. and global standards (CSA, IEC and UL). They are among the industry’s most versatile transformers.

FEATURES• Dual bobbin design with insulating shroud• Meets global safety standards• Quick disconnect connection

SPECIFICATIONS• Frequency: 50/60 Hz• 25 to 175 VA• Sec voltage: 5 to 230V

APPLICATIONS• Oil/gas equipment• Conveyor ovens• Heat exchangers• Music equipment

Triad’s VPT Series features a compact toroidal design, which is cost effective and effi cient with higher power density and reduced magnetic fi elds. They are approved to UL 506, CE: EN 61558-1, and CE: EN 61558-2-6 with Class B insulation for use up to 130°C

FEATURES• Isolated dual primary and secondary coils• Class B (130°C) rated insulation• High effi ciency

SPECIFICATIONS• Frequency 50/60 Hz• 25 to 2500 VA• Sec voltage: 6 to 230 V

APPLICATIONS• Sound reproduction• Power system equipment• Production equipment

Triad’s VPL Series is similar to the VPP Series, but with chassis mounting and leads. The VPL sets the industry standard with European style split bobbins. These leaded devices meet all international safety agency standards.

FEATURES• Low inter-winding capacitance requires no electrostatic shielding• 3500 V isolation between primary and secondary• Compact footprint

SPECIFICATIONS• Frequency: 50/60 Hz• 5 to 56 VA• Sec voltage: 5 to 36V

APPLICATIONS• Building & plant equipment• Lighting• Temperature controls• Material handling

Triad’s VPP PC Mount World Series™ Transformers are an advanced line of more than 40 quality transformers. They are perfect for board level applications requiring the added safety and security of insulating shrouds over the windings. They are also UL 5085-1 & 2/3 recognized.

FEATURES• Dual bobbin construction• Insulating shroud meets UL V0 fl ammability specs• No electrostatic shielding needed

SPECIFICATIONS• Frequency: 50/60 Hz• 2.5 to 56 VA• Sec voltage: 5 to 36V

APPLICATIONS• Battery charging• Spa controls• Soft drink machines• Security access and control

For info or to purchase, go to:www.TriadMagnetics.com

D E S I G N S P OT L I G H T M AG N E T I C S F O C U S

M A G N E T I C S M A G N E T I C S

TRIAD’S 50/60 H Z WORLD SERIES TRANSFORMERS

With dual primaries, -C2 Split Pack™ Transformers are the only device of their type with TUV approval and are UL 5085-1&3 recognized. They utilize a Class F 155°C insulation system and can be used in myriad applications requiring inherently/non-inherently limited transformers.

FEATURES• Split bobbin design• No electrostatic shielding required• Use in series, parallel or separate circuits• High isolation between secondaries

SPECIFICATIONS• Frequency 50/60 Hz• 1.1 to 36.0 VA• Nominal secondary voltage: 5 to 56V• HIPOT dielectric: 4200 Vac

APPLICATIONS• Commercial food & beverage equipment• Motor speed controls• Industrial controls• Timers

C2 PC MOUNT SPLIT PACK CLASS 2/3

With dual primaries, -C2 Split Pack™

With more than 70 years of transformer design and manufacturing experience, Triad now off ers over 500 diff erent power transformers that are available off -the-shelf from the industry’s leading distributor network. In today’s shrinking world and global market, it is essential to design products that can be delivered worldwide. For this reason, Triad has designed one of the industry’s most complete off erings

of international power transformers. Triad’s World Series Transformers range in power from 1.1 VA to 10 kVA. Most have confi gurable 120/240V primaries with output voltages that range from 5.0 to 240 Vac. They’re UL recognized and TUV tested to IEC global safety standards. They can be customized to your requirements and are backed by our world class service.

Triad’s VPM Series offers output power up to 10 kVA and are UL recognized and CE certifi ed for medical applications. They feature toroidal construction with dual secondaries, allowing for both series or parallel connections. Faraday and fl ux band shield maintains low leakage current and low stray fi elds, respectively.

Triad_2pgSpreadConv_F.indd 2-3 9/28/18 8:57 PM



5 Steps to Secure Electrical Devices in a Connected World

By: Max Wandera, CISSP, GLSC, Director of the Cybersecurity Center of Excellence at Eaton

By 2020, an estimated 31 billion devices will be connected to the Internet

Last year, businesses were expected to spend an esti-mated $964 billion on IoT devices. (Source: Gartner)

Moving forward, analysts forecast that connected devices and the data they generate will continue to grow exponentially. By 2020, an estimated 31 billion devices will be connected to the internet. (Source: IHS)

As customers deploy more intel-ligent and connected solutions, it is essential to trust and verify that the technologies they are relying on are designed, built and tested to proven engineering practices.

For years, Eaton has maintained strict procedures at every stage of the product development process. The main goal of our approach to product security is to advance safety while protecting the avail-ability, integrity and confidentiality of electrical systems.

Industry first cybersecuritycollaboration with UL We began collaborating with UL earlier last year to drive the de-velopment of new cybersecurity standards for power management products. Our mutual efforts are helping drive the development of common criteria for assess-

ing products to ensure they meet industry standards and reduce cybersecurity risk. Through rigor-ous cybersecurity processes and having the first lab approved to participate in the UL Data Accep-tance Program, we hope to set a precedent for developing products and systems that comply with the most stringent standards and expectations for safe, secure power management.

Further, our work with UL is help-ing establish measurable cyberse-curity criteria for network-connect-ed power management products and systems. As we introduce

more intelligent and connected systems, this collaboration will continue to provide the industry with robust standards, testing methodologies and technologies designed to help device manu-facturers build tru

Date post:	22-Jun-2020
Category:	Documents
Upload:	others
View:	3 times
Download:	0 times

0 3 5 ) . & 3 - Power Systems Design · operating conditions, and then order free evaluation...

Documents