From: Paul
Date: Mon, 22 Jan 2001 19:12:13 +0000
Subject: Re: [TSSP] ready for some measurments
Hi All,
Once again I've got behind with my emails, so apologies for late
replies.
Bob wrote:
> am about ready to do some measurements on my two secondary.
> I was thinking of measuring the secondary placed on a milk crate
> above a concrete floor with a carpet on top. The room is about 10'
> x > 10' x 8' with a sloping ceiling up to 12'
Are you able to lay down some tinfoil on top of the carpet as a
ground plane, out to a radius of a metre say? This should make the
Q factor much less dependent on factors external to the coil.
For example, my coil has a wound AC resistance of around 25 ohms,
and I've a Q getting on for 700 under favourable conditions outside.
On Saturday I brought the coil in from it's storage place in a spare
chicken shed (the shed is in danger of collapsing) and put it up in
the living room. The coil's far too big to use for indoor
measurements, but just for fun I measured the Q at a mere 70 above the
carpet/stone floor. Adding a tinfoil groundplane of modest radius,
say about 1/4 of the coil height, sent the Q up to 95. I then took
loads of foil and hung it from the walls etc, all around the coil to
form a kind of curtain - a partial one - I had about 30% coverage of
the walls. The Q was then 200. This goes to show that if the E-field
is left to find its own way through the walls and floor down to the
earth below, the resulting loss in the earth return path completely
dominates the Q factor. This makes it hard to get repeatable Q
factors and also hinders accurate prediction of the resonant
frequencies. There's not a hope of making measurements on my coil
indoors - it only just fits in the room, see
http://www.abelian.demon.co.uk/tmp/pics.html
but with a 21" secondary, laying out sufficient foil with the aim of
attracting directly to it the bulk of the external E-field would seem
to be well worth the effort. I've had very little success modeling
coils which don't have a well defined ground plane.
> how should I ground everything...
No special arrangements necessary if you're using a ground plane.
The instruments remain grounded through the mains supply earth and the
tinfoil is earthed to the instrument ground through the screen of the
coax which supplies the base drive.
> standard CB car aerial...Will this be OK as the scope input?
I use a similar aerial. You might need a resistor to ground. Without
this the pickup of 50Hz can be troublesome.
> Should I try and lower the input impedance of the signal genny?
It's output resistance needs to be small, say 1%, compared with the
total AC resistance of the coil and ground return path. You might
need to pad the output down to 5 ohms or less, ie
| coil
|
o-------[50 ohms]------------------------------|
|
generator [5 ohms]
|
o-------------------------------------
| ground plane
-----------------------------
/ / / / / / / / / / / / / / /
I'd usually recommend a lower shunt resistance but your small drive
voltage doesn't leave much room to manoeuvre.
To begin with the first thing to do is to get the coil set up in a
repeatable test position and measure the first three resonant
frequencies f1, f3, and f5, and the Q factor at f1 - all without a
topload. If you can send me details of the secondary, and let me know
how far the start of the winding is above ground, I'll setup a model
of the coil.
> Does anyone have the list archives stored anywhere
I keep an archive of tssp traffic, although I haven't raised enough
enthusiasm to make it available as a searchable web archive. It's
always hard piecing things together from a series of list postings and
I tend to think that it's better to make useful info available as web
pages rather than expect a reader to sift through meandering list
discussions.
Cheers All,
--
Paul Nicholson,
Manchester, UK.
--
Maintainer Paul Nicholson, paul@abelian.demon.co.uk.