[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: Essex magnet wire



Original poster: "Dr. Duncan Cadd by way of Terry Fritz <twftesla-at-uswest-dot-net>" <dunckx-at-freeuk-dot-com>

Hi Gary!

>Thanks for the suggestion about shunt resistance.  My thought is that
since
>the 1994 and 1997 Heavy Soderon coatings were different, the problem
was not
>just outside the dielectric. If water actually soaked into the
dielectric,
>it could easily explain the observed losses.  If the surface became
>conductive, there should be corona in spots, with radio noise (like
power
>line insulators in damp and dirty conditions) which I did not
observe.

That's a fair comment.  However, conductivity is a relative thing, the
current may be too low to give these symptoms.  Also, if the two wire
coatings were different, maybe also their polar constituents are
different which would affect adsorption.  I'm willing to concede it
may in fact make little odds whether the water is absorbed or
adsorbed, the point is you've measured a real difference due to
humidity, the problem as always is finding the explanation . . .

>But is there some other type of test I could run which would
differentiate
>between the effects?


Right off I can't come up with anything, but . . .

I've had another think (hazardous, as Marc has found ;-)

It could be that you don't need conductivity to account for all the
loss/impedance change, though as Malcolm comments I'm sure it
contributes something.  There is an alternative/addition I should have
thought of before.  The dielectric loss, tan delta, power factor (or
whatever) of water may be enough to account for this.  Think of
microwave ovens, or rf timber-dryers.  The reason they work is because
of the losses associated with moving water molecules around i.e.
dielectric loss.  If the wire coatings adsorb or absorb water, the
point is there will be water molecules in the closest proximity to a
region where the electric field is undergoing extremely vigorous
variation, the dV/dt is going to be big!  The permanent dipole on the
water molecule will not only glue it to the surface of a wire with a
polar insulation coating, but also ensure that it tries to follow
changes in electric field, irrespective of whether the molecule is
adsorbed or absorbed.  There are three basic loss mechanisms possible
for a material in a varying electric field:

1) Movement of electrons in the material
2) Movement of ions
3) Dipole reorientation

(1) is very low loss because electrons move rapidly and consume little
energy in moving.  If this is the case, the dielectric constant K and
the square of the refractive index of the material to visible light
are equal.  So it is for polystyrene.

(2) and (3) are very lossy because the masses being moved are heavy in
comparison with electrons.  In these cases, K is not equal to the
ri^2.

So with water K = 80 (very variable depending on impurities) and the
ri = 1,33 - i.e. the dielectic loss is via a lossy mechanism, in this
case dipole reorientation.  The dissipation factor is around 0,05 at
1Mc/s.  That's 100x worse than a good TC primary cap dielectric, and
means that 1/20 of the energy which is applied in one cycle disappears
as heat (dissipation factor = energy dissipated/energy applied, per
cycle).  Given the field intensity variations at the surface of the
secondary wire, maybe even a thin layer of water will result in
something measurable.  It amounts to an extra load which the primary
has to drive, perhaps an extra loss in the secondary self-C (with
apologies to Paul!).

Perhaps that helps account for the impedance changes, I can't offhand
think of any other mechanisms whereby water would change anything
other than conductivity or dielectric loss.

Dunckx