Fifty years of Electromagnetic Theory.


Fifty years of Electromagnetic Theory.

Ivor Catt. 20 March 2010



About classical electrodynamics

The 109 Experiment

The dramatic events of the last two weeks call for a discussion of the historical progression to the present.


In Ferranti in 1960, the late Gordon Scarrott suggested that the interference between logic signals travelling from the main computer to an additional memory box (about the size of a piano) containing three thousand more (40 bit) words of memory might be caused by mutual inductance. Previous to his remark, the consensus was that digital signals interfered with each other by mutual capacitance – capacitive coupling. Previous computers used thermionic valves, with high voltages and low currents. Our computer, “Sirius”, was the first transistorised computer, using low voltages and heavy currents, making mutual inductance more significant.


While in Dataproducts Corp. in Los Angeles in 1964, Art Cappon came to the company to promote integrated circuits manufactured by his company Motorola in Phoenix, Arizona. After a conversation he decided that Motorola needed to get me to investigate interference in digital systems. Motorola made the fastest (ECL 1nsec) logic on the market. I was offered a job and refused. However, when Data Products fired me two months later, I told Motorola that I had changed my mind, sold my Los Angeles house, and took my family to Phoenix.


Motorola Phoenix were expert in semiconductors electronics and integrated circuits, but lacked expertise in the field of electromagnetism, which is where I came in. On a printed circuit board, a signal travelled six inches in a nanosecond, so perhaps a major problem was to arise soon with their 1.35 nsec logic gates.


IBM was making a major computer for NSA (National Security Agency), part of the Pentagon. This contained a small high speed memory, the “Ballman Scratchpad Memory”, at the centre of a hierarchy of memories. This memory was to be 64 words, 8 bits per word, with an access time of 20 nsec and a cycle time of 20 nsec. IBM said they could only achieve 35 nsec, so the decision was made to subcontract the task of a nine month project to a specialist Integrated Circuit company, Texas Instruments, to make a partially populated model. However, there were complaints that TI always got such contracts, so it was decided to give a split contract to TI and Motorola.


At the time, 1964, an integrated circuit package 1cm square contained only two memory bits. With a delay of 1 nsec every six inches, the whole memory had to be compressed into a cube less than one foot cube. The daughter boards plugged into a mother printed circuit board which has thirteen layers of copper, alternate signal wires and voltage planes. Manufacturing such a board was expensive, and it was imperative that we knew in advance what was the interference between two parallel signal lines in one of the middle planes.


I read the literature, particularly Jarvis, reference 11 in my 1967 paper , and found that the interference depended on signal rise time. The faster the rise time, the greater the interference until it would be larger than the original signal. I doubted that this way we could have a voltage amplifier, and started to research independently of the literature. I soon found that the interference was a flat topped pulse, not a spike related to the active line’s signal rise time. However, the reality was confusing.


By chance, I met the late Ken Johnson, the best researcher in Ferranti, whom I had last seen thousands of miles away, walking down the corridor in my new company. I grabbed his and told him my problem. He replied that there were two signal modes, and gave me them incorrectly. However, the key was the idea of two signal modes, later to be called "Even Mode" and "Odd Mode" .


Using only Faraday’s Law of Induction and the law of conservation of charge, I developed (and published) the theory of why a TEM Step travelling down between two parallel conductors can have only one voltage/current ratio (Zo) and one velocity. It is now at Appendix I , "The Interconnection of Logic" and "Properties of a Transmission Line" .


Then using only Faraday’s Law and the Law of Conservation of Charge, but extending it to mutual inductance and “mutual capacitance”, I mathematically proved that only two modes could travel down between a symmetrical double pair of conductors. This is worked out at Appendix II and  "Crosstalk in digital systems" .


What I did not prove, mathematically or otherwise, was that it was physically possible to superpose the two modes, even and odd. Of course, the photographs I took of oscilloscope traces 1 , 2 , clearly showed the two modes superposed, so there seemed to be no problem.


It took me 46 years to realise that there was indeed a problem . ( See the smoking gun .) Still, nobody except myself has noticed. Faraday’s Law does not permit the superposition of two modes. The mathematics made no distinction between a case where superposition was permissible and a case, like Faraday’s, where superposition was not. So why did the use of Faraday’s Law falsely lead me to a conclusion which the photographs 1 , 2 indicate is valid, if indeed Faraday’s Law was contradicted? Independently I have shown that Faraday's Law is faulty. Throughout the Faraday experiment, TEM Waves were involved, and never was there isolated electric field, electric current or magnetic field. This is just one example of the way that in a highly physical, not mathematical, subject like electromagnetic theory, the mathematics is merely a metaphor for physical reality. At other points the products of the mathematics break down, for instance as demonstrated by "The Catt Question" . Of course, this problem disappears if we migrate from classical theory to Theory C , which excludes electric current anyway, and so completely bypasses "The Catt Question". Under Theory C , electric charge and electric current are the physically non-existent results of mathematical manipulation of the Energy Current electromagnetic field. Faraday’s Law is a complex development from the starting point of the Energy Current TEM Wave, and not the starting point from which the TEM Wave is developed, in the traditional way. For more than a century, the cart has been before the horse. Since Faraday's Law was developed from slowly changing fields, it is dubious, as I have shown.


This month, March 2010, I asked my co-author David Walton whether, like me, he had ever realised the implication of Faraday’s Law excluding the superposition of two fields, whereas the photographs 1 , about which he was expert, showed that such a thing existed. (The third traces in the photographs 1 show a situation which Faraday’s Law makes illegal.) He said that he too had not realised this. I also asked him the question of why Energy Current, the central feature of Theory C , demands symmetry in its environment, "Even Mode" and "Odd Mode" . Both have a symmetrical appearance to the conductors which are guiding them. He agreed that it was not clear to us. The answer will be buried in Figure 30 , where Energy Current continues to the right in a vacuum but also slowly, and minimally, penetrates north and south into nearly perfect conductors. (It would not penetrate a perfect conductor at all.) We know that the TEM field will approach a perfect conductor at right angles. It has to, to remain TEM. Perhaps this is the constraint which limits the possible modes in a four wire system to two, "Even Mode" and "Odd Mode" . This requires more thought.


In March 2010 David Walton and I agreed that the illegal (according to Faraday’s Law) situation was already known to us decades ago when two TEM pulses from opposite directions travelled through each other in a coaxial cable. However, we agreed that under classical theory, the case in the diagram of "Even Mode" and "Odd Mode" travelling in the same space in the same direction but upside down to each other was more grotesque than in the case of crosstalk.


Questions for Classical Electrodynamics


The four major problems are;

Displacement Current.

The Catt Question.

Faraday's Law.

Pictures deriving from Crosstalk Theory.



Displacement Current.


This paper points out an oversight which has continued for a century. This is discussed in my article at


Displacement Current.

Our article in December 1978 pointed out that electric charge entering the plate of a capacitor did not immediately desire to traverse the space between the capacitor plates. (Bleaney is wrong when (s)he writes that the field between the plates is uniform.) After entering the capacitor plate from the input wire, the charge first has to spread itself across the plate. Only then can it express a desire to traverse the space between the plates.

This desire led to "Maxwell's leap of genius", Displacement Current. Maxwell himself, and all those who followed him and worshipped him, failed to notice that after entering the capacitor plate from the input wire, the charge had first to spread itself across the plate. This intermediate step has been ignored by all text books and lecturers. Since we pointed it out in December 1978 , it has been ignored for a further quarter century. All of today's text books are written as though this problem, of charge spreading out across the plate, remains unnoticed, like the Emperor's nakedness.


The spreading out of electric charge across the capacitor plate is real electric current, and must cause real magnetic field, according to Ampere's Rule or the Biot-Savart Law. The alleged genius of Maxwell was that the notional electric current, called by Maxwell "Displacement Current", was invented to produce magnetic field and so lead to key conclusions. Since the key (and only) purpose of Displacement Current is to cause magnetic field, it is unacceptable that there continues to be no discussion of the magnetic field which must be caused by the much more real electric current as the electric charge spreads out across the plate from the incoming wire.


The archetype for this kind of myopia is Professor W H G Lewin.  His lectures, including that on Displacement Current, are celebrated worldwide. He has been given prizes for lecturing excellence. Looking at his lecture, we can learn a great deal. Early on he draws a uniform electric field in the capacitor, but later on he says; “There is also a current going up on these plates .... “. He draws an electric current travelling along the capacitor plate at right angles to the horizontal current, which means that the horizontal electric field in the capacitor cannot be uniform. He refuses to discuss with me the magnetic field generated by this electric current. (Compare with Walter still makes time to reply to every single e-mail he gets from his internet fans .) However, he gets closer than other luminaries to the crisis we pointed out in December 1978 . No other lecturer or text book writer (including Heaviside) admits, or notices, that there is such a sideways current flow. Two other things are notable in his lecture. First is the bemused look on the faces of the students. Second, we see the arcane mathematics which causes this bemusement, and convinces the students that they are not capable of becoming expert in the subject. Of course, his mathematics is not as Byzantine as that in Wikipedia and elsewhere.




The Catt Question.

When a TEM Step travels down a coaxial cable at the speed of light for the dielectric, negative charge must accumulate on the bottom conductor to terminate the electric field. That charge cannot reach the required point in time.




Faraday's Law.

In Faraday’s famous experiment, a TEM step enters the primary of the transformer. When it has reached half way across the transformer, the change in magnetic field in the primary should cause a voltage to be registered to the far right in the voltmeter, but it measures no voltage because no signal has yet reached it.




Pictures deriving from Crosstalk Theory.

I have only just realised that the Pictures deriving from experiment show that there can be two distinct electric fields and two distinct magnetic fields, that is, two energy currents, at one point in space at the same instant of time, which defies Faraday’s Law.

He agre



Crosstalk (Noise) in Digital Systems

Pages 1 , 2 , 3 , 4 , 5 , 6 , 7 , 8 , 9 , 10 , 11 , 12 , 13 , 14 , 15 , 16 , 17 , 18 , 19 , 20 , 21 , some of which is in two of my books. The argument starts at page 30 of one book , and at page 4 of the other book , continuing on page 55 . Here in figure 9.2 we see “a very narrow pulse introduced at the front end of the active line. If there were no parallel passive line nearby, this pulse would travel down the active line (at the speed of light for the dielectric) more or less unchanged,” in a TEM mode. “However, as the other two traces show, the presence of the passive line caused the original narrow pulse to break up into two similar pulses.”

He agre



The Role of the Luminary

How do "Questions for Classical Electrodynamics"  appear to accredited luminaries – Professors of Electronics, text book writers, journal referees and the like? The challenge they face is formidable. They are deeply immersed in a number of red herrings;


Wave-Particle dualism is the Kiss of Death to an attempt to grasp the discussion. Particles have to be removed from our consciousness.


A major barrier to their grasping the real subject is the wrong theory for a TEM Wave , "The Rolling Wave" , which we can be sure all of them adhere to. This makes it extremely hard for them to grasp the idea that the same amount of energy is travelling through the red square and the blue square in the diagram. In fact, they do not even know the diagram. (If they did, they would draw it in their text books. The diagram is very important, because the same field pattern occurs with L, C, R and Zo.) I found it in only one text book fifty years ago, and cannot find it on the www. The only place where I can find it in what is available today is in Figure 2 of my own book. However, here we only see one of the two necessary sets of lines to illustrate the all-important curvilinear squares . (I find “curvilinear squares” on Google, but they refer to water flow and paid-for electronics articles.) They might well know the pattern made by iron filings above a magnet, but this does not mean that they are familiar with curvilinear squares , and their association with the Poynting Vector. Wikipedia does not have the diagram, and neither do other www pages on “The Poynting Vector”. Familiarity with the diagram is necessary. I have only recently realised that this field pattern is not available to a professor or text book writer. Lacking that picture, and also thinking that E causes H causes E "The Rolling Wave" , it is almost impossible for them to think clearly about the second, third and fourth Questions above.


No professor or text book writer has ever considered two pulses from opposite directions overlapping in a coaxial cable as they pass through each other. This is a very early case which repudiates Einstein's  "The Rolling Wave" model for the TEM Wave. Another well known case which repudiates it is the case of white light, multi-frequency, where According to "The Rolling Wave"  model for a TEM Wave the positive change of magnetic field for one colour must be causing one E field at the same time in the same place as the negatively changing magnetic field in another colour of light is causing an opposite E field. For a century, there has been plenty of unnoticed evidence which finally led to the greater absurdity of comparing Faraday's Law with the photographs deriving from crosstalk experiments , where we clearly have two electric and two magnetic fields at the same point in space. This evidence does not require migration from the standard Theory N to Theory H or Theory C.


The idea that a TEM Wave must be sinusoidal is pervasive. To confirm this, do a Google search for “TEM Wave” or “Transverse Electromagnetic Wave”. If a sine wave is imposed on any of the above four Questions, it submerges the Question in confusion. Of course, it is difficult to see how a non-sinusoidal TEM Wave could propagate according to "The Rolling Wave" theory. That is, "The Rolling Wave"  excludes the possibility that one logic gate can communicate with the next! It excludes the most fundamental element in digital electronics. Classical Electrodynamics , trapped in "The Rolling Wave" , doggedly ignores digital electronics, which is more than 95% of electronics today.