From: Paul
Date: Sun, 19 May 2002 01:30:27 +0100
Subject: Re: [TSSP] Racing arcs
I've set up the simulator to compute both the radial and the vertical potential gradients (E-field strengths) at the coil surface, and the results are in animated form, http://www.abelian.demon.co.uk/tssp/cmod/ Now in addition to the existing animations which show secondary volts, secondary current, and primary current, we now have another set which show vertical gradient, radial gradient, and primary current. The gradients are plotted in units of kV/cm. The vertical gradients obviously allow for the voltage non-uniformity but they don't allow for any 'corrugation factor', so the actual peak vertical fields may be a little stronger than shown. The radial fields are computed by assuming all the coil charge is on the outer surface of the winding, which is not really the case, of course. Therefore it will over-estimate the radial field by up to a factor of two. Offseting this a little is the fact that the program doesn't take account of the spacing ratio when calculating the surface charge density, so this will underestimate the peak radial surface gradient by approximately a factor of the spacing ratio. If it were felt that 80% of the coil charge was on the outer surface (reasonable), and a spacing ratio of 0.8 was in use, then the two approximations cancel out. The radial field is calculated taking into account both the secondary and the primary voltage distributions, although the program assumes a linear rise per turn of the primary, which is obviously wrong for a spiral, but the results should still be in the right ballpark. It seems that the contribution to the radial field by the primary is not very great, and only really shows up right at the bottom of the coil. Most of the surface gradient seems to the radial due to the coil voltage with respect to ground, rather than vertical due to the coil voltage gradient. The total field strength at any point on the coil could be as high as sqrt( Evert^2 + Eradial^2), but this is not necessarily the case, because the greatest radial fields will appear at the points (A) in http://www.abelian.demon.co.uk/tssp/tmp/bp190502.gif whereas the greatest vertical fields will occur in the troughs (B). Thus it's probably incorrect to add them vectorially. -- Paul Nicholson, --
Maintainer Paul Nicholson, paul@abelian.demon.co.uk.