[Home][2015 Index] Re: [TCML] Glass Plate Capacitor [Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: [TCML] Glass Plate Capacitor



Hi Alex,

Sounds like an interesting coil!

Glass plate or glass bottle capacitors have a long history in HV applications. Tesla himself used heavy glass water bottle immersed in oil for his large Colorado Springs system. Soda-Lime window glass has excellent dielectric strength (~13 kV/mm) and a relatively high dielectric constant (between 7 and 8). Unfortunately, soda lime glass is also fairly lossy when used as a dielectric at Tesla coil frequencies. Its loss tangent ranges from about 0.02 - 0.04, making it about 300-600 times as lossy as modern HV capacitors made using Low Density Polyethylene (LDPE) or Polypropylene film. Its dielectric loss is actually about 50% higher than Mylar (PET), which is not recommended for Tesla Coil use because of its excessive losses. Also, glass becomes more electrically conductive when heated via ionic conduction. This can actually lead to overheating, thermal runaway, and dielectric failure, particularly for glass capacitors used in higher power systems.

Single-strength glass plates are about 3/32" thick, or about 2.4mm. The transformer will deliver a nominal peak voltage of 7kV, and during operation, your capacitors will see a equivalent stress of ~14 kV (RF peak-to-peak) during operation. Using 13 kV/mm, the DC strength of each plate would be about 31 kV, so you should be OK (at least for short-term operation).

However, if your plate capacitor is configured as a tightly-packed sandwich, the interior plates will become hotter than the outer ones due to the longer thermal path to ambient air. Assuming your HV transformer delivers about 500 watts (depending on your ballast inductor), your capacitor may need to dissipate an average of 10-20 watts of heat in operation - and possibly twice this if you're really pushing your transformer. If the plates were all in free air, convection cooling would likely be sufficient to prevent overheating. However, extended operation with the plates sandwiched together may cause the interior plates to overheat and fail.

Some other changes you may want to consider:
1. Make sure you have a safety gap across your rotary gap to prevent any gap misfires from allowing resonant rise (at mains frequency) from overvolting your capacitor or HV transformer.

2. Consider increasing the spacing between primary and secondary windings or their relative position (i.e., move the primary so that it is lower relative to the secondary). Either will reduce the coupling coefficient(k), but will likely require retuning the system. A lower k is actually desirable with spark gap coils, which typically operate best with k in the range of 0.1 - 0.2. This will likely fix your primary-secondary breakdown problems as well.

3. Consider adding either a small corona ring (1/2" copper or aluminum tubing) above the top of your secondary winding to prevent the winding and coil form from being damaged by corona and spark tracking. This will still allow you to keep the "look" of your coil.

4. Add a larger toroidal topload to further improve performance. However, this will significantly change your coil's appearance.

BTW, the first Tesla Coil that I built (when I was a pup way back in the late 1950's) used double-strength glass plate capacitors driven from a 15 kV 30 mA NST. The caps generated horrendous amounts of ozone, and lots of noise from the aluminum foil rattling against the glass as they charged and discharged. They also generated the most beautiful blue glow around the edges of the foil plates, with lots of little longer bluish-white spark tendrils. Unfortunately, most plates eventually failed, perhaps at defects or bubbles in the glass. The breakdown and arc-through usually fractured the plate.

If you get a chance, please post some links to images of your system. Also, do you remember the title of the book where the coil was described?

Play safely and best wishes,

Bert
--
Bert Hickman
Stoneridge Engineering, LLC
http://www.capturedlightning.com
***********************************************************************
World's source for "Captured Lightning" Lichtenberg Figure sculptures,
magnetically "shrunken" coins, and scarce/out of print technical books
***********************************************************************

b alex pettit jr via Tesla wrote:
Hello All, I have been in the process of getting an old build
operational again.I constructed this RSGTC back in the mid '80s based
on info from a book published in 1920.  It is of classic design in
thatthe Secondary is wide and short  by current day designs ( 8" dia
x 15" hi ) and has no top load - just a small 1"  brass ball on a
short post. It creates a very large brush discharge from the top turn
and equal length sparks from the ball.It looks like a true antique as
the components have nicely finished wood bases and boxes. It was
built using a surplus military grade 5KV 90 ma transformer and is
throttled by a tapped inductor of my design on the AC input.
Thecapacitor is a 2" stack of  'single strength' window glass and
aluminum foil whichwas immersed in hot beeswax such that the edges of
the glasspanes are sealed ..  The foil comes to within 1/2" of the
edges of the glass. The cap measures  0.025 uF.  Input 5KV max.  ARSG
runs at 360 BpS. The coil resonance measures 290 KHz and this agrees
with the JavaTC analysis. Coupling 0.3 k I have Never attempted full
power operation as there were always pri to sec arc-overs.I finally
fixed this ( after 30 years !  ) by sleeving the upper turn of the
primary with plastic tubing. Question : Whats the probability that
the Capacitor withstand full 5 KV transformer output ?
Regards,AlexOrlando FLKK4VB

_______________________________________________ Tesla mailing list
Tesla@xxxxxxxxxx http://www.pupman.com/mailman/listinfo/tesla




_______________________________________________
Tesla mailing list
Tesla@xxxxxxxxxx
http://www.pupman.com/mailman/listinfo/tesla