From: Paul
Date: Sat, 23 Jun 2001 15:50:34 +0100
Subject: Re: [TSSP] shortcoil measurements
Finn Hammer wrote: > I have remeasured the short coil (fh1) with the new counter... Thanks Finn. Can you confirm that the start of the winding is 1.0 m above the ground plane, ie the same indoor setup that you had before? > F1 368.78 > F3 868.03 > F5 1278.25 > F7 1692.12 > F9 2104.17 This is a particularly good set of results for assessing attempts to calculate the contribution of the dielectric. The effect of 5.5mm of PVC is quite dramatic. A similar effect is seen with Kurt's Sk-38cm coil, but the material is less well defined and is too lossy for the higher modes. Notice how quickly the mode frequencies settle into a uniform interval, the value of which is almost entirely determined by Cint, which in turn is dominated by the PVC dielectric. The f1, which really wanted to be at about 460 kHz, has been pulled down significantly by the external capacitance, but other than this, Cext is relegated to the role of a minor correction for the higher mode frequencies. To demonstrate this, it would be a good idea to remeasure this coil at say 10cm base height, so that Cext is increased somewhat. Only f1 should change significantly, with f5 and above remaining pretty much the same. Meanwhile, I'm faced with bringing the laplace solver out of retirement in order to do a proper job of accounting for the former dielectric. So far we've seen that a rough first-order correction gives about the right amount of change, but we need to do a proper calculation to demonstrate that we really understand what's going on when the former thickness is a significant fraction of a higher mode quarter-wavelength. Since the short-range Cint under these conditions is relatively independent of the surroundings, I think we can try to get away with a one-off use of the laplace solver to derive a general correction formula, rather than permanently building the laplace multigrid into the model. > Hope they are of some use. Undoubtably are, although I think it may take a few months to pin this stuff down. Perhaps you can let us know whether the higher mode frequencies are indeed fairly insensitive to the surroundings of the coil? > With regard to the other coil, the one for the current profile. I will > make it very big, perhaps on one large barrel, 600 mm dia. Would need a lot of copper for 60cm diameter. I'm still thinking about what we can do with the 20cm diameter, depends really on what we can find to register the current. If small enough, then a small coil would suffice. > You talked about inserting a current censing device, small and self > contained, into the series tapping points. The only current censing > device I can think of is a resistor, what did you have in mind? I'm open to suggestions. It only needs to indicate a current threshold, so something which lights up at a certain current would do. A resistor sounds like a good start, just need to think of something to put across it. Cheers, -- Paul Nicholson, Manchester, UK. --
Maintainer Paul Nicholson, paul@abelian.demon.co.uk.