From: Paul
Date: Wed, 21 Feb 2001 20:44:10 +0000
Subject: Re: [TSSP] Problems with high elevation/small radius
Metlicka Marc wrote:
> ... since my top turn curls into the center of the bare coil
> for topload mounting, is this causing any added capacitance?
Good point. I would have guessed very little, and before seeing
your figures, I would have been comfortable with ignoring it.
> i first "QUICKLY" hooked my gen. radio bridge up to the
> coils 2 and 3, over my ground plane on the milk crate.
> i found that with the wire curled to top mount screw (nylon)
> it did give a .8nf and .5nf higher c then with the wire extending
> straight up.
OK, I guess you mean pF rather than nF. These capacitances you've
measured are more than I would have guessed at.
> i then hooked coil 2 up for f1 readings. today, with wire
> curled to screw, the f1 is 191.5 khz, with wire straight up,
> 193.3 khz.
Hmm, that's quite a difference, just under 1%. The capacitance
difference will be the square of that, and comes out to 1.9%,
and since your equivalent shunt capacitance is 27.2 pF, thats
an increase of just over 0.5pF, which nicely agrees with the
size of your direct C measurements.
Thanks Marc, that's an enlightening observation of just how little
it takes to introduce a significant disturbance of the resonant
frequency. If this extra C was presented to a smaller coil, of the
size at which we are having trouble with, it would reduce the
frequency by 5%. We've also seen similar shifts to the frequency
of Terry's small coil when the terminal hardware is fitted.
The wire curls in to the center? Roughly how long is this extra
piece of wire? Does it curl in slowly, making more than one
additional turn, or does it curl in rapidly?
BTW, this means that the comparison for coil #2, un-probed,
mm2: bare d=0.108m h/d=9.97 sr=0.81 b/h=0.31 turns=1700
measured modeled error
f1 276.9kHz 273.8kHz -1.1%
f3 711.8kHz 692.7kHz -2.7%
is now out of date, since your true Fres will be around 1.7% higher
than 276.9 kHz, around 281.6 kHz, so the model will be -2.8% in
error, which is a bit high. We might need to take a closer look
at #2 to figure out this discrepancy.
Cheers,
--
Paul Nicholson,
Manchester, UK.
--
Maintainer Paul Nicholson, paul@abelian.demon.co.uk.