Re: Designing BIG Secondary Coil

["Tesla list" <tesla-at-pupman-dot-com>]

Original poster: "Paul Nicholson by way of Terry Fritz <twftesla-at-qwest-dot-net>" <paul-at-abelian.demon.co.uk>

I'll add my bit to this,
> the best secondary I have wound to date... thickwall sewer pipe.

When comparing resonant frequencies of secondaries with accurately
predicted values, if we don't take account of tube material
dielectrics, the error in Fres can be as much as several percent when
the wall thickness becomes a noticeable fraction of the coil radius. 

It's a bit tricky calculating the E flux through the material but we
can put an upper limit on the typical loss just on the basis of the
frequency offset.  

Say the actual Fres is 10% lower than expected, so we have an extra
20% E-flux that we attribute to the tube material.  That means we can
say that 20% of the circulating energy passes through the material.

(These are rather higher values than would be found in typical coil,
so we are erring on the side of caution with these estimates).

The fraction of energy lost per cycle is then 20% times the loss
factor of the material, and we can put an upper limit of the Q
factor due to this, eg

 Material     Loss factor     Loss factor    Q factor
              (material)      (effective)    (upper limit)
  PTFE          0.0002         0.00004      160,000
  PE            0.0005         0.0001        63,000 
  PVC           0.01           0.002          3,100
  ABS           0.03           0.006          1,050
 
These figures suggest that loss in typical tube materials is not a
major issue.  Cellulose/paper/cardboard seems to be a little more lossy
than ABS, but not much.  I've noticed when looking in tables that some
of the stuff that can be mixed in with fibre type materials can have
very high loss factors, which may support some of the doubts about
sonotube.  I'd stay with regular plastic materials and avoid any sort
of composite or fibre. 

I built a BIG coil for max Q factor, thick wire, etc.  The Q factor
can be quite good, on a good day!  The measured Q comes in around
100-300 according to mood.  If I allow for a measured ground
circuit resistance, the intrinsic Q of the coil is around 750, and
the value predicted on AC wire resistance alone is around 1400.  This
is 2mm diam wire, 760 turns, 0.6m diameter by 1.6m length on PE former,
resonating at 92kHz and loaded down to around 65kHz.  I suspect we
reach a limit where the coupling to the environment around an un-
shielded coil like a TC becomes the limiting factor.

When it's not carrying voltage it sits in the barn and makes a nice
VLF antenna with a low frequency corner of 15 Hertz.  In this
application, ground cct resistance is not involved, nor is proximity
loss, so the full Ldc/Rdc ratio of the coil is exploited.

 http://www.abelian.demon.co.uk/tmp/p1.jpg

Coilers and Lowfers both get to wind some BIG coils!
--
Paul Nicholson
--