From: Paul
Date: Wed, 10 Jan 2001 11:40:05 +0000
Subject: Re: [TSSP] Final solution...great chance for it
Boris, Bodlovic's paper arrived in the post this morning. Many thanks for taking the trouble to send it - its quite large! Its certainly going to take some time to study this, so I can't comment much at this stage, but a few things are apparent from a quick glance through... He uses a transmission line model for each turn, whereas we gather a bunch (eg 1 to 8) of turns together and model them with a single LCR element. So he will be able to examine much faster transients - he refers to solutions on the 'microstructural' level which I thought was a good way to describe it. He refers to experiments involving transients with 42nS rise time and his intention is to examine some anomalous distortion of these fast transients. The analysis appears to ignore capacitance between non-adjacent turns, and uses a uniform distribution of the lumped parameters so as to obtain a tri-diagonal capacitance matrix and the usual forward and reflected waves within each turn - these parameterised by the turn number. He proceeds with a spatial frequency modal analysis, doing some fancy stuff with the L and C matrices to couple the transmission line turns and the whole thing seems to reduce to an eigenvalue problem. He does make some very interesting comments in conclusion. He says that the velocity of the low frequency modes (ie the ones we use) is determined largely by the external C and the effective inductance. The higher frequency mode velocities he says are more influenced by internal C, and that the effects of mutual inductive couplings are decreased in the higher modes due to opposite directions of currents in nearby turns. I'm pleased to see that these comments are in agreement with the handwaved statements in the last paragraph of section 5 in pn2511. There's some interesting stuff happening with the matrices in the middle of the paper, it must in some way be equivalent to the manipulations done within tsim to couple the elements into a network - the formalism appears elegant and incomprehensible so I've got some studying to do! Thanks again Boris, and I look forward to receiving your own theory notes from Malcolm. Cheers, -- Paul Nicholson, Manchester, UK. --
Maintainer Paul Nicholson, paul@abelian.demon.co.uk.