From: Paul
Date: Sun, 05 May 2002 10:13:28 +0100
Subject: Re: [TSSP] Secondary voltage stress factor
Looking at the voltage 'red lines' in pn040502 we can compare the height along the coil at which the voltage reaches, say 100kV. Roughly, k=0.10: height 25% k=0.20: height 20% k=0.27: height 15% So we have some opportunity for increased likelyhood of strikes from the coil outwards. The most significan thing is that on going from k=0.1 to k=0.27, the primary has to be placed closer to the coil, and quite a bit higher up. Even without the increased voltages, the primary proximity alone would be enough to cause problems. Another factor that I want to consider is what happens when a discharge occurs from the topload. The sudden loss of charge from the topload appears to the secondary as a step transient applied to the top, and this transient would be expected to propagate down the coil. I decided to model this for the case of the in-tune k=0.20 coil. The waveforms for the first 10uS are shown in http://www.abelian.demon.co.uk/tssp/pn040502/tfsm1-h1d.wave.gif The discharge occurs at 7.1uS - just as the topvolts reaches its peak. One very odd thing is that the resulting step is so small. This may be a problem with the model - I expected a step going all the way to zero volts and then recovering, but we see here only 1/6th of the topvolts is lost. The animation for this is in http://www.abelian.demon.co.uk/tssp/pn040502/tfsm1-h1d.anim.gif and you can see a strong pulse travel down the coil to the base, to be reflected back to the top. As you can see from the voltage red line, there is a considerable percentage increase in the voltage near the bottom of the coil. If the model is a little broken and that pulse is in fact much bigger than shown, there is a 'potential' for trouble if you excuse the pun. Please can we have some captured traces of topvolts in a discharging TC so that I can test this part of the model? Meanwhile, I'll try to figure out if that step size is wrong. -- Paul Nicholson, --
Maintainer Paul Nicholson, paul@abelian.demon.co.uk.