Issue 110 February 2014 - EMC Information Centre · 2014-04-01 · Conference. This year the...

ISSN 1748-9253

Know your StandardsBy John WoodgateSee page 12

EMCUK 20147th & 8th Octoberwww.emcuk.co.ukSee page 5

the journal Issue 110 February 2014

More than 40 years of EMC T&M experience

Best in EMC testing:More speed.More insight.More intelligence.The ¸ESR is the new test receiver for standard-compliant EMC testing.❙ Save time The ¸ESR time domain scan makes EMC measurements faster than ever before, allowing you to dedicate more time to your main mission.❙ Find everything The realtime spectrogram and the persistence mode seamlessly display the spectrum so you can detect and analyze sporadic or hidden emissions and their cause.❙ Work fl exibly The ¸ESR combines a test receiver and full-featured spectrum analyzer in a single instrument, giving you all the tools you need for your measurement task.❙ Simplify measurements The ¸ESR combines functionality with clear structure for easy navigation on the convenient touchscreen.

Want to know more?Convince yourself – visit www.rohde-schwarz.com/ad/esr.

Please follow us on Twitter: Rohde_SchwarzUK

9608_ESR_210x297_e.indd 1 17.08.12 7:42 Uhr

The EMC Journal February 20143

What’s In This Issue

www.theemcjournal.com www.emcgoggles.com www.emcuk.co.uk www.emcacademy.orgEvery effort has been made to ensure that the information given in this journal is accurate, but no legal responsibility is accepted for any errors, omissions or misleading statements in that information cause by negligence or otherwise, and no responsibility is accepted in regard to the standing of any firms, companies or individuals mentioned or for any advice given by them.

News and Information

Banana Skins

John Woodgate’s Column

Know Your Standards

Product Gallery

The many benefits of HDI (microvia) PCB technology By Keith Armstrong, Cherry Clough Consultants

A Passive Culprit? By Richard Marshall, Richard Marshall Limited

FCC Proposed Rule Changes By Michael Derby, ACB Europe

Advertisers Index

29

27

18

14

12

10

8

5

EditorialThe EMC Journal continues

In December 2013, after 109 issues, Nutwood UK Ltd called it a day on the EMC Journal. This magazine has served our engineering community with valuable information, news and articles for very nearly 20 years. The EMC Journal has been a central communication tool for the EMC industry, not just here in the UK, but all over the World. Congratulations to Alan, Pam and Lynne for all your excellent efforts in the creation and delivery of this incredible magazine. I wish all three of you a very happy and fulfilled retirement.

Over the Christmas period, I thought of the impact of the end of the EMC Journal and I felt that it had become such a valuable information resource to our

industry that someone had to pick it up and allow it to continue. Therefore, my company, EMC Goggles Ltd, decided to take on the EMC Journal. I met with Alan and we talked through the details. Alan, Pam and Lynne have been extremely helpful in getting me going and, being new to journalism, I am thankful for their assistance. I have also received the backing of our regular advertisers and contributors and I am thankful to them also.

So it’s with pleasure that EMC Goggles provides you with Issue 110 of the EMC Journal. Enjoy!

John DaviesEditor, The EMC Journal

31

the journal

ar europeIn Europe, call ar United Kingdom +44 1908 282766 • ar France +33147917530 • ar Deutschland GmbH +49 6101 80270 0 • ar Benelux +31 172 423000

Copyright © 2014 AR. The orange stripe on AR products is Reg. U.S. Pat. & TM. Off.

www.arworld.us

Radiant Arrow® Bent Element Antennas

We bent the rules and advanced the science of log periodic antennas.

Up to 75% smaller, lighter, and more compact to fit in

smaller chambers.

Our SMA connectorized modules provide output powers up to 30 watts CW Class A and 50 watts CW Class AB

The solid state alternative to TWTA’s.Improved harmonics, lower noise, superior linearity and reliability and now featuring 1,000 watts CW.

All-in-one test systems for any EMCapplication, DC to 50 GHz. Our systems make testing more efficient, accurate, and affordable.

The new 16000A225 amp. It covers 10 kHz to 225 MHz and delivers 16,000 watts of power and we’re not stopping there.

To Complement Our 4 To 18 GHz Designs. Output Power Up To 200 Watts CW.

New Hybrid Power Modules

1 to 2.5 GHz Solid State Amplifiers

Lighter, more portable, and up to 50% smaller. Available with all the power you need .7 to 4.2 GHz and up to 1200 watts.

“S” Series Amplifiers

Integrated Test Systems

16,000 Watts of Pure Power

700 MHz To 6 GHz Single Band Amplifiers

These self contained systems driven by our proprietary software; simplify calibration, testing, troubleshooting and reporting. They allow accurate, sensitive and repeatable measurement in one unit. And did we mention incredible speed with outputs up to 150 watts and 400 MHz.

For the first time you can go from 0.7 to 18 GHz with the reliability of solid state in a single housing. Numerous models with various output powers help you reduce the footprint and cost of separate instruments.

RF Conducted Immunity Test Systems

Dual Band Amplifiers

The new series of laser-powered electric field analyzers have an extremely high sample rate and can preciselymeasure pulsed electric fields in the microsecond range. They represent a whole new way to more accurately measure modulated and CW electric fields in both conventional and reverberation chambers and allow the user to see the modulation envelope.

This incredible system cuts RF Radiated Immunity testing from days down to hours by testing multiple frequencies simultaneously, reducing product development cost and time to market.

This 18 GHz EMI receiver changes the way you think about emissions testing. Data is more accurate and test time is reduced. Testers have a suite of new mitigation tools.

Field Analyzer

Multi-Tone Tester

Precision DSP Receiver

Lots of suppliers claim to have “The Best Products”, but do they have any proof to back it up? AR has been proving it for over 43 years. We’ve built a reputation for reliable products that go the distance… and then some. Products that are more powerful, faster, smaller, and more efficient. Products that outlast, outperform,

and outrun any of our competitors. And if that wasn’t enough, every one of our products is backed by our comprehensive worldwide support and no nonsense warranty.

Proof of Performance

To learn more, visit us at www.ar-europe.ie or, for technical support, call +353 61 504300.In UK, contact AR UK, www.arukltd.co.uk or call +44 1908 282766 ISO 9001:2008

Certified

EMC Journal UK_Proof of Performance.indd 1 2/3/14 12:00 PM


News

Front Cover

Hero image, Syfer, page 17Circle top, SARLite, page 18Circle middle, Yokogawa, page 16Circle bottom, Teseq, page 16

The EMC JournalFree to readers worldwide

February 2014 – Issue No. 110 Published every other month

First Issue March 1995

Editorial & Publishing Director:John Davies

[email protected]

EMC Goggles Ltd26 Rumney Walk

LlanyrafonCwmbranNP44 8RN

www.theemcjournal.com

www.compliance-club.com

EMC Goggles Ltd provides classroom and laboratory training as well as product and compliance consultancy to companies worldwide. Director, John Davies, has been involved in EMC since 1990 and is a UKAS technical assessor, a member of British

Standards EMC committees and appointed UK expert to the IEC/CISPR. John is also Secretary of the EMC Test Labs Association.

CEM BCSThe CEM 2014 Ninth International Conference on Computation in Electromagnetics is a call for papers conference for the presentation of the latest research into computational techniques in electromagnetics. Held in London, UK between 31 March - 1 April 2014 it is one of the flagship events of the IET Electromagnetics Network. The event presents a unique opportunity to both industry and academia to hear about a wide array of aspects and applications of numerical modelling based on computational techniques.

“We bring together the communities of researchers engaged in methods and techniques of computational electromagnetic,

engineers facing the challenges of hazards and EMC/EMI and designers of low frequency as well as high frequency devices, in fields as diverse as electrical machines and power apparatus at one end of the scale to antennas, waveguides and optical instruments at the other. The emphasis of the conference is on networking and allowing exchange of information and experiences regarding the modelling and simulation of electromagnetic phenomena on one hand, and needs and expectations of industry on the other. Keynote presentations and introductory review talks will address these issues to facilitate discussion.” Professor Jan Sykulski, CEM 2014 Chairman.

In addition to being able to present research posters and network with peers, conference authors will see their work published in the conference proceedings and submitted for inclusion into IET Inspec and IEEE Xplore. Successful authors will also be invited to submit their papers for publication in three special issues of post-conference journals –IET Science, Measurement and Technology (SMT), ACES Journal and Archives of Electrical Engineering.

Please visit www.theiet.org/cem2014 for more information, or email [email protected]

New Rittal Catalogue 34Rittal’s new Catalogue 34 provides the essential product guide for industrial and IT solutions. Over 700 pages outline latest developments and acts as a comprehensive information resource about available product types, sizes, specifications, software and services, as well as a source of advice and support for customers.

Providing the complete solution for enclosures and enclosure systems, power distribution, climate control and IT Infrastructure across industry and IT applications, Rittal’s latest edition of the Catalogue turns the spotlight onto added value for customers.

Crystal clear navigation allows users to track down the right product in minimal time, make their final selection with ease, find out straight away what accessories are available and then place an order.

The catalogue is available now as hard copy, CD or download. To order your copy simply e-mail [email protected] or call 01709 704105.

EMC UK 2014 Exhibition & Conference

It’s business as usual at EMC UK this year, on Tuesday 7th & Wednesday 8th October 2014 at Newbury Racecourse, as with previous years there

will be the two day Training & Technical Conference.

This year the exhibition and conference will be organised by EMC Exhibitions & Training Ltd. Should you wish to exhibit or submit a paper for the conference then please contact Alan Warner on 01202 885399 or Email [email protected] the subject headings will be available shortly.

The full program & exhibition detail will be in future editions of the EMC Journal.

The EMC Journal February 2014 6

News and informationMPE appoints new German distributor for its EMC, EMP & TEMPEST filters & feedthrough capacitors

MPE Ltd of Liverpool, UK, the long-established designer and manufacturer of EMC, EMP and TEMPEST filters and feedthrough capacitors, has appointed Electrade as its new distributor covering the whole of Germany for both standard and customised products. This will include local stockholding in Germany of many standard product lines. MPE’s products fill a gap in the portfolio of Electrade, who in turn maintain a technical team with a high level of engineering expertise and understanding. They can successfully promote the MPE brand in markets where it is less well known. Electrade’s broad customer base comprises 40% mainstream commercial such as facilities management, process control, public utilities and transport, 30% defence and aerospace, and 30% information and communications technology (ICT) and the medical marketplace.

Established in 1992, Electrade, based at Graefelfing near Munich and with offices in Frankfurt-am-Main and Hannover, handles sales of high-technology products such as electronic materials, connectors, sensors, transducers and instrumentation from leading manufacturers which neatly complement MPE’s solutions in many areas. Paul Currie, Sales & Marketing Director of MPE Ltd, explains: “MPE has recognised that, whilst its product sales into German EMC test houses remain buoyant, there are many more applications and market sectors which MPE products can address yet remained untapped. MPE has therefore actively sought a new partner capable of penetrating such new areas, and in Electrade I am confident that MPE has found the ideal partner.” Dieter Muehlberger, Vice-President & co-owner of Electrade, comments: “Electrade have often been asked for high-performance EMC and EMP filters, but have previously had no solution to offer. The MPE product range provides Electrade with not only a solution for this requirement but also a technology which is very complementary to our other product offerings. Accordingly we look forward very much to working with MPE in the coming years.”

Left to right: Dieter Muehlberger, Vice-President of Electrade GmbH and Erich Thomich, General Manager of Electrade GmbH, MPE’s new distributor for Germany. They are holding, respectively, a high-performance MPE powerline filter and an MPE brochure.

Left to right: John Jephcott of MPE Ltd and Dieter Muehlberger of Electrade GmbH signing the distributor agreement for Germany

Dieter Muehlberger, Vice-President of Electrade GmbH, MPE’s new distributor for Germany

For further details of MPE’s products and services, contact Paul Currie, Sales & Marketing Director, MPE Ltd, Hammond Road, Knowsley Industrial Park, Liverpool, L33 7UL, U.K. Tel +44 (0)151 632 9111. Fax +44 (0)151 632 9112. Cell +44 (0)7850 200 705. Email [email protected]. Website www.mpe.co.uk

Accredited calibration service now available for electric field

strength metersLink Microtek, the leading UK supplier of EMF safety equipment, is now able to offer an accredited calibration service for instruments measuring electric field strength. This follows the recent accreditation by DAkkS, the German national accreditation body, of the Pfullingen-based calibration laboratory operated by Narda Safety Test Solutions GmbH, for whom Link Microtek acts as representative in the UK and Ireland.Covering frequencies ranging from 9kHz to 18GHz and electric field strengths from 2V/m to 300V/m, the Narda accreditation is particularly relevant to measurements that are carried out for the purpose of ensuring human safety.

The accreditation certificate confirms that the laboratory meets the ‘general requirements of competence for test and calibration laboratories’ as set out in EN ISO/IEC 17025:2005. The calibration method itself is described in IEEE Std 1309-2005, which is the globally recognised standard for calibration of electromagnetic field sensors and probes. Each instrument is supplied with an accredited calibration certificate providing the customer with details of the measured values and their associated measurement uncertainties, as well as additional information about the calibration that has been performed. This certificate is recognised worldwide by all the institutions listed by ILAC (International Laboratory Accreditation Cooperation) in its Mutual Recognition Arrangement.

Non-accredited calibration continues to be available from Link Microtek’s own in-house laboratory in Basingstoke, which offers a fast-turnaround calibration and repair service on all types of Narda personal monitors, survey meters and probes. For further information, visit www.radhazonline.com.

AR New AppointmentAR UK Ltd is pleased to announce the recent appointment of Rob Hersom as Technical Sales Engineer.

Rob joins the company from QinetiQ and has a wealth of EMC experience in the commercial and military/aerospace sectors

having worked at EADS Astrium as an EMC Test Team Leader and both RFI and TRaC Global as a Senior Test Engineer. Rob’s background and broad range of experience in RF & EMC applications, standards and testing methodology will make him a valuable member of the sales team at AR UK.


Accreditation for Serbian EMC laboratoryIdvorsky Laboratories, in Belgrade, Serbia, has achieved Serbian accreditation for its EMC testing services. The Accreditation Body of Serbia is a member of the EA and ILAC MRA.

The laboratory offers a wide range of EMC testing services as well as product life support, EMC/RF consultancy, troubleshooting, prototype improvement, research & design of antenna and microwave circuits using specialized CAD tools as well as electrical safety testing. Laboratory Manager, Sasa Jorgovanovic, said, “Achieving accreditation for our EMC testing services is a testament to the hard work of all our staff and I am very proud to be leading such a talented group of dedicated professionals.”

For more details:visit www.idvorsky.com or email [email protected].

News and information

EMC for Luminaires … the ongoing struggle!The Scenario….New technologies can lead to more efficient products.But some can be electrically noisy, causing serious radio interference.

The Problem….LED lamps are a classic example.Whilst many have been designed to be ‘quiet’, there are many that do cause interference well above legal limits(EN55015). These have led to consumer complaints, product withdrawals and action by Trading Standards, withconsequent damaged reputation for the supplier.

The Solution….Always test products for EMC. Low cost EMC emissions test equipment can be used in-house to quickly andeasily measure the emissions and ensure compliance with the standards.

Why take the risk? In-house testing is the logical and cost effective answer.

Laplace Instruments Ltd has developed ALECS*, a complete integrated system designedspecifically for the EMC testing of Luminaires according to EN 55015.* Automated Luminaire Emissions Compliance System

See www.laplace.co.uk or call +44 (0) 16 92 40 20 70


Banana Skins...Editor’s note: I receive many potential Banana Skins, and no doubt they are just the tip of the EMI iceberg. Keep them coming – but please don’t be disappointed if your contribution doesn’t appear for a while, or at all.

812 An EMI-related safety accident in a Gypsum mine

A Gypsum mine bought some Intrinsically Safe handheld radios on 80MHz FM (principally because of their IP rating and rugged cases).

They accessed the mine via a surface drift which had a large cable winch installed at the top to assist the ascent and descent of Landrover personnel carriers and wheeled excavators up the 30deg incline. The winch was operated by push buttons on a cable pendant, the winch supervisor held the pendant whilst attaching or detaching the winch cable to the vehicle and there had been some incidents where the operator was nearly pulled into the winch drum.

So one handheld radio was fitted into a box and controlled the winch via tone filters, detectors and relays, the winch supervisor wore another radio fitted with up/down and stop/go buttons coupled to tone generators, and a portable unit was thrown into each vehicle as it coupled on to the rope.

The idea was that the winch operator and vehicle units had both to be generating a ‘go’ tone before the winch would move. All designed by an ‘electronics expert’ on the company staff.

The problem was that the tone detectors were simple band-pass audio filters followed by a diode detector and also that the nearby RAF station used 80MHz AM for their airfield security.

Then came a slight change in local BBC Radio 2 frequency, intermodulation between the RAF and Radio 2 frequencies landed on top of the winch receiver frequency, this is an FM radio so it would output lots of white noise in this situation, noise that would pass straight through the tone filters and cause the relays to operate.

To be really unsafe it required the winch operator to be cleaning the winch cable on the main drum at the same time as the RAF transmitted, but it did eventually happen – with awful consequences.

Of course it was the fault of the radios and the HSE Inspector took a lot of convincing otherwise, but eventually the system at the Gypsum mine was issued with a prohibition notice...but the mining company did not think to apply it to all of their installations, until it happened again.

(Kindly provided by Peter Burne, on the 13th September 2013.)

813 EMI Named as Cause in Ferry Crash Lawsuit

A 2011 ferry crash could be the result of inadequate electromagnetic shielding.

BC Ferries is suing a German manufacturer of ship control systems in connection with a 2011 incident in which a ferry rammed a dock in Nanaimo, British Columbia, Canada.

The British Columbian ferry operator is seeking [1] at least $4 million in damages from SAM Electronics GmbH for the Dec. 20, 2011 crash at Duke Point, which injured seven passengers and nine crew members and required several months’ worth of repairs.

In its court document, BC Ferries alleges that an isolating amplifier in the bow propulsion pitch control system, which controls the angle of the propeller blades, was not properly shielded against electromagnetic interference, resulting in the crash. In addition, the controls for the equipment were difficult to decipher and no audible alarm to warn the crew of danger was available.

An investigative report [2] released by Canada’s Transportation Safety Board following the accident also said that the equipment malfunctioned, but added that the problem in the propulsion controls was missed because crew members did not follow proper procedures for testing the equipment before docking.

The ferry struck the Duke Point dock at a speed of approximately 5.6 knots, resulting in damage to both the vessel and the dock. The ferry was reportedly out of service for 23 days, while the dock was closed for repairs for three months.

BC Ferries has since implemented a variety of new systems and operating procedures designed to prevent similar accidents and better prepare the crew in the event of a problem.

[1] https://s3.amazonaws.com/s3.document cloud.org/documents/997138/civil-claim-b-c-ferries-vs-sam-electronics-gmbh.pdf

also https://www.documentcloud.org/documents/997138-civil-claim-b-c-ferries-vs-sam-electronics-gmbh.html

[2] http://www.tsb.gc.ca/eng/rapports-reports/marine/2011/M11W0211/m11w0211.pdf

(From a report of the same name by Aliza Becker on 12/23/2013 in Interference Technology’s eNews magazine,

www.interferencetechnology.com/ferry-crash-emi.)

814 EMI filters pass individual equipment tests, but resonate when in a system

Recently, a power system manufacturer and a customer (a major aircraft prime contractor to the military) collaborated in resolving an interaction problem in a complex system. A custom airborne PDU incorporated an EMI filter designed to meet the aircraft’s EMI attenuation specification with adequate margin. The filter passed its lab test easily. The individual aircraft equipment items met their individual EMC requirements.

The combined aircraft equipment system showed some excessive EMI noise that should easily have been suppressed to within specification when buffered by the PDU filter. When the two systems were mated, the equipment was expected to meet the EMI specifications readily. Surprisingly, the EMI increased drastically and at a low frequency of about 45kHz (and at higher multiples of that frequency).

Individual systems were, one by one, cycled on and off to determine which was affecting the pattern. When the aircraft computer was shut down, the EMI disappeared. The aircraft computer had its own EMI filter to reduce the noise from its SMPS.

The investigation disclosed that the aircraft computer EMI filter was resonating with the power distribution unit’s EMI filter at the problem frequency. The filter engineer was able to redesign the PDU EMI filter in the field and to move its resonance to a point where it would not resonate with the aircraft’s computer filter.

These types of problems can occur whenever there are significant modifications to a complex system, and they require due diligence from the design team.

(From an article entitled “EMI considerations for Power Distribution Units” by James F. McNulty, of Marway Power Systems, in Interference Technology’s 2003 Annual Guide, www.interferencetechnology.com)

815 Low voltage LED string triggersshop’s burglar alarm

There is a new comment on the post “The energy-saving LED bulb that switched off the radio”.

http://conversation.which.co.uk/energy-home/led-bulb-radio-interference-dab-test/

Author: Dave D. Comment:

Interesting that this convo should resurface again: yesterday I went to my local electrical


retailers for a few oddments and noticed that he had his managed alarm system engineer in, tinkering with the controls.

As I always have a chat with the proprietor, anyway we got talking and after a while the alarm engineer came and asked to speak to Phil. The conversation went along the lines of:

“When did you first get these false alarms?”...

“About 3 weeks ago, about midnight”...

“Does it happen every day?”...

“Yes”...

“about the same time?”...

“Yes” ...

“And always the same zone?”...

“Yes”....

“The shop windows?”...

“Yes”...

The engineer went away for a while and was poking around in the shop window and then came back to ask:

“It looks like you’ve recently had the lights changed over the window?”...

“Yes, that was about 3 or 4 weeks ago”...

“What sort did you have before?”...

“Fluorescent tubes”...

The engineer went away again, and then came back.

“I think I’ve solved it, do your window lights automatically switch off late at night?”...

“Yes, about 11:30ish”...

“I think it’s these new spotlights you’ve had in, the driver unit is near to the alarm cables and I think it’s causing interference when it switches on and off”...

“Why doesn’t it set the alarm off when they switch on then?”...

“What time do they come on?”...

“About 9:00 in the morning”...

“And are you usually here before they come on?”...

“Oh yes, about half an hour before”...

“So the alarm is not set when the lights come on ?”...

“Ah, I see what you are saying!”

Engineer and shopkeeper then locked shop door briefly, switched off window lights, set alarm and switched on window lights... sirens blaring, lights flashing.

So it seems that the LED driver unit for the little low voltage LED track of lights over the window causes EMF (or EMI) triggering burglar alarms too.

(Kindly supplied by Richard Marshall, of Richard Marshall Ltd., www.design-

emc.co.uk, on 5 November 2013. Which? magazine says “see all comments on this post here: http://conversation.which.co.uk/energy” but this is a list of hundreds of posts and I found it easier to go straight to http://conversation.which.co.uk/energy-home/led-bulb-radio-interference-dab-test – Editor.)

816 Satellites Need Better Shielding from Space Weather

Scientists are investigating the effects of extreme space weather on geostationary satellites with the intention of better preventing electronics failure.

In a new study published in Space Weather, researchers from the Massachusetts Institute of Technology (MIT) say that solar flares, geomagnetic storms and other forms of electromagnetic radiation may be to blame for up to 26 failures in eight geostationary satellites owned by London-based telecommunications company Inmarsat that took place over 16 years of operation.

Geostationary satellites orbit at the same rate as the Earth’s rotation, which allows the satellites to maintain a constant location relative to the planet throughout their lifetime while providing access to television, Internet and communication services. Designed to last for up to 15 years, the satellites are heavily shielded to protect sensitive electronic components from solar radiation; however, say MIT researchers, over time radiation can penetrate the shielding and affect the performance of these components.

“If we can understand how the environment affects these satellites and we can design to improve the satellites to be more tolerant, then it would be very beneficial not just in cost, but also in efficiency,” Whitney Lohmeyer, a graduate student in MIT’s Department of Aeronautics and Astronautics, said. Lohmeyer and Kerri Cahoy, an assistant professor of aeronautics and astronautics, are working together to evaluate how sensitive satellite components are to the weather conditions in space, and how these conditions contribute to satellite failure.

Results from the study indicated that the

majority of the Inmarsat satellite failures overlapped with periods of high-energy electron activity during declining phases of the solar cycle. The researchers believe that this particle flux may have accumulated in the satellites over time, creating internal charging that damaged the amplifiers responsible for strengthening and relaying signals back to Earth. While most satellites carry back-up amplifiers, says Lohmeyer, over time this supply may run out.

“Once you get into a 15-year mission, you may run out of redundant amplifiers,” she said. “If a company has invested over $200 million in a satellite, they need to be able to assure that it works for that period of time. We really need to improve our method of quantifying and understanding the space environment, so we can better improve design.”

Today, engineers design satellites using radiation models to predict how much radiation a satellite in a particular orbital path may be exposed to during its lifetime. But these radiation models aren’t perfect, says Cahoy.

“Space weather is a lot more dynamic than models predict, and there are many different ways that charged particles can wreak havoc on your satellite’s electronics,” she said. “The hard part about satellites is that when something goes wrong, you don’t get it back to do analysis and figure out what happened.”

As users continue to demand more capabilities, engineers will need to ensure that increasingly complex satellites remain adequately protected from solar radiation. Understanding the connection between space weather conditions and the effects on satellite components will help guide these design improvements.

(From “Satellites Need Better Shielding from Space Weather, Say Researchers” by Aliza Becker in Inter ference Technology’s eNews magazine, 09/17/2013, www.interferencetechnology.com/satellites-need-better-shielding-from-space-weather-say-researchers. See more at: http://web.mit.edu/newsoffice/2013/space-weather-effects-on-satellites-0917.html.)

Banana Skins

Banana Skins are kindly compiled for us by Keith Armstrong.If you have any interesting contributions that you would like included, please send them together with the source of the information to: [email protected]

Although we use a rather light hearted approach to draw attention to the column, this in no way is intended to trivialise the subject. malfunctions due to incorrect EMC procedures could be life threatening.


John Woodgate’s ColumnSlow-down?The BSI EMC committees for Product and Basic standards met in January but the previous meetings were in April 2013. Some years ago, these BSI committees were meeting up to four times a year. So does that mean that EMC standards work is reaching maturity? It doesn’t seem so; CISPR 14-1 and -2 and CISPR 15 are being fully revised, while the saga of CISPR 32 and 35 is reaching at least the end of phase 1. And CISPR 16 sections are almost as dynamic as ever. So the reduced frequency of meetings must probably be put down to the economic climate and one of its more obvious consequences – anyone who has a job has far too much work to do to give a high priority to standards meetings.There has been some growth in the use of electronic communication – IEC and CENELEC require some documents to be distributed via their web sites and there are facilities for commenting and discussing, but there seems quite a lot of reluctance to use them. BSI also has a communication web site for committee members, with a host of facilities but again, many seem to be little used. I base that on the fact that only on very few occasions have I received an advice to go to a feature other than a committee’s plain database of circulated documents.

Some standards-making bodies do not like plain email communication, even using comprehensive official address lists. Could this be a matter of not feeling ‘in control’? Of course, such communication can’t be stopped, and efforts to do so might drive it underground, with an increase in the incidence of the main drawback of the system – that not everyone who should see a communication might in fact see it, maybe due to a changed email address, an omission from the mailing list or simply the email black hole that lurks in cyberspace and swallows email without trace on random occasions.

The IEC set up open round-table meetings to discuss new approaches to standards-making, but the uptake seems to have been less than expected. Apart from the current reluctance to allocate time to standards meetings in general, it may be that most people don’t see what a ‘new approach’ would be and why it might be helpful. As it is, most methods of preparation seem to be explored, from having one person prepare a full first draft, through preparation by ‘small groups’ of two to ten to (usually inefficient) preparation by a larger group.

Even preparation of different parts of the text by several people is undertaken, but this is not a good idea, because it can be difficult to ensure lack of duplication and preservation of consistency and completeness, and different writing styles can be evident and confusing. Following initial preparation, the commenting and voting procedures seem to work quite well, although the continuing pressure to work faster is becoming counter-productive, as predicted, because errors that persist even to the publication stage are increasing.

Is there anyone there?It is notable that committee membership lists include names of people who have never been known to attend a meeting or submit comments on documents. Some standards bodies have rules that require repeated inactivity to be reported and investigated, but the situation persists. It is clear that not everyone who has an interest in a standard can dash around the globe attending one-day meetings, but the absence of comment submission is much less easy to explain away. National standards committees also seem to have different philosophies about participation in international work. To preserve full participation status, they are now required by some international bodies to comment and vote, but some achieve this by ‘supporting’ every comment document and voting positive on every occasion. This latter can be pernicious, as enough positive votes can result in a publication that attracted large numbers of adverse comments from National Committees that actually looked in detail at the document.

How to write a standardPeople hate, love or just tolerate the special language in which standards are written. But standard writers have a large body of editorial rules to observe. Unfortunately, many either don’t know that or disregard it. This is extremely inefficient, because it often results in hundreds of editorial comments from National Committees, only some of which are consistent with each other and with the actual rules. We even see comments proposing to ‘correct’ text that conforms to the rules. It would be good if the editorial rules were consistent among standards bodies, but they are not, even to the extent that what one body insists on is prohibited by another. Clearly, this can confuse people who draft standards for more than one body. and isn’t consistent with the spirit of standardization.Many standards bodies have a Microsoft Word template that sets the text styles and page layout, but while some require committee members who engage in drafting to use the template, others reserve it for use only by their own staff. Unfortunately, problems now occur because people are using different vintages of Word, from 2003 to 2013, and different geographic versions – those used in Asia have to cope with double-byte characters which can have unexpected effects on other features.

Getting the structure of a draft in the approved form is particularly important where a draft in support of a New Work proposal is being produced. The proposal is far more likely to be approved if the draft ‘looks like’ a standard rather than some personal notes. It is also important to observe the rules about the use of auxiliary verbs:

shall – expresses a mandatory provision;

will – is best avoided unless it really expresses the future tense, as it can be confused with ‘shall’;


should – is considered to express recommendation, even if it doesn’t, as in ‘Should this be the case....’;

may – is considered to express permission, even if it doesn’t, as in ‘The second value may be greater or less than the first...’;

can – expresses possibility.

An insidious trap is using ‘may’ or ‘should’ in a Note, which many standards bodies now do not allow. Suitable substitutes are ‘might’ and ‘could’, but if permission or recommendation is indeed expressed, then the text should not be a Note.

Clearly, it’s not possible to reproduce all the editorial rules here, but as a guide, the first four clauses (not ‘sections’ or ‘paragraphs’) of a draft could look like this:

1. ScopeExactly what is covered, and maybe what is not, if that is not self-evident. No requirements to be included.

2. Normative referencesUse the latest version of the introductory text set by the standards body and format the entries correctly.

3. Terms and definitionsDetailed formatting rules often apply, also the way in which the definition texts are expressed, such as single phrases with no verb and no initial article. A recent case involved a ‘definition’

of 170 words – not a definition but a life history!

4. GeneralProvisions, generally not quantified, that apply to the whole subject of the standard, and need to be understood in order to interpret subsequent clauses correctly.

Then follow further clauses containing detailed, usually quantified, provisions. Some material, which would break up the flow of the main text, may be placed in Annexes, which can be normative (provisions using ‘shall’) or informative. But it is very desirable to avoid ‘annexorrhoea’ – putting almost everything in an Annex. There have been recent examples of this, such as a standard with six clauses and twenty-two annexes.

Next timeThere will be some standards meetings to report and comment on, and perhaps further philosophical issues will become prominent. Note that this time, neither PLT nor Smart Grid have been mentioned.

J. M. Woodgate B.Sc.(Eng.), C.Eng. MIET MIEEE FAESHonFInstSCEEmail: [email protected]:www.jmwa.demon.co.uk© J.M.Woodgate 2014

Output on phase angle


Know Your StandardsOverview of CISPR publicationsOnce a year, CISPR circulates to National Committees a document giving details of progress on all of the standards under its aegis. Included is a table of all CISPR standards, with summary data, and the table below is based on it.

The information in this table is subject to frequent change, so references should be confirmed before use.

CISPR publications and responsible committees

Publication CISPR xx-x-x Subject Latest issue/

amendmentMaintenance date

(earliest)Committee

CISPR11 Industrial, scientific

and medical equipment2009 2014 B

12 Engines 2009 2012 B13 Radio and TV –

emissions2009 2012 I

14-1 Appliances – emission 2008 2014 F14-2 Appliances – immunity 2008 2014 F15 Lighting 2008 2015 F16-1-1

Measuring instruments and fixtures

2010 2013 A16-1-2 2006 2012 A16-1-3 2004 2012 A16-1-4 2012 2012 A16-1-5 2012 2011 A16-2-1

Methods of measurement

2008 2013 A16-2-2 2010 2013 A16-2-3 2004 2013 A16-2-4 2003 2011 A16-2-5 2008 2016 A and H16-3 Technical reports 2010 2013 A16-4-1

Uncertainties, statistics and limit modelling

2009 2012 A16-4-2 2007 2012 A16-4-3 2006 2010 A16-4-4 2007 2014 H16-4-5 2006 2012 A17 Filters 2011 2020 ATR 18-1

Overhead power lines and HV equipment

2010 2015 BTR 18-2 2010 2015 BTR 18-3 2010 2015 B19 Emissions from

microwave ovensWithdrawn B

20 Radio and TV – immunity

2006 2012 I

21 Interference due to impulsive noise

Withdrawn (see CISPR 25)

D

22 ITE – emissions 2008 2012 I23 Limits for ISM

equipmentWithdrawn (see CISPR TR 28)

H


Publication CISPR xx-x-x Subject Latest issue/

amendmentMaintenance date

(earliest)Committee

CISPR24 ITE – immunity 2010 2013 I25 Vehicle equipment –

emissions2008 2012 D

TR 26 Emissions from electric traction systems

Project abandoned at CDV stage, in 2000

2013 proposal to revive did not receive enough National Committee support

B

27 Not issuedTR 28 Guide on limits for

ISM emissions1997 2012 B

TR 29 TV picture quality assessment

Withdrawn I

TR 30-1 Emissions from electronic ballasts for fluorescent lamps

2012 2016 F

TR 30-2 Emissions from electronic ballasts for other discharge lamps

2012 2016 F

31 Database on radio services characteristics

2012 2015 H

32 Multimedia equipment – emissions

2012 2013 I

33 Not issued34 Not issued35 Multimedia equipment

– immunity2014 2016 I

IEC 61000-6-3 Generic emission standard

2011 2014 H

IEC 61000-6-4 Generic emission standard

2011 2014 H

More from OttawaLast time (despite information to the contrary, it actually wasn’t the last time, of course), all the outcomes of the CISPR and TC77 standards meetings in September were not available, but they are now, and there have been further developments.

Generic standards

IEC 61000-6-3A CD for Amendment 2 was circulated and as a result a second CD is expected early in 2014.

IEC 61000-6-4Exactly as for its partner above, a CD for Amendment 2 was circulated and as a result a second CD is expected early in 2014. Both are expected to be published at the end of 2014.

CISPR 11CISPR 11 is a quite complex standard, with a very wide scope. A CDV for one fragment of an amendment is promised early in 2014. Another fragment will also be brought to the CDV stage and the two fragments will be combined for FDIS voting. Two more fragments were issued as draft CDVs in late 2013,

but the real CDVs will be issued early in 2014. Yet another fragment was approved at the CDV stage in mid-2013. All this work will result in a sixth edition of the standard, probably by the end of 2014.But it doesn’t stop there. New proposals include:• Measurements at 5 m separation and applicable limits;• Introduction of Fully-Anechoic Room (FAR) tests;• Wireless power transmission.

CISPR 14-1A CD for a full editorial revision (to become Edition 6) and a separate one for a technical revision have been circulated. A combined CD, taking into account extensive National Committee comments, will be produced.

CISPR 14-2A CD for a full revision was circulated, followed by a second CD. No move to CDV is indicated yet.

CISPR 15Two Interpretation Sheets to Edition 8 were issued. For the next amendment, two fragment CDs have been issued, including the ISHs and technical changes. ADC (Document


for Comment) for a full revision, resulting in Edition 9 was circulated. Its purpose (as merely an outline) was not clearly explained, so it attracted many comments. A second DC will be circulated, which is unusual.

CISPR 13Amendment 1 to Edition 5 is at the CDV stage

CISPR 18A first set of three maintenance CDs is expected from CISPR B early in 2014.

CISPR 20No longer in the Programme of Work of CISPR I

CISPR 22No longer in the Programme of Work of CISPR I

CISPR 24The issue of a CDV for changes to this standard (to be superseded by CISPR 35) has been delayed, at least until some time in 2014.

CISPR 26Not to proceed; see the Table above.

CISPR 32

Fragment amendments for both Edition 1 and Edition 2 are currently at the CDV stage. It is probably unprecedented for

two editions to be under development at the same time.

CISPR 35The FDIS was circulated early in January 2014, but considering the volume of comment on the CDV, the outcome of the second stage voting must be considered uncertain.CISPR 32 and CISPR 35 perforce have much common text; there would be huge problems if they were inconsistent. So it is proposed to have a separate maintenance process for this common text.

Next meetingsWorking Groups, Maintenance Teams and Task Forces have meetings throughout the year, and some higher-level committees have two meetings a year. It is useful to know the plans as they can indicate deadlines for submitting comments and new proposals.CISPR and its sub-committees will next meet in Germany in October 2014.

Next timeThings are always changing, so no doubt there will be development to report, but at present it is not clear what they will be.

J. M. Woodgate B.Sc.(Eng.), C.Eng. MIET MIEEE FAESHonFInstSCEEmail: [email protected]:www.jmwa.demon.co.uk© J.M.Woodgate 2014

PRODUCT GALLERYSARLite: Smart, Compact and Fast SAR Measurements

The Microwave Vision Group (MVG) is proud to introduce SARLite, a device that gives you access to a new and exciting experience in speeding up your SAR measurements.

SARLite simplifies and accelerates the traditional SAR measurement method. With its 265 probes and 651 sensors in a 3D matrix design, an area greater than 20 x 20 cm area can be measured in only a few seconds. In addition, the enclosed wideband tissue simulant gel enables a wide range of frequency measurements.

The SAR (Specific Absorption Rate) is the quantity characterizing the absorption of electromagnetic wave energy from any type of radiating device such as a mobile phone,

laptop, or a tablet PC into the human head or body. SAR measurements are traditionally performed by using an electric field probe encased in a phantom (artificial head) filled with a liquid. The dielectric properties of the liquid are representative of head or body tissues. The probe is moved inside the phantom by a robot, making it possible to obtain discrete E-field values. These values are averaged in a volume corresponding to a cube weighing 1g or 10g. SARLite can be used alone or as a complement to your traditional SAR measurement systems to pretest your equipment, better

preparing it for certification.

There is no robot in SARLite, so there are no robot servicing needs. Probes and the enclosed tissue simulant gel are well protected inside the case. This means no evaporation and no need to refill or replace the liquid.

SARLite allows you to measure the E-field values emitted in order to tune electromagnetic radiation at any stage in R&D and on any mobile device.

Tel: +44 (0)1942 296 [email protected]


PRODUCT GALLERYYokogawa obtains ISASecure® EDSA certification for ProSafe®-RS safety instrumented system

Yokogawa’s ProSafe®-RS safety instrumented system has obtained the ISASecure® Embedded Device Security Assurance (EDSA) certification

Cyber crime and cyber warfare are on the rise worldwide, and are growing ever more sophisticated. Recent prominent cases of industrial sabotage and espionage have escalated these concerns. Companies in the oil, petrochemical, power, and other industries are increasingly vigilant about such intrusions, and are being careful to check a product’s cyber security features before purchasing it. Yokogawa has obtained the ISASecure

EDSA certification for the ProSafe-RS safety instrumented system to assure customers of its high reliability.

The ISASecure program has been developed by the ISA Security Compliance Institute (ISCI) with the goal of accelerating the industry-wide improvement of cyber security for industrial

automation and control systems (IACS). It achieves this goal by offering a common industry-recognised set of device and process requirements that drive device security, simplifying procurement for asset owners and device assurance for equipment vendors. The ISASecure EDSA certification has three elements: communication robustness testing (CRT), functional security assessment (FSA), and software development security assessment (SDSA), and is based on the IEC 62443-4 standard.

Nobuaki Konishi, vice president of the Systems Business Division

in Yokogawa’s Industrial Automation Platform Business Headquarters, comments: “Yokogawa is continually striving to provide its customers with optimum security solutions for their control systems by developing highly secure systems and instruments and providing operational support services. Yokogawa aims to strengthen its relationships with customers by demonstrating the reliability of its security solutions.

Tel: +31 88 464 [email protected]/eu

Stand-Alone Generator for Slow Damped Oscillatory Wave Testing from Teseq

The NSG 3040-SOW tests single phase equipment up to 270 V and 16 A

Teseq, a leading developer and provider of instrumentation and systems for EMC emission and immunity testing, has introduced an easy-to-use, stand-alone generator for slow damped oscillatory wave testing in single phase equipment up to 270 V and 16 A. The new NSG 3040-SOW is in compliance with current testing standards IEC/EN 61000-4-18 and ANSI

C37.90.1.

Ideal for use by power stations, substations, electric meter manufacturers, relays and switch manufacturers as well as EMC testing labs that already use Teseq equipment, the NSG 3040-SOW enables users to perform over-testing at test levels up to 4.4 kV.

The NSG 3040-SOW features unique capabilities including selectable reduced source impedance as well as higher

pulse rates which enable users to test under conditions that are closer to reality, ensuring their product will perform as intended in the real world.

Designed to complement Teseq’s line of NSG 3040 and NSG 3060 series of EMC test generators, the NSG 3040-SOW features the same user-friendly 7” high contrast, color touch-screen interface featuring superb graphics to enable quick and user-friendly set-up of test procedures.

The generator can obtain quick, reliable results of standardized tests with a few clicks using the integrated Test Assistance (TA) function. The NSG 3040-SOW offers inputs supported by an integrated keyboard or thumbwheel with additional keys for sensitivity adjustment.The new generator features Windows-based control software that simplifies test programming and compilation of complex test sequences composed of multiple waveform types.

Technical Specifications• Test voltage range from 0.2 kV to 4.4 kV• Oscillation frequency of 100 kHz and 1 MHz• Selectable source impedance of 200 Ohm and 150 Ohm• Pulse repetition rates exceed standards requirement: 1 MHz pulse: From 1/s to 600/s, default is 400/s 100 kHz pulse: From 1/s to 120/s, default is 40/s• Ethernet port for external PC control• Approximate weight –25 kg (55 lbs)• Dimensions – 449 mm (17.7”) wide x 226 mm (8.9”) high x 565 mm (22.2”) deep

Tel: 0845 074 0660 [email protected] www.teseq.com

The new NSG 3040-SOW stand-alone generator from Teseq


PRODUCT GALLERYUL achieves two new GlobalPlatform qualifications

Leiden, The Netherlands; 19 December 2013: UL is pleased to announce the qualification by GlobalPlatform of two new Collis Compliance Test Suites: GlobalPlatform UICC SCP81 Extension and GlobalPlatform UICC Memory Management Extension, which further enhances UL’s compliance test suite portfolio.

The Collis Compliance Test Suite for GlobalPlatform UICC SCP81 Extension implements the official test suite developed for the GlobalPlatform Compliance Program. The test suite ensures the compliance of a UICC that implements GlobalPlatform Card 2.2.1 Amendment B (Remote Application Management over HTTP) technology.

The Collis Compliance Test Suite for GlobalPlatform UICC

Memory Management Extension implements the official test suite developed for the GlobalPlatform Compliance Program. The test suite ensures the compliance of a UICC that implements GlobalPlatform Card 2.2.1 Amendment C (Cumulative Delete and Cumulative Granted Memory) technology.

Maxim Dyachenko, Service Line Manager Test Tools at UL Transaction Security, added:

“Over the past years, smart card technology has changed considerably. Many smart card applications in various industries rely on GlobalPlatform technology for personalization, key management and security. Adding new GlobalPlatform configuration specifications keeps our existing portfolio of certified GlobalPlatform test suites up to date with the GlobalPlatform Compliance Program, empowering our customers to use the latest versions of these test suites.”

Addi t iona l t es t su i tes that complete the Collis GlobalPlatform Compliance Test Suite portfolio are the test suites for GlobalPlatform UICC Configuration, GlobalPlatform UICC Contactless Extension, GlobalPlatform Mapping

Guidelines, GlobalPlatform SWP and HCI, GlobalPlatform Basic Financial Configuration, and GlobalPlatform ID Configuration.

GlobalPlatform has extended its compliance program to provide a formal framework that enables the creation and qualification of test tools that can now be used to confirm the compliance of a card product to the association’s Card Specification and different configurations.

www.ul-ts.comwww.globalplatform.org

Modelithics Global Models™ for Syfer’s Ultra-Low ESR MLCCs

Modelithics Inc. has added Syfer’s Ultra-Low ESR range to their component modelling library. These models allow design engineers to accurately simulate the performance of Syfer’s Ultra-Low ESR Series multilayer ceramic capacitors (MLCCs) in circuit simulations. They are the first addition to Syfer’s Modelithics Component Model Library and are fully integrated with a range of software applications including

Agilent ADS, Agilent Genesys, and AWR Microwave Office.

Modelithics Inc. is the industry leader in providing simulation models for RF, microwave and millimeter-wave devices. The company offers comprehensive measurement and modelling services, as well as highly accurate, scalable measurement-based models that integrate seamlessly with popular electronic design automation

(EDA) tools. Modelithics Global Models™ reliably track how each component’s performance will change with substrate characteristics and automatically scale all substrate-sensitive parasitics with nominal part value, enabling a single model to generate accurate simulations of a complete component family.

“Simulation-ready models like Modelithics Models can save designers considerable amounts of time and money on new designs,” says Steve Hopwood, senior Applications Engineer at Syfer. “For many applications where economical, high performance is required these new Modelithics Models for our Ultra-Low ESR capacitors will be beneficial to RF and microwave designers. In particular this range of high frequency capacitors is ideal for use in RF power amplifier assemblies, mobile communications, satellite, GPS, automotive, matching networks, and wireless LAN applications.”

Modelithics President and CEO, Larry Dunleavy, added “We are very pleased to welcome Syfer as a new Strategic Modelithics Vendor Partner (MVP) and we look forward to helping designers get great design results with these models”. The Syfer Ultra-Low ESR capacitor range utilises a very stable High Q ceramic material to give excellent low loss performance in systems over a wide temperature range from -55°C to +125°C. To offer increased reliability with superior mechanical performance Syfer’s FlexiCap™ termination system option is available along with various other termination options. RoHS-compliant finishes are standard. Parts are available taped and reeled as well as bulk packaged.

Tel: +44 (0)1603 [email protected]

CST – COMPUTER SIMULATION TECHNOLOGY | www.cst.com | [email protected]

Components don’t exist in electromag-netic isolation. They influence their neigh-bors’ performance. They are affected by the enclosure or structure around them. They are susceptible to outside influ-ences. With System Assembly and Mode-ling, CST STUDIO SUITE helps optimize component and system performance.

Involved in emc/emi analysis? You can read about how CST technology is used for EMC/EMI analysis at www.cst.com/emc.

If you’re more interested in microwave components or signal integrity analysis, we’ve a wide range of worked applica-tion examples live on our website at www.cst.com/apps.

Get the big picture of what’s really going on. Ensure your product and compo-nents perform in the toughest of envi-ronments.

Choose CST STUDIO SUITE – Complete Technology for 3D EM.

Make the ConnectionFind the simple way through complex EM

systems with CST STUDIO SUITE


The many benefits of HDI (microvia) PCB technologyKeith Armstrong (www.cherryclough.com)

IntroductionI first wrote about High Density Interconnect (HDI) printed circuit board (PCB) Technology in this Journal in 2005 [1], and later in Chapter 7.5 of my book on PCB Design for EMC [2].

In the intervening 8 years there have been many significant developments in this important technology, and many useful new publications on it, so I felt it was time to revisit the subject.

It is generally found that designing for good electromagnetic compatibility (EMC) at circuit and PCB level automatically achieves excellent signal integrity (SI) and power integrity (PI), and is the best way to achieve the lowest overall-cost-of-manufacture with the shortest time-to-market – the modern recipe for success in electronic product manufacture. This approach to improving competitiveness and financial success underlies almost everything that I write.

So, my focus in this article is on the use of HDI board technology to reduce the overall cost of manufacture required to achieve good SI, PI and EMC and get new products to market as quickly as possible.

What is HDI and how does it save cost?HDI PCB technology is also known as ‘microvia’, and sometimes as ‘sequential build-up’ or simply ‘build-up’ technology. In this article I shall call it HDI, following the approach taken by the IPC standard that defines it [3].

It is based on the use of via holes of 150µm (0.006 inch) diameter or less, i.e. holes smaller than can be drilled mechanically, which can go between adjacent layers in a PCB without needing to go through all of the layers. These are generally called ‘microvias’, and they can be ‘buried’ or ‘blind’.

Microvias are created by laser-drilling and through-plating the vias between each adjacent pair of copper layers separately, before they are laminated together to create the finished PCB. When the end result is a via hole that only connects between internal copper layers, it is called a buried microvia. When the result is a via that connects to one outer layer of the PCB and not to the other, it is called a blind microvia. Traditional drilled and plated all the way through the PCB vias (i.e. through-hole plate, THP) can also be used along with buried and blind microvias.

These basic features are shown in Figures 1 and 2. Figure 1 shows a typical 6-layer HDI substrate used for MCM (multi-chip module, [4]) and POP (package on package, [5]) types of integrated circuits (ICs).

Figure 1 : Some features of original, basic HDI (microvia) PCB technology

Figure 2 : Some features of modern microvia technology

The way that microvias are created results in them always being closed off at one end, or both, which means they don’t ‘steal’ solder during reflow soldering. Just as when using copper-filled or copper-capped THP vias, this allows via-in-pad layouts to be used, which saves PCB area and is very good for the PCB’s decoupling performance because it reduces the inductance of the decoupling current loop. However, if blind microvias are not totally filled or capped with copper, care must be taken to ensure the little bubbles of air they contain don’t harm the solder joint when they expand during reflow soldering.

A problem with the commonplace through-hole plate (THP) PCB technology is that each via hole penetrates through every layer, and so tends to make routing more difficult. It is often found that once the number of THP PCB layers required to be able to fully route a board reaches 10, the associated large number of via holes causes diminishing returns for the routing.

Each additional THP layer that is added provides fewer routing paths than the one before, so it is often found that if 10 layers aren’t sufficient, the number of board layers rapidly increases to 16, 18, or more.

But microvias only penetrate the layers they are required to, and so each additional HDI PCB layer permits as much routing as every


previous layer, no matter how many layers are used.

Because of the above issues, HDI technology can achieve twice the number of pins per unit area than THP and so reduce the area of PCB required by 40%. It can also significantly reduce the number of layers required where a THP design would require 10 or more layers (also making the board much easier to lay out) [6]. Recent advice from Mentor Graphics was that a high-density 18-layer THP PCB would only need 10 layers if realised using HDI.

HDI techniques help to make the smallest, lightest, and least power-hungry products, and can be found in a wide variety of common products, such as cellphones and even some toys. Some high-reliability or harsh-environment products use HDI because microvias are inherently more mechanically robust than THP vias. The basic standard for the design of HDI PCBs is IPC-2315 [3], and there are many other IPC standards on HDI technology.

Some of the early articles on HDI are [7], [8] [9] [10] [11] and [12]. Coomb’s book on PCBs [13] is claimed to provide a good introduction to HDI technology, and [14] is probably the modern bible on HDI.

Improved routing under BGAsFigure 3 shows the typical ‘dog bone’ layout required when using THP for a ball grid array (BGA) type of IC package. Compare this with the much simpler layout permitted by via-in-pad techniques shown in Figure 4, using copper-capped THP vias.

Figure 3 : Typical THP dog-bone pad patterns under a BGA

Figure 4 : Much easier routing under a BGA by using via-in-pad THP

Copper-capping increases the bare board cost by about +10%, when purchasing in volume. Copper-capped prototypes can save cost by not applying the copper-capping process, instead manually filling the

vias that will be capped with high-temperature solder (being careful to maintain a flat-enough surface for reflow soldering the BGA).Figure 5 shows how using HDI via-in-pad under a BGA allows microvia holes to be moved around within the confines of the pads, to ease the routing on the buried PCB copper layers underneath the IC.

Figure 5 : HDI under a BGA means the vias don’t have to be on a regular grid

Figures 6 and 7 show an example of an HDI substrate of an IC, in this case used mostly as a pin-spreader between the contact pads on the semiconductor die itself and the much greater spacing of the contact pads intended for soldering to a PCB. This HDI substrate also carries a number of special-purpose decoupling capacitors, including reversed aspect and interdigitated types (see [15] and [16]).

Figure 6 : The Xilinx Virtex-7 FPGA

Figure 7 : Example of the HDI substrate used in the Xilinx Virtex-7 FPGA


A great deal of very useful and detailed information about HDI board routing for BGAs with ball-pitches from 0.8mm to 0.25mm pitch (actually, 0.25mm pitch is for flip-chip devices) is given in [17], but this document does not permit me to replicate anything from it in without specific authorisation from Viasystems, which unfortunately I don’t have time to obtain before the publication deadline for this article.

PI, SI and EMC problems with THP solved by HDIA comprehensive review of SI benefits of HDI can be found in [6], and [18] addresses its PI benefits.

A big problem for the PI and SI of BGA devices is achieving low-enough impedance in their power distribution networks (PDNs), which is at least partly due to the difficulties of achieving ‘solid’ 0V and Power planes underneath them when using THP.

A perforated 0V plane under a BGA also means less shielding due to the image plane effect (see Chapter 4.1 in [2]) – making it more likely the board-level shielding (BLS), or much more costly enclosure shielding will need to be added to achieve adequate EMC.

The best that can be done with THP board technology is to use suitable track-and-space (track-and-gap) layout rules to achieve a complete mesh or grid over the area covered by the BGA’s solder pads.

Figure 8 shows the bad effect of not using suitably small track-and-space rules for the pitch of the BGA, whilst Figure 9 shows the benefits of using appropriate rules, both taken from [19].

Figure 8 : Example of partial 0V plane meshing under BGA with large gaps caused by using THP with too-large track-and-space

rules

Figure 9 : Example of partial 0V plane meshing under BGA, when using appropriate track-and-space rules for THP

To be able to create complete meshed 0V and Power planes under a BGA with ball pitch of 1mm needs 175µm (0.007 inch) track-and-space rules, whereas BGAs with pitches 0.8 - 1.0mm need 100µm (0.004 inch) and 0.5 - 0.8mm pitch needs 50µm (0.002 inch) track-and-space.

Often, when I suggest using less than 175µm track-and-space PCB rules, I get the automatic reaction that this will not be allowed because it will increase the BOM cost. Leaving aside the fact that the BOM cost is irrelevant, and what matters is the overall cost of manufacture (see [20]), this reaction shows that people are not keeping up with modern PCB technology.

Not keeping abreast of PCB technology contributes to losing competitive advantage, because for several years now, 100µm volume-manufacture has cost no more than 175µm, and 50µm volume-manufacture has only cost a little more (though we usually had to shop-around Chinese board manufacturers to get this). As far as prototypes are concerned, some UK PCB manufacturers can do 50µm track-and-space (e.g. Merlin Circuit Technology).

But even properly-meshed areas of 0V or Power plane are still not very good for PI, and so are far from being good for EMC.

However, planes in HDI technology boards are automatically solid copper sheets (except where there are any THP or microvias passing through them without connecting to them) and so give the best possible PI and EMC for the type of stack-up.

In fact, this aspect of HDI was one of the original reasons for the development of this PCB technology. One of the things that makes viewing contemporary films and TV series from the 1990s more amusing than their Directors intended, is the way the actors use their state-of-the-art-at-that-time cellphones as if they are really cool, when from our perspective in 2014 they look stupid holding brick-sized electronics with external antennas to their ears. (No doubt future generations will think our iPhones and Blackberries look pretty uncool too, since their cellphones will be modules the size of an aspirin that are implanted at birth and communicate directly with their brains!)

Anyway, to get back on track, as cellphone manufacturers tried to reduce the size of their products from bricks to something that would easily fit in a pocket or handbag, and the component industry responded by making 0201 and 01005 sized passives, quad-flat pack and BGA ICs, the routing density on the reduced-size PCBs became so high that the number of THP vias required reduced the 0V and Power plane layers to flimsy, high-inductance meshes. These meshes were of little help in preventing the phones’ 2 Watt microwave radio transmitter signals from interfering with their microphone and microprocessor circuits.

So one of the main reasons why HDI PCB Technology was developed at all, was to provide 0V (so-called ‘Ground’ or GND) and Power plane layers that were continuous low-inductance sheets of copper having good characteristics even up to 1900MHz.

Figure 1, 2 and 10-17, show cross-sections of a great many types of HDI board construction, clearly showing that the microvias do not cause perforation of a plane layer unless they are designed specifically to pass through it without connecting to it. (There is no point in showing a plane view to compare with Figures 8 and 9 above, because it would simply be a blank sheet of copper.)

But even where microvias do perforate a plane layer, they are so very small that the gaps required to prevent them from shorting-out to the plane (i.e. ‘antipads’) are very much smaller than would be the case for a THP via hole, causing much less increase in plane inductance and corresponding degradation of SI, PI and EMC characteristics.


EMC benefits of HDI These include…• via-in-pad reduces decoupling inductances and so pushes the inevitable undesirable resonances associated with decoupling capacitors to higher frequencies

• The shorter traces have their lowest resonant frequencies increased, so only become efficient ‘accidental antennas’ at higher frequencies• The smaller PCBs have their lowest resonant frequencies increased, so only become efficient ‘accidental patch antennas’ at higher frequencies• The shorter traces might not need to be treated as matched transmission lines, for a given value of signal distortion (overshoot, ringing, intersymbol interference, etc.)• HDI 0V and Power planes aren’t as heavily perforated as those of THP PCBs so they have lower impedances (hence lower PCB emissions and better immunity). • Other benefits of these much-less-perforated planes include a more constant return path inductance for improved Z0 control of transmission-line traces (better SI and EMC); increased inherent shielding because of the improved image plane effect, and better shielding between the circuits on the top and bottom sides of a PCB (e.g. lower crosstalk between them).

The above are all very valuable for modern ICs, especially where ball grid array (BGA) devices are used – when the use of THP PCB technology results in severe plane perforation under and around the device – exactly where good EMC and good SI require the most solid 0V and Power planes, as discussed above.

HDI technology makes it much easier to use modern small IC package styles, such as...

• Miniature or Micro BGA

• DCA (direct chip attach)

• Flip-chip

• CSP (chip scale packaging)

• Tape Automated Bonding (TAB)

These small ICs, and the smaller PCBs they allow, can generally be made to have excellent SI and EMC because their thinner packages place them in closer proximity to the 0V plane in the PCB, so the image plane effect is more powerful. Also, the smaller bond wires and lead frames means they are less effective as accidental antennas at frequencies below their first resonance, and their first resonance is at a much higher frequency, so these devices tend to emit less from their bodies.

However, a problem with these very small devices is that they allow much higher switching edges and higher-frequency noises to be conducted into the PCB’s power distribution and signal traces, which can worsen emissions considerably. But they can generally be made to have better EMC than the larger devices they replace, if all of the advanced PCB techniques described in [2] are applied where needed.

HDI suppliers and costsThe usual assumption is that HDI will increase the BOM cost, making a product less competitive, but there are two reasons why this is usually wrong.

Firstly, a May 2000 IPC survey found that HDI bare-boards could be purchased in reasonable quantities for the same price as THP. More recently, [21] says that THP boards needing more than 8-10 layers should be able to reduce layer numbers and cost when made in HDI (e.g. a THP board needing 18 layers to route might only need 10

layers in HDI).

But even where its bare-board price is higher, HDI’s advantages in size, power consumption and especially EM performance could reduce a product’s BOM cost – even for boards that would need 8 or fewer THP layers.

Secondly, focussing on the BOM cost instead of the real-life cost of manufacture is a commonplace economic error (see [20]).

I remember the same complaint about increased BOM costs being raised when double-sided PCBs were first proposed, then again when THP was developed, then again when 4-layer THP PCBs were starting to become necessary, and yet again when 4 layers were no longer enough. The change to HDI from THP is no different. Where miniature BGAs with 0.5mm (or less) pitch, chip-scale, DCA, TAB or flip-chip devices are to be used – HDI is probably an essential technology for getting EMC-compliant products with low warranty costs to market quickly, and selling them at competitive prices.

Advanced types of microviasSince I first wrote about HDI PCB Technology in 2005 [1], there have been many significant developments in the technology.

Viasystems [22] list the following types of advanced microvias, cross-sections of which are shown in Figure 10.

Figure 10 : Some examples of advanced microvias from Viasystems [22]

A) Stacked MicroVias (SMV®) or Second Generation Microvias:Allows increased routing on multiple layersProvides solid copper plate eliminating potential solder voidingImproves Current Carrying Capability & Thermal ManagementProvides a Planar surface for BGA (Via-in-Pad)

B) Deep Microvias (DpMV™):Provide additional dielectric material & small geometry featuresImproved Impedance performanceProvides RF Microvia solutionsProvides a solid copper plateImproves Current Carrying Capability & Thermal ManagementProvides a Planar surface for BGA (Via-in-Pad)

C) Deep Stacked MicroVias (DpSMV™):Provides additional dielectric for RF applicationsMaintains small geometries on multiple layersImproved SIProvides a solid copper plateImproves Current Carrying Capability & Thermal ManagementProvides a Planar surface for BGA (Via-in-Pad)


Mentor Graphics provides a truly excellent summary of modern microvia techniques in [23], which includes the summary of modern microvia types in Figure 11.

Figure 11 : Examples of modern microvia types, from Mentor Graphics [23]

Originally, to connect through 2 or more board layers microvias had to be staggered, and examples of these can be seen in Figures 1 and 11. But Figure 11 also shows ‘skip’ and ‘stacked’ vias, which correspond with the advanced microvias in Figure 10 above. Stacked vias can also be seen in Figure 2. These advanced types of microvias help save board space where it is required to connect through two or more layers using microvias.

In [23] Mentor Graphics also describes the classifications in IPC 2315 [3] for certain defined ‘Types’ of HDI construction, as shown in Figures 12, 13 and 14.

Figure 12 : IPC Type I HDI board construction

Figure 13 : IPC Type II HDI board construction

Figure 14 : IPC Type III HDI board construction

IPC Type I HDI uses both microvias and THP vias in a structure consisting of a laminated core with no more than a single microvia layer on either/both sides. The microvias used are all ‘blind’, which are less costly to produce than ‘buried’ microvias.

IPC Type II HDI construction uses microvias, buried vias, and may also use THP vias (like Figures 1 and 2 above)

IPC Type III HDI construction uses microvias, buried vias, and may also use THP vias.

IPC Type IV, V and VI HDI constructions are also defined by [3], but are not described by [23] because (it says) “they are more expensive to fabricate and are probably not necessary for large dense PCBs with BGA breakout and routing challenges”.

There are many on-line sources of detailed information on advanced modern HDI PCB design and manufacture other than [22] and [23], for example [24] and [25].

Note that the stackups in Figure 12, 13 and 14 all use an FR4-type of core material with either FR4 or polymer prepreg ‘build-up’ layers containing microvias and/or THP vias. The FR4 and similar lo-cost woven dielectrics with THP vias are used to keep costs low, which seems to be a basic assumption in [23].

However, the use of different layer materials with their different temperature coefficients and rates of moisture absorbance (FR4 is hygroscopic) means that differential forces, strains, will exist between layers in the stack-up, and delamination is a real possibility – especially where temperature or humidity vary greatly in a repetitive cycle.


It is always possible to use the same material for every layer (which should therefore be a polymer, liquid crystal or other non-woven material to suit the microvias), although at extra cost.

This would bring less likelihood of delamination, and (because FR4 is a very lossy material with variable dielectric constant due to its glass weaves) it would also benefit the SI of high-speed, high-frequency, or high-data-rate signals (see Chapter 6.5 in [2]).

Also, using buried or stacked microvias to reach inner board layers means – as discussed above – using fewer layers and possibly saving overall cost (at least, once the number of THP layers reaches or exceeds 10).

So it seems to me that the use of FR4 layers with THP vias in Figures 12, 13 and 14 is probably only the lowest-cost approach where there are few such layers.

Using microvias throughout and avoiding as many THP vias as possible, seems to offer the most cost-effective approach overall – when SI, PI and EMC are taken into account, and when the total number of THP board layers would have to equal or exceed 10 to fit all the routing in.

Good HDI stackups for EMC, PI and SIVery important and significant benefits can be obtained from using a given order of 0V plane, Power plane, and signal layers in a PCB stack-up.

For the purposes of this discussion, I’ll leave aside the issues discussed at the end of the last section, about the overall cost benefits of using microvias on all board layers and avoiding as many THP vias as possible, but of course they apply whatever the stack-up.

Mentor Graphics’ guidance in [23] recommends the stackups shown in Figure 15.

Figure 15 : Good HDI stackups for EMC and PI, from [23]

Figure 16 shows an example of the use of one of the stack-ups in Figure 15.

Figure 16 : Example of a good HDI stackup for EMC and PI, from [23]

Having the outer layers (top and bottom) as 0V (i.e. GND) planes with Power (i.e. VCC) planes adjacent to them provides good inherent shielding and a distributed interplane capacitance immediately below the devices.

The interplane capacitance provides good decoupling above 300MHz because it suffers very little from the series inductances caused by the lengths of the via holes that connect the ICs to these two planes (see Chapter 5.3 of [2]).

The 0V planes on the outer layers of the board are of course perforated by gaps for the solder pads for the devices mounted on it. But modern large ICs, such as BGAs, generally have close to 50% of their pins dedicated to 0V, so the number of 0V plane perforations is reduced. Maximising the meshing of the remaining 0V plane perforations – as discussed above – is strongly recommended.

As long as there are no traces routed on the outer layers (e.g. microstrip traces) – that is, all the routing is done on inner layers connected to the device solder pads by via-in-pad – BLS can be used to deal with situations where the perforations in the outer 0V plane layers weaken the board’s inherent shielding by too much.

This is the basis of the “totally shielded board assembly” approach described in Chapter 4.6 of [2], which can avoid the need for any shielding at the level of the enclosure (and is the way that most/all cellphones are designed these days).

The stackups in Figures 15 and 16 also benefit from placing IC decoupling capacitors on the same side of the board as the devices they serve. As well as providing the best decoupling, this will minimise the number of 0V vias underneath the device, causing fewer plane perforations and easing the board’s routing. Of course we will still need some 0V vias under the devices to improve signal return paths for good SI and EMC, and we will also need a grid of 0V vias linking all the 0V planes together to push the cavity resonances between them to higher frequencies than we care about (see Chapters 4.2.8 and 5.3.6 of [2] for more details on suppressing the cavity resonances inside PCBs).

[23] says “if you manage your return paths with an appropriate but not excessive number of GND vias under the BGA, then there really isn’t a downside to using this method”.

These stackups benefit from the use of stacked microvias, although they will add to the cost. And these stackups are also an opportunity to reduce board size, product weight/size, and overall manufacturing costs by using:


• embedded capacitors, e.g. [26]• embedded pull-up and termination resistors, e.g. [27]• embedded surge/transient protection, e.g. [28]

As well as, or instead of, embedded discrete capacitors such as the GRU series in [26], these stackups can benefit significantly by increasing the interplane decoupling capacitance by using very thin dielectrics between the 0V and Power planes, ideally less than 50µm (0.002 inch).

The main problems with using such thin layers of FR4 or other prepregs are associated with quality and yield, so it is often better to use proprietary “embedded capacitance” board materials, for example [29] [30] [31] and others.

These are double-sided PCB laminate cores which can offer up to several nF per square centimetre, made from very thin (as little as 3µm, 0.00012 inch) but tough materials that withstand most PCB manufacturing processes. Good manufacturers supply such laminates as having passed individual tests at 500Vdc or more, to guarantee they have no short-circuits.

Using such high-capacitance core laminates it is possible to eliminate the use of all decoupling capacitor (excluding ‘bulk’ decoupling capacitors of 10µF or more).

The resulting savings in component cost, reductions in board area, and improvements in board assembly yields can overcome the added cost of using proprietary embedded capacitance laminates, to help achieve an even lower overall cost of manufacture, plus a smaller, lighter product.

This is yet another example of how using more costly, advanced PCB manufacturing techniques can reduce overall costs and increase profitability – but I meet many people who seem to believe that the way to increase profitability is to use the cheapest of everything, and who automatically ignore anything that might increase the bare-board cost. Real life is much more complex than such a simplistic approach can cope with, as I hope I show in my articles from time to time.

It is commonplace to route traces against either 0V or Power plane layers, especially where controlled characteristic impedance is required. But large ICs (especially BGAs) generally require multiple voltage supply planes which are often created by splitting Power plane layers into different voltage regions. I described this in a recent article [32], where I pointed out that it is bad for SI, PI and EMC if traces cross plane splits.

Where stackups such as those in Figure 15 are used to obtain good SI, PI and/or EMC, this issue could well be significant.

Quite often, devices and their traces can be organised to lie in the region covered by their associated Power plane, but perhaps the easiest solution is to sandwich all split Power planes between two solid 0V planes, as described in [32]. This has the added benefit of increasing the value of the embedded interplane decoupling capacitance.

Figure 17 shows my interpretation of a PCB stack-up for which I am indebted to Lode Decapmaker from Belgium. Based on the stack-ups in Figure 15, it uses a proprietary embedded capacitance between 0V and Power plane layers, with 0V planes on the outside of the board to provide shielding.

Figure 17 : Lode Decapmaker’s suggested stack-up

The power pins use standard microvias to perforate the outer 0V planes and reach the Power planes on the next layers in, and the signal pins use skip or stacked vias to reach the signal routing layers on the next layer in from the power planes.

Decoupling capacitors (other than ‘bulk’) are not required because of the use of high-capacitance-per-unit-area embedded dielectric between the 0V and Power planes on each side, for example 3M’s ECM type C4003, which provides 6.2 nF/square cm (40.01 nF/square inch), which is only 3µm (0.00012 inch) thick.

Lode makes the very reasonable suggestion that using such a very high capacitance embedded dielectric would reduce the SI and EMC problems caused by traces crossing power plane splits – reducing the need for an additional 0V plane layer as discussed above, or for additional components in the return path (see Chapter 4.4 of [2]).

For clarity, I have exaggerated the sizes of the microvias (they would generally be between 0.5 to 0.25 the drawn size, where the BGA drawn has a 0.5mm to 1mm ball pitch), and I have also exaggerated the thickness of the embedded high-capacitance dielectric.

To correspond with the figures in [23] that I have copied above, I have drawn Lode’s suggestion showing FR4 laminations with THP vias for signal routing, inside the outer pairs of microvia layers (although Lode never suggested this).

However, as discussed earlier, delamination is a worry when using such mixed dielectrics, and when the number of THP layers exceeds 10 or so, cost may be saved by realising them using microvia layers instead. Better EMC performance may be achieved at lower cost by using microvia layers throughout, with even smaller numbers of layers.

HDI PCB design issuesHDI requires a different approach to PCB layout, and some PCB design techniques may not always be able to be used. There are now many PCB manufacturers using HDI. In May 2000 the IPC found 62 HDI manufacturers worldwide, and in May 2008 there were 32 HDI manufacturers in the UK alone.

Manufacturing techniques can vary between different HDI board suppliers, and may need different PCB layout techniques. So we should always ask an HDI PCB supplier we are thinking of using what PCB design rules and restrictions they require to be followed, to see if they would cause us any difficulties for SI, PI or EMC.

Issues with Quality ControlAs for most things in engineering, if we want low warranty costs as well as low BOM costs, we have to be cleverer – in this case by conducting highly-accelerated life tests on proposed mixed-dielectric stack-up(s) from proposed supplier(s), to see which ones will be


reliable enough.

There is a potential problem with this scenario, though, because suppliers know that when they are asked for samples, those samples will be tested thoroughly, but that production volumes are often not tested at all when received by a manufacturer.

So, even when a mixed-dielectric stack-up passes tests that simulate its worst-case environment(s) for the required number of years, it does not mean that production quantities would – and so appropriate batch-checking at Goods In is required. There are standard rules for how many random samples must be tested to achieve the specified Acceptable Quality Level (AQL), for example [33].

Samples should be taken from the batch of received bare boards and subjected to highly-accelerated life testing followed by peel strength tests. Only batches that pass these tests are permitted to be used in production.

Of course, if suppliers know that this is what you will be doing, they will think twice before trying to slip low-quality boards into your manufacturing process, but you still need to do the batch testing anyway.

Similar QA issues arise over controlled-impedance traces; see Chapter 6.4.4 in [2] and [34].

References

[1] “Advanced PCB design and layout for EMC, Part 7 – Routing and layer stacking, including microvia technology”, in Issue 54 of the EMC Journal, September 2005, pp 28-40, available from: www.compliance-club.com/Archive.aspx

[2] “EMC for Printed Circuit Boards, Basic and Advanced design and layout techniques”, Edition 2, Nutwood UK December 2010, ISBN 978-0-9555118-5-1, full colour graphics throughout, available only from www.emcacademy.org/books.asp.

This book is not available via Amazon or other distributors, who might indicate that it is out of print when in fact it is not because it is printed on demand.

This 2nd Edition is identical to the 1st Edition except for size and format – so if you already have the 1st Edition there’s no need to buy the 2nd!

[3] IPC-2315, “Design Guide for High Density Interconnects & Microvias”, and a number of other standards and guides on HDI, Build-up, and Microvia PCB technology, including IPC-2226, IPC-4104, IPC-6016 and IPC-9151, can be purchased on-line from IPC at www.ipc.org

[4] “Multi-Chip Module”. http://en.wikipedia.org/wiki/Multi-chip_module

[5] “Package on Package”, http://en.wikipedia.org/wiki/Package_on_package

[6] “HDI’s Beneficial Influence on High-Frequency Signal Integrity, Parts 1 and 2”, Happy Holden, available via Mentor Graphics’ index at www.mentor.com/techpapers/fulfillment (search for HDI), or direct from: www.mentor.com/products/pcb-system-design/techpubs/hdi-s-beneficial-influence-on-high-frequency-signal-integrity-part-1-7404, and www.mentor.com/products/pcb-system-design/techpubs/hdi-s-beneficial-influence-on-high-frequency-signal-integrity-part-2-7405

There are many other useful papers on HDI available from this Mentor Graphics index, if it is searched for ‘HDI’.

[7] “Signal Integrity and HDI Substrates”, Dr Eric Bogatin, The Board Authority, Vol 1(2), A supplement to Circuitree Magazine,

June 1999, page 22, available via the list of articles and technical papers at www.bethesignal.com, or direct from www.bethesignal.net/bogatin/bts022-signal-integrity-p-167.html?cPath=23

[8] “Cost-Effective Use of Microvias”, Charles Capers, Printed Circuit Design, March 2003, Vol. 20 Issue 3, p14, http://connection.ebscohost.com/c/articles/9223337/cost-effective-use-microvias

[9] “Microvias and RF – ready for 10GHz?”, Ron Neale, Editor, Electronic Engineering, August 2000 pp 59-62, https://getinfo.de/app/Microvias-and-RF-ready-for-10GHz/id/BLSE%3ARN082408538

[10] “The via squeeze”, Charles L Lassen and Mark V Christensen, IEEE Spectrum, Volume 36, Issue 10, October 1999 pp 36, 38-41, ieeexplore.ieee.org/iel5/6/17247/00795606.pdf‎

[11] “Deep Microvias in Next Generation System Design”, Leigh Eichel, International Cadence Users Group Conference 2003, www.amphenol-tcs.com/doc?id=36‎ (may also be available via: www.teradyne.com/prods/tcs/resource_center/whitepapers.html)

[12] “How To Get Started in HDI With Microvias”, Happy Holden, Mentor Graphics Technical Paper Series, www.mentor.com/pcb/resources/overview/how-to-get-started-in-hdi-with-microvias-6c012699-5d73-4596-aeec-0ce7de663a3d

[13] “Printed Circuits Handbook”, by Clyde Coombs, Edition 6, October 2007, ISBN: 9780071467346, https://www.mcgraw-hill.co.uk/html/0071467343.html

[14] “The HDI Handbook”, by Happy Holden, John Andresakis, Eric Bogatin, et al, www.hdihandbook.com

[15] “Low Inductance Chip Capacitors, InterDigitated Capacitor (IDC)” www.avx.com/docs/Catalogs/liidc.pdf, and “IDC Low Inductance Capacitors (RoHS), 0306/0612/0508 IDC (InterDigitated Capacitors)” from www.avx.com/docs/catalogs/w2lw3l.pdf

[16] “8-Terminal Type Low ESL Capacitor Ideal for Power Supply Decoupling of High-speed Operation IC”, from http://psearch.murata.com/capacitor/lineup/lla/ and http://psearch.murata.com/capacitor/lineup/llm/

[17] “Solutions Beyond Limits” Viasystems Group, Inc., IPC Northwest Design Council , July 26, 2012, http://dcchapters.ipc.org/assets/pnw/presentations/20120726_microbga.pdf

[18] “Power Integrity Effects of High Density Interconnect (HDI)”, by Happy Holden, available via Mentor Graphics’ index at www.mentor.com/techpapers/fulfillment (search for HDI) or direct from: www.mentor.com/pcb/resources/overview/power-integrity-effects-of-high-density-interconnect-hdi--a4c6125f-12b7-4a4d-9d51-1323cc4a8552

[19] “PCB Layout Recommendations for BGA Packages”, Lattice Semiconductor Technical Note TN1074, September 2013, www.latticesemi.com/~/media/Documents/ApplicationNotes/PT/PCBLayoutRecommendationsforBGAPackages.pdf?document_id=671

[20] “BOM cost, and profitability”, Keith Armstrong, in the EMC Journal, Issue 82, May 2009, pp 32-34, www.compliance-club.com/Archive.aspx

[21] “Microvias in Printed Circuit Design”, Kevin Arledge and Tom Swirbel of Motorola Land Mobile Products Sector, www.eetasia.com/ARTICLES/2000FEB/2000FEB24_SMT_ICP_AN.PDF?SOURCES=DOWNLOAD

[22] “HDI and Advanced HDI”, Viasystems Group, Inc., www.viasystems.com/technology/hdi.html


[23] “HDI Layer Stackups for Large Dense PCBs”, by Happy Holden and Charles Pfeil, Mentor Graphics, July 2007, http://communities.mentor.com/mgcx/servlet/JiveServlet/downloadBody/1128-102-1-1183/hdi%20_layer_stackups_for_large_dense_pcbs.pdf

[24] “HDI Printed Circuit Boards”, NCAB Group, www.ncabgroup.com/wp-content/uploads/2012/01/hdi_presentation_110913.pdf

[25] “HDI”, by Kenneth Jonsson and Bo Andersson, NCAB Group, www.dnu.no/arkiv3/HDI%20-IPC%20presentation_Norge%20090924-2.pdf

[26] “PCBs With Embedded Components Emerge for Capacitors”, introduces the GRU series of embedded capacitors, www.murata.com/products/article/pdf/ta10d1.pdf

[27] “Plated Additive Resistor Technology”, describes MacDermid’s ‘M-Pass’ resistive layer technology, http://aept.ncms.org/presentations/09%20MacD%2001’30’2003.pdf

[28] “Shocking Rules and Material Remove ESD Risk in Allegro PCB Smartphone Designs”, by Team Allegro on June 27, 2012, www.cadence.com/Community/blogs/pcb/archive/2012/06/27/shocking-rules-remove-esd-risk-in-allegro-pcb-smartphone-designs.aspx

[29] “Buried Capacitance® Technology”, www.sanmina.com/pdf/solutions/pcbres/buried_capacitance_technical_0106.pdf, and www.sanmina.com/pdf/solutions/bc.pdf

[30] “Reduce PCB Impedance, Noise, and EMI and Simplify PCB Layout”, describes 3M™ Embedded Capacitance Material (ECM), h t t p : / / s o l u t i o n s . 3 m . c o . u k / w p s / p o r t a l / 3 M / e n _ G B /EmbeddedCapacitanceMaterial/Home/

[31] “Faradflex” from Oak-Mitsui, www.faradflex.com and www.oakmitsui.com/pages/advancedtechnology/faradflex.asp

[32] “Suppressing ICS Which Have BGA Packages and/or Multiple Power Rails”, Keith Armstrong, EMC Journal Issue 107, July 2013, pp 24-28, available from www.compliance-club.com/Archive.aspx

[33] “What is the “AQL”, and When is it Applicable?” in “Quality Inspection Tips – Practical advice for importers sourcing in Asia”, by Renaud Anjoran, 28 November 2011, www.qualityinspection.org/what-is-the-aql/

[34] “Printed Circuit Board (PCB) Test Methodology”, Intel Corporation, Revision 1.6 January 2000, download.intel.com/design/chipsets/applnots/29817901.pdf

World class experience in shielded chamber and absorber design, manufacture, installation, testing and maintenance

• 3m compact, 3m, 5m & 10m solutions

• Antenna measurement chambers

• Emissions & immunity

• High-performance shielded rooms for EMC, Tempest, EMPP and MRI

• Masts, turntables & controllers

• Mode stir

• High-performance absorbing materials

Powered by and

Tel: +44 (0) 1942 [email protected]

Absorbers & Chambers Made to Last.

Your One Stop Shop for Anechoic Chambers and Absorbing Materials


A Passive Culprit?By Richard Marshall, MA, CEng., FIEE, FIET, FInstP, Richard Marshall Limited

In the 1970s I drove a VW Passat estate with a medium wave radio. On the way home from work I always listened to the news on this radio and noticed that when I drove into my car port a loud buzzing noise became audible over the last few yards. Why could this be? The car port was made of wood and asbestos with negligible shielding effect. The only nearby electrics were an incandescent security light with low voltage remote switching.

I resolved to fix this, and the resulting investigation proved most instructive, as is set out below.

Fig. 1 30 Ox Lane, Harpenden

Figure 1 shows the front view of my house as it is today. The trees were rather smaller in the 70s but otherwise things were much the same. The Passat was normally parked in the carport at the extreme left, and since this was open to the road I took the security precaution of providing lighting that is controlled both by a local passive infrared sensor and also by a remote

switch by the front door of the house. Figure 1 shows that the remote switch uses DC borrowed from the secondary of the front door bell transformer circuit via a bridge rectifier to send a momentary 15 Vdc signal via a red and black pair of wires across the front of the house to operate a timer relay in the carport. The mains supply to the Car Port Lamp and its PIR sensor comes from the rear of the building. The radio signal is strong: It comes from the Brookmans Park transmitter only 11 miles distant. The effect was equally present when the lamp was on or off but disappeared when the front door bell transformer was switched off. How could this set-up possibly interfere with my car radio?

Fig. 2 The problem circuit

The circuit in Figure 2 is passive – and really very simple. The preliminary tests had shown that the electronics in the PIR sensor and the timer relay were not contributing to the problem. The relay box and the lamp in the carport were however only a couple of metres from the car radio aerial, and it did seem that coupling between the aerial and the red/black cable to the house was involved and the buzz appeared to be 100Hz or thereabouts.

Only after some days of musing about this whilst driving home from work did I realise that the diodes of the bridge rectifier could act as switches and that when the instantaneously-applied AC was less than about 1.5 volts the low voltage red to black cable would be isolated from the house ground connection, whilst for the rest of the mains cycle there would be an effective radio-frequency ground connection. Accordingly, the problem circuit might be acting in common mode as an extension antenna, capacity coupled to the car radio antenna and with its remote end disconnected from ground when the AC supply voltage crossed zero – that is 100 times per second. Was that feasible? How could it be fixed? Think about it before you turn the page.


It seemed feasible to me so I added a capacitor to bypass the rectifier bridge at high frequencies, as in Figure 3.

Fig. 3 The problem solved

This was completely effective. Whatever ‘extension antenna’ the wiring added was now remotely grounded all the time, and so not amplitude-modulating RF signals.

I then realised that the 1930s radios that I dismantled as a teenager commonly had a capacitor of about 10 nF between their chassis and one wire of the mains supply ‘to avoid modulation hum’. (For user safety, that ought to have been a 4.7nF class Y capacitor, but such items were not then available!). In those cases the culprit switching function must have been provided by the valve rectifier, the path to earth being via the inter-winding capacitance of the radio’s mains transformer. Modulation hum is still of importance to vintage wireless enthusiasts [visit: www.vintage-radio.net/forum/showthread.php?t=2935], but otherwise the term seems to have fallen out of use. Nowadays we would use the term ‘intermodulation’ and reference to Wikipedia will provide a host of twenty-first century examples.The essential characteristics of modulation hum are that it results from a non-linearity causing intermodulation between the mains and an RF signal, and so appears when a signal is tuned in on a receiver that detects amplitude modulation. Off-tune, the receiver is quiet.

What I had inadvertently created – and then destroyed by adding the capacitor – was an antenna that received an innocent radio transmission, modulated it with harmonics of the 50Hz mains, and re-radiated it as an interference source. An ‘Interfering Transponder’.

I could have solved the problem by buying a new car with an FM receiver, but fitting a 10 nF capacitor was a cheaper solution. Good EMC design is always cost-effective.

But that was 40 years ago: Why is this relevant today?

My experience was with a low-voltage rectifier. Mains voltage rectifiers are commonplace. I have recently had easy access to a total of five wall warts or in-line power supplies. They are all

designed for 2-wire ac inputs with a bridge rectifier as the first component at the mains input interface. One was an electronic transformer intended for an electrician to fit in conjunction with an LED lighting installation: (the use of that component as a part of a kitchen downlighter installation seems certain to cause trouble to a nearby Medium-wave portable radio). The others were all plug-in consumer items. All 5 were intended for use in applications where there existed an output cable that the bridge rectifier could RF-connect to the supply at twice the mains frequency, so they are all potential ‘Interference Transponders’.

From the standards viewpoint, to define test methods and limits for interference transponders would require a simultaneous test of immunity and emission. There does not appear to be sufficient evidence to justify this, but certainly the phenomenon discussed here is one that should be remembered when troubleshooting EMC problems. It probably explains many instances of ‘mains hum’ that presently go undiagnosed. It would be nice to build an up-to-date data base.

Richard Marshall Ltd was founded in 1996 to undertake RF and EMC-related electronic design, consultancy and training.Richard can be contacted by:Email: richard.marshall @ iee.org / Tel: 01582 460815Web: www.design-emc.co.uk© Richard Marshall 2014

www.kemtron.co.uk+44 (0) 1376 348115 · [email protected]

RFI /EMI shielding gaskets& components

Kemtron-EMC 128x88.qxd:Layout 1 2/8/13 14:10 Page 1


FCC Proposed Rule ChangesBy Michael Derby – Senior regulatory review engineer and Director of ACB Europe.

I suspect that very few of you have ever sat down with a warm cup of cocoa in a comfortable chair to read, cover to cover, the latest edition of the FCC Rules. Even chapters 2, 15, 22 and 24 which are probably the most familiar to you, remain unread in their entirety. It’s fair to say that the FCC rules are not an easy read and not a compelling page turner, but it’s an important read and one that many of us need for our work. Myself, I read at least one or two sections of the FCC rules, multiple times per day. Like you, I have never read them all.

The FCC rules are not like an ETSI or CENELEC standard, which may involve a small team of engineers constantly amending and working on the next issue. The FCC rules are not easy to change and changes are a significant event. Significant changes are rare indeed.

In 2013, the FCC issued several proposals for changes to their rules. For the purpose of this article, I am going to focus on only one of these proposals because I think it is the most relevant to us. Also, this article attempts to condense a long FCC document into a short space, so some details will be abbreviated.

In February 2013, the FCC issued a Notice of Proposed Rule Making (NPRM). Document FCC 13-19, ET Docket No. 13-44. It means that they announced a proposed change to their rules. Our industry was given a comment period which has since expired and the FCC are now reviewing the comments.

Here is a link to the document:

h t t p : / / t r a n s i t i o n . f c c . g o v / D a i l y _ R e l e a s e s / D a i l y _Business/2013/db0215/FCC-13-19A1.pdf

On this occasion, the changes are significant to TCBs, Test Labs and Manufacturers. The document proposes amendments to FCC Parts 0, 1, 2 and 15, with small changes to Part 68. For most of us, Parts 2 and 15 are the most applicable sections that we use on a daily basis.

The main topics covered by the proposal are:

The TCB Program; including the certification process, post market surveillance, assessing TCB performance and TCB accreditation.

Test Laboratories; including test lab accreditation, accreditation bodies and test site validation.

Measurement procedures; including the main testing standard for Part 15 devices and updates to the listing of measurement procedures.

Let’s start with changes to the TCB program and how that will affect us all. Historically, certifications of radio equipment for the USA were done directly with the FCC. Since the creation

of the TCB program over 10 years ago, it has been possible to obtain a certification from the FCC, through a TCB. In recent years, 98% of products are FCC certified through a TCB whilst only 2% of devices are certified directly with the FCC.

The TCB is able to certify most devices but there is a list called the “TCB Exclusion” list, found in FCC KDB 628591. TCBs are not permitted to certify devices which appear on this list, or using test methods on this list. The small number of devices and test types shown on this list is actually what represents the 2% of devices still certified directly with the FCC.

There is also a “Permit But Ask” (PBA) list, found in FCC KDB 388624. This is a list of devices that the TCB is permitted to certify but only whilst holding the FCC’s hand through the process. Basically, the FCC has a look at the application before the TCB is permitted to complete the certification.

The first proposal in the NPRM is that the FCC will no longer accept applications submitted directly to them and therefore the TCBs will process all FCC certifications. Of course, this means that there will no longer be any TCB Exclusion list. The proposal is to create a whole new process called the “Pre-Approval Guidance Procedure”, which is actually very similar to the present “Permit But Ask” (PBA) process. The plan would be to take all of the items from the present TCB Exclusion list and the present PBA list and put them all together on the Pre-Approval Guidance Procedure. As the name suggests, it would be a process for the TCB to complete the certification whilst holding the FCC’s hand. In reality, it would appear to be a slightly modified version of the present PBA process.

Some of you will have found that the FCC’s own lab sometimes requests a sample for testing before certification, such as for a 5 GHz WLAN which requires a DFS test. This is known as “pre-Grant sampling” and it could still be part of the process under the Pre-Approval Guidance Procedure.

The next proposal is a small one but could have interesting implications. There is a proposal that each application shall include a signed declaration from the manufacturer that the application is accurate. This may sound obvious and an unnecessary step but it could suddenly become important in cases where a test lab or agent has falsified documentation within the application. By providing such a letter, the manufacturer may be inclined to closer examine the documentation his test lab or agent is submitting to the TCB. It could also give the FCC someone to blame in the event of data falsification.

The next issue to affect the TCB is the dismissal of a Grant. You may understand that when a TCB completes an FCC certification, the application is not automatically checked by the FCC. The TCB completes the certification and the device is Granted. (This is one of the many differences between the


FCC and Industry Canada). A dismissal occurs when a problem is later discovered with a certification and the Grant must be removed. Also, there are many cases where a manufacturer may choose to dismiss the certification of one of their own products. Presently, only the FCC can dismiss an FCC Grant. Under the new proposal, the TCB would be permitted to dismiss a Grant that they issued. As you can imagine, the TCB would not be permitted to dismiss a certification completed by a different TCB. The ‘Grantee’ or manufacturer can appeal to the FCC if they feel the TCB was wrong to dismiss the Grant and the FCC would have the power to overturn the dismissal. Similarly, the FCC would still have authority to dismiss certifications completed by any TCB.

The TCB would not have the authority to deny an FCC ID or block a Grantee Code.

Next we move on to market surveillance but stay within the TCB program. A TCB must perform audit surveillance assessments each year on 5% of the devices they certify. Presently the role of auditing is officially the responsibility of the FCC and the TCB is instructed from within a KDB. The NPRM proposes that the audit surveillance requirements will become officially a TCB duty, codified within the FCC rules. The FCC could still request audit samples directly from the manufacturer and this would require communication between the FCC and TCB to ensure we do not all perform multiple audits on the same device. Note that if a sample is not sent for audit, punishments are applied and this is a common cause for the manufacturer to dismiss their own certification; if they have been asked to send a sample but they never actually sold or marketed the device and they do not wish to build one just for auditing. There may also be a change to the surveillance reporting timescales, which presently states that the TCB must send a report to the FCC once per year.

The next issue affecting the TCB is the assessment of their ability and performance. I am often asked why a TCB can make mistakes but not be punished by the FCC. The reality is that the FCC rules do not allow for the FCC to punish the TCB, and the FCC does like to follow their own rules! Presently, the only power the FCC has over a TCB is to withdraw the designation of the TCB, which is a very severe punishment and could easily lead to loss of employment for many people. The NPRM proposes a scale of punishments to befit the crime and the intention would be to fix the problems rather than put a company out of business. Suggested punishments include, for example, the FCC’s ability to ‘set aside’ applications of a problem type from the TCB, or force the TCB to use the Pre-Approval Guidance Procedure on certain applications until the TCB can prove they have fixed their problems. Similarly, the TCB could retain their designation but not be permitted to certify certain types of device.

There are some text changes proposed in the rules, to meet the latest accreditation documents for TCBs, such as ISO 17025, 17065 and 17011.

Finally, the issue of TCB designation or recognition is discussed. Presently the process for becoming a TCB is different for TCBs within the USA, to those outside of the USA. There is a proposal to change this, such that all TCBs are recognised in the same way and all are recognised directly

by the FCC.

Now let’s move on to changes which affect the test lab. The first one is associated with the test standard that is probably used more than any other when testing for the FCC; ANSI C63.4. Presently, the FCC rules state that only ANSI C63.4-2003 is permitted for use. Of course, we see from the FCC’s website that the 2009 version of the standard is also acceptable but it’s not in the rules. The proposal then is to insert ANSI C63.4-2009 into Part 15 of the FCC rules and remove the 2003 version. Please ensure you read the 2009 version because there are differences! As an additional comment, ANSI C63.4-2009 states that for measurements above 1 GHz, a test site compliant with CISPR 16-1-4: 2007 is acceptable, or the use of a square of RF absorber is acceptable. Note that the FCC has indicated that they will only allow the CISPR-1-4 compliant sites, not the use of a square of absorber. In addition to the above, ANSI C63.10 would be added to the FCC rules and this is the standard to use for testing Part 15 transmitters.

With the above point in mind, there is a proposal that the FCC should be permitted to update the measurement procedures listed within their own rules more easily, without the need to go through the process of another NPRM each time a standard is updated.

The next issue to affect the test labs is accreditation. Presently, you do not need an accredited test lab to test any device for the FCC which requires Verification or Certification. Only DoC testing requires a test lab accreditation. (Note that the accreditation required for DoC testing must be recognised and listed with the FCC, not just with the lab’s accreditation body). The proposal in the NPRM is that 17025 accreditation (and recognition by the FCC) would become mandatory for Certification testing also. This should not be too much of a hurdle for any established test facility in Europe or the USA but it could spell problems for small, unaccredited, testing companies or manufacturers who perform their own measurements. It also causes problems for any lab located in a region which does not have an MRA with the USA, such as China, since they cannot get their accreditation recognised by the FCC. The proposal in the case of non-MRA area labs is that the lab could resolve the problem by getting accreditation with an accreditation body that is acceptable to the FCC. The FCC will then codify the criteria for determining the acceptability of an accreditation body for accreditation of test labs.

It should be noted that presently there is an option to list a test lab with the FCC to Part 2.948. Test site listing is also covered within the scope of recognised accreditation, so the FCC proposes to remove the test site listing option for Certification testing. The requirements of Part 2.948 should remain within the FCC rules to assist accreditation bodies in evaluating test sites and to demonstrate suitability of a test site for Verification testing.

In addition to all of the above, there would be some general housekeeping and clearing up of outdated references within FCC Part 15.

Finally, I mentioned that there would be some minor changes to Part 68. These changes are typically to clear up references to outdated standards or documents.


So, what happens next? Well, with the comment period closing in the summer of 2013, we sit and wait for the FCC’s next move. But sadly it’s not that simple. The last time I spoke with the FCC on this subject was October 2013 and the feedback was this: Instead of announcing their new intentions or implementing these rule changes now, the FCC plans to issue a new NPRM in 2014, with similarly significant changes proposed. There will be a comment period, naturally. Following the comment period of this new NPRM, the FCC will review all the comments gathered from both NPRMs and then announce some the rule changes which they predict will bring significant changes to the entire authorisation system! Interesting times ahead.

Thank you for your time and please feel free to contact me for further information: E: [email protected] Website: www.acbcert.com

Post script:

OK, this has nothing to do with the NPRM but I feel it is useful information. Let’s call it a “Michael’s Top Tip”. It is relevant here because of the discussions above regarding the DoC authorisation route, compared to the Verification or Certification authorisation routes.

Many of you will be familiar with this symbol:

There is a popular misconception that this symbol simply means “I’ve done some FCC stuff”, but that really is not the

case. This is the FCC’s Declaration of Conformity Logo. So, if the device is authorised through the DoC route, then the logo must appear on the label. Note that the FCC rules dictate which authorisation route applies to each type of device. If the device does not have a valid DoC associated with it, this logo must not appear on the device. By placing this logo onto the label of a device, the manufacturer is stating that the DoC authorisation route applies to their device and the full DoC process has been correctly applied. That’s a little more than just “doing some FCC stuff”. ☺

ACB is a global approvals company, specialising in radio equipment certifications and authorisations. ACB has offices worldwide, including the USA, UK, Netherlands, Finland, China and Taiwan. ACB has been a TCB since the beginning of the FCC’s TCB program and operates independent certification services for test labs and manufacturers.

ACB Europe is led by Michael Derby as the senior regulatory

review engineer and Director of ACB Europe. Michael has been with ACB since November 2007. Michael is a TCB for the FCC, an FCB for Industry Canada and a Notified Body for the R&TTE and EMC Directives. Michael is an active member of the R&TTECA and the TCB Council. He is the liaison between the two organizations and is presently the Chairman of the TCB Council Board of Directors. Michael is also responsible for ACB’s training activities, including webinars, workshop seminars and bespoke training courses. Michael’s past experience includes product development, testing and certification for test laboratories and manufacturers. Michael provides worldwide services to ACB’s global customers from his office in the Hampshire area of the UK.

Advertisers IndexListed below are the Advertisers in the current issue showing the page number where the company’s advertisement appears, together with their web address and email

A R Europe page 4www.ar-europe.ie [email protected]

CST page 17www.cst.com [email protected]

Kemtron page 28www.kemtron.co.uk [email protected]

Laplace Instruments Ltd page 7www.laplace.co.uk [email protected]

Microlease Back coverwww.microlease.com

Rainford page 26www.mvg-emc.com [email protected]

Rittal page 5www.rittal.co.uk [email protected]

Rohde&Schwarz IFCwww.rohde-schwarz.com [email protected]

Syfer page 16www.syfer.com [email protected]

Telonic page 11www.telonic.co.uk [email protected]

Teseq page 15www.teseq.com [email protected]

UL page 16www.ul-ts.com [email protected]

Yokogawa page 15www.yokogawa.com [email protected]

Book your session by scanning this QR Code

Issue 110 February 2014 - EMC Information Centre · 2014-04-01 · Conference. This year the...

Documents

Transcript of Issue 110 February 2014 - EMC Information Centre · 2014-04-01 · Conference. This year the...