…. The Government will have to learn the hard way that it is not a good idea to leave officials to dictate the law, without checking whether they know what they are talking about” – Sunday Telegraph, 13july03, p12
Graham Elvis, in his letter in EW May03, points to a key issue. “A simple example is that of the humble relay, operating once or twice per shift, its un-suppressed contacts can cause havoc among state-of-the-art gizmos but the EMC directive completely ignores it!”
Here we see the fatally flawed tradition of EMC, coming out of analog work, when applied to digital systems. In the case of EMC’s old stamping ground, radio interference, that relay would merely have upset one spoken word broadcast during a whole eight hour shift. Failure to migrate to digital electronics is an added flaw in already flawed EMC magic.
In the same issue of EW, “Name and address supplied” puts his finger on further key issues. “For each type of product there is a committee of the great and the good from government departments, interested companies and ‘EMC Magicians’”. Only big “interested companies” could afford the time, and would not object to a bureaucratic way to drive out their small competitors, in Europe and abroad.
I would like to give examples from the past. In 1962, Ferranti’s Head of the Drawing Office, who had never done any logic design and never would, was the man adjudged important enough to go on the committee which would devise the British Standard for logic diagrams. He was so important that he did not consult any of us, who were meanwhile busy actually doing logic design and drawing logic diagrams. The result was a ridiculous British Standard which meant that all our logic diagrams had to be redrawn three times larger, so increasing his empire of draughtsmen. This ridiculous British Standard for logic diagrams soon disappeared without trace. There are limits to British patriotism.
One of the minor issues at the core of the matter was whether a logic diagram should emphasise the circuit or the logic. De Morgan’s Theorem tells us that the same circuit can do an AND function or an OR function. The problem is, should the logic diagram tell us how the logic designer viewed the role of the particular circuit? The British Standard got it wrong, while the US MIL STD got it right by siding with the logic designer. Regardless of any other committees, US MIL STD 806B became the de facto world standard, including in Britain
In around 1980, Paul Borrill came to one of my private seminars on digital hardware design. Much later, he became head of the IEEE Standards Committee developing the new IEEE bus for interfacing a microprocessor with other circuits. He was anxious to get the bus circuit design right according to my theories on the matter. However, he found that nobody else on his Committee grasped the electronic principles involved. He urged me to become a member of his committee, in Denver Colorado. Why should I, if I had to pay my air fare from England to Denver periodically? (However, I did send a number of forceful letters of support for him to show to his committee.) On his own, he was in permanent battle with all the other members of a technically ignorant committee, who we can assume were all very important. Borrill was anxious to introduce a high speed serial port, and just about succeeded, but only as an alternative to the parallel port.
As the microprocessor world length increased from 4 bits to 8 bits to 16, these numskulls held onto the idea that an n bit word travels down n wires (ignoring the teleprinter precedent). So the packages got ever larger and more expensive, merely to accommodate the added pins. As a result of this committee and ignorance elsewhere, we now have these god-awful packages with multiple pins through each of which data flows at a very slow rate via multiple buses. This enormously increases the cost, size and unreliability of today’s PC. Only now, decades later, is the industry at last moving towards high speed serial linkage between units. (One advantage of serial connection is that failure is total, and so easy to find.) If you think I only mean data pins, then remember that both data (pixels) and control (frame sync.) come down one wire from your roof, and your TV occupies a tiny area on a chip to separating them out. I should mention that in Motorola in 1966, my standard pulse in the high speed logic that I designed was 2nsec rise time, 2nsec flat top and 2nsec fall time, giving a comfortable 100Mb rate down one wire. Today, I am in correspondence with someone who says his chip outputs have a rise time of 200psec. Thus, you can see the massive data rate achievable 40 years ago and even more today down one pin (plus return), and why our microelectronic packages bristling with large, expensive, unreliable pins are so ridiculous. This pin constipation is a direct result of anonymous, ignorant committees evading discussion of the principles involved.
The reason for my involvement in the matter today is that these committees have the temerity of make it a criminal offence if we ignore their standards. The failure of our industry to organise the setting of standards competently means they have to be ignored, and some of us have to be jailed, unless I gain enough support from you to identify and go after those who try to defend the EMC standards. “Name and address supplied” writes; “…. The prosecution only has to prove that the law has been broken ….”. However, there are many ways to wipe out the power of a law based on fantasy and ignorance, if I get your support.
Ivor Catt 21may03
EMC is Attainable
In his letter to EW with the above title in June 2003, Ian Darney missed the point of the circuit diagram of my revolutionary Mains Filter in my article in March 2003.. We all know that if someone switches on any equipment, it sends other digital equipment barmy. We hopefully use the mains as both a supply of pure water, and as a sewer. This scandal has been steadfastly ignored by our profession, and perhaps our Editor would like me to write an article about it. My co-authors and I did not discuss it in our private seminars on Digital Hardware Design, or in our articles in Wireless World.
The problem with the mains filters, on each computer in my room full of computers in West Herts College, indispensable for my computers to function properly, was the steady 50Hz current they created into the earth line leaving the classroom.
My 1979 book “Digital Hardware Design”, p84, says that a typical mains filter uses 1uF capacitors.
A mains filters has two 1uF capacitors connected directly from L to E – in my view an unnecessarily dangerous practice. With 240v across them, Ohm’s Law tells us that this results in an earth current of 240 x 50 x 1u = 12ma per capacitor. With twenty computers and their mains filters in the computer room, this added up to an earth current of 0.48 amps, enough to kill a very healthy student.
Ian Darney then writes; “His objective could have been to reduce the amplitude of the switch-on current surge created by the mains filter to a level below the threshold of the earth leakage detector, without reducing the efficiency of the filters. Instead, he chose to waste time by trying to change the regulations.”
I did both. My revolutionary mains filter, illustrated in my March 2003 article, which was manufactured for me at that time, shielded the L from the G, and so reduced the earth current by a factor of 100, since the voltage drop between N and G is less than 1% of the voltage between L and G. I also forced the man in the IEE responsible for the British Standard, which seemed to require an earth leakage detector in all rooms used by students, to admit that his Standard was only advisory, not mandatory, and so could be ignored. As we will do with EMC regulations, I got documentary evidence of his concession. Similar documentation will be presented to Prosecuting Counsel who are attempting to prosecute when we ignore European EMC Regulations, and to the jury.
I do not see the connection between the page 5.2 Ian recommends on his website http://www.iandarney.pwp.blueyonder.co.uk/ and the present discussion.
Ivor Catt 22may03
Eima Burdd, in his EW June 2003 letter with the above title, writes; “The application of proper EMC to this situation illustrated the problem and enabled a proper cure to be effected – and none of this with reference to the “bible” of Don White. ….”
It appears that Burdd is arguing for proper design practice, and so agrees with me, see the last sentence on p45 in my March 2003 article, which he thinks he is attacking. I wrote; “To be relevant, [EMC magicians] must get involved in the design process; or else leave the whole problem to the existing design engineers. ….” This closely matches Burdd; “Some say that the above work was just part of correct engineering practice – and indeed it was ….”
Other joke names appearing in Wireless World appear at the end of this document.
Ivor Catt 22may03
EMC – A Flawed Discipline
Michael O’Beirne, in his letter with the above title, doubts that a 1964 vintage Line Printer used massive current to simultaneously print a whole row of As
Indeed it did. A row of 120 (or 200) hammers were simultaneously projected at the paper, with behind it inky tape, and behind that a rotating drum with a row of As etched on it. Whether the printer printed a row of 120 characters or 200 characters is lost to history, which explains the slight confusion in my article. However, the notes which I retain from 40 years ago tell me the rest. They show 3, not 6, amps going through each ten ohm hammer, leading to a total of 360 amps or 600 amps, depending on whether the line printer could print 120 or 200 characters in a row. However, I think the paper was narrower than normal computer printout, which tends to indicate only 120 characters, or 360 amps.
I retain fascinating photographs of oscilloscope traces. Using a Tektronix 535 scope with Type O Operational Amplifier plug-in unit (integrator), I used the back EMF of the hammer coil to show its physical movement (plotted against time) towards the drum, recoiling from the drum, and then bouncing on the back stop. I wanted to pulse the hammer coil again (creating an electronic “stop”) just before it hit the back stop, to speed up recovery, but this was banned. The damping down of the hammer after a print trajectory was the domain of mechanical engineers with their damping back stops, and turf wars prevented an electronic contribution, which could have massively speeded up the line printer and improved the life of the back stops.
[Circuit diagram and scope photograph appended, sent by slomail in return for sae and £1 to Ivor Catt, 121 Westfields, St. Albans AL3 4JR, England.]
Ivor Catt 22may03
EMC – A fatally flawed article?
John Blythe, in his letter with the above title to EW, June 2003, writes; “Too much rudeness about GEC (and Weinstock). At least he ran a fairly stable, profitable company for many years (and so was able to provide the shilling of convenience Mr. Catt was willing to take). …. GEC (Marconi) was all but destroyed and the pension funds of thousands have been damaged (perhaps for ever) by some who thought they knew better ….”
Blythe is completely wrong, as are the technology-free journalists who write about the spectacularly rapid collapse of GEC. In 1980 I was so concerned about the state of
Weinstock’s GEC that the Attorney-General arranged a meeting for me with the chief procurement officers in the MoD in Whitehall, see http://www.electromagnetism.demon.co.uk/gamoe.htm , which was written at the time. I was in the employment of GEC when I blew the whistle.
In 2001, long before GEC collapsed, I put onto my website at http://www.electromagnetism.demon.co.uk/19161.htm ; “I have written that it is obvious that Weinstock, head of GEC, was not operating a major serial scam against the taxpayer because he accompanied the Queen alone in her carriage on the way to Epsom races while his wife accompanied the Duke of Edinburgh alone in the next carriage.” GEC was an empty shell in 1980, lacking middle management and lacking technical expertise. Both cost money, and Weinstock fired any divisional manager who did not show improved profit for the next six months. Today’s journalists have failed to check back to what was written at the time about the undermining of GEC by Weinstock. “GEC operated a number of ‘defence’ weapon ‘design’ scams …. “ [Ivor Catt, Ewmay03, p46.] The collapse of the Cold War undermined the scale of these scams, and so GEC’s collapse was inevitable, long before the maligned Lord Simpson arrived on the scene. All GEC had was a money mountain. Simpson had to try to buy in management and technical competence. But read http://www.theregister.co.uk/content/53/20435.html “both Lord Simpson and John Mayo are accountants by profession” At, http://news.bbc.co.uk/1/hi/in_depth/uk/2000/newsmakers/1527551.stm we read; “The name Marconi once stood for all that was great about British inventiveness. Now, it is the symbol of how badly things can go wrong. Guglielmo Marconi, who gave the company its name, was the co-inventor of the radio. He must be turning in his grave.”
Like a pack of wolves, technically ignorant journalists have got the date wrong. As engineers who were employed by GEC (except for Blythe) know, Weinstock turned GEC into an empty shell decades before Simpson arrived. Weinstock covered up using “defence” scams against the taxpayer. Simpson failed to restart the Cold War, the only way to save GEC, unless he bought up real companies, which he tried to do. How could the accountant Simpson tell that the dot.com bubble might burst? It was no more wowsy than “defence”.
I remember when GEC’s Technical Director called all the “engineers” (?milkmen?) together in GEC Chaucer House Portsmouth. He said; “I have good news for you. We have trebled our spend capability.”
To explain. I have it on good authority that “defence” scams were cost plus 14%. GEC would be more profitable if it could “spend” more taxpayers’ money. We “engineers” on the Tornado and Tigerfish projects were there to absorb the 100%, leaving GEC with the 14%. A good reason for employing contract engineers would have been if contract agencies paid off a west country front company set up by MoD staff in return for the MoD giving them the job of recruiting contract engineers for GEC. The collapse of the USSR would cause all such ruses to collapse.
Ivor Catt 22may03/ 1 june03
Joke names attached to letters published in Wireless World (Partial list).
Ouida Dogg, Wireless World December 1982, April 1983, January 1985 and October 1985
F U Weaver-Mowes, February 1985
Ivor Catt, May 1985; “I feel I must assert that I really exist. I was in no position to protest when my parents devised my ridiculous name. ….”
Oliver Fish, October 1985
Ivan L. E. Fant, March 1986
R. Petzeratt. (Date?)