Return to List

Ghosts 

   From time to time, ghosts come back to haunt me, arising from articles I have written in the past.  For example, the Editor recently forwarded to me an interesting query about “The MOS controlled thyristor”, an article which appeared in these pages in the September 1993 issue.  Other such ghosts have surfaced from even further back; my first contribution to these pages - a Design Idea entitled “Two for the price of one” - was back in about 1970, a useful circuit although I no longer have a copy of it as published. It’s not just from articles either. Occasional queries arrive, triggered by people reading my books, particularly the one titled “Oscilloscopes”†.  This was first published in 1981, is now in its fifth edition, and has never been out of print.

   These “ghosts” are always interesting, and originate from readers just about anywhere, France, India, Australia, the USA among other places, all come to mind. They often set me thinking even further back over a career in electronics spanning half a century, to days - for example - spent servicing AI10 units.  Airborne Interceptor Mk10 was an airborne radar which saw service during the Second World War, still in service with the RAF’s Gloster Meteor NF11 night fighters in the mid fifties, although by then hopelessly out of date.  The American “doorknob pentodes” in the IF strip were unobtainable by then, explaining the pilots’ routine complaints of low sensitivity.

   A couple of years later found me at The Central Research Laboratories of The G.E.C., in the days when that company’s motto was “Everything Electrical”.  The labs were much later renamed “The Hirst Research Centre”, and have now disappeared off the face of the earth, to be replaced by houses, shops and what-have-you.  It was, in its heyday, an extraordinary institution, where Ph.D.s were two a penny, and a carefree university-student atmosphere pervaded the buildings. There, at one stage, I had access to a diffusion furnace, and stayed behind one evening to make myself a silicon rectifier.  This was for a mains power supply for my home-made battery portable radio, using DK96 series valves with their 25mA filaments.  They were happy days, when petrol cost four shillings (20p) a gallon, and - following WWII - things could only get better.  Some of the extraordinary activities and characters at the CRL of the GEC in those days would doubtless be of interest to, and probably even strain the credulity of, a later generation, but space dictates that any such reminiscences must await another occasion.

ISBN 0 7506 4757 4   Newnes/Butterworth-Heinemann                                Ian Hickman

Published in the April 2004 issue of Electronics World, this leader prompted the following letter from a reader in Bristol,  published the following month.

  Ghost in the loft

With regard to Ian Hickman's leader article, I've got another ghost for him. Until very recently I was using the Practical Wireless 'Purbeck' oscilloscope that he designed in 1978. I built it when I was 16, using the Watford Electronics kit, and apart from one or two minor repairs it has been working well for the past 25 years. Unfortunately the custom mains transformer went open-circuit in the high voltage secondary, and I have decided to buy a commercial scope with much better performance. I don't think I can bear to bin it though - it will probably sit in the loft for the next 25 years!

 

Who knows your whereabouts? 

  In Edinburgh, you can buy time on a parking meter with your mobile phone.  You call the number displayed on the meter (with caller I.D. enabled), and a voice response system asks for the meter’s I.D. number. This enables the parking system server to identify your location and activate the meter. The latter now lets you choose how long you want to pay for and prints out a ticket for you to place on view in your car. It also instructs the server to charge that amount to your credit card, or to a special account.     

  Obviously, in the process, the system knows just where you are.  But this is not the only, or even the earliest system for locating you.  Someone calling the U.K. emergency number 999 from a fixed line may be unable, or not in a fit state, to give his or her exact location.  But the origin of the call can be traced via the exchange(s) involved, and the person thus located. In the United States, E911 (the “Enhanced 911 mandate”, passed 1996, revised 1999) requires all cell-phone operators to install facilities, able to locate a mobile caller dialling the 911 emergency number, to within 50 to 100 “meters”, by 2005 (is the U.S. going metric at long last?). 

   In Europe, wireless network operators are already required by E112 to be able to locate a caller making a call to the emergency number 112.  However, there is currently no accuracy specification, and most wireless network operators will simply return the location of the cell via which the call was set up - leaving a “fix” which could cover hundreds or even thousands of metres.  The GSM system is now spreading in the U.S., and mobile network operators there use uplink time difference of arrival.  This depends upon the mobile being received via at least three base-stations, and the system may presumably have to instruct the mobile temporarily to transmit at higher power, to reach enough base-stations.  The dominant U.S. mobile technology, like GSM, also uses TDMA, and uplink time difference of arrival technology is appropriate there also.  CDMA (code division multiple access) is a different problem as uplink time difference of arrival is not appropriate, and many CDMA mobiles have a GPS function built in. This returns good position information if the user is outdoors, but less accurate if indoors or in a heavily built up area.  These systems are designed to identify a caller’s location in an emergency, but in principle could be used by the police continuously to track any suspect, or by national security services for the same purpose, building up a record of an individual’s movements up to the present time.

   There are also other ways in which one’s location, either current or at some time in the past, could be determined.  RFID (radio frequency identification) tags are set to become ubiquitous.  The Gillette Company of  Boston, USA, proposes to buy up to five hundred million or so tags, to mark its razors and packs of blades.  These tags are already incorporated in product being sold there, in U.K. and Germany, and the resulting improved inventory management is expected to save billions annually.  The tags, read by scanners, will provide records giving details of the time of the sale and outlet, but will remain in the product and be accessible in principle thereafter.  Similar tags could appear hidden in the hem of clothes, the binding of books, car tyres and almost any merchandise you can think of.  The tags cost tens of pennies today, this will be just a few pence each before long.  Almost any purchased item will be able to identify the whereabouts of the purchaser either in the past or in some cases, currently, creating - or breaking - an alibi.  Even the humble credit card will leave a trace whenever used, creating a record of what and when you bought what where.  Japan is well ahead of the game, and before long there you will be able to opt for personalised targetted advertising.  Knowing where you are, the system could ring your mobile to alert you that the shop you are approaching sells your favourite brand of chewing gum.

   Some of these means of telling where you are, are obviously beneficial, even potentially life-saving.  But others can be expected to raise anxious representations from civil liberty groups.  For more details on this story, see the July 2003 issue of “Spectrum”, the journal of the IEEE. 

                                                                                                           Ian Hickman        

Published in the September 2003 issue of Electronics World

 

A Late Developer 

   RFID is surely a technology that is about to explode and proliferate into every sphere of modern life.  I have been keeping a weather eye on this topic for about fifteen years, since involved, while with Plessey/Siemens, in a study into the feasibility of using Radio Frequency IDentification tags to keep track of children in theme parks.  A practical problem - several get lost every day and one or two a year get lost permanently - it is a difficult application for the technology, due to the range required.  In retailing, a tag is only required to operate over a short range, a few metres at most. 

   One of the problems holding back widescale application of RFID tags has been that of standardisation.  It would be a great disadvantage if Europe, the US and Asia went their different ways, with mutually incompatible systems.  Fortunately international co-operation is on its way, driven by the needs of industry and commerce.  For keeping track of pallets for the transfer of goods, EPCglobal Inc. is developing the proposed the EPCglobalG2 standard.  This uses a 96 bit format, and tags using this format are already available, with read and write operating ranges up to a few metres, while operation at UHF permits multiple tags to be read/written per second without collisions.  

   I don’t have details to hand of the allocation of the 96 bits to headers, serial numbers etc. but 2^96 is a seriously large number, about  7.922816244 ´10^ 28!  Major concerns, both commercial and governmental, are mandating the use of RFID technology, with Walmart, Tesco, US D.O.D and others requiring its use in their supply chain and logistics systems as early as 2005.  And in the EU, new regulations will require the use of tags to keep track of individual animals such as beef herds etc. with the system already proven by pilot trial schemes. Tags complying with ISO 11784/85 were placed subcutaneously in each animal, allowing it to be tracked at various points during its lifetime, and after. Knowing where the beef the housewife buys comes from, should boost her confidence - and sales - in the view of past catastrophes such as scrapie, BSE, foot and mouth disease, etc. 

   Another burgeoning RFID application is smart tickets, which can be loaded with credit like a phone card, and debited automatically, in accordance with the journeys undertaken by the holder.  Card readers on the various forms of transport could simplify matters by enabling a traveller to make a journey involving both buses and trains, without the need for a through-ticket or separate tickets for each stage.  Already trialled in Perth, Australia, it is shortly to be tried out on the Seattle mass transit system. 

   Other applications abound, and Jo(e) Public will have to get used to tags appearing everywhere, whether he or she likes it or not.  For instance, tags in clothes may cause the washing machine to inform you of the right programme to use for each, while tags on food packs may enable the smart fridge to warn you when you are running low on butter, or eggs, or milk - - -.  In time it will become as difficult to buy a “dumb” fridge or washing machine as it is now, for those with a “keep it simple - less to go wrong” philosophy, to buy a car without central locking. 

   Much more on RFID tag technology and applications can be found in the Philips publication “ON THE MOVE, Identification News”, Volume 6 Issue 3, September2004.  Philips’ World News, Volume 13 Number 3, September 2004 also covers the topic, with other useful and interesting articles such as Bluetooth/WLAN, intercommunication, hands-free car phones, class D amplifiers and electronic ink, among others. 

                                                                                                Ian Hickman

Published in the November 2004 issue of Electronics World

 

 Specialisation and availability

When I was an undergraduate, in the late 50s and early 60s, all engineering students (with the exception, if I remember rightly, of the aeronauticals and the chemicals) took the same course in the first year. So we all studied thermodynamics, strength of materials, dynamics of fluids, theory of structures and metallurgy, as well as electrical engineering. This broad syllabus was both interesting and stimulating, and some has stuck permanently. If ever I had to calculate the flow versus height over a given Vee-notch weir (can't remember off hand whether I need Francis' formula or d' Arcy's formula), I could still do so as the appropriate textbook is residing awaiting (improbable) use, in the technical section of my library. In the later years of the course, the content of the maths curriculum was identical to that of the Honours Maths syllabus, except that we made do with “it can be shown that...” rather than proving all those results in full mathematical rigour.

   The situation is very different nowadays, since the frontiers of knowledge have expanded so enormously. Such a broadly based foundation is considered an unaffordable luxury. It is now thought that, to be useful to an employer and hence secure a job, a new graduate must specialise - not in electrical, or some other branch, of engineering. Not even in light versus heavy electrical engineering, but in a smaller field still - telecommunications for example. Presumably the heavy electricals have also partitioned themselves into ever smaller specialisms, while various universities, some redbrick, some even newer, offer specialised degrees in a variety of arcane subjects, from the infamous 'media studies' to others as bizarre as African tribal art.

   So does the newly qualified telecomms (or whatever) engineer emerge with exactly the right toolkit of knowledge to fit into a research or development job like a round peg in a round hole? The answer of course is no. The syllabus is never really up to date: when I graduated in 1961, there were plenty of valves in the syllabus, but no semiconductors. I gather that syllabuses are not quite as far behind the times as that nowadays, but the technology moves on remorselessly and the new graduate, assuming he is lucky enough to find an employer, is likely to find himself faced with a fairly steep learning curve, despite the best efforts of universities. The inadequacies of the syllabuses are generally recognised, even by politicians, so they must be glaring indeed! For example, the UK government (or, to be more specific, the Department of Trade and Industry) sponsored over the last few years, the Radio Frequency Engineering Education Initiative, in an attempt to alleviate the dearth of competent RF engineers face by an industry busy with Bluetooth, IEEE S02.11a, b, c, and g, HomeRF, WLANs of all sorts, etc. Having been involved with the implementation of the Initiative at my nearest University, I know from first hand that current electronics courses are completely deficient in practical hands on work with real live components and circuitry - it is so much easier to sit a student down in front of a PC and tell him to work through the exercises with SPICE.

Not surprising, then, that employers are always looking for engineers already experienced in whatever happens to be their particular line of product development. Some firms, only the largest and then not always, are willing to take on the raw new graduate, and let him build up his experience base to the point where he is positively contributing to the firm's advancement. But a glance at the recruitment ads in the 'freebies' (controlled circulation magazines) shows that most firms seek to obtain their experience staff by recruiting from other firms, who then, naturally, feel forced to poach engineers in their turn.

The problem is, of course, the low esteem in which engineers are held; not by the general public - who cares about that - ­ but by employers. And let's be honest, esteem is a quality that can be accurately measured - in pounds sterling.

 

Ian Hickman

Published in the January 2003 issue of Electronics World

This page last updated - see homepage.

Return to List