TSSP: List Archives

From: Paul
Date: Sun, 10 Jun 2001 08:39:18 +0100
Subject: Re: [TSSP] Interesting article on Medhurst, Wheeler, modeling,etc.

Morning Kurt,

Kurt Schraner wrote:

> Wire diameter, bare Cu   1.0 mm
> Wire diameter, insulated 1.2 mm (cotton spun)
> Diameter of former d = 0.383 m
> Length of former       0.501 m
> Former's wall thickness 8.0 mm (Material: hard paper, ~ pertinax)
> Winding length    h = ~ 0.44 m (cotton insulation is compressible!)

> Number of turns     N = 346 (counted manually)
> Aspect ratio h/d = 0.44/0.384 = 1.15
> Spacing ratio sr = 346*1/440 = ~ 0.786
> Height above ground b = ? m (not yet available)

I'll assume it is at 0.5m

> Electric data (Multimeter & LCR-Meter):
> DC Resistance R = 9.2 Ohm
> Inductance L = 28.12 mH @ 1 kHz   Q=17.3
> Inductance L = 28.23 mH @ 120 Hz  Q=2.29
 
Thanks for the info.  At h/d=1.15 this coil is well worth a closer look.
I'll set about predicting the response.  Let's not worry about running
it outside for now - since we are most interested in the higher mode
frequencies, f3, 7, 9, etc.  I'm coming to understand that these are
much less sensitive to the surroundings.

> I'm aware, the wall thickness of the former is really fat.

Yes, it is.  For that reason it may help to determine whether we must
allow for the material.  Indications so far are yes, so we expect to
get a similar pattern of errors at the higher modes.

> My small problems are:
> - enough free space (...why this offer, for free access to an empty
> industrial-hall is never arriving ;-)) ?
> - My low-Z Amp (=Terry's design) is in a steel-metal-box of about
> 24x16x(h=)12cm: how should I "hide it" from the base feed of the coil
> (Dig a hole into the ground? Move it away by 2m ?).

Yes, just keep it a couple of metres away.

>          a)                         b)
>      o       o                   o       o
>      o       o                   o       o
>      o       o                   o       o
>      o       o                   o       o
>      o---.   o                   o       o
>          ¦                       ¦       
>          ¦ 0.4mm Wire            ¦ 0.4mm Wire
>          ¦                       ¦ 
>          .---------------        .------------------
>             Coax?   Low Z Gen.      Thin Wire?   Low Z Gen.
>          .---------------        .------------------ 
>          ¦                       ¦
> --------------------------   -------------------- Ground

>  
> While b) is easier for me, I think your recommendation was more like a
) 

(a) would be my recommendation if you had a 50 ohm or above signal
source, because the coax adds a predictable amount of shunt capacitance
to the feed. However, with a very low Z source such as Terry's, this
line  capacitance can safely be neglected, so (b) is fine.   For an
elevated coil, eg base height greater than coil length, I think it's
important to approach the coil with a thin wire riser, as you have
shown in (b), to minimise the disturbance to the end capacitance. 
When the base is closer to the ground plane, the much larger Cext
which results would comfortably swamp the slight wire capacitance.
So use (b).
 
> Ground will be a wire mesh (like chicken wire),
>  with ~ quadratic meshs of sidelength 13mm.

That's fine - I think!  This whole business of ground planes needs some
investigation in order to find out whether it really matters.

> I'm waiting the next non-rainy day, to perform the measurements
> outdoors.

Yes, outdoor measurements are better for one reason: I don't know how
well the program accounts for walls and ceiling.  If the predictions
are in error, there is always the suggestion that the Cext might be
wrong.  Outside - there is no such argument and the results are more
equivocal.

Cheers,
--
Paul Nicholson,
Manchester, UK.
--


Maintainer Paul Nicholson, paul@abelian.demon.co.uk.