From: Paul
Date: Tue, 26 Dec 2000 19:19:18 +0000
Subject: Re: [TSSP] pn2511 is baloney
In reply to Terry's comments... 1. Good point. I've converted tesla to Tesla throughout all docs. 2. Good point. As you say, its not sufficient just to ignore self inductance and resistance. Circular current paths must be prevented too. However page 2 is probably not the right point to mention it, since the aim here is to introduce the capacitances, so instead I've raised this point in section 3, adding a paragraph at the end. 3. The self capacity to infinity is a well defined concept, although often a cause of confusion. I've included a mention of it, partly for completeness but also so that the text and analysis is not limited to resonators operating with the earth as one terminal but can be applied to a coil in free space with a limited ground plane. Not that anyone intends operating a coil sticking out the doors of the Shuttle payload bay, but it may help with the conceptual understanding of the resonance and the earth's non-essential role in it. Anyway, if I didn't include it, someone would say 'what about the cap to infinity then?'. You're not alone in being uneasy about the Cinfinity. I've found an eloquent exposition on the subject in an essay by Fred Erickson, available in postscript from http://www.ttc-cmc.net/~fme/spheres.11-03-99.ps.gz which deals with the Cinfinity of spheres, and of the earth. If windows doesn't support the gnu compression, let me know and I'll make available a decompressed version of this excellent work. 4. Agreed. I've added a final diagram to section 2 summarising the resonator capacitance in terms of the physical capacitance functions that have been introduced. Let me know if its on the right lines. 5. Thanks. It's important that section 4 comes across clearly as the diff equs are the gateway to all that follows. The approach to developing the diff equs by examining an infinitesimal segment of conductor is conventional and will be familiar to elec eng undergrads. It's the longitudinal terms that are peculiar to the solenoid and make things hard to calculate. 6. The blocks of vertical whitespace occur because I told the typesetter not to allow text to flow past the diagrams, so it cannot float the diagrams to achieve a better fit with the page breaks. I'll get around to numbering the diagrams at some point, so that I can release them from their fixed positions wrt the text. 7. Thanks. A potentially doubtful part is in the last para of sect 5, in which I handwave an explanation for the uniform intervals between the higher overtones by suggesting that an in-coil wavelength is small enough to span an approximately uniform patch of Cint. Perhaps Boris may have some comment to make on this. If it remains doubtful I'll take it out, because we haven't really studied this enough yet to pronounce on it. 8. Agreed. I've inserted a VI graph for a more normal coil (your test coil, as it happens. The h/d=1 example is Malcolm's 272 turn coil at high elevation). See if you think that does the trick. 9. Yes, I could box the significant formulae. I tried it on a couple of sections and it looked odd because too many of the equations were boxed. It might be better to add some sort of appendix at the end as a summary of definitions and useful formulae. I need to experiment. 10. I've introduced a mention of eddy coupling to both topload and groundplane at the end of section 3, and then promptly dismissed it as being negligible. Whether this is really the case remains to be seen. So far it doesn't seem to be troublesome in comparing simulations with measurements. This is a relief as I havn't been able to make much process with quantifying it, as I said I would try to do when we were looking at your low Q factors some time ago. Terry, Dealing with your comment 8 forced me to notice a major mistake in sect 6. I'd stated that the current maximum coincided in position with the voltage inflection point. That's almost true for the h/d=1 case but by h/d=3 the two have noticeably separated. I've added a paragraph commenting on this and explaining the reason why. This is just the kind of defect that a stiff review of the document will help to identify, so I'm pleased to see the process is working. Don't know why I've been saying the two maxima coincided. It looked a bit that way from the more prominent examples, so I just accepted it without really going into it. Frightening how easy it is to take something for granted. Terry, thanks for your valuable comments - pn2511 is now noticeably better as a result. I hope the others on the list will follow your example. I've applied all the changes mentioned above, except for 9, I'll need to experiment a bit there. http://www.abelian.demon.co.uk/tssp/pn2511.ps version is now 0.1c, 527 Kb. I'm afraid that renders the PDF copy out of date. If you can handle gzip compression, you can download http://www.abelian.demon.co.uk/tssp/pn2511.ps.gz instead at 132 Kb. > I hope this is of help. Certainly was - Cheers, -- Paul Nicholson, Manchester, UK. --
Maintainer Paul Nicholson, paul@abelian.demon.co.uk.