From: Paul
Date: Sat, 19 Oct 2002 08:10:49 +0100
Subject: Re: [TSSP] F A N T C: - It's NEW! - and needs some TSSP input
Hi Bart, > nothing I could readily identify directly to the applet > not working. I doubt NS could have broken Java, therefore I suspect that in geotc.js, references to the applet within the document, such as document.vprof, document.iprof, document.geodraw, are always null. Perhaps NS have changed the definition of their document objects and properties. Perhaps there's a new approved way for the script to locate applets in the browser's memory. This would account for why it doesn't work and the lack of error messages. >> c) E and H field at a distance from the coil; After more thought, the E would be fairly accurate, ie would take account of shielding by walls, floor, etc. The H would be less accurate because we could not take account of eddy currents induced in the various electrodes of the system and surroundings. >> d) distribution of EMF in secondary due a current in primary; This would be an easy one to do. It could be added to the existing S,V,I output; Altering things a little, it could have columns S,I,V,E, in which S is position (0..1) along sec, I is current at that position, V is voltage at that point induced by 1 amp in sec, E is voltage at that point induced by 1 amp in pri. Thus, in a real coil, during primary notches the voltage would be V * Ibase, while during secondary notches the voltage would be E * Ipri, and in between it would be some mix of the two varying sinusoidally with the beat envelope. If no primary was specified, the E column could be dropped, or its values set to zero. Additional runtime would be about the same as computing the M value. Alternatively, it could have its own function, making a 4th category of calcs. That would enable lots of primaries to be tried without having to compute the whole resonance as an unwanted side effect - 10 times quicker. >> a) DC resistance of coils We still have an unresolved anomaly here, a slight offset in the errors. >> e) location of highest surface field (breakout point); I'm having doubts about this one, perhaps geotc doesn't have the spatial resolution to get the breakout point correct. In marginal cases it could get it very wrong. True also for tssp, but with geotc things would get marginal much sooner. More thought needed, but I guess these are things for the next revision. The problem isn't so much coding them as testing them. Only (a) would be 'easy'. -- Paul Nicholson, --
Maintainer Paul Nicholson, paul@abelian.demon.co.uk.