Re: HV Cables

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

Original poster: "Jim Lux" <jimlux-at-earthlink-dot-net> 


----- Original Message -----
From: "Tesla list" <tesla-at-pupman-dot-com>
To: <tesla-at-pupman-dot-com>
Sent: Sunday, November 09, 2003 6:41 AM
Subject: HV Cables


 > Original poster: "Ian McLean" <ianmm-at-optusnet-dot-com.au>
 >
 > Hi all,
 >
 > I am wondering what is the general consensus out there amongst coilers as
to
 > what makes the best cable for HV.

Depends on application and voltage...
 >
 > For some time, I have been using RG-8 coaxial (Thick Ethernet), with the
 > outer braided shield removed.

Thick Ethernet cable is NOT RG-8, nor RG-213, although it is of similar
size.

This has excellent dialectric strength and is
 > rated for more than 35kV.

It is NOT rated for 35 kV.  It does in fact, hold off 35 kV in actual use,
but that's a big difference from rating.  The various half inch family
coaxes are rated for 5kV AC RMS.

I believe it is, or has been in the past, been
 > widely used as HV transmission cable?  I managed to get a reasonable
length
 > of used cable from an ex-employee when they moved their building trunk
cable
 > over from this to fibre.

Indeed, miles of RG-8, RG-11, RG-213, RG-214, etc. have been used for HV
cable.  I use it WITH the shield still on (better control of the field in
the dielectric and the added safety of having a grounded metal shield
between the HV and other stuff, so that if there IS a breakdown (and there
will be), it goes to the shield, not to something else.

 >
 > However, the thick, heavy, yellow RG-8 coax is becoming very difficult to
 > obtain and expensive, and I am wondering about alternatives.  I have
looked
 > through Tesla archives, I did not find much, except that someone mentioned
 > the use of RG213 coaxial, but according to the spec's, this cable is only
 > rated for 3.7kV RMS.  It is also not solid core, but multi-stranded.


 >
 > I have been toying with two ideas:
 > 1) Just buy some Neon cable.
 > 2) Using thick guage enamelled copper, with my own insulation around it (a
 > few layers of heatshrink maybe).
 >
 > I believe, from my research, that for HV, esp. at relatively low current,
 > solid copper conductors are best (comments welcomed).  I have tried
 > multi-stranded copper (high current DC) before for 15kV-at-60mA (NST ouput),
 > but the HV tends to "buzz" inside the cable, and you lose most of the
power
 > to what I believe must be coronal losses within the strands of the cable.

The best HV wire is wire made for HV from the big mfrs like Rowe-Talley,
Caton, Tyco, Belden, Alpha, etc.

Home brew HV wire (bare wire with your insulation wrapped around it) will
have corona problems, unless the conductor is very large in diameter.

Solid conductors work better than stranded, because the radius of curvature
is greater.   Guideline for wire radius to avoid corona is 70 kV/inch
minimum, a 2:1 margin is better. This would imply that for NST voltages
(20kV), you'd want a wire radius of, say, 1/4" (or 1/2" diameter)... that's
pretty big, so you make do with a smaller conductor, and trust in the mfr to
have done a good job encapsulating it.

Real HV wire has a semiconducting layer that the inner conductor is imbedded
in to provide a larger apparent diameter for the conductor to reduce corona
inside the wire. Field management is very important.  Most failures occur in
small voids in the insulation, where a small discharge occurs, destroying
the insulation.  Then the void is a bit bigger, and the field concentrates
even more, causing yet another small breakdown, etc.

Coax works fairly well as HV cable, although not actually rated for it,
because the surrounding shield makes the field in the dielectric quite
uniform, and because they have to control the number and kind of voids in
the dielectric to insure that the impedance is stable and well defined.
(obviously, cheap coax is worse than good quality).  However, new fancy low
loss coax with air or foam dielectrics are actually worse as HV cable,
because non-homogenous dielectric results in field concentrations.