From: Paul
Date: Fri, 28 Jul 2000 09:35:19 +0100
Subject: [TSSP] Update on progress
Update on progress I've been doing quite a bit of work on the laplace solver program tlap.c. The reason it has been giving the wrong answers is that the relaxation gets terminated too early. This is most evident when the program is run with a single electrode and ground. The outgoing flux from the electrode is always higher than the flux arriving at the boundaries. Running with a longer relaxation time fixes this and increases the external capacitances by around 20%, while the internal capacitances remain about the same. The end result is a much better estimate of the f1 resonance. I've also been dabbling with the top toroid, which is not used in the v0.1 software. The original intention was to use tlap to derive an equivalent lumped capacitance to earth for the toroid, but it seems that there is also quite a bit of capacitance from the toroid to the coil, contributing to the internal capacitance, and therefore presumably to the dispersion. Therefore I'll be adding another array to describe this extra distributed capacitance, and some extra code in tsim to introduce this to the model. Might be a little while before this is released to the group, as I'm under orders to finish a CW driver for my coil so that my partner can hold her 'electric barbecue' while it is still summer! I have at least solved one mystery, to my own satisfaction at least, which was one of the reasons this simulation project began. A CW driver was exciting the coil not only at f1, but also f3, causing a problematic 3rd harmonic ripple in the drive current. Now I can see that, with top capacitance acting to increase f3, while dispersion is acting to reduce it, the two effects may balance, with f3 ending up at or near the 3rd harmonic of f1. Thats all for now. Feedback welcome! Regards, -- Paul Nicholson, Manchester, UK. --
Maintainer Paul Nicholson, paul@abelian.demon.co.uk.