Creepage increases streamer length - food for thought <was: Trying this at home>

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

Original poster: "Hall, Dale by way of Terry Fritz <twftesla-at-qwest-dot-net>" <Dale.Hall-at-trw-dot-com>

Sue, Scott & List,

We may all be utilizing the "creepage" phenomena to produce corona and
streamers !

  just that most prefer the 

    gas - gas (Air) XYZ dielectric dynamic boundaries commonly available
however inconsistent
          rather than the 

    gas - solid XY axis dielectric boundary (more consistent but with usage
constraints)

          some consider this "cheating" 
          but isn't Air gas 'control' necessary to be fair (Not cheat) ?

          Or must we design TC's robust enough to 
             perform well into widely variant loads, 
             i.e. our typical uncontrolled gas mixture - Air ?

I have used the "solid - gas dielectric boundary" (creepage) phenomena 
  for many years with my single shot battery powered DC TC's to achieve
  consistent 38" Air discharges (blue-white) over glass or hard plastic at
4.9J per bang.

I chose this "air discharge over a solid dielectric boundary"

  as method of streamer discharge CONTROL, 

    (concentrates charge at the dielectric boundary)
    (influencing likely path which increases streamer length compared to
gas only)

  to bring consistency to streamer observation w/repeatable lengths for
electrical evaluation.

The dynamics of air discharge presents extreme variance (turbulence, Air
gas mixture),
  challenging the observer/researcher's control. 


Some assumptions:
1.) Use low bangs per second (~<1 to ~10)
    to avoid surface heat build up of the solid to not evolve its material
as another gas
    (SiO2 doesn't break down easily to participate but is fragile)

    (note any carbon tracking on SiO2 Sec form is due to wire insulation)
    (enamel, polyU, insects, Not the glass !)

2.) The desired arc/streamer is produced within and from the 
    gas constituent "Air" portion above the solid boundary 
    evidenced visually by a bright blue-white arc/streamer.

Other visual arc/streamer color variations indicate 
 participation of the solid dielectric (yellow, orange, red, etc.)
 altering the control value (but not aesthetics of arc viewing)
 or a gas concentration of other than normal N2 O2 + derivatives
    Air gas type (neon = orange, etc.) 

Here are some thoughts as to an enabling mechanism:

The boundary between any two dielectrics (especially great as solid to gas)
presents 

a decreased reactance Xc = 1/(2PiFC) via a virtual capacitance

at any dielectric boundary 

creating a preferential charge path

  (compared to smaller magnitude differential Z variations in/from 
   our typical case - gas air turbulence/gas combination around the topload)

initially at the PRI / SEC ring-up frequency = ~200Khz in the pre-corona stage,

then at much higher corona oscillatory frequencies as corona develops,

  (higher freq lower Xc)
  (exposing more preferentially lower resistance paths - LEADERS ??)

leading to higher current streamer discharge at yet different oscillation
frequencies

  (I've scoped corona/streamers between ~1MHz to ~30MHz)

supplied (fed) by the topload (charged) to ground capacitance.

:) :) :) :) :) :) :) :) :) :) :) :) :) :) :) :) :) :) 

Interesting speculation: the whole process of leader growth

may be driven by this mechanism by the dielectric dynamics of N2 O2 O3 etc. 

within the topload to ground, gas dielectric, capacitive environment.

It is not this simple, streamers also possess inductance and complex paths.

Anyone up for a potential doctoral subject 
 to develop, test with controls and present
 the 1st order LCR Air Discharge Model Mechanism ?

Regards, Dale
Redondo Beach, CA

-----Original Message-----
From: Tesla list [mailto:tesla-at-pupman-dot-com]
Sent: Tuesday, July 03, 2001 2:23 PM
Original poster: "S.Gaeta by way of Terry Fritz <twftesla-at-qwest-dot-net>"
<sgtporky-at-prodigy-dot-net>
Sent: Monday, July 02, 2001 8:29 PM
Subject: Re: Along the lines of "Trying this at home"

*snip*
> my coil puts out 12' ( foot) long streamers....  I wouldnt touch it with a
> 20' fiberglass pole
> ...   in a previous test I placed a 4' long glass tube near the toriod of
> my small coil ( which
> puts out 1 - 4' long streamers) and at low output ( normally 1.5'
> streamers) the streamer
> traveled across the tube and over to the grounded wire. Goes to show how
> "non insulative" glass
> is at higher voltages...
> Scot D

Hi Scot,
In the electronics industry this is known as "creepage" and does not reflect
that the material is a poor insulator. This happens when two different
dielectrics are in contact with one another ie: glass and air. The flashover
tends to travel along the surface of the insulator and happens at a much
lower breakdown voltage than what is rated for either of both dielectric
materials. This is why they put ridges on high voltage insulators. It
increases the creepage distance as well as the breakdown voltage.

You can actually increase the streamer length of your coil, using this
phenomina to your advantage! Next time, see what happens when you use an 8
foot fiberglass or PVC pipe. Don't do this indoors, the fumes are awful, and
use only scrap material, for you will eventully have a carbon track on your
pipe, deeming it no longer an insulator anyway!

Cheers, Sue