design world/ee network - wireless & rf handbook

April 2015designworldonline.com

WirelessHandbook

& RF

150401_BOMM_EEW_US_Snipe.indd 1 3/30/15 11:57 AM

EE_Wireless_May_Cover_Vs3.indd 1 4/27/15 11:49 AM

150401_WLSG_EEW_US.indd 1 3/30/15 11:46 AMDigiKey_EE_4-15_ad.indd 1 4/23/15 3:50 PM

USA: +1 480-961-1382 Europe: +32 9 235 3611 www.rogerscorp.com/rogmobile*Based on studies conducted by Rogers Corporation.

Next Generation

RO4835 Laminates™

With improved oxidation resistance compared to traditional hydrocarbon ceramic materials.

Today’s requirements for higher MTBF’s (mean time between failure) and improved long term thermal reliability challenge RF engineers as they design circuits operating at higher frequencies and increased power levels. Up to now, the only option has been costly PTFE materials.

Introducing Rogers next generation of RO4000® materials, RO4835 laminates, available in all of the thickness and copper confi gurations you are used to working with. Accelerated aging tests reveal RO4835 laminates have approximately 10 times improved oxidation resistance compared to traditional hydrocarbon ceramic laminate materials.* For designs where printed-circuit oxidation is a concern, try RO4835 materials.

Visit www.rogerscorp.com to learn more about Rogers’ full lines of high-frequency laminates, including next-generation, oxidation-resistant RO4835 laminates.

IMPROVED

OXIDATION

RESISTANCE

www.rogerscorp.com/rogmobile

See us at IMS Booth 3640

Rogers_EE_4-15_ad.indd 1 4/23/15 3:52 PM

2 DESIGN WORLD 4 • 2015 analogictips.com

LELAND TESCHLERExecutive Editor@DW_LeeTeschler

Wireless charging for your coffee maker

TODAY’S average teenager has never seen a dial telephone. It looks

as though cords that plug into ac outlets will soon look just as strange to high schoolers. That’s because wireless charging may become ubiquitous enough to make ac power cords obsolete for most kinds of electrical equipment rather than just being a novelty for personal electronics.

“Eventually your kitchen appliances sitting on the countertop won’t have cords. The idea of plugging your blender into the wall will seem silly. Ditto for construction workers and wireless tools. In EVs, wireless charging will be like cup holders.—you won’t be able to sell vehicles without it.”

So said Argonne National Laboratory Principal Electrical Engineer Theodore Bohn. He came to this conclusion from his work in Argonne’s Transportation Technology R&D Center. There he tests vehicular wireless charging systems to see if equipment from different manufacturers will work with each other.

Interestingly, Bohn thinks wireless charging will become common in EVs long before it starts to pervade the rest of society. The reason is that the Society of Automotive Engineers is hammering out standards for wireless EV charging. But no standards govern the wireless charging schemes now available for consumer electronics. That make the whole field

a little like the Wild West; there are several different technologies used. Some employ inductive techniques, others use magnetic coupling, still others depend on resonant

circuits. There is no interoperability among any of them. Moreover, Apple, the 800-lb gorilla of consumer electronics, has chosen not to adopt any of the

techniques proposed so far. “Apple doesn’t use any of them and wants to do its own thing,” said Bohn.

Meanwhile, Bohn thinks the SAE standard for wireless vehicular charging could be ready by the end of 2016. But there is still a lot of work to do.

“Interoperability is the biggest technical challenge,” he said. “Each company has its own IP that it tries to advocate. For example, Qualcomm has a double-D-shaped coil, other suppliers have configurations that look like solenoids, and still others want the pick-up coil to be flat and circular.”

And there is a lot of variability in the performance of the systems Bohn is testing. “Some of these designs work well when they are aligned, but are miserable when misaligned by even a few centimeters. Of course, it’s difficult to park a 20-ft vehicle to a spot within centimeters. Even some vehicle navigation systems can’t self-park cars to that accuracy. The throughput for some wireless charging systems really falls off quickly with any misalignment because they are highly tuned,” Bohn explained.

One factor slowing the process is that manufacturers can’t decide on parameters in a reference design that all systems must interact with. “The standard must require a certain amount of efficiency and throughput under set conditions. For example, if you exit the vehicle and its height rises by 30 mm, you want it to still charge acceptably. Or there might be snow melting and the parked vehicle gets 30 mm closer to the charging pad. These things really happen—people park poorly and on top of debris. They are issues the committee is working through,” Bohn said.

All this makes it time consuming to come up with specs that will be bullet proof in the real world. But SAE standards will be the reason cars will have wireless charging long before coffee makers or food processors do.

WIR

ELES

S &

RF

anal

ogic

tips

.com

In EVs, wireless charging will be like cup holders—you won’t be able to sell vehicles without it.

Editorial_EE_April_Vs2.LT.MD.indd 2 4/23/15 3:57 PM

The Smart Choicefor Small Spaces

Coilcraft is the brand engineers trust most when specifying tiny inductors for compact devices

WWW.COILCRAFT.COM

®

Boost the performance and battery life of your smallest designs 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

You can get started by using our suite of web tools to quickly locate the perfect inductors for your design. Compare and analyze multiple parts based on your true 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_EE_4-15_ad.indd 3 4/23/15 3:58 PM

The Microchip name and logo and the Microchip logo are registered trademarks of Microchip Technology Incorporated in the U.S.A. and other countries.All other trademarks are the property of their registered owners. © 2015 Microchip Technology Inc. All rights reserved. 4/15 DS00001761B

www.microchip.com/wireless

The RN4020 is a Bluetooth® 4.1 Low Energy Module for designers who want to easily add low-power wireless capability to their products. It provides a highly integrated solution designed with an easy-to-use ASCII interface for complete configuration and communication over UART. The RN4020 will reduce your development cycle with:

• Simple ASCII UART interface• Selectable on-board Bluetooth profiles• Microchip’s Low-Energy Data Profile (MLDP) for custom serial data• Scripting for standalone analog and digital data collection and transmission• Bluetooth SIG and regulatory agency certification*

*Check Microchip’s website for latest Regional Government Regulatory Certifications

Adding Bluetooth® Low Energy to Your Application Just Got Easierwith the RN4020, a Complete Bluetooth 4.1 Low Energy Module

Microchip_EE_ad 4-15.indd 4 4/24/15 2:12 PM

Gore_EE_4-15_ad.indd 5 4/23/15 4:02 PM

02 Wireless charging for you coffee maker

10 UPDATES

14 How to spec inductive coils RF designers should know a few things about the way coil-

winding houses fabricate coils.

18 How to sniff out EMC problems It pays to do preliminary studies before sending products off

to the test lab for electromagnetic compatibility problems.

22 Analyzing millimeter frequencies with external harmonic mixing External mixers can help extend the capabilities of signal

analyzers so these instruments can handle frequencies

ranging into several hundred gigahertz.

26 Cellular amps boost mobile and M2M applications Cellular signal amplifiers help bring reliable communications

to remote locales, but there are subtleties to their deployment

that can trip up the uninitiated.

30 Better materials for better antennas Thanks to better materials and novel processing,

antennas today can be lighter, smaller and easier to

make than their predecessors.

36 Testing wireless devices for electromagnetic compatibility Wireless devices must function in an increasingly hostile RF

environment. Here are a few tips on setting up a test regime

that can uncover problems.

40 Basics of RF switches RF & microwave switches route signals through transmission

paths with a high degree of efficiency. Four fundamental

electrical parameters characterize how RF & microwave switch

designs perform.

44 Wide-band instruments in the 5G Era Just a decade ago, signal analyzers with 20 MHz of

instantaneous bandwidth were considered state-of-the-art.

Within a decade, signal analyzers with no less than 2 GHz of

bandwidth will likely be entry-level devices.

48 ADVERTORIAL

56 Top Products

59 Top Forum Members …on the Radar

analogictips.comWIRELESS & RF | CONTENTS


14 18

30

26

Contents_EE_APRIL_Vs3.indd 6 4/24/15 12:52 PM

© Keysight Technologies, Inc. 2015

US A: 800 829 4444 CAN: 877 894 4414

Download our white paper Implementing a Flexible Testbed for 5G Waveform Generation and Analysis at www.keysight.com/find/5G-Insight

Keysight TechnologiesImplementing a Flexible Testbedfor 5G Waveform Generationand Analysis

White Paper

Keysight Authors: Greg Jue, Applications Development Engineer/Scientist Sangkyo Shin, Product Marketing/App Development

HARDWARE + SOFTWARE + PEOPLE = 5G INSIGHTS

In just a few years, the fth generation of wireless communications will be a reality. It will allow data to be transmitted up to 100 times faster than today’s 4G networks. But getting to that speed won’t be easy. You’ll need genuine insights to overcome enormous technical challenges. We can help. We have the industry’s rst and most comprehensive 5G software library. It can signi cantly streamline design feasibility because it incorporates an iterative design sequence with every piece of Keysight 5G test equipment.

There’s a 5G momentof discovery out there.We’re here to help you nd it.

Designed for testing 5G simulation to veri cation

Software platforms and applications that work seamlessly across our 5G instruments

Incorporate iterative design and rapidly move betweenstages of your 5G development ow

Industry’s rst and largest 5G library

Keysight N9040B UXAsignal analyzerwith 89600 VSA softwareand M1971E smart mixer

Keysight 89600 VSA software

Keysight N5247A PNA-X microwavenetwork analyzer, 67 GHz

Keysight MIMO PXI test solutionM9381A PXI VSG and M9391A PXI VSA - Up to 8x8 phase-coherent MIMO measurements

Keysight DSOZ634AIn niium oscilloscopewith 89600 VSA software

Keysight N5152A 5-GHz/60-GHz upconverterKeysight N1999A 60-GHz/5-GHz downconverter

Keysight M8190A arbitrarywaveform generator

Keysight M9703A high-speed digitizer/wideband digital receiver

Keysight N7608B Signal Studiofor custom modulation

Keysight 5G Baseband Exploration Library for SystemVueIndustry’s rst 5G ExplorationLibrary for researchers

Keysight E8267D PSG vector signal generator


Keysight_EE_4-15_ad.indd 7 4/23/15 4:15 PM


WIR

ELES

S &

RF

anal

ogic

tips

.com

DESIGN WORLD does not pass judgment on subjects of controversy nor enter into dispute with or between any individuals or organizations. DESIGN WORLD is also an independent forum for the expression of opinions relevant to industry issues. Letters to the editor and by-lined articles express the views of the author and not necessarily of the publisher or the publication. Every effort is made to provide accurate informa-tion; however, publisher assumes no responsibility for accuracy of submitted advertising and editorial information. Non-commissioned articles and news releases cannot be acknowledged. Unsolicited materials cannot be returned nor will this organization assume responsibility for their care.

DESIGN WORLD does not endorse any products, programs or services of advertisers or editorial contributors. Copyright© 2015 by WTWH Media, LLC. No part of this publication may be reproduced in any form or by any means, electronic or mechanical, or by recording, or by any information storage or retrieval system, without written permission from the publisher. Subscription Rates: Free and controlled circulation to qualified subscribers. Non-qualified persons may subscribe at the following rates: U.S. and possessions: 1 year: $125; 2 years: $200; 3 years: $275; Canadian and foreign, 1 year: $195; only US funds are accepted. Single copies $15 each. Subscriptions are prepaid, and check or money orders only.

Subscriber Services: To order a subscription or change your address, please email: [email protected], or visit our web site at www.designworldonline.com

POSTMASTER: Send address changes to: Design World, 6555 Carnegie Ave., Suite 300, Cleveland, OH 44103

EDITORIAL

Editorial DirectorPaul J. [email protected]@dw_editor

Managing Editor Leslie Langnau [email protected]@dw_3Dprinting

Executive EditorLeland [email protected]@dw_LeeTeschler

Senior EditorMiles [email protected]@dw_Motion

Senior EditorMary [email protected]@dw_marygannon

Senior EditorLisa [email protected]@dw_LisaEitel

Assistant Editor Michelle [email protected]@wtwh_Michelle

Contributing EditorAimee [email protected]

Contributing EditorMike [email protected]@DW_MikeSantora

NEW MEDIA/WEB/

BUSINESS DEVELOPMENT

Web Development ManagerB. David [email protected]@wtwh_webdave

Web Development SpecialistPatrick [email protected]@amigo_patrick

Integrated Media SpecialistJohn [email protected]@wtwh_jhansel

Video CoordinatorJoshua [email protected]@wtwh_josh

Video InternJay [email protected]@wtwh_Jay

Business Development ManagerPatrick [email protected]@wtwhseopatrick

Online CoordinatorJennifer [email protected]@wtwh_jennifer

Director, Audience DevelopmentBruce [email protected]

AccountingJill Steinbuch [email protected]

GRAPHICS

Director, Creative ServicesMark [email protected]@wtwh_graphics

Graphic Designer Matthew [email protected]@wtwh_designer

Graphic Designer/Production Coordinator Margaret [email protected]@wtwh_Meg

Traffic ManagerMary [email protected]

MARKETING

Marketing ManagerStacy [email protected]@wtwh_Stacy

Marketing CoordinatorCarli [email protected]@wtwh_Carli

Marketing CoordinatorNicole [email protected]@wtwh_Nicole

Media and Events CoordinatorHeather [email protected]@wtwh_Heather

Social Media SpecialistJen [email protected]@wtwh_jenFo

llow

the

who

le te

am o

n tw

itter

@D

esig

nWor

ld

CO

NN

ECT

W

ITH

US!

2011, 2012, 2013, 2014

2014 Winner

Crain’s Cleveland Business Fast 50 2014

WORLDA DESIGN WORLD RESOURCE

WTWH Media, LLC

6555 Carnegie Avenue, Suite 300

Cleveland, OH 44103

Ph: 888.543.2447

Fax: 888.543.2447

Staff_EE_April_Vs1.indd 8 4/24/15 4:48 PM

800 453 6202

>> Accelerate your system design productivity at ni.com/labview-platform

©2013 National Instruments. All rights reserved. LabVIEW, National Instruments, NI, and ni.com are trademarks of National Instruments. Other product and company names listed are trademarks or trade names of their respective companies. 11217

NI LabVIEW is the only comprehensive development environment with the

unprecedented hardware integration and wide-ranging compatibility you need to meet

any measurement and control application challenge. And LabVIEW is at the heart of

the graphical system design approach, which uses an open platform of productive

software and reconfi gurable hardware to accelerate the development of your system.

Infi nite Designs, One Platformwith the only complete

system design environment

LabVIEW system design software offers unrivaled hardware integration and helps you program the way you think–graphically.

12117 Infinite Designs Ad_US.indd 1 6/25/13 12:01 PMNational Instruments 8-14.indd 9 4/24/15 12:16 PM

analogictips.comWIRELESS & RF | UPDATES


DARPA asks for comm chips that consume zero power

Magnetic induction lets earbuds go wireless

he Defense Advanced Research Projects Agency (DARPA) is looking for ways of developing wireless, event-driven sensing schemes where the electronics remain dormant—effectively asleep yet aware—until an event wakes them. Dubbed N-ZERO,

the program wants to exploit the energy in the right signal signatures while rejecting noise and interference.

Basically, DARPA wants to find ways of drastically cutting the quiescent current that circuits consume while waiting for a specific event or activity.

“Our goal is to use the right signal itself to wake up the sensor, which would improve sensors’ effectiveness and warfighters’ situational awareness by

drastically reducing false alarms,” said Troy Olsson, DARPA program manager.

The goal is to use less than 10 nW during the sensor’s asleep-yet-aware phase. Specifically, N-ZERO seeks to extend unattended sensor lifetime from weeks to years. Alternatively, N-ZERO could also reduce battery size for a typical ground-based sensor by a factor of 20 or more while still keeping its current operational lifetime.

N-ZERO intends to initially concentrate on improving capabilities for sensors used

for RF, electromagnetic, acoustic, and inertial detection and analysis. DARPA said that if N-ZERO is successful, the resulting technology could similarly untether the Internet of Things, the ever-expanding global network of wirelessly connected devices, projected to reach 30 billion by 2020.

ear Field Magnetic Induction (NFMI) technology is being used to field truly wireless earbuds that stream wireless audio from ear to ear.

The technique, recently demonstrated by NXP Semiconductor, uses the NXP NxH2280 NFMI based radio transceiver. Existing earbuds use wired connections largely because sending a stereo audio stream toward two distinct earbuds is not possible with the 2.4-GHz technology used in a standard Bluetooth A2DP profile, which supports only point-to-point connections. Forwarding a high-quality audio stream from one ear to the other using reasonable

power levels is notoriously difficult using 2.4-GHz technology because most of the signal is absorbed by the human body tissue.

Near Field Magnetic Induction (NFMI) overcomes the difficulty. NFMI systems are designed to contain magnetic field energy around the communication system so it does not radiate into free space. The near-field power density attenuates at a rate proportional to the inverse of the range to the sixth power, or -60 dB/decade. The short range involved in NFMI (about 1.5 m) also creates a private network, making it is much less susceptible to interference than 2.4-GHz transceivers. The NFMI carrier frequency is

By advancing state-of-the-art sensing capabilities for national security through N-ZERO, DARPA could help make the Internet of Things more efficient and effective across countless scenarios and environ-ments, thus transforming the way people live. —Troy Olsson, DARPA program manager

T

N

REFERENCESN-ZERO announcement: darpa.mil/NewsEvents/Releases/2015/04/13.aspx

Short news_EEApril_Vs3.indd 10 4/24/15 3:49 PM

analogictips.com UPDATES | WIRELESS & RF

analogictips.com 4 • 2015 DESIGN WORLD 11

Wi-Fi, IoT apps looking at better links through steerable antennas

arbled Wi-Fi reception caused by factors such as multipathing and reflections could increasingly become non-problems thanks to new techniques that adapt

reception to changing RF environments. One recent development in this area is an active steering IC, complete with an embedded processor for

G

13.56 MHz and has a wavelength (λ) of 22.1 m. The crossover point between near-field and far-field is at approximately λ/2π or 3.52 m for 13.56 MHz.

NXH2280 is NXP’s second generation fully-integrated, single-chip NFMI radio transceiver. Operating from a 1-V supply, it consumes 1.5 mW when audio is streamed in the earbud application. The chip supports a maximum bit rate of 596 kbps, and can be configured to operate either stand-alone or in conjunction with an MCU. An embedded ultra-low-power CoolFlux DSP handles audio processing algorithms. The network stack supports up to 15 devices, low latency communication (< 5 msec) and multiple simultaneous audio and data streams in receive and transmit mode.

REFERENCESNXP Semiconductor nxp.com

Multiple Input Multiple Output (MIMO) applications. Developed by Ethertronics Inc. and dubbed the EtherChip EC482, it uses what’s called active steering to help signals navigate multiple walls and ceilings as might separate a router from a Wi-Fi device. Algorithms on the chip’s processor monitor RF link performance to generate up to four radiation patterns and select the optimal antenna for the best performance. The operating frequency range is 100 MHz to 7 GHz.

The chip can work in conjunction with a special antenna designed for providing multiple radiation patterns from which the chip can choose depending on the conditions. The antenna employs a multi-layer isolated magnetic dipole (IMD) comprised of an IMD element positioned above a ground plane, along with conductive

elements, slot regions and capacitive elements. The range of frequencies covered is determined by the shape, size and number of elements in the physical configuration of the components.

The IMD element confines the electromagnetic currents on the antenna to boost efficiency and limit frequency de-tuning as a function of change in surroundings. The distance between the ground plane and the IMD element or the conductive element is varied to give a desired frequency characteristic. The IMD element and conductive element can be offset from one or more additional IMD elements to get a desired bandwidth.

REFERENCESEthertronics Inc. ethertronics.com


analogictips.comWIRELESS & RF | UPDATES


Chipless RFID tags could eliminate barcodes

esearchers at Monash University in Austrailia are developing chipless RFID tags that will someday eliminate the need for barcodes on inexpensive consumer products such as

plastic water bottles.Monash Dept. of Electrical and Computer Systems

Engineering Prof. Nemai Karmakar said his research team has developed fully printable tags for metals and liquids including water bottles and soft-drink cans. The tag can be printed with an inkjet printer and can be read once attached to reflective surfaces such as metal cans and water bottles. Until now, reflective media have tended to interfere with the technology.

Karmakar said the team was believed to be the first to develop fully printable chipless RFID tags on paper and plastics, and the technology could revolutionize the multi-billion dollar RFID market.

“However, it can still contain a high amount of data and information. The main challenge that we have overcome is to transfer the technology to paper and plastic while retaining the required printing resolution. Uniquely, the 60-GHz millimeter-wave tag can handle printing errors and surface variations. It’s very promising indeed in its ability to revolutionize the multi-billion dollar RFID market.”

Karmakar said the chipless RFID tag could also work in temperatures above 80º C and at cryogenic temperatures.

R

A millimeter-wave chipless RFID tag devised by researchers at Monash University.

The new chipless RFID technology is a high-data-capacity millimeter-wave barcode system that operates at 60 GHz. It is much smaller than any other commercially available chipless RFID tags,

— Nemai Karmakar, Monash University Professor

The researchers, based in the Monash Microwave, Antenna, RFID and Sensor Laboratory, recently received a $90,000 grant from Xerox to fund further development work.

REFERENCEMonash University monash.edu/news/show/chipless-tracker-could-transform-barcode-industry


analogictips.com UPDATES | WIRELESS & RF


Better understanding of electromagnetism could bring antennas-on-a-chip

niversity of Cambridge researchers in the UK say they have figured out one of the mysteries of electromagnetism, and that their discovery could lead to the design of antennas small enough to be

integrated into an electronic chip. Writing in the journal Physical Review Letters, the

researchers proposed that electromagnetic waves come not only from the acceleration of electrons, but also from a phenomenon known as symmetry breaking. The idea could help identify the points where theories of classical electromagnetism and quantum mechanics overlap.

Radiation from electron acceleration has no counterpart in quantum mechanics, where electrons are assumed to jump from higher to lower energy states. The thought is that observations of radiation resulting from broken symmetry idea could help link the two fields.

The new theory could also help explain the behavior of dielectric antennas. “In dielectric aerials, the medium has high permitivity, meaning that the velocity of the radio wave decreases as it enters the medium,” said Dhiraj Sinha, the paper’s lead author. “What hasn’t been known is how the dielectric medium results in emission of electromagnetic waves. This mystery has puzzled scientists and engineers for more than 60 years.”

Working with researchers from the UK’s National Physical Laboratory and an antenna maker called Antenova, the team studied thin films of piezoelectric materials and found that at a certain frequency, these materials become not only efficient resonators, but efficient EM radiators as

well. The researchers claim the reason for this phenomenon is a breaking of the symmetry of the electric field associated with the electron acceleration.

Symmetry breaking can also apply in cases such as a pair of parallel wires in which electrons can be accelerated by applying an oscillating electric field. “In aerials, the symmetry of the electric field is broken ‘explicitly’ which leads to a pattern of electric field lines radiating out from a transmitter, such as a two wire system in which the parallel geometry is ‘broken’,” said Sinha.

“If you want to use these materials to transmit energy, you have to break the symmetry as well as have accelerating electrons—this is the missing piece of the puzzle of electromagnetic theory,” said Cambridge Professor of Engineering Gehan Amaratunga. “I’m not suggesting we’ve come up with some grand unified theory, but these results will aid understanding of how electromagnetism and quantum mechanics cross over and join up. It opens up a whole set of possibilities to explore.”

REFERENCE: Dhiraj Sinha & Gehan Amaratunga, Electromagnetic Radiation Under Explicit symmetry Breaking, Physical Review Letters, 114, 147701 (2015)

University of Cambridge cam.ac.uk/research/news/new-understanding-of-electromagnetism-could-enable-antennas-on-a-chip#sthash.Ai8iFZDJ.dpuf

U

Researchersusedathinfilmofpiezoelectricmaterial with interdigital electrodes where only one side of the electrical terminals are excitedwithothersidefloatingfreelytogetan asymmetric excitation. Under symmetric excitation, the resonator behaves like a conventionalfiltertransferringenergyfromthe input to the output section.


14 DESIGN WORLD 4 • 2015

How to spec inductive coilsMICHAEL HARPER Coil Experts, LLC

It pays RF designers to

know a few things about

the way coil-winding

houses fabricate coils.

INDUCTIVE coils are ubiquitous in any kind of RF or wireless circuit. But some of the tiny coils

that characterize RF work can be challenging to produce. Custom coil designs often get sent to commercial coil winding houses, such as Coil Experts. These winding houses can create devices with highly repeatable qualities. However, not all coil winders are the same. Here are a few things designers might want to know about working with coil winders before beginning the coil design process.

First the basics: Designers must specify more than just the inductance of the coil they need. A commercial coil winder will want to know the general dimensions of the coil including the length-to-width ratio, the coil size and shape, the type and size of the wire, the dc resistance, how the leads should be configured, the type of wire insulation, and so forth.

Fortunately, the equations for calculating many of these factors are well-known and available in handbooks. The coil-design process itself tends to be iterative, with the designer starting with desired qualities and ascertaining through repeated calculations whether the design envisioned is something that works in the real world.

Coil winders can produce tiny coils like this one, which consists of ultra-fine copper wire wound on a mandrel.

WIR

ELES

S &

RF

anal

ogic

tips

.com

Photo: Josh Jones

Coil_EEApril_Vs4.indd 14 4/24/15 3:49 PM

Crystek_EE_4-15_ad.indd 15 4/23/15 4:39 PM


State-of-the-art coil-winding capabilities can produce devices such as these air coils wound from ultra-fine platinum wire for a medical application..

The parameters specified for coils generally depend on the application area. The usual range of inductance for RF coils is in the microhenry range. Coils made for ac and RF circuits have inductance requirements, but perhaps no specification for their dc resistance. In contrast, coils destined for use in dc circuits generally have dc-resistance requirements.

One of the main parameters the designer specifies is the number of turns in the coil. Of course, the number of turns determines the coil inductance to a high degree. A point to note is that the coil winder won’t give a tolerance specification for coil inductance; a batch of coils that all use the same length of wire and the same winding pitch will all provide the same inductance to a close degree. Thus, winding pitch and manufacturing process control basically sets inductance tolerance.

Most coils are wound on round forms, and round coils are the easiest to make. That is because the constant pull on the wire during coil forming puts less stress on the wire than is the case when making coils that are square, rectangular or triangular. The wire experiences a small shock as it is

pulled around each corner when making these latter shapes. But this limitation may make it hard to realize such shapes with ultra-fine wire, and the tight process parameters needed for fabrication may be beyond the means of some coil winders. Nevertheless, non-round coils may sometimes be necessary to obtain special electromagnetic qualities.

There are also practical limits to the size of the coil. A general rule of thumb is that some coil-winding houses can fabricate coils down to 1-mm diameters. That said, it’s best not to assume smaller devices are impractical. RF designers constantly push the state-of-the-art, so the only real way to ascertain whether a form factor is unfeasible is to discuss it with the coil house.

The same can be said for wire diameter. The RF designer always specifies the wire size. Electrical current is the primary factor that determines minimum wire diameter. Typically, RF coils are low current devices. Their wire size usually ranges from ultra-fine (58 to 50 AWG) to very fine (49 to 40 AWG) to fine (39 to 30 AWG). Clearly, coil fabricators can work

WIR

ELES

S &

RF

anal

ogic

tips

.com

Photo: Josh Jones



analogictips.com

with surprisingly small wire diameters, considering that 58 AWG (0.00991 mm) wire is so delicate that it can break if someone breathes on it too hard.

In cases where an ultra-fine wire would be adequate to handle the current involved, designers often use a larger wire simply for mechanical strength and easier handling. For example, the strength of the leads is often a major concern. Ultra-fine wire leads are quite fragile, so designers might specify thicker wire.

Wire size and coil dimensions also impact costs. The finer the wire, the more fragile the coil. And fragile coils often can only be handled with tweezers under magnification. They are usually packed in some sort of egg carton tray. The more fragile the coil, the more delicate the handling and packing procedures. Delicate handling and packing costs more.

Small coils also may only be practical in relatively high volumes. That’s because microcoils are difficult to make without the use of specialized machines. Only coil winders that deal in relatively high volumes are likely to have the necessary coil winding machines. Similarly, not all coil winders have the facilities to handle micro coils. Older generic coil winding machines often can’t provide the necessary accuracy.

Finally, RF designers also specify the insulation used on the coil. Selection criteria for coil insulation typically includes the level of heat expected and whether there will be mechanical or chemical stress. Biocompatibility also becomes an issue in all medical applications.

REFERENCESCoil Experts, LLCcoil-experts.com

Coil Experts uses the new Model 101 Micro Coil Winder, a bench top machine designed to wind micro coils. These are typically used in MRI imaging, navigation, RFID and other similar applications. The Model 101 is a product of Machine Control Specialists, Inc.

WIR

ELESS

& R

F



How to sniff out EMC problems

An engineer in a relatively small company usually must rely

on experience and tribal knowledge to design a product that is electromagnetically compatible (EMC) with other equipment. Many designers and manufacturers don’t have the luxury of their own in-house RF test lab with an EMI-proof chamber equipped with expensive RF test-gear. That is why it is estimated that more than 50% of products fail the first time through an approved EMC testing facility. And failing is expensive. Retest costs are high and a retest may push back project schedules and market introduction dates.

However, pre-compliance testing has now become much more affordable. Such “early sniffing” can provide a good idea of problems before the fix gets expensive. Standards vary by country, but common EMC regulations for the U.S. are described in FCC Part 15, with subsections depending on whether or not the product is a consumer item. For Europe’s CE mark, EN55011 is the common standard, while some products have even stricter requirements.

Do-it-yourself, homegrown bench testing is becoming more necessary to head off problems passing the required emissions

standards. It’s expensive to book an approved, certified test facility to ensure your new product meets EMI

emissions limits. But it’s even more costly if your product fails and must go back to the bench to

fix radiation issues, then return for a retest. Fortunately, it is now quite

inexpensive to purchase an RF

ALAN LOWNEPresidentSaelig Co. Inc.

It pays to do preliminary studies before

sending products off to the test lab for

electromagnetic compatibility problems.

Typically near-field sniffer probe sets (above left) include magnetic (H) field and electric (E) field passive, near-field probes and perhaps an extension handle to provide access to remote areas in larger units. The probes provide a fast and easy means of detecting and identifying signal sources that could prevent a product from meeting federal regulatory requirements. Transverse Electromagnetic (TEM) or Crawford cells (below right) generate accurate electromagnetic waves over a frequency range of up to several megahertz. EM waves generated in the cell propagate in transverse mode and have the same qualities as a plane wave. TEM cells generate a consistent electromagnetic field for testing small RF devices. An external test signal applied through the input port of the TEM cell generates a plane wave test field inside the cell. The radiation field emanating from a device-under-test sitting in the cell can also be detected through the port using an EMI receiver.

WIR

ELES

S &

RF

anal

ogic

tips

.com

SAELIG_EEApril_Vs3.indd 18 4/24/15 3:50 PM

analogictips.com

4 • 2015 19within 3 DAYSBuilt and Shipped

BUILT ANDBUILT AND

LISNs(lineimpedancestabilizationnetworks)arebasicallylow-passfilterstypically placed between an ac or dc power source and the device-under-test to create a known impedance and to provide an RF noise measurement port. An LISN also isolates the unwanted RF signals from the power source. LISNs can help get a handle on conducted emissions for diagnostic and pre-compliance testing.

spectrum analyzer with associated near-field probes or antennas, so you can get a first look at basic EMC/EMI problems before they do too much damage. Spectrum analyzers are now affordable—prices of quite sophisticated bench-top units have dropped dramatically in the last few years. They are even available in the form of a USB thumb drive, which can be connected to a Windows PC or tablet.

A set of sniffer probes—which look a bit like bubble wands for kids—can quickly find both the sources of problem radiation and help gauge the success of proposed fixes. The probes can disturb the field being measured, bringing added capacitance in proximity to an unwanted oscillator. Experience will reveal how useful and valuable this EMI tool can be.

Other useful tools for in-house testing include TEM (Transverse Electro-Magnetic) cells for radiated emission and immunity testing, and line impedance stabilization networks for conducted-emission testing of dc-powered equipment. Economical wideband amplifiers can boost the sensitivity of commercial spectrum analyzers or digital oscilloscopes using the scope’s FFT spectrum analyzer setting—but oscilloscopes are often not sensitive enough to provide much useful information.



WIR

ELES

S &

RF

anal

ogic

tips

.com

Shielded enclosures—either bench-top metal boxes or quick-erecting portable EMI tents—are useful for keeping out ambient radiation and are well within the economics of most companies. The enclosure construction employs multiple layers of conductive silver/copper/nickel RoHS-compliant materials. These portable EMI test enclosures provide high RF/EMI attenuation for a variety of EMI-quiet applications. Their average shielding effectiveness is up to -98.9 dB at commonly used cellular and Wi-Fi frequencies. They erect quickly and are collapsible. External aluminum tent supports can be either fixed-frame or E-Z Up assemblies, so they can disassemble quickly when not in use.

Often, an internal or external vestibule is included as part of the design so personnel can enter and exit during tests without compromising the results. Tent options include size and orientation, use of through-plate connectors with integrated high performance filters for ac power configurations up to 100 A, communication connectors that include GigE and USB 3.0, and so forth. Tents even have their own ventilation and air-conditioning, a white ESD lining, and hardened, motion-detecting LED lighting.

SNIFFER PROBESWhether used inside shielded tents or on a bench-top, small sniffer probes for near-field testing can help discern EMI problems. You can often quickly isolate the source of EMI emissions using hand-held H-field and E-field probes. Near-field magnetic (H-field) and electric (E-field) probes can help home in on radiation problems in circuit layouts, cables and shielding. H-field probes use a conductive loop to detect magnetic fields produced by circuit board signals or switching power supplies. The probes produce a voltage corresponding to the magnetic field detected in front of the loop.

To find emissions on individual pins or PCB traces, use E-field probes in direct contact with circuit traces. It may be useful to evaluate emissions with more than one size of H-field probe. Kits often include several probe sizes.

Openings in enclosures and shielding cans can let emissions escape and cause current flow in surrounding metal enclosures. EMI gaskets or robust soldering techniques may reduce the offending signals. You can use an H-field probe to compare fields “before and after” inside an enclosure, or to gauge the RF energy a cable picks up from the source because of poor connector shielding. A near-field probe will help identify a cable acting as a radiating antenna. Current probes positioned around the cable can measure common-mode current that can cause unwanted emissions. A dipole antenna will reveal far-field emissions and is useful for checking trial fixes.

It can be challenging to modify a layout or circuit to eliminate an unwanted radiator. Typical methods of improving EMC include changing the clock speed, reducing rise-times and adding capacitors or inductors for filtering. Judicious changes to PCB layouts can also help with EMC. Usual approaches include moving a trace to a sub layer instead of the top or bottom layer, beefing up the grounding scheme, adding a shielding can, or redesigning the PCB as a multilayer board with ground planes on the outermost layers.

Another useful technique employs a low-EMI clock oscillator. This is one that slowly moves its center frequency by about 1% or so. An example is the Euroquartz EQHM series of low-EMI oscillators. They reduce EMI radiation using spread spectrum technology to modulate the output frequency with a low frequency carrier, spreading the peak energy of the output frequency and its harmonics over a wider bandwidth. This can

Benchtop spectrum analyzers have become more economical in recent years, making it possible for many engineering departments to run their own EMC tests prior to submitting products to third-party testing organizations. An extreme example is the Triarchy Technologies spectrum analyzer built into a USB dongle. It has an advertised frequency range of 1 MHz to 5.35 GHz and costs $599 online.



analogictips.com

eliminate the need for many expensive EMI protection aids such as PCB ground-planes, metal shielding, ferrite beads, RF gaskets and EMI filters.

A gap in shielding cans will also cause unwanted emissions. Changing the shielding configuration or soldering points can reduce the radiation. Typical problems that might be found include bypass capacitors too far away from ICs, poor power and ground tracks that allow ringing, and gaps in shielding. When RF emission emanates from power lines, adding an inductor or ferrite bead in the power track may be all that’s needed. Clock lines are often

another emission source at low frequency, so it’s best to avoid long PCB clock traces on an outside layer.

After you’ve done all your home-work, keeping unwanted emissions small and within limits, the product can go to the RF test lab with a high probability of success and assurances you’ve probably saved your company a lot of money!

REFERENCESSaelig Co. Inc., saelig.com



WIR

ELES

S &

RF

anal

ogic

tips

.com

External mixer

Low band path

High band path

Preselector

Sweep generator

322.5 MHz

LO 3.8 to 8.7 GHz

To externalmixer

4.8 GHz

3.6 GHz 5.1225 GHz 322.5 MHz 22.5 MHzAnalog or digital IF

300 MHz

Display

Analyzer input

Waveguide input

IF out

IF in

Analyzing millimeter frequencies with external harmonic mixing

SHORT-RANGE connections are increasingly making use of millimeter-wave signals.

Application examples include satellite-to-satellite links, point-to-point radios, secure communications, and wireless audio and video connections. Measurements of millimeter-wave signals between 30 and 300 GHz often depend on the use of an external mixer. The mixer becomes the front end of instruments such as microwave signal analyzers, bypassing the input attenuator, preselector and first mixers. This economical technique can extend measurements to 325 GHz and beyond.

The external mixer uses a harmonic of the analyzer’s local oscillator (LO) to down-convert the input signal. The LO output of the analyzer is sent to the intermediate frequency (IF) input of the mixer. The analyzer sets the outgoing LO signal to a frequency that will mix down the signal of interest to be within the analyzer’s IF range.

CHERISA KMETOVICZKeysight Technologies(formerly Agilent Technologies electronic measurement business)

External mixers can help

extend the capabilities of

signal analyzers so these

instruments can handle

frequencies ranging into

several hundred gigahertz.

A block diagram shows how an external mixer would connect with a signal analyzer to accurately measure signals at higher frequencies.

KEYSIGHT_EEApril_Vs3.indd 22 4/23/15 5:05 PM


WIR

ELESS

& R

Fanalogictips.com

InputImageRF 6-RF 6+RF 8-RF8+

Harmonic mixing tuning lines

LO frequency (GHz)

7 7.5 8 8.5 9 9.5 10 10.5 11

10.36

10.47

62.500

8+

8-

6+

6-

61.855

80

75

70

65

60

55

50

IF frequency = 322.5 MHz

This method has two important tradeoffs. First, external mixing has the potential to reduce amplitude sensitivity because there is no preamplifier; however, the resulting sensitivity may still be quite good. Second, it may increase system phase noise because harmonics of the LO typically have greater phase noise than that of the fundamental.

Typically, a signal analyzer that supports external mixing has one or two additional connectors on its front panel: an “LO out” port that routes the analyzer internal first LO signal to the external mixer, and an “IF in” port that accepts the mixer IF output.

An extension of this capability is a “smart mixer” that improves the measurement performance and functions available through external mixing. Using a built-in USB link, each mixer can identify itself by model and serial number. It can also provide information such as harmonic number, LO path-loss data and conversion-loss calibration data.

To further simplify test setups, the latest analyzers have a single diplexed port that sends an LO signal that ranges

from 3 to 14 GHz and receives a 322.5 MHz IF signal from the external mixer. With this wide difference in frequency, a single coaxial interconnect cable can carry both signals.

EXTERNAL MIXING PRODUCTSAn external mixer typically uses higher harmonics of the signal analyzer’s first LO; in some cases, the first LO frequency is doubled before being sent to the external mixer. Use of higher fundamental LO frequencies helps reduce mixer conversion loss, which is inescapable but correctable on the analyzer display through digital compensation.

The LO and its harmonics mix with all signals present at the input, including the desired input signal and all out-of-band signals. This generates mixing products that can be processed through the first IF along with valid signals.

Most signal analyzers use a tunable preselection filter, or preselector, to attenuate unwanted signals before they reach the mixer. An external mixer that lacks a preselector will produce false responses or images that appear on the analyzer display.

A plot of signal frequency versus LO frequency includes the products of 8+/8- (blue & green) and 6+/6- (yellow & gray) mixing.



WIR

ELES

S &

RF

anal

ogic

tips

.com

A function called “signal identification” can deal with these unwanted artifacts.

An example will illustrate how this works. Keysight analyzers have an IF of 322.5 MHz. Assume the hardware is a 50-GHz signal analyzer connected to an unpreselected external mixer that covers 50 to 75 GHz (V-band) and uses “six minus” (6–) mixing, which is the lower of the two responses produced by the sixth harmonic of the 322.5 MHz IF.

If the measured signal is at 62.5 GHz, the correct response will appear at that frequency and the companion mixing product will appear at an indicated frequency of 61.855 GHz, which

is below the real response by 645 MHz (twice the 322.5 MHz IF).

Using the fundamental behavior of mixing, a signal analyzer can identify the real signal and reject the false component. A technique called image shift retunes the LO frequency on alternate sweeps. The retuning is given by the equation (2 x fIF)/N where fIF is the IF frequency, N is a given harmonic of the LO, in this case, -6. Image shifting shifts the Nth harmonic by twice the IF frequency, thereby causing the desired signal to appear in the center of two overlapped display traces while a single image component appears in each trace above or below the correct signal.

An extension of this technique is the “image suppression” method. This mode takes two sweeps using the minimum hold function, which saves the smaller value of each display point from each sweep. The first sweep uses the normal LO tuning value, while the second sweep offsets the fundamental LO frequency by (2 x fIF)/N. As with image identification, the correct harmonic will appear at the same place on the display, producing a high value in each sweep. Because the unwanted images will produce a high value in one sweep and a low value in the other, only the low will display.

It is important to note that the two signal-identification methods produce correct frequencies but yield less-precise amplitude values. Once the correct frequency component has been identified,

With the image shift technique, the two sweeps each produce the desired signal in the center of the display and images above and below the real signal (blue and yellow traces, respectively).

Through use of image suppression, the real signal remains and the minimum hold function replaces false peaks with the lowest value appearing at that point on the trace to produce a clean display, as evident in this example.



WIR

ELESS

& R

Fanalogictips.com

the next step is to turn off the signal-identification mode and zoom in on the signal by narrowing the measurement span.

To ensure accurate amplitude measurements, it is necessary to first enter the calibration data for the external mixer. Normally, the mixer manufacturer will provide this data in tabular form with conversion loss in dB at a number of frequency points across the mixing band. These values are entered into a correction table in the signal analyzer, which then uses the data to compensate for mixer conversion loss.

The smart mixers described here remove the chance of human error by automatically transferring conversion-loss information into compatible signal analyzers through a USB connection. This calibrates the signal analyzer reference level at the input of the external mixer.

Even when working at higher frequencies, the signal analyzer must also handle a variety of measurement and analysis tasks. They include making application-specific measurements such as adjacent-channel power (ACP), noise figure and phase noise; performing digital modulation analysis defined by industry standards such as LTE, 802.11 or Bluetooth; and performing vector signal analysis of highly complex signals. Many signal analyzers have these capabilities built in or offer them through

“measurement apps” integrated into the instrument interface.

The complementary cumulative distribution function (CCDF), which shows power statistics, is another measurement capability built into many of today’s signal analyzers. CCDF measurements provide statistical information such as the percentage of time the instantaneous power of a signal exceeds the average power by a certain number of decibels. This information is important in the design and development of power amplifiers, which must handle instantaneous signal peaks with minimum distortion within designs intended to minimize cost, weight and power consumption.

Other examples of signal analyzer measurements include occupied bandwidth, third-order intercept (TOI),

harmonic distortion and spurious emissions. Here, fundamental instrument settings, such as center frequency, span, and resolution bandwidth, depend on the specific radio standard to which a device is being tested. Many modern signal analyzers have the necessary settings stored in memory, saving time and ensuring accurate results.

Another example is measurement of noise figure, which directly affects the sensitivity of receivers and other subsystems. Although dedicated noise-figure analyzers are available, some signal analyzers handle these measurements as an optional capability. This option provides control of the noise source that drives the input of the DUT and also includes the firmware needed to automate the measurement process and display the results.

PHASE INFORMATION Digital modulation techniques use phase and amplitude information to carry more baseband data in a limited amount of spectrum and period of time. That is why today’s signal analyzers handle phase and amplitude data, which is essential to understanding the in-phase and quadrature (I/Q) components of complex modulation schemes. Fortunately, this information is retained when using external mixing.

Complete demodulation and analysis requires vector signal analysis software. This enables essential measurements, such as error vector magnitude (EVM), which is a key metric for the quality of digitally modulated communication signals. Of course, pure amplitude measurements are still needed to determine signal characteristics such as flatness, adjacent-channel power levels and distortion.

All in all, external mixing techniques can be used to extend the frequency range of proven microwave signal generators and vector network analyzers. This saves time and effort, and helps ensure meaningful measurement results.

REFERENCESKeysight Technologies keysight.com



WIR

ELES

S &

RF

anal

ogic

tips

.com

Cellular amps boost mobile and M2M applications

AS WE ALL KNOW, cellular radio networks use base stations to cover a specific area. Rural areas have fewer base stations.

So cellular coverage in these geographies can be weak or drop out completely between cell sites. Urban areas have different problems: There are often too many users on a base station, which slows data connections and leads to call failures.

Radio waves are absorbed by many substances—examples include earth, vegetation and most building materials. Thus, it is not uncommon to experience signal loss while surrounded by areas where RF signals come through just fine. Signals often drop out, for example, in ravines, canyons and valleys, as well as on flood plains and other areas shadowed from the nearest base station by hills, buildings or both.

The key point is that these network details are all managed on the base station side of the radio link. The mobile station (handset, tablet, data router and so on) is assumed to have a fixed level of performance that is determined by the combination of its radio and antenna performance.

The amount of received signal strength determines whether the connection is stable. Signal strength is easier to assess with data routers than with handsets because routers often provide a more granular measurement than the typical three or five-bar signal-strength graph on a cell phone. But it can be challenging to assess signal strength accurately without a calibrated instrument like a spectrum analyzer. The problem is that readings on uncalibrated instruments can be affected by small variations in component tolerances, which can make different copies of the same device yield different measurement values.

Temperature variations can have the same effect. At different temperatures, an instrument can report varying measurement values though the signal level is identical. Errors due to device offset, unit-to-unit variations and temperature can easily add up to 30 dB. That is the difference between no connection and a solid one. It is advisable to make measurements with the same device, kept at the same temperature, and to assume that a reading could be ±10 dB. Nevertheless, with proper care, one can still make relative measurements of signal strength.

DERMOT O’SHEAPresidentTaoglas Antenna Solutions

Cellular signal amplifiers help bring reliable communications to

remote locales, but there are subtleties to their deployment that

can trip up the uninitiated.

The Penta-band cellular Barracuda Outdoor antenna is an example of a cellular antenna designed for long distance coverage. It has a UV-resistant coating and a fiberglass housing so it can mount oudoors, on poles or on walls with a bracket. The antenna is often used with a booster amp in remote areas or difficult RF environments indoors and outdoors.

Taoglas_EEApril_Vs4.indd 26 4/23/15 5:14 PM

200 Broad Hollow Rd., Farmingdale NY • 631-249-0001 • fax 631-249-0002 • www.batteryholders.com

What are the chances of shock or vibration? Who is the target user and how much dexterity can they be assumed to have? Whether you design defibrillators or glucose meters or some other product, we offer progress reports, prototypes, and design testing to fulfill our pledge of excellence to you.

IS YOUR DEVICE STATIONARY OR PORTABLE?

Memory_Protection_Vs3.indd 27 4/23/15 5:17 PM


WIR

ELES

S &

RF

anal

ogic

tips

.com

Ditto for mobile applications. An antenna buried under the dash or in a metal trunk has a poorer link path than that of an antenna mounted outside the vehicle. A high-quality external antenna can make a difference.

Sometimes, improvement of the radiation path involves moving the antenna away from the radio device, usually with a coaxial transmission line. In general, the thinner the transmission line, the more loss per unit length. The greater the length, the greater the loss. The coaxial loss, in extreme cases, can be enough to cancel out the advantages of a better antenna location. A thicker, heavier, more expensive coaxial line can minimize this loss unless, of course, it’s impractical for installation reasons.

An in-line, bi-directional amplifier can help overcome coaxial losses. In the context of cellular communications, this is commonly referred to as a cellular booster and is a device that contains an amplifier for transmit and/or receive. The amplifier resides on the antenna side of the coaxial link back to the radio. A properly designed device will have similar gain in both the transmit and receive directions because both paths experience the same coaxial loss. RF connections to booster amps always have a 50-Ω characteristic impedance and normally use standard coaxial connectors.

For the receive path, it is best that the incoming signal be amplified before it sees the extra coaxial loss between the antenna and the amplifier and/or between the amplifier and the radio receiver. In contrast, there is no value in having the amplifier at the receiver end of the coaxial line. Adding an amplifier after the coaxial line boosts not only the signal, but also the noise associated with the cable link. The actual signal-to-noise ratio (S/N) gets a little worse because the amplifier itself adds a small amount of noise to whatever it amplifies.

In the transmit path, the signal coming from the radio’s transmitter is much higher than the receive signal. As such, the small amount of added noise from the coaxial loss has little impact on the transmit signal’s S/N ratio. The amplitude of the transmit signal dwarfs any added noise. Thus, on the transmit path, the goal is to compensate for the actual losses and get the transmit signal back up to optimal strength when it goes into the antenna.

It should be noted that the power output of a cellular device is limited by government regulations. Thus, booster amplifiers are not allowed to amplify output

The CSB.01 cellular booster is an example of a modern, high-performance, microprocessor-controlled, bi-directional RF amplifier for the North American 850 MHz cellular and 1900 MHz PCS frequency bands.

There are various ways to handle poor coverage locations. Clearly, the antenna height and transmit power on the network side are fixed. In rare cases, it may be possible to move obstacles out of the link path. But the equipment at the end of the link is usually the only part of the system one can control. To limit the scope of this discussion, we’ll assume the performance details of the radio device are fixed. All cellular devices typically have the same transmit power and similar receive sensitivity metrics. This leaves the selection of antenna, the location of the antenna and the connection from the antenna to the radio as avenues for improving link performance.

The antenna choice usually boils down to either omni-directional or directional radiation patterns. Omni-directional antennas are best for mobile applications or those involving more than one base station. Fixed applications may be able to use a directional antenna to improve the link.

There is a practical limit to variations in antenna performance. An omni-directional antenna might have as little as 2 dBi gain at the band of interest, while a reasonably sized directional antenna would likely have a maximum of 8 dBi. Unfortunately, hills in the way could drop the signal level by 20 dB, while trees could have an impact of 10 dB or more.

Luckily, antenna height is controllable. A 40-ft telephone pole or 60-ft small tower will overcome local obstacles, especially trees. Antenna height is typically the easiest thing to adjust that makes the biggest difference for link quality.



WIR

ELESS

& R

Fanalogictips.com

power to a point where it exceeds levels proscribed by law. So, on the transmit side, the booster is there merely to overcome cable losses.

Similarly, amplifiers designed for the U.S. can work in any country that uses the same frequency bands. Canada, Mexico and some South American countries use some of the same bands as the U.S. Those countries, however, require their own government approvals. So commercial amplifiers usually target specific countries and sometimes even specific carriers.

Cellular booster amps must amplify in two directions simultaneously. Luckily, the transmit and receive signals are at different frequencies. So by using filters, the transmit and receive signals are separated and amplifiers for each mode can operate at the same time. The effectiveness of these filters is one measure of a booster’s quality. Boosters with poor filtering have problems with cellular network protocols, like CDMA, that transmit and receive at the same time.

The last part of the system is the power supply. The amplifier is an active element, so it needs power. The transmit amplifier, in particular, uses about the same amount of power as the radio transmitter that it is amplifying, so it needs a similar power source. Power can be delivered to the amplifier over the coaxial cable using bias tees. Alternatively, separate power can be supplied at the booster side of the cable.

The output of the power supply must be filtered to ensure the amplifier works properly. This filtering can take place inside the booster or using external filter elements. It is common to see this filtering done externally to lower the cost of the booster when it is used in fixed (meaning not vehicular) applications where clean power is easily available.

CELL BOOSTER CONSIDERATIONSIf the amplifier loses power, in many cases it can block the radio signals.

Thus, if the booster amplifier fails, it can actually prevent the radio link from working in a situation where it otherwise would. As an example, consider a typical cellular router mounted in a bus.

The radio mounts behind the driver with the other electronics. There are 30 ft of coaxial cable and an associated loss to the back of the bus where the antenna mounts. A booster mounts just inside the bus next to the antenna, compensating for the coaxial losses back to the radio. The bus is constantly moving, and, in many places, it receives a strong network signal.

If the amplifier fails for some reason, the bus no longer can connect, even when it’s in a good coverage area, because the amplifier blocks the signal between the antenna and the radio. There are cellular booster products designed as a fail-safe if their power or electronics go out for some reason, maintaining a path between the antenna and radio.

Another item to consider is amplifier gain. Not all commercial products provide the same amount of gain; it varies product to product. Typically you will see commercial amps providing 12 to 25 dB of gain with some vendors offering two versions of a given amp, one high gain, the other low.

The amount of gain you need is directly proportional to the loss in your coaxial system. As long as the gain is more than the coaxial loss, you’re fine. The only downside of too much gain is that it could cause problems if the system were to get close to a cell tower, as might happen in a mobile application. Some amps (like those from Taoglas) will sense when received signal levels are getting too strong. They will then switch themselves out.

Whenever there is more than 6 dB of coaxial loss, a cellular booster can add to total system performance. As long as the amplifier gain equals or exceeds the losses in the coaxial cable, the booster is sufficient. Any additional gain only hurts the ability of the amplifier to work in strong signal

areas. If total coaxial loss is less than 6 dB, a booster is not likely to make a noticeable difference in performance. The greater the coaxial losses, however, the bigger the impact of an antenna-side amplifier.

Also important: Cellular amps must be designed for the specific bands of interest and, in some cases, for the specific modulation used. The distinctions here can get a little tricky. For example, one new Taoglas product handles the 850 MHz and 1900 MHz bands, and it does so with amplifiers linear enough to support OFDM as used in LTE signaling. That said, it does not support the 700, 787 or 1700/ 2100 MHz bands most commonly used for LTE in the U.S. by AT&T and Verizon. Sprint intends to deploy LTE on 850 and 1900 MHz bands. So a logical question is whether the device can

“support LTE.” It can, but only on those bands, so only for Sprint.

Therefore, the distinctions are not really a matter of 3G versus 4G versus 5G, so much as a matter of what bands are supported, what carriers use those bands and what technology the carriers use on those bands.

There are other issues that help differentiate commercial booster amps. Competitive factors include power supply robustness, the ability of the device to fail in safe or functional mode, mounting options, robustness of the enclosure, carrier and government certifications, customer service and quality of documentation.

Finally, it is worth noting that the FCC changed the rules for cellular booster amplifiers in February of 2013. Rules now require the devices to be much more intelligent so they cannot cause any network issues. These features are mandatory on all legal cellular booster products in the U.S. It’s imperative to get a product that complies with these new rules.

REFERENCESTaoglas Antenna Solutions taoglas.com



WIR

ELES

S &

RF

anal

ogic

tips

.com

Better materials for better antennas

ANTENNAS often don’t look the way they used to. The expanding consumer wireless industry has

fueled rapid advances in injection molding and selective metallization technologies. Wireless antennas have swiftly evolved from printed circuit board and stamped metal designs to molded parts metallized with conductive inks. The high-volume manufacture of wireless antennas has quickly matured these technologies for the consumer market.

Several industries can benefit from this technological maturity. Injection molding of composites as well as processes for selective metallization offer paths for creating compact 3D antenna geometries, saving space and weight, minimizing part count, and even reducing aerodynamic drag for flight and automotive applications. Such 3D antenna designs can be mechanically robust and can withstand harsh environments.

However, some consumer-grade technology doesn’t work well in the harsh environments associated with such areas as aerospace and defense. Here, antenna materials must withstand vibration, mechanical

KATHLEEN FASENFESTSenior Electrical Engineer, Antenna Products TE Connectivity (TE) Aerospace Defense & Marine

Thanks to improved materials and novel processing, antennas today

can be lighter, smaller and easier to make than their predecessors.

Antennas can be embedded into composite enclosures as is the case with this wire access point for in-flight entertainment systems devised by TE Connectivity.

shock, temperature and weather extremes, sometimes for decades. And low-loss metallization and substrates are needed because high-performance applications must minimize termal losses that boost heat and degrade antenna sensitivity.

Structurally, antennas for super-tough settings often make use of machined aluminum parts and thermoset radomes. These fabrication methods are effective but costly. Machined aluminum parts are typically heavier than plastics and require more fabrication time. Plating is commonly used to selectively metallize surfaces. This plating typically involves hand-trimmed masks. The masks make it difficult to maintain the tight tolerances necessary

TE_Connectivity_EEApril_Vs4.indd 30 4/23/15 5:26 PM

Make your mark.

Hand your RF ceramic filter, frequency control and thermal management needs to CTS. Whether it’s wireless infrastructure, telecom, industrial or aerospace, we have the solutions to help you get it done.

www.ctscorp.com

PMS 2925

PMS 541-theirs

PMS 313

Visit us at IMS 2015 • Booth #2930

CTS_EE_4-15_ad.indd 31 4/23/15 5:28 PM


WIR

ELES

S &

RF

anal

ogic

tips

.com

TECHNIQUE ADVANTAGES DISADVANTAGES

Machined aluminum chassis • Well-established technology • Good thermal heat dissipation

• Long machining time • Weight

Hand-built thermoset radomes • Well-established technology • Large structures relatively easy to fabricate

• High touch-labor costs • Process repeatability due to touch labor • Prices are relatively constant regardless of production volumes • Shapes are typically smooth, and small features are difficult to create without secondary machining after molding

Selective plating using hand-trimmed masks

• Well-established technology • High conductivity of traces

• Difficult to achieve consistent parts • Tolerances often do not meet requirements for high frequency applications

Antennas on printed circuit boards

• Well-established technology • Multi-layer boards relatively easy to achieve • Easy component integration onto RF traces

• Embedded antennas not the ideal shape • Designs limited to two-dimensional structures • Conformal antennas difficult to build

for high-frequency applications. Similarly, thermoset radomes not only have a high labor content, but often need secondary machining to realize small features.

Antenna elements are commonly etched onto printed circuit boards (PCBs). PCB antennas allow easy fabrication through proven processes and sometimes can be integrated on the same board with matching circuits and other components. The main drawback is that these antennas are limited to 2D structures. Two-dimensional antennas may not efficiently use all the available space within an enclosure, and therfore, not provide optimal gain and bandwidth.

Recent advances in materials and fabrication technologies now make possible improved antenna designs that are smaller and weigh less, have high integration and are less expensive. Key innovations influencing next-generation antenna designs include composite materials and selective metallization processes. The resulting 3D antennas are mechanically robust and can withstand harsh environmental conditions. And they don’t bust the budget.

New generation composite materials offer two main benefits. They can be tailored to specific applications by careful selection of fillers; and they are injection moldable. Molding can create intricate and precise shapes with high repeatability.

Fillers are the secret sauce that make custom properties possible. Some useful fillers include carbon fibers, glass fibers, hollow microspheres, graphene, carbon nanotubes and foaming agents. Several critical antenna properties can be adjusted by careful application of fillers:

Dielectric Properties. Glass fibers boost the dielectric constant of most composites, reducing antenna size when these composites serve as substrates. Composite materials can also be engineered to provide “designer” dielectric constants through the addition of various filling materials, such as hollow glass microspheres, conductive particles or foaming agents.

Loss Tangent. A high loss tangent offers absorptive properties. Control of the loss tangent lets designers tailor RF absorption. Conductive-fiber fillers can absorb energy radiated in unwanted directions. This can raise antenna gain in a

Strengths and weakness of traditional antenna manufacturing techniques



desired direction by cutting interference. The same direction-specific absorption can be used to shield electronics near antennas.

Lighter Weight. Conductive composites typically weigh 30 to 40% less than aluminum parts. Carbon fiber composites may also be sturdier than aluminum counterparts because they are not easily dented or deformed and won’t corrode.

High Conductivity. Carbon-fiber composites are addressing the need for lightweight, economical, mass-produced, electrically conductive parts to fabricate ground planes or EMI shields. For antenna applications, uses of carbon fiber composites range from ground planes to enclosures.

Newer metallization processes use several techniques to get costs out and improve performance. Key processes here include laser direct structuring (LDS) and conductive coatings. Both are selective metallization processes that can repeatedly make traces and fine structures as small as 100 μm. They are suitable for creating 3D antennas and traces on molded parts.

LDS is a three-step process. First, the substrate is molded in a standard thermoplastic molding process. Next, a laser etches the part to expose a specialized plating additive in the polymer resin. Finally, the substrate goes into an electroless nickel plating bath, where the plating adheres only to the laser-activated plastic.

LDS BENEFITSOne benefit of LDS is that its metallization patterns can change without a mold or mask redesign. A quick update to the laser path allows modification of the artwork design. Because the metallizaion is created using standard plating techniques,

conductive losses are low. This makes LDS a good choice for high-frequency applications.

Conductive coatings make it possible to metallize arbitrary materials—including composites—for creating conformal antennas on nearly any shape. These 3D selective metallization processes can be applied to a wide range of substrates—including plastics, chemically resistant composites, glass, ceramic and metals—and handle a temperature range from -75 to 260° C while resisting corrosion.

High-quality conductive coatings are durable enough to withstand shock, vibration, fluids and salt spray to the levels typically required for aerospace and defense applications. Conductive coatings are compatible with molded or machined parts, so the same process can be used for low-volume prototyping and production.

Composites offer a great approach for building moldable, inexpensive antenna substrates that can be mass produced. These substrates mold into arbitrary shapes and even include mechanical mounting provisions, such as retention ledges and snap features, for easy assembly. The technology offers antenna engineers design flexibility not available with traditional substrate materials.

Composites and selective metallization allow higher levels of antenna integration. The aerospace industry, for example, is developing a highly modular approach to decentralized avionics known as the Mini-MRP, or Mini-Modular Rack Principle. One type of module is a wire access point for in-flight entertainment systems. A composite enclosure can not only include an embedded antenna, but connector shells, shielding and other features.

Selective metallization through conductive coatings offers a good way of creating circuit

Modular antenna arrays, such as this device fabricated by TE Connectivity, increasingly exhibit use of composite materials and conductive coatings in their construction.

analogictips.com

WIR

ELESS

& R

F



WIR

ELES

S &

RF

anal

ogic

tips

.com

traces on composite parts that would more typically be etched using standard circuit board techniques. The conductivity of some coatings can approach the conductivity of bulk copper, adding minimal loss to the circuit while enabling more economical manufacturing. However, this selective metallization process particularly excels when applied to the manufacture of 3D circuit topologies. For example, it can make practical 3D RF couplers and direct-circuit connections to antennas.

Particularly in antenna assemblies for defense or aerospace applications, aluminum parts can account for a large portion of the total assembly weight. And the fabrication of aluminum parts entails significant machining time. Composite ground planes offer 30 to 40% weight savings over traditional aluminum parts. They are easy to manufacture and often cost much less than aluminum equivalents. Composite ground planes can be conductively coated if necessary to improve electromagnetic interference (EMI) shielding, grounding or protection from lightning strikes.

Similarly, traditional radome manufacturing historically involves tedious and slow hand

layups of multiple material layers. The process becomes quite difficult for intricate radome shapes. Recent advances in long glass fiber and continuous glass fiber composites offer a means for eliminating the need for hand layups. Radomes can be thinner and lighter-weight when they are injection molded. The ability to mold strong, lightweight radomes represents a significant change in the economics of antenna design and fabrication. When necessary, 3D selective metallization can provide lightning diversion or frequency selectivity features to these radomes.

All in all, conductive coatings, injection-molded composites and selective metallization with conductive coatings combine to offer electrically and mechanically robust antennas and arrays in conformal, lightweight form factors. These advances in materials, processes and technology enable the size and weight reductions needed for next-generation antenna designs.

REFERENCESTE Connectivityte.com

One example of how conductive coatings can be flexibly and precisely applied to composite substrates is this wideband patch antenna fabricated by TE Connectivity.


Have a design challenge and/or a design solution?

Visit the EDABoard.com Power Electronics Forum with over 7,000 discussions on “power electronics.” Visit EDABoard.com Today

edaboard.comA Design World Resource

House_ad_EE_ad.indd 35 4/24/15 12:29 PM


WIR

ELES

S &

RF

anal

ogic

tips

.com

Testing wireless devices for electromagnetic compatibility

IT IS increasingly critical to test and evaluate wireless devices for electromagnetic interference (EMI) and to ensure they comply

with electromagnetic compatibility (EMC) rules. EMI can cause minor problems (a blip on a monitor; an accidental alarm that is ultimately no cause for concern) or more serious issues. In medical devices, for example, EMI can interfere with a pacemaker or interrupt a ventilator. Testing and evaluation to mitigate EMI and ensure EMC compliance helps designers make sure new devices pose minimal risk of failing or interfering with other devices in the area.

Government regulating bodies, such as the Federal Food and Drug Administration, require certain products under their purview be tested for EMC compliance before being sold commercially. Manufacturers may run EMC tests themselves or do so through third-party organizations. Thorough EMC testing will examine two things: whether electromagnetic emissions radiated from a device affect other equipment and electronics in the area, and whether a device is susceptible to interference from other devices and other electromagnetic disturbances in its operating environment.

Additionally, devices that incorporate radios will have more regulatory requirements because they communicate as well as emit radio frequencies (RF). The result is often additional testing and approvals from spectrum regulators.

EXAMPLE: FDA WIRELESS COEXISTENCE TESTING It is useful to review some of the standards that govern EMC and EMI tests. In the medical device area, for example, the applicable standard is IEC 60601 published by the International Electrotechnical Commission. Compliance with IEC60601-1 has become a requirement for commercializing electrical medical equipment in many countries. But some countries have their own specific requirements as well.

In the U.S., the standard doesn’t apply in nursing facilities, which are considered environments providing professional healthcare. Devices typically mandated to use the standard include oxygen concentrators, body-worn nerve and muscle stimulators, beds, sleep apnea monitors, and associated battery chargers prescribed for use at home. In vitro diagnostic devices, such as blood glucose meters, are covered by another standard called IEC 61010.

In addition to the EMC testing specified in the IEC 60601-1-2 standard, there’s additional FDA-mandated testing when radios are incorporated into medical devices.

NICHOLAS ABBONDANTEChief Engineer for EMCIntertek Group

Wireless devices must

function in an increasingly

hostile RF environment.

Here are a few tips on

setting up a test regime

that can uncover problems.

Intertek_EEApril_Vs3.indd 36 4/23/15 6:01 PM

analogictips.com

Often called co-existence testing, these tests evaluate devices in real-world settings to try to determine scenarios for interference and how a device will react. These real-world scenarios include how the device functions in the presence of emissions from cell phones, cordless phones, Bluetooth devices, Wi-Fi, radio-frequency identification (RFID) and other devices in the area.

Based on the intended product environment, other coexistent devices and test levels may vary. For example, a home device might be tested against cell phones, cordless phones, Wi-Fi and common household RFID devices, while a device intended for use in a hospital would check cell phones, monitoring devices, metal detectors, MRIs or walkie-talkies.

In addition to wireless coexistence, the FDA is also concerned about cybersecurity. Manufacturers of medical devices must be prepared to explain

how they address security, especially in the context of any radios that have been added. For example, if you add Bluetooth functionality to your device, can any other Bluetooth device pair with it and take control?

COEXISTENCE TESTING STANDARDS DO NOT EXISTStandards for wireless coexistence testing, such as ANSI C63.27, are still under development. The latest edition of the medical device EMC standard, IEC 60601-1-2 4th, includes a form of coexistence testing involving simulation of actual radios, but this has not taken the place of FDA wireless coexistence testing. An FDA guidance document, published on Aug. 14, 2013, titled Radio Frequency Wireless Technology in Medical Devices—Guidance for Industry and Food and Drug Administration Staff can be a valuable resource when faced with the specter of coexistence testing.

EMC testing for medical devices resembles the processes that run for consumer electronics like cell phones or televisions. But testing can be more specific and stringent in the medical device industry given the serious consequences of device failures. Because the space is different, there are some things you should keep in mind when it comes to EMC verification.

Taking advantage of experts with EMC knowledge can make the development of any device easier. Independent third-party testing groups can function as a second set of eyes to help ensure a product’s safety and performance. Additionally, third-party groups often have experts on hand who know the space, are familiar with regulatory requirements and have experience that may help inform the testing and development of products.

Manufacturers should also consider the ways their product could impact



WIR

ELES

S &

RF

anal

ogic

tips

.com

Thefigureshowsagenericsetupforcoexistencetesting.Theequipmentundertest(EUT)isplacedinachamber,andaninterferingdeviceorsimulator,suchasasignalgeneratorwiththeappropriatemodulationpersonalityoraWi-Fiaccesspoint,isplacedinthechamberwithit.TheEUTandtheinterferingdevicearearrangedsothattheinterferingsignalcouplestotheEUTinaworst-casefashion.Signalgeneratorsasinterferingdeviceshavetheadvantagethattheinterferencelevelcanbecontrolledwithouthavingtorepositionthem,movethemfurtherawayoroutoflineofsight.AcompanionEUTdeviceiseitherplacedinthechamberaswell,orinanauxiliarychamberasshowninthediagram,andawantedcommunicationslinkisestablishedbetweentheEUTandthecompaniondevice.Controlofthelinkbudgetisachievedthroughtheuseofvariableattenuationatthebulkheadconnection,orbyothermeanssuchasdegradingthecouplingtotheEUT,orinotherconfigurations,movingthecompaniondevicefartherawayoroutoflineofsight.ThewantedRFlinkisthenmonitoredfordegradationinperformanceorerrorsintransmitteddataduetothepresenceoftheinterferingsignal.Coexistencetestingisusuallyperformedwithanensembleofinterferingdevices,tosimulatedifferentthreatsintheintendedusageenvironment.

EUTCompanion Device (RF Link)

Bulkhead connection with variable attenuation

Interferer

devices or be affected by incoming EM fields. While it may sound dramatic or even problematic, testing for the worst-case scenario is essential. Is it possible that a malfunction of your device could result in serious consequences, like injury or death? Illustrating you have considered these scenarios and planned against it will go a long way with the regulatory process.

It may sound obvious, but manufacturers should also be prepared to submit EMC testing data. Because standards for coexistence testing are scarce, test data demonstrates that you have done due diligence, thought ahead and tested against multiple scenarios.

In addition to these suggestions, there are several other considerations for manufacturers when transmitters are being incorporated. Be prepared to get modules tested and approved. Depending on the device, radio modules may need review and approval in addition to the device itself. Even changing antennas may be outside the scope of modular approval and such changes must be considered with verification testing. Care should also be taken to ensure modules are installed in a way that conforms to RF exposure requirements. Consider geography. Frequency bands are not universal. If your

device is offered in different countries, you will need to take this into consideration as you test it. Of course, different countries have different regulatory requirements. Global regulatory approvals vary and can often be complex. Requirements in the U.S. differ from those in the EU. Some countries will accept reports or approvals from other areas, some will not.

In addition, devices that communicate are considered to be radios and, as such, fall under the applicable radio testing requirements of your region. Devices that do not communicate, but simply transfer energy, are thought to emit RF energy to perform a task and will not have the same testing requirements.

REFERENCESIntertek Group intertek.com

33 Billion Internet Devices by 2020; Four Connected Devices for Every Person in World. Strategy Analytics. October 14, 2014. https://www.strategyanalytics.com/default.aspx?mod=pressreleaseviewer&a0=5609. Accessed 2/25/15


Laird_EE_4-15_ad.indd 39 4/23/15 6:03 PM


Basics of RF switches

SEVERAL electrical parameters are associated with the

performance of RF and microwave switch designs, but four are considered to be of fundamental importance to the designer because of their strong interdependence: isolation, insertion loss, switching time and power handling.

Isolation is a measure of how effectively a switch is turned off. It’s the attenuation between the input and output ports of the circuit. Insertion loss, or transmission loss, is the total power lost through the switch in its “on” state. Insertion loss is the most critical parameter to a designer because it may add directly to the system’s noise figure. Switching time

is the period a switch needs for changing state from “on” to “off” and “off” to “on.” This period can range from several microseconds in high-power switches to a few nanoseconds in low-power, high-speed devices. The most common definition of switching time is the time measured from 50% of the input control voltage (TTL) to 90% of the final RF output power. Power handling is the maximum RF input power that the switch can withstand without any permanent degradation in electrical performance.

RF and microwave switches can be categorized into two primary groups: electromechanical (EM) relay switches and solid state (SS) switches. There are several design configurations possible that can range

TIM GALLA Product Manager Active RF/Microwave ComponentsPasternack Enterprises

RF and microwave switches route signals through transmission paths

with a high degree of efficiency. Four fundamental electrical parameters

characterize how these switch designs perform.

An example of an SP12T EM switch using 12 distinct SMA female coaxial connectors.

WIR

ELES

S &

RF

anal

ogic

tips

.com

Pasternack_EEApril_Vs5.indd 40 4/24/15 9:25 AM


WIR

ELESS

& R

Fanalogictips.com

from single-pole/single-throw (SPST) to single-pole/sixteen-throw or higher (SP16T), where one input can switch between 16 different output states. Transfer switches are double-pole/double-throw (2P2T) designs. They have four ports with two possible switch states and have the capability to switch a load between two sources.

EM RF switches are usually larger assemblies because they incorporate a series of coils and mechanical contacts. As with ordinary relays, electrically energized coils move the relay contacts. EM relay switches have low insertion loss (< 0.1 dB), high isolation (>85 dB), and can switch signals at speeds in the milliseconds. Major benefits are they can operate down to dc and up into millimeter wave frequencies (50+ GHz), and are not susceptible to electrostatic discharge (ESD). They can handle high power levels (up to several thousand watts of peak power) and have no video leakage.

There are some operational Issues to be aware of when it comes to EM RF switches. Their

standard operating life can be limited to about one million

cycles, and

One example of a high-rel RF switch is the Pasternack PE71S6064. It is an SP2T EM design that operates from dc to 40 GHz and is guaranteed for 10 million cycles.

50-ohmtermination

Indicatorterminals

Actuator

Powerinputs

RFinputs

VccRTN

TTL E1E2

Switch schematic

the assembly can be sensitive to vibrations. Operating life refers to the number of cycles the EM switch will complete while meeting all RF and repeatability specifications. Applications that need a higher operational lifetime can make use of higher quality or high-rel EM switches, which offer exceptional reliability and performance, with operating lives up to 10 million cycles. The longer life comes from a more ruggedly designed actuator and transmission link that has been optimized for magnetic efficiency and mechanical rigidity. They are also designed to withstand more stringent environmental conditions for sine and random vibration and mechanical shock in accordance with MIL-STD-2002.

For example, Pasternack Enterprises offers standard EM RF switches with operating lives of one million cycles, as well as high-rel EM RF switches that will last from 2.5 to 10 million cycles. One such device is the model PE71S6064, an SP2T high-rel switch design which operates from dc to 40 GHz and is guaranteed for up to 10 million cycles.

Examples of typical features found on electromechanical RF switches can be found on the schematic of a PE71S6064 SPDT device. It has a 28-Vdc latching actuator and separate contacts for indicating the position of the switch. It also incorporates 50 Ώ terminations on its unused ports.

Pasternack_EEApril_Vs5.indd 41 4/24/15 3:59 PM


An example of a PIN diode switch is the Pasternack PE7167. It is a SP4T design that operates from 500 MHz to 40 GHz and has a 100 nsec maximum switching speed. In solid-state switches, PIN diodes basically function as variable resistors whose resistance is controlled by the dc bias.

GHz and has a 100-nsec maximum switching speed with up to +20 dBm input power handling.

All in all, SS RF switches are more reliable, last longer and switch faster than EM versions. So they are preferred in applications where fast switching speed and high reliability are critical. Applications characterized by wide frequency coverage down to dc and a need for low insertion loss are candidates for EM RF switches, with high-rel versions preferred if longevity is absolutely essential.

Designers should also be aware of additional features related to these switch designs. 50-Ω resistive loads are one example. Any unused open transmission line in a switch circuit has the possibility to resonate at microwave frequencies. Resonance could cause power to reflect back to the active source and damage it, especially in a system that works at 26 GHz or higher where isolation drops off considerably. Many transmission lines are designed to have a 50-Ω impedance so RF switches that incorporate 50-Ω resistive loads reflect little energy.

EM RF switches are categorized as terminated or unterminated. In terminated versions, the selected path is closed when all paths are terminated with 50-Ω loads, so all current is cut off or isolated. Incident signal energy is absorbed by the terminating resistance so none reflects back. Unterminated switches do not have 50-Ω loads. So the impedance match must take place at some other part of the system to reduce reflections. But unterminated switches have the benefit of lower insertion loss.

Also important with EM RF switches is the armature relay mechanism. When the coil energizes, the induced magnetic field moves the armature coils, which open or close the contacts. Some models are non-latching and have a normally-closed initial position. The force of a spring or magnet holds the switch closed while no current flows. Normally-closed devices are useful for applications where a switch must return to a known state if power is lost.

Other models have latching mechanisms. They have no default position and maintain the last position without power applied. Latching relays are useful where power consumption and

In contrast, solid-state RF switches have lower package profiles and are usually physically smaller than EM versions because the circuit assembly is planar and contains no bulky components. The switching element is either a high-speed silicon PIN diode, a field-effect transistor (FET), or integrated silicon or FET MMIC (monolithic microwave integrated circuit). These switchers are integrated onto a circuit board assembly with other discrete chip components like capacitors, inductors and resistors.

Switches that use PIN diode circuits can handle more power, and FET-based switches usually have faster switching speed. Of course, solid-state switches have no moving parts, so their operating life is infinite. They have high isolation levels (60 to 80+ dB), ultra-fast switching speeds (<< 100 nsec), and the circuit assembly resists shock and vibration quite well.

Other factors to consider with SS RF switches include their insertion loss. It is not as good as with EM versions and they are of limited use at low frequencies, with operation only down to the kilohertz range (they do not operate down to dc). This limitation comes from the inherent carrier lifetime properties of the semiconductor diode.

Also, SS RF switches are more sensitive to ESD. Their power handling capability depends on the switch design and connector type as well as operational frequency and temperature exposure. Some PIN-diode switch designs can handle power levels up to a few thousand watts of peak power, but there is a tradeoff with slower switching speed. An example is Pasternack’s model PE7167, an SP4T PIN diode switch design that operates from 500 MHz to 40

WIR

ELES

S &

RF

anal

ogic

tips

.com



WIR

ELESS

& R

Fanalogictips.com

RF common

DC bias 1

RF filter

DC blockDC bias 2

RF filter

RF 2RF 1PIN diodes

Simplified schematic — SPDT PIN diode switch

dissipation are an issue. The coil for a contact consumes power only for an instant while the relay switches off.

Some models have a fail-safe mode that always returns the RF path to the de-energized position when there is no voltage applied to the coil. But this feature requires that voltage be continuously applied to the coil to maintain the energized position. This sort of continuous energization can cause a lower mean time between failure (MTBF) than with latching switch designs.

Another feature to consider in EM RF switches consists of a set of auxiliary dc contacts linked to the coil that switches the RF paths. The aux contacts normally control indicators or pilot lights signaling the position of RF paths. They may also be used to give status information to an external control system.

SWITCH DETAILSSS RF switches can be categorized as absorptive or reflective. Absorptive switches use a 50-Ω load termination in each output port which results in a low VSWR (voltage standing wave ratio) in both on and off states. In the terminated port, the terminating resistance absorbs the incident signal energy that would otherwise reflect from an unterminated port. When there is a signal on the input that must propagate through the switch, the open port is disconnected from the termination so all the signal energy can propagate through. Absorptive switches are used in applications where it’s important to minimize reflections back to the RF source.

In contrast, reflective switches have no termination resistors. So their open ports have a lower insertion loss.

Reflective switches go into applications where high off-port VSWR isn’t critical. The impedance match is provided at another point in the system.

Another important feature to consider with SS switches is the driver circuit. Some SS switch designs have integrated drivers with TTL logic states available for specific control functions. The TTL driver supplies all the necessary currents to ensure diodes have either reverse or forward bias voltage.

RF EM or SS switches come in a variety of package sizes and connector configurations. Most coaxial switch designs use SMA connectors for operation up to 26 GHz. Designs that operate up to 40 GHz typically use 2.92-mm or K connectors. Designs that work up to 50 GHz use 2.4-mm connectors, while designs of up to 65 GHz employ 1.85-mm connectors.

Switches with waveguide ports are widely used for high-power communication signals covering microwave and millimeter-wave frequency bands. They give the lowest possible insertion loss. Coaxial switch designs that handle more power (up to several hundred watts of CW power) might use larger N-Type or TNC connectors. Package styles can range from commercial grades which are not environmentally sealed, to high-rel grades which are hermetically sealed to withstand harsh conditions.

REFERENCESPasternack Enterprises pasternack.com

An example of a single-pole double-throw (SPDT) RF switch uses PIN diodes as switching elements and passive components that decouple the RF and dc bias signal paths. The RF common port might be connected to the system antenna in a typical application with RF posts one and two connected to transmitter and receiver. The PIN diode functions as an RF resistor having a resistance controlled by the magnitude of the diode forward-bias direct current. DC bias current can typically adjust the RF resistance of a typical PIN diode over three or more orders of magnitude. When the diode is biased off, it has a high impedance approximating an open circuit.



WIR

ELES

S &

RF

anal

ogic

tips

.com

Wide-band instruments in the 5G Era

CLEARLY there’s a trend toward signal analyzers supporting wider instantaneous

bandwidth. The trend is primarily driven by advances in off-the-shelf, analog-to-digital converter (ADC) technology and wireless standards. But the benefits of faster ADCs reach far beyond the wireless industry. Improvements in off-the-shelf ADCs now let test equipment makers address the needs of a broad spectrum of industries, but particularly those of aerospace and defense.

One can understand how wireless communications have helped drive signal analyzer technology by reviewing the rapid rise in channel bandwidth across today’s modern wireless standards. For example, an AMPS communication channel (1G cellular) consumes around 30 kHz of bandwidth for one-way communication (60 kHz for full duplex), a GSM channel (2G) consumes 200 kHz, and a UMTS channel (3G) consumes 5 MHz.

The widespread development of 802.11ac devices has also had an impact. This WiFi networking standard defines high-throughput wireless local area networks (WLANs) on the 5 GHz band. It has expected multi-station WLAN throughput of at least 1 Gbit/ sec and a single link throughput of at least 500 Mbit/sec. This is accomplished by, among other things, a wider RF bandwidth (up to 160 MHz). A few years ago, this standard was ahead of the capabilities available in RF signal generators and analyzers. As a result, many test and measurement vendors accelerated their development of wider bandwidth instruments just to support the bandwidth requirements of 802.11ac in a timely manner.

Looking ahead, the next major milestone for RF test equipment is the ability to test fifth-generation cellular devices. Instrumentation makers will also respond to researchers’ use of advanced software-defined radio tools to actively prototype 5G-candidate technologies, such as massive MIMO (multiple input, multiple output), GFDM (generalized frequency division multiplexing) and millimeter wave communications. The potential use of wideband

DAVID HALLPrincipal Product Manager RF and Wireless CommunicationsNational Instruments, Corp.

Just a decade ago, signal analyzers

with 20 MHz of instantaneous

bandwidth were considered state-

of-the-art. Within a decade, signal

analyzers with no less than 2 GHz

of bandwidth will likely be entry-

level devices.

NI_EEApril_Vs4.indd 44 4/24/15 4:02 PM

analogictips.com

The $300,000 ScrewContact one of our EMC and Thermal Management experts to learn how Nolato helped one customer save $300,000 annually on this screw. EMC & Thermal Management Solutions

[email protected] | 651-955-9505

millimeter-wave signals most likely will need RF test equipment able to offer 2 GHz of bandwidth by 2017 or 2018 to support a 2020 deployment.

By any standard, 2 GHz of instantaneous bandwidth in an RF signal analyzer would be a major landmark. If such an instrument existed, it would be an incredibly useful tool for bandwidth-hungry applications, such as radar-pulse measurements and spectrum monitoring.

Moore’s law is one reason why the industry is going to get to 2 GHz of bandwidth. The Law, of course, is an observation that transistor density on an integrated circuit doubles every two years. But CPUs and FPGAs are not the only technologies that have benefited from exponential improvements in IC transistor density. ADC sample rates are following a similar trend. Consider the maximum available sample rate of 12-bit ADC technology over time. Ever-

faster 12-bit ADCs boost the dynamic range available for analyzing frequency domain signals, so they are an effective proxy for the bandwidth capabilities of RF signal analyzers.

It is interesting to note the technical progress that firms making ADCs foresee. For example, officials at Analog Devices Inc., a company that produces about half of all ADCs sold commercially, have said they can see 12-bit ADCs pushing to 10 GS/sec over the next two to three years, and 14-bit ADCs pushing to 2.5 GS/sec in that same time frame. They also said 14 to 16 bits at 10 GS/sec is on the horizon, though they feel technical breakthroughs would be needed for these converters to come to fruition.

In 1975, a 12-bit ADC with 2-μsec settling time (approximately 500 kS/sec, though not an exact corollary) was considered state of the art. Today, the fastest sampling 12-bit ADCs are hitting

NI_EEApril_Vs4.indd 45 4/24/15 9:37 AM


WIR

ELES

S &

RF

anal

ogic

tips

.com

rates exceeding 2 GS/sec—a feat that’s powering some of the widest bandwidth signal analyzers in the industry.

Based on the current rate of development, 12-bit converter technology will soon be able to give RF instruments instantaneous bandwidth in the multigigahertz range and boost today’s gigahertz-bandwidth oscilloscopes to even higher resolutions.

COMING TO WIRELESS INSTRUMENTATIONEngineers will soon be using exciting new measurement approaches and techniques ushered in by next-generation RF signal analyzers (and even oscilloscopes). In radar design and development, for instance, instruments with better bandwidth and signal-processing capabilities should soon make it possible to design more advanced

radar prototypes. In high-volume manufacturing tests, instruments will be able to acquire ultra-wideband signals in a single shot. This should help test engineers easily capture data from multiple wireless devices in parallel for faster multi-site test configurations.

In many respects, the bandwidth limitations of yesterday’s RF signal analyzers

drive some of the test techniques we use today. Now that we’re in the middle of a bandwidth revolution, we need to consider how wider bandwidth will empower the test techniques of tomorrow.

REFERENCESNational Instruments, Corp.ni.com

Based on the current rate of de-velopment, 12-bit converter tech-nology will soon be able to drive RF instruments to multigigahertz of instantaneous bandwidth.

A plot of ADC sample rate versus time shows how wireless communications system bandwidth continues to grow at an exceptional rate.

Sample rate vs. time

NI_EEApril_Vs4.indd 46 4/24/15 4:01 PM

Design your products using the LARGEST CATALOG IN THE INDUSTRY. Visit www.RencoUSA.com/getcatalog to get your copy today.

321.637.1000

World’s largest selection of Coils & Transformers

GENETIC ENGINEERINGRenco Components Building Blocks of

Better Products

Parts shipped from USA or Hong Kong. @RencoElec

Visit us at APEC Booth #406

Renco_EE_ad.indd 47 4/24/15 9:54 AM


Analog switches solve many problems within a cell phone

nalog switches have been available since the ‘60s as a component for system designers. With the advent of ASICs and ASSPs, many designers have

not experienced the real value of these devices. Today’s designers may be familiar with the venerable MC14066, a workhorse that has been around for 30 years or more. The device consists of 4 independent switches that provide bilateral capabilities when “ON” and nearly infinite impedance when “OFF.”

The 4066 was done in 9.0 µm metal gate CMOS. It has a voltage rating of 3.0 to 18 V, and when switched “ON,” a resistance of ≈500 Ω. The smallest package is a 14 pin TSSOP with approximately 32 mm2 of board space occupied. If today’s designer is not aware of the strides made in analog switches, many of the issues associated with the old metal gate part have been solved with today’s sub-micron silicon gate CMOS in tiny transistor-like packages.

Enter the single gate solution: ON Semiconductor now offers a low-voltage, single version of the 4066 in a one-gate package. The MC74VHC1G66DFT2 is a single switch (SPST) occupying less than 4.5 mm2 in a tiny SC70/SC88A package. The device is specified from 2.0 to 5.5 V and offers < 25 Ω resistance when turned “ON” and almost infinite impedance when turned “OFF.” Interestingly, the device can pass/stop either a digital or analog signal. Digital signals get passed with < 1.0 nsec delay, and very nearly no change in the signal. Analog signals get passed with less than 0.1% distortion and the device has a –3.0 dB point of > 100 MHz. Because of its utility, ON Semiconductor has elected to add several more devices to the portfolio including SPST, SPDT, dual SPST, DPDT, 2:1 Mux and dual DPDT functions. These are all available in tiny packages from 2.1 x 2.0 mm to 3.0 x 3.0 mm.

The following problems are presented to illustrate the use of analog switches in cell phone applications.

A

WIRELESS & RF

ADVERTORIALanalogictips.com

ON_Semi_Advertorial_EEApril_Vs4.indd 48 4/24/15 4:08 PM


WIRELESS & RF


PROBLEM: To design an oscillator with two frequencies. The purpose of this is to illustrate how an analog switch may be used to switch in a different crystal and have a single input to a device. The purpose might be to slow a microprocessor or DSP down to conserve power, yet still function. In this application it is assumed that the MCU has two inputs for a crystal. The NLAS4599 SPDT switch selects one of two crystals electronically under control of the MCU/DSP. Only one tiny SC70 type package is needed. The diagram doesn’t show power and ground to the part for simplicity.

PROBLEM: To cut power consumption on a system using a PLL. A PLL operating at 1.0 GHz or more, can often draw more than 50 mA continuously. If the design is such that the frequency is constant for a period of time, say, for several minutes at a time (or longer), it is possible to use a sample and hold output where the PLL charges a capacitor, hold the charge for a period of time (for example, 50 msec) and go back and reclose the loop. The amount of droop from nominal voltage will depend on the capacitor used and the time the circuit is held open loop.

PROBLEM: Improving lock time of a PLL by changing the time constant, having a fast “attack” and long hold time constant. An analog switch can be used to change either the resistor or capacitor that makes up the time constant for a PLL. Changing the time constant allows the loop to get close to lock quickly, then go to a long time constant for maximum PLL noise rejection.



WIRELESS & RF


DEVICE FUNCTION PACKAGE NLAS4501 1−SPST SC88A, TSOP−5

NLAS4599 1−SPDT SC88, TSOP−6

NLAS3157 1−SPDT SC88

NLAS323 2−SPST, Pos EN US8

NLAS324 2−SPST, Neg EN US8

NLAS325 2−SPST, 1 Pos, 1 Neg US8

NLAS1053 1−2:1 Mux US8

NLAS4592 2−Independent SPDT Micro−10

NLAS44599 2−Independent DPDT QFN

SPST = Single Pole Single Throw, SPDT = Single Pole Double Throw, DPDT = Double Pole Double Throw

SC88 is a 5/6 lead package that has a 2.1 x 2.0 mm footprint

TSOP−5/5 is a 5/6 lead package with a 3.0 x 2.0 footprint

US8 is an 8 lead package with a 3 x 2 mm footprint

Micro−10 is a 10 lead package with a 3 x 5 mm footprint

QFN−16 is a 16 lead package with a 3 x 3 mm footprint

DESIGN FOR TESTABILITY Cell phones are complete communication systems built into tiny housings. They are comprised of I/O, memory, a CPU and firmware. With firmware becoming such a large part of the design, it is crucial that the designer be able to enable software debugging, and create a system that can be analyzed for field problems. The analog switch permits the designer to re-route some of the I/Os to create a new path for analysis. For purposes of illustration, let’s say that two I/Os need re-routing to be tested by an external tester. Let’s make the assumption that the designer cannot afford to give up two I/O pins. Analog switches then make the perfect solution. They introduce nearly zero delay time, cause almost no distortion to the signal, and all the configuration to be switched by an external pin.

The NLAS4599 is a Single Pole Double Throw (SPDT) switch that will allow the normal configuration to be wired through. If the control pin is grounded through a resistor, when an external plug is presented, it needs to take the (assumed) 3.0-V supply and route it out to the control pin, and then it will have the two I/O pins available to use. The result is only 8.0 mm2 of board space used, with near zero delay and no degradation to the signal. When the external connector is plugged in, the enable changes the circuit over and the two I/O pins are routed from their normal position within the circuit, to be used as external pins for testing.

CONCLUSION ON Semiconductor offers more than a dozen analog switches in tiny packages varying from SC70 and TSOP−5 to a 3.0 x 3.0-mm QFN (Quad Flat No-lead) 16 pin device. (The complete list is shown below.) The ideas shown here and many more can be implemented by one or more of these devices.

ON Semiconductor onsemi.com


SEMICONDUCTORS

Power SupplyIn Package (PSiP)NN3100xA Series Panasonic’s new NN3100xA Series is a single channel, synchronous DC-DC step down regulator PSiP that features a small 8.5x 7.5x 4.7mm QFN package. The NN3100xA Series achieves efficiencies of greater than 94% with low power consumption and low heat generation .

Relays & Connectors • Capacitors • Circuit Protection • Electromechanical • SensorsIndustrial Automation • Resistors & Inductors • Semiconductors • Wireless Connectivity

na.industrial.panasonic.com 1-800-344-2112

Panasonic brings strategic innovations toour customers’ product development process. We provide the technology and engineering resources to enable manufacturers to plan and build world-class solutions to meettheir customer needs.

RELAYS

Latching AC-CutoffPower RelayADZ Series

Panasonic’s ADZ Series is available in a horizontal and vertical configuration and has a low profile to facilitate space saving designs. Featuring a low oper-ating power and twin-contact design, the ADZ Series provides lower contact resistance, and a broader operating temperature range (-40 to 85°C).

ELECTROMECHANICAL

ASQM1 SeriesTurquoise Stroke SwitchMiniature IP67 Stroke Switch with High ReliabilityPanasonic’s new ASQM1 Series is 45% smaller than conventional products, allowing for usage in smaller spaces. Features a sliding contact structure for silent actuation and high contact reliability, the ASQM1 Series is perfect for low level load switching. The IP67 rated enclosure ensures stable operation.

WIRELESS CONNECTIVITY

Place and Play Bluetooth® Low Energy ModuleWith a 5mA Current ConsumptionThe PAN1740 Series is Panasonic’s next generation “nanopower” Bluetooth Low Energy module with reduced form factor, significantly lower powerconsumption, embedded BLE stack and profile. The 9.0 x 9.5 x 1.8 mm footprint of the PAN1740 Series makes it one of the smallest BLE modules available.

Panasonic_EE_ad.indd 51 4/24/15 10:28 AM


WIRELESS & RF


Motivation for RF integration

hile CMOS technology has made great strides in its ability to fabricate radio frequency (RF) circuitry, many RF chip

designers have yet to take advantage of this capability. After long relying on more expensive technologies, such as silicon germanium (SiGe) and gallium arsenide (GaAs), RF designers who transition to the latest RF CMOS processes gain the enormous advantages of full system-on-a-chip (SoC) integration.

Proliferation of a wide range of wireless applications, from cellular phones to wireless audio/video products, has created enormous opportunities for the semiconductor industry. Although major components in traditional wireless systems have long been fabricated in a variety of CMOS and compound semiconductor process technologies, recent advances in RF CMOS have made it the technology of choice for these applications. RF CMOS is contributing greatly to the successes of wireless products in the marketplace.

THE WIRELESS LANDSCAPE By increasing performance while decreasing cost, wireless semiconductor devices have been key drivers for improving productivity and quality of life for the last two decades. Making phone calls from virtually any location and exchanging text messages and emails has become a way of life. Instant wireless access to the Internet and location-based services has become affordable and will become ubiquitous in many parts of the world over the next few years. The digital home is being realized with access to high-quality audio and high-definition video anywhere in and around the home—all connected wirelessly. Very-high-speed local and metropolitan wireless networks promise to make the knowledge-based workforce more productive than ever before.

The technologies that make these capabilities available will continue to drive costs down. RF CMOS process technologies are helping to make these products affordable.

Figure 1 is an illustration of an RF system. The baseband processor or the controller, depending on the application, is a system-on-a-chip device that typically contains one or more CPUs, memories, complex logic and analog circuits to make the device function as a cellular telephone, wireless LAN, and so on.

W

Fujitsu_Advertorial_EEApril_Vs3.indd 52 4/24/15 4:07 PM


WIRELESS & RF


FUNCTION PROCESS TECHNOLOGY ADVANTAGES DISADVANTAGES Baseband/Controller CMOS High integration level, good analog

devices, relatively inexpensive, proven evolutionary technologies, multiple foundry sources

Not suitable for high-power appli-cations

RF Transceiver Silicon GermaniumBiCMOS

Very high frequency capability, high drive capability, bipolar devices

Costly, two to three process nodes behind logic, CMOS

Power Amplifier Gallium Arsenide Extremely high performance, very high drive capability

Very expensive

Table 1: Advantages and disadvantages of process technologies

THE RF SYSTEM A typical RF system includes the following basic components: • antenna • RF filter • low-noise amplifier • mixer • oscillator • phase-locked loop • frequency synthesizer • analog-to-digital and digital-to-analog converters • power amplifier

SYSTEM-IN-PACKAGE VERSUS RF INTEGRATION Most RF systems incorporate several integrated circuits (plus other components) that are fabricated using different process technologies. These processes are optimized for their specific functions. To lower the overall bill of materials (BOM) and reduce the form factor of the end product, engineers always want to integrate these functions into the fewest number of chips possible.

The digital baseband processor or controller is normally implemented in advanced CMOS process technologies—perhaps in the 90- or 65-nm nodes. Although many RF CMOS SoCs have been made, the RF transceiver is still typically manufactured using a SiGe BiCMOS process. GaAs or BCDMOS are among the processes of choice for the power amplifier, which requires high-current drive and linear characteristics.

One way to reduce component count on circuit boards is to put two of the chips into a single package. This system-in-package (SIP) approach reduces the component count, board footprint and power consumption, but not the system cost. The more complex assembly procedure and higher package cost may actually increase the BOM.

A better approach is to integrate the RF transceiver and the digital baseband processor into a single SoC. The power amplifier is

likely to remain a separate IC in most cases due to its unique functional requirements. Integrating the RF functions with the predominately digital baseband/controller chip has the benefits of lowering the BOM as well as reducing power consumption and the system form factor.

WHY RF CMOS? The RF system’s digital baseband processor is usually the system’s largest chip that includes an embedded CPU, millions of logic gates and large blocks of memories. The RF transceiver requires a smaller die and contains up to thousands of transistors and many of the passive components. The power amplifier consists of a handful of large transistors that provide high-current drive to send RF signals through the antenna.

SiGe has been the process of choice for the transceiver for its excellent high frequency capability and high-current drive. SiGe achieves switching frequencies that are about as high as CMOS that is two generations ahead. However, SiGe is an expensive process technology. It also lags behind the process geometries of CMOS by two to three lithographic generations.

Figure 4 shows that advanced RF CMOS processes have switching speeds (represented by transistor cutoff frequency fT) that are fast enough for most of the



WIRELESS & RF


wireless technologies in use today or in development. Future generations of CMOS processes will be even faster, making extremely-high-frequency (EHF) systems that operate between 30 to 300 GHz affordable.

In addition to switching speeds, transistor architects have also been able to improve MOSFET parasitics and oxide quality to achieve good 1/f noise performance that is essential for low-noise, high-gain amplifiers and low-distortion oscillators. Production engineers are able to tightly control manufacturing to minimize device mismatch for high-quality circuits. These transistor enhancements allow engineers to design high-gain, low-noise RF circuits with excellent dynamic range.

PASSIVE COMPONENTS While characteristics of active devices are critical for amplifiers and oscillators, equally important is the quality of passive components such as resistors, inductors, capacitors and varactors, which are used for filters, mixers, dividers, and so forth. The passive components fabricated in advanced RF CMOS technologies are as good as or better than those found in traditional RF processes.

RF CMOS technologies have special process modules for passive devices to optimize the characteristics for RF designs. Polysilicon resistors are available with values ranging from several ohms per square (silicided) to one thousand ohms per square (unsilicided). Thick copper wires, up to 3 µ, are available for inductors to achieve high inductance per unit area yet with Q factors in excess of 10 for most common carrier frequencies.

In addition to standard MOS capacitors, RF CMOS commonly offers metal-in-metal (MiM) capacitors with unit capacitance from 1.0 to 2.0 fF/µm2. MIM capacitors can be stacked to produce twice the unit capacitance in the same die area at the expense of additional process steps. Another way to build linear capacitors is using stacked metal fingers to form metal-oxide-metal (MOM) fringe capacitors. A stack of four to five metal layers can produce unit capacitance comparable to that available in MIM capacitors.

CMOS COST ADVANTAGE SiGe wafers typically cost three to four times more than comparable CMOS wafers of the same lithographic node. Adjusting for switching frequency (since SiGe is faster), SiGe is still about 30 to 50% higher than CMOS at comparable performance levels. Clearly it is more economical to integrate the RF circuits in CMOS than to integrate logic in SiGe.

PROCESS DESIGN KITS Digital designs can be synthesized from RTL using readily available EDA tools. Analog and RF circuits, on the other hand, must be custom designed using components from a wafer foundry’s process design kit (PDK). Contents of the PDK include schematic symbols for circuit entry for both active and passive devices, SPICE models for circuit simulation, layout technology files, and parameterized cells and physical verification command files for design rule check, layout-versus-schematic verification and parasitic extraction.

Many RF and analog designers prefer to run their own test structures and develop their own design kits due to the inadequacies of foundry-provided PDKs. The benefit is that engineers are able to extract electrical parameters using the exact layout styles proven in previous RF designs. However, it takes several calendar quarters to design and layout the test structures, fabricate a test chip, extract electrical parameters and develop device models before the actual circuit design can begin. The risk of missing the market window due to the long design cycle is high.

To help customers shorten their product development time, some wafer foundries invest in comprehensive RF-optimized PDKs with rich device selections and highly accurate models. Some design kits feature scalable, surface-potential-based transistor models, such as PSP, for better accuracy. PSP models provide more accurate noise modeling and avoid the problem of discontinuity at high-order derivatives found in common threshold-based BSIM4 models.



WIRELESS & RF


Rather than furnishing only a simple collection of predefined elements, a high-quality PDK provides tool kits for designers to automatically generate the complete device layout and accompanying models for the exact design requirement. For example, an inductor tool kit enables designers to easily generate the complete layout and device models for a 3-nH inductor. This process is far more efficient than having to construct a custom inductor to obtain the proper value or selecting one from a pre-defined collection that includes only 2- and 4-nH inductors.

Another capability that improves both design quality and productivity is statistical analysis. Sophisticated PDKs thus provide statistical analysis tools that allow engineers to explore process variation spaces to optimize the design and make appropriate tradeoffs.

Depending on the experience of the design team, RF circuits may require three to four cycles to tune to perfection. However, a good PDK with accurate device models allows designers to minimize time-

consuming and expensive silicon re-spin to achieve quicker time to market and lower product development cost.

SUMMARY Advances in process technologies and circuit design techniques, along with sophisticated RF design kits, have made CMOS technology the platform of choice for wireless designs. Using RF CMOS to fabricate highly integrated, single-chip solutions enables products that are smaller, more affordable, more power efficient and have a longer range than previously possible. Proliferation of applications such as cellular communication, navigation systems, personal and local wireless networks, and wireless audio/video links will help drive revenue for the entire semiconductor industry.

For more information on Fujitsu Manufacturing Services capabilities, please visit the company website at: fujitsu.com/us/services/edevices/ microelectronics/sms/ or send an email to [email protected]


WIRELESS & RF | TOP PRODUCTS


analogictips.com

Low-jitter clock generator supports JESD204B

Microwave RF assembly calculator

A NEW IC provides a low-power, multi-output, clock signal with low-jitter, low phase noise clock distribution.

The AD9528 provides JESD204B-compatible subclass 1 SYSREF and deterministic latency clocking signals and supports a variety of options for SYSREF signal generation. The most basic is a simple buffer function wherein a user-provided SYSREF signal is fanned out to the SYSREF output pins. When provided with an external SYSREF source, the chip is also capable of synchronizing the SYSREF outputs to the clock outputs being generated internally, which is necessary to achieve accurate deterministic latency.

The AD9528 is also capable of generating the SYSREF source internally. The chip supports both continuous signal SYSREF generation and “n-shot” pulse generation.

When connected to a recovered system reference clock and a VCXO, the AD9528 generates 12 low-noise outputs with a range of 1 MHz to 400 MHz, and two high-speed outputs at up to 1.25 GHz. The frequency and phase of one clock output relative to another clock output can be varied by means of a divider phase-select function that serves as a jitter-free, coarse timing adjustment in increments that are equal to half the period of the signal coming from the VCO output. The SYSREF signals each have additional phase offset capability making it easy to dial-in the optimal arrival time at each target device.

Analog Devices, Inc. — analog.com

AN ONLINE tool called the Gore

microwave/RF assembly builder is used to calculate insertion loss, VSWR and other parameters of microwave/RF assemblies for different cable types. The calculator is particularly useful when the initial cable type is unknown and needs to be specified independent of the connector.

The main Calculator section enables the user to select a Market (Aerospace and Defense, Spaceflight, Test & Measurement) and Category (such as General Purpose, Ruggedized and Phase Stable, etc.) from the dropdowns to view cable types and their specs. The user then enters

data in fields for frequency, length, temperature and altitude to narrow the results in the main Cable Types table. The Multiple Cable Review tool allows the user to select up to three cables to be compared on a single, detailed results screen.

The calculator includes a conversions page that has most of the everyday conversions, including distance, frequency, power, temperature, VSWR/return loss and weight. A glossary is also included that features explanations for many of the terms used in the online calculator.

W. L. Gore & Associatesgore.com

MIKE SANTORAContributing Editor

Products_EEApril_Vs2.indd 56 4/24/15 4:12 PM

TOP PRODUCTS | WIRELESS & RF


analogictips.com

RF switch with high isolation and linearity

Rugged bench top RF amplifiers Doherty PA reference design for small cell wireless

AN RF switch called the F2912 combines low insertion loss, high isolation and

linearity making it a candidate for base stations (2G, 3G and 4G), microwave backhaul and front haul, test equipment, CATV headend, WiMAX radios, wireless systems and general switching applications.

The F2912 supports features that include:• Frequency range of 300 kHz to 8 GHz• Low IL of 0.4 dB, providing low path

loss without compromising isolation performance.

• Isolation of 60 dB at 2 GHz to reduce signal

leakage between adjacent RF port paths.• OIP3 of +64 dBm to reduce

intermodulation distortion.• P1dB of 30 dBm, providing a 1 W

compression point, ensuring rugged operation for a variety of applications.

• 3.3 and 1.8 V control logic consistent with common FPGA and microcontroller logic levels.

• Operating temperature range of -55 to 125°C for high reliability in harsh thermal environments.

Integrated Device Technology — www.idt.com

A LINE OF rugged, portable

bench top amplifiers is designed to meet MIL-STD-202F environmental test conditions for humidity, shock, vibration, altitude and temperature cycle. These features make them suitable for use inside high traffic test labs in industries such as aerospace, defense, optical, industrial, telecom and R&D. They also cover wide frequency bands up to 40 GHz.

This latest release of RF amplifiers includes four models covering multi-octave bandwidths from 1 to 40 GHz and exhibit flat gain response. These bench top amplifiers offer up to 60 dB small signal gain with high dynamic range, a noise figure of 5 dB and output P1dB compression power ranges from +10 to +22 dBm. Additionally, these portable and compact amplifiers have an internal AC voltage power supply of 115-120 VAC at 60 Hz, an operating temperature range -40°C to +85°C and allow a storage temperature of -40 to +100°C. The 20 GHz modules use SMA female

connectors, while the 40 GHz versions utilize 2.92 mm connectors.

The new ultra-broadband portable bench top RF amplifiers use a single AC voltage supply with internal voltage regulation, fuse protection, and are designed with front panel access with an on/off switch and RF input/output connectors. No export license is required for these products.

Pasternack Enterprises — pasternack.com

THE CDPA35045 asymmetric Doherty PA reference design for the 3.5 to 3.7 GHz band provides increased

wireless system capacity for both licensed wireless carrier services and unlicensed public use, such as Wi-Fi. Providing 10 W average output power and predistortion correctability, Doherty PA design uses 30-W CGHV27030S and 15-W CGHV27015S GaN HEMT devices, which can operate with either 50 or 28 V drain supplies.

The CDPA35045 asymmetric Doherty PA combined with radio signal processing IP provides base station designers with a design that supports the development of new small cell wireless infrastructure equipment, significantly reducing design time and enabling faster product time to market.

Cree Inc. — cree.com

Products_EEApril_Vs2.indd 57 4/24/15 12:14 PM

AD

IND

EXan

alog

icti

ps.c

omKey Account Manager Jim [email protected]@jpowers_media

Regional Sales Manager Tom [email protected]@wtwh_Tom

Regional Sales Manager Courtney [email protected] @wtwh_CSeel

Regional Sales Manager Neel [email protected]@wtwh_ngleason

Regional Sales ManagerMegan [email protected]@wtwh_Megan

Business DevelopmentMichelle [email protected]@mflando

SALES

VP SalesTodd [email protected]@wtwh_ttidmore

VP of EE Product Development Mike [email protected]

VP, Business DevelopmentMichael [email protected]@mrference

General ManagerTodd [email protected]@wtwh_todd

VP, Business DevelopmentDavid [email protected]@wtwh_david

LEADERSHIP TEAM

Publisher Mike [email protected]@wtwh_memich

Managing DirectorScott [email protected]@SMMcCafferty

EVPMarshall [email protected]@mmatheson

Acopian ................................................................... 19

Allied Electronics, Inc. .............................................BC

Coilcraft ..................................................................... 3

Crystek Corporation ................................................ 15

CTS Corporation ..................................................... 31

Digi-Key Corporation ..................................Cover, IFC

IXYS ........................................................................IBC

Keysight Technologies .............................................. 7

Keystone Electronics Corp. .................................... 60

Laird ......................................................................... 39

Memory Protection Devices, Inc. ............................ 27

Microchip Technology Inc. ........................................ 4

National Instruments Corp. ...................................... 9

Nolato ...................................................................... 45

Panasonic Corporation ............................................ 51

Renco Electronics, Inc. ............................................. 47

Rogers Corporation .................................................. 1

Saelig Company, Inc. ............................................... 37

Teledyne LeCroy ...................................................... 58

W. L. Gore & Associates, Inc. .................................... 5

8 Channel12-Bit High Definition

Oscilloscopes

Who’s doin’ that!

teledynelecroy.com/hd4096

TLEC-HDO_8000-9x5.25-halfpage-ad.indd 1 1/27/15 4:09 PM

Follo

w th

e w

hole

team

on

twitt

er @

Des

ignW

orld

CO

NN

ECT

W

ITH

US!

Ad Index_EE_4-15_Vs1.indd 58 4/24/15 4:49 PM


WIR

ELESS

& R

Fanalogictips.com

“Well-Known Member” and 10-year Electro-Tech-Online veteran, RadioRon’s start in radio electronics began in the 1960s with a fascination in short wave radio. Ron worked for ten years at Motorola developing mobile data radios at a time when pushing high-speed data through a narowband mobile radio was something new.

He left Motorola to help start Sierra Wireless, designing portable and mobile wireless data modems for cellular networks. He tried retirement but found himself pulled back into RF work because it “was too much fun to pass up.” He retired again after a few years at a startup called Fastback Network working with 6-GHz antennas and radios.

Ron enjoys writing, so he feels it is a natural fit to spend some of his spare time explaining things to “the youngsters on ETO.” He finds giving advice to the student and hobbyist members interesting and, “… along the way, I’ve stumbled into learning a lot of things I didn’t know or didn’t remembers from many of the other experts on ETO.”

Seven-year EDABoard “Advanced Member” Neazoi is a computer and telecommunications engineer living and working in Greece. He is the editor of a Greek microwave website, and also owns and runs a personal amateur electronics website (qrp.gr). Neazoi is an official operator of the UPsat ground station in Greece as part of the QB50 project, an international network of 50 miniaturized satellites for multi-point, in-situ measurements in the lower thermosphere and re-entry research.

“Forums like EDABoard are an efficient way of discussing with other engineers and enthusiasts who share the same interests. I am impressed how many people will help others in the forums with problems they encounter without expecting any financial benefit from it. For me, this selfless help is the spirit of collaboration toward a better future.”

EDAboard “Advanced Member” Jiri Polika’s areas of expertise include microwave, millimeter wave and satellite technology, as well as radio astronomy. He has enjoyed participating in the last five years on EDABoard offering advice because, as a circuit and system designer, he understands quite well the complex

problems his colleagues face. One piece of advice he does

stress with engineering students before they participate on forums is to more thoroughly study their problems and understand the simple basics before they ask questions. Jiri educates himself on the forums, as well.

AIMEE KALNOSKASContributing Editor

MEMBER PROFILES

“RADIORON” AKA RON VANDERHELM “NEAZOI”

JIRI POLIVKA

Top Forum Members …on the Radar

WHEN we see engineers passionately and actively sharing their knowledge with fellow engineers online for the benefit of

thousands of members, we feel a thank-you of some sort is in order.That’s why we’d like to recognize some of the frequent contributors to our

two online engineering communities, EDABoard.com and Electro-Tech-Online.com (ETO). These communities of forums, blogs and articles are virtual knowledge libraries for individuals trying to solve engineering problems.

Our moderators enthusiastically contribute to microwave- and RF-related discussions offering advice based on years of engineering experience. We encourage you to get to know them, see what makes them tick, and share in the conversations online.

I’ve stumbled into learning a lot of things I didn’t know or didn’t remember from many of the other experts on Electro-Tech-Online.

I am impressed how many people will help others … For me, this selfless help is the spirit of collaboration toward a better future.

I learn many useful

things by checking

questions and answers.

edaboard.comA Design World Resource

Electro-Tech-Online.comA Design World Resource

TopContributors_EE_APRIL_Vs3.indd 59 4/24/15 11:52 AM

It’s what’s on the InsIde that counts®

E L E C T R O N I C S C O R P.

www.keyelco.com • (718) 956-8900 • (800) 221-5510

zERO Ohm SmT jumPERS

• True Zero ohm jumper • Designed to replace 0603, 0805 and 1206 “Zero” ohm resistor chip sets without changing board designs or layouts • For high current applications • Low Impedence • Low profile • Manufactured from Copper

with Silver plate • Supplied on tape and reel • Compatible with vacuum pick & place assembly systems • Other chip set sizes available upon request • Request Catalog M60.2

™

DW THiNK Dusk+ZJumpers_1-15.indd 1 12/9/14 3:47 PMKeystone 1-15.indd 60 4/24/15 11:54 AM

ID(cont)

TC=25°C(A)

RDS(on)

maxTJ=25°C

(Ω)

Ciss

typ(pF)

Qg

typ(nC)

trr

(ns)PD

(W)

RthJC

max(°C/W)

PackageStyle

VDSS

max(V)

PartNumber

POWER

D

SG

IXYS’ Surface Mount Power Device (SMPD) TechnologyLose The Weight! Put Your Design On A DIET!

Lighter weight, more power (ultra- low profile, energy efficient, and rugged)

www.ixys.com

CONFIGURATIONSBuckBoostFull-bridgeHalf-bridgePhase legSingle

APPLICATIONSDC-DC convertersBattery chargersSwitching and resonant power suppliesDC choppersTemperature and lighting controlsMotor drivesE-bikes and electric and hybrid vehiclesSolar invertersInduction heaters

SMPD ADVANTAGESUltra-low and compact package profile(SMPD: 5.3mm height x 24.8mm length x 32.3mm width)(Mini SMPD: 5.3mm height x 16.8 length x 29.5mm width)Surface mountable via standard reflow process(Available in Tape & Reel packaging) Low package weight (SMPD: 8g, Mini SMPD: 5g)Up to 4500V ceramic isolation(DCB)Low package inductanceExcellent thermal performanceHigh power cycling capability

EUROPEIXYS [email protected]+46 (0) 6206-503-249

USAIXYS [email protected]+1 408-457-9042

ASIAIXYS Taiwan/IXYS [email protected]@ixyskorea.com

For more parts, visit www.ixys.com

SMPD IGBTs

IC25

(A)

VCE(sat)

maxTJ=25°C

(V)

Eoff

typTJ=125°C

(mJ)

RthJC

maxIGBT

(°C/W)

Configuration Package styleVCES

(V)Part Number

32925023

1.83.52

3.1

1.73.550.86

-

10.311.131.25

BoostCopacked (FRED)

Half-bridgeFull-bridge

SMPD-BSMPD-XSMPD-ASMPD-A

12001200600

2500

IXA20RG1200DHGLBMMIX1Y100N120C3H1MMIX2S50N60B4D1MMIX4G20N250


SMPD Power MOSFETs

6001026324

0.00130.02

0.0430.29

40000280001860019000

590335250310

100200250300

830570520500

0.180.220.240.25

SMPD-XSMPD-XSMPD-XSMPD-X

40300500

1000

MMIX1T600N04T2MMIX1F160N30TMMIX1F132N50P3MMIX1F44N100Q3

SMPD-X

Mini SMPDSMPD-B

FEATURES• Low on-state voltages VCE(sat)• Optimized for high-speed switching (up to 60kHz)• Short circuit capability (10µs)• Square RBSOA• Positive thermal coefficient of VCE(sat)• Ultra-fast anti-parallel diodes (Sonic-FRD™)• International standard packages

APPLICATIONS• Battery chargers• Lamp ballasts• Motor drives• Power inverters• Power Factor Correction (PFC) circuits• Switch-mode power supplies• Uninterruptible power supplies (UPS)• Welding machines

EUROPEIXYS [email protected]+46 (0) 6206-503-249

USAIXYS [email protected]+1 408-457-9042

ASIAIXYS Taiwan/IXYS [email protected]@ixyskorea.com

POWERwww.ixys.com

PartNumber

IXXH30N65B4IXXH110N65C4

IXXN110N65B4H1IXXK160N65C4IXXX160N65B4IXXK200N65B4

VCES(V)

650650650650650650

IC25TC=25°C

(A)

65234240290310370

IC110 TC=110°C

(A)

30110110160160200

VCE(sat)max

TJ =25°C(V)

22.352.12.11.81.7

t typ

TJ=150°C(ns)

10043

10557

160110

Eoff typ

TJ=150°C(mJ)

0.60.771.41.3

2.362.54

RthJC max

(°C/W)

0.650.170.170.160.160.13

Configuration

SingleSingle

Copacked (Sonic-FRD™)SingleSingleSingle

Package Style

TO-247TO-247

SOT-227BTO-264

PLUS247TO-264


TO-264

High power densities!

Low gate drive requirements!

TO-247Hard switching capabilities!

Temperature stability ofdiode forward voltage VF

SOT-227B

Highly Ecient Low On-State Voltage IGBTs650V XPT™ Trench IGBTs

IXYS_EE_4-15_ad.indd 1 4/24/15 11:57 AM

© Allied Electronics, Inc 2015 ‘Allied Electronics’ and the Allied Electronics logo are trademarks of Allied Electronics, Inc. An Electrocomponents Company.

1.800.433.5700

When the Clock Is Ticking, Count on Allied

Order Up to:

10 PM ETSame-Day Shipping

8 PM ETNext-DayDelivery

Out of Time?

OCT14 Shipping Ad (DW)_Design World 4/17/15 11:53 AM Page 1

Allied_EE_4-15_ad.indd 1 4/24/15 11:58 AM

design world/ee network - wireless & rf handbook

Documents