Jump to content

Talk:Ivor Catt/Archive 12

Page contents not supported in other languages.
From Wikipedia, the free encyclopedia
Archive 5Archive 10Archive 11Archive 12Archive 13Archive 14Archive 15

Is Catt a hoaxer- or just a deliberate obfuscator?

Catt wrote the Catt Anomaly to present the conventional view as being a hoax. He included the electric current question about where the charge comes from to entangle the unwary. I think he also included all the false clutter about displacement current flowing when the voltage is steady to obfuscate the picture. He certainly did not fully grasp the conventional picture, in which displacement current requires a variation of voltage with both distance and time, because he shows displacement current as operating when there is no such variation, thus defying Maxwell's equations. It worked! Experts discussed the question, without pointing out the most major errors in the diagram and text. The initial impetuous for the electrons to accelerate is from the electric field causing the voltage to vary longitudinally (along the conductor). Two transverse mechanisms come into play to sustain the logic step current as guided by two conductors: electromagnetic radiation exchange due to net charge acceleration, and magnetic field energy exchange once the charges have a constant speed (where the voltage is steady 10 volts or whatever). The acceleration of electrons at the front of the logic step is due to the variation in the potential of the electric field from 0 to 10 volts. Once the front has passed the electron by at light speed, the magnetic field created by the motion of the electrons (i.e., the electric current) causes the motion of charge in the other conductor, and vice-versa. Nigel 172.213.66.114 18:44, 9 March 2006 (UTC)

Catt's arguments always remind me of one of Martin Gardner's articles from Scientific American, where he looked at the Trisection of The Angle problem. Apparently, a perennial annoyance for professional mathematicians has been (if it is not still) the amateur who thinks that he has found a way of geometrically dividing an arbitrary angle into three exactly equal parts, using only a straight edge and a pair of dividers. The problem is that someone came up with a definitive proof that this is impossible in the general case as long ago as 1836. However, the proof requires the use of work from different branches of mathematics which the amateurs have not studied. Consequentally, when they got back letters saying "Sorry, your proof must be wrong, but we really don't have the time to find your mistake," they reacted in much the same way that Catt always does, claiming "conspiracy" and so on. (See for instance [1].) -- Kevin Brunt 19:40, 9 March 2006 (UTC)
I tend to agree with the sentiments here although one must be careful about dimissing all so called crank ideas. Take Frank Whittles jet engine, which he was told by the RAF reaserch experts just couldnt work. There must be many other examples. However, I dont think therse much doubt amongst engineers that the idea of capacitor and TL equivalence is correct. THe exact details have not been defined yet, but there is so much experimental evidence that it is foolish to deny it. As I said anyone with a length of coax a fast scope and step generator can see the wavfoems exactly as Catt shows in his article. A lot of people would have seen them before (as I have) in a lab setup but just ignored them as an unimportant and underirable artifact of what they were trying to do. Those who deny the Heaviside expalnation of a charging capacitor need to propose an alternative thoery to fit the facts.--Light current 17:42, 10 March 2006 (UTC)
Much of the issue with Catt is that he is an "Engineer", rather than a "Scientist", from the days when the distinction was much greater than it is now. I suspect that his lecturers at Cambridge tended to be a little dismissive of "theory", because "reality" always has something else up its sleeve. This shows, for instance, in his rejection of the "Laplacians" of the Telegrapher's Equations, while he uses the solutions of the equations for a step pulse in a resistanceless TL. He presumably thinks that the results are based on observation rather than theory. -- Kevin Brunt 19:40, 9 March 2006 (UTC)
His story about his time at Ferranti may also be indicative. Googling reveals that the real problem with the development of the Sirius computer was a lot more fundamental than a crosstalk issue with the explansion of the memory. Ferranti took Manchester University's computer designs and "commercialised" them. The Sirius project was an in-house exercise that may well have been oversold to a newly-appointed head of division, who was unaware of the some of the lurking factory politics. The Sirius team were in direct competition with the people working on the academically-derived product lines, and some of Catt's anti-academic prejudices are explicable if he picked them up when he was working at West Gorton. -- Kevin Brunt 20:34, 9 March 2006 (UTC)
The reason that I was Googling for the Ferranti Sirius in the first place was that I was wondering what its architecture is. It has been described as a successor to the Pegasus, which was a serial machine, (ie, its "memory buses" were one bit wide and its memory and "accumulators" were shift registers - it added two numbers together by shifting them into a one-bit adder, and shifting the result out of the other side. My estimate is that I would be surprised if it could outperform a Sinclair ZX Spectrum.) Catt referred to learning to program on the Pegasus in his "Boolean Castles" article in Electronics World, where he even included the flow diagram for a square root routine from the Pegasus manual. (I suspect that he was unaware that the algorithm the routine uses is referred to as "Newton-Raphson" since it was invented by Newton and his friend Raphson.) The Pegasus was a scientific "number-cruncher" rather than a commercial machine. Catt's New Scientist article and his later writings seem to suggest that he regards computing as being about crunching numbers rather than the shunting around of data that characterises the commercial side. His NS article (of 1969) is largely a complaint that during the 1960's the scientific and commercial designs largely merged to produce machines such as the IBM 360, which were capable of doing both roles. The 1960's also did something else. The IBM 360 was officially the "IBM SYSTEM\360" - not merely a computer, but a computer with a full-blown operating system, and associated "utility programs". In creating the SYSTEM\360 IBM discovered that the difficulties entailed in getting the computer hardware to work are trivial in comparison to getting the software to do likewise. OS\360 was a fiasco that had the potential to bankrupt IBM, and the fallout from it had a lot to do with the evolution of the computer away from the purely hardware aspects which are the domain of the electronic engineer, and into the era of the "computer system" which is more the domain of the "computer scientist". I'm being a little imprecise here, I know, but whatever the precise definition is, I don't think that Catt understands that a computer is not just the hardware, and that the hardware is more and more the easy bit. -- Kevin Brunt 20:34, 9 March 2006 (UTC)

The Newton-Raphson method was taught in A-level maths in 1990 when I did it. I don't see why it would be around in Newton's time and in 1990, but not in 1953(?) when Catt did his two maths A-levels getting grade As. I agree with Catt strongly on computing, that a computer should be designed at the hardware level to do the function designed. If Microsoft had designed fancy visual software to waste the hardware of the computers that put men on the moon in 1969, they wouldn't have got the rocket off the ground. Software is only the hard part now because the hardware architecture has ossified. Some kind of Kernel Machine technology will probably be needed for efficient voice input/output and live video interpretation by computer to allow practical robots, where the whole concept of keyboard/screen is superfluous. I don't think people will want or get much more powerful PC hardware or software in the future, because with wireless laptops it will just drain the batteries faster. Windows XP is five years old and that shows it is the end of the road for the PC [2], and once you can store 100 hours of movies on the hard drive further advances are just superfluous. The next big advance will have to be something totally new. Nigel 172.213.66.114 22:51, 9 March 2006 (UTC)

Did you hear the true story of the old Cambridge professor who was a lecturer on electricity? It was a warm day, and a student dozed as the guy droned on. Eventually the professor raised his voice and, staring straight at the student, asked: "now what is this thing called electricity?" The student in a panic replied: "sorry, I did know, but I've forgotten." Professor: "Gentlemen, let us lament the greatest tragedy in history, Mr X here has discovered the secret to physics, but he has since forgotten it!" 172.213.66.114 22:51, 9 March 2006 (UTC)

I belive Catt is not a hoaxer in the accepted sense of the term. Ialso do not beleive he is a deliberate obfuscator. I believe however, that he may be an unwitting obfuscator if he is not prepared to answer the valid criticsims of his theories. Catt is Catt - we have to live with it!--Light current 19:27, 10 March 2006 (UTC)
LC, your most recent comments are very close to the heart of the issue. I don't think that anyone is disputing the demonstration that the application of a voltage to one end of an unterminated TL through a high source impedance results in a stepwise increase in the voltage. Catt et al appear to have got the maths right and it looks entirely plausible. The problem is that although the result is interesting, it does not support the vast superstructure of speculation that has been built upon it. Catt, Davidson and Walton are using the description of a TL in terms its characteristic impedance and velocity of propagation, yet they deny the validity of the "Laplacian" differential equations (the Telegrapher's Equations) from which that description is derived. It is no coincidence that the mathematical step function used in the derivation is still known as the Heaviside Step Function. Oliver Heaviside not only knew about the problem of the "long, thin" capacitor, he knew that it could be solved by treating the capacitor as a long line of infinitesimal capacitors. -- Kevin Brunt 20:45, 10 March 2006 (UTC)
Agreed that moving on beyond the TL/capacitor equivalence is where most of the problem lies. I think even Nigel would agree with us on that one. In fact I think he said so earlier.--Light current 01:18, 11 March 2006 (UTC)
The same applies to "energy current". Catt's rejection of charge/current flow is inconsistent with the standard explanation of how (for example) a lead-acid accumulator cell works, as it connects the two halves of the chemical "redox" reaction through the transfer of electrons via the conductor. Without an alternative explanation, Catt's Theory C is utterly incompatible with the fundamentals of chemistry. Heaviside's version ("Theory H") does not have this problem, because it is postulating that the charge motion is a result of the electric and magnetic fields, rather than the other way round, but is not denying the association. My quarrel with Nigel over the kinetic energy of the electrons is very closely related to this. The KE is a consequence of the motion of the electrons (ie it is related to current, and the magnetic field); Nigel ignores the concept of the electrons "touching" and so is ignoring the transmission of the forces which cause the electrons to move. This touching/transmission of force is clearly associated both with the electric fields between the electrons and (at the macroscopic level) with the applied voltage. We thus tie together voltage/current with force/motion, and end up with a description that relates the electrical concepts with mechanical equivalents. What is more, we have a description which has voltage and current simultaneously in play. Catt's Theory C deals only with the voltage steps in the TL, and Nigel's KE argument ends up with the voltage in one model and the current in another! -- Kevin Brunt 20:45, 10 March 2006 (UTC)