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

Re: design for new capacitor



Original poster: "D.C. Cox" <resonance@xxxxxxxxxx>


One inch total creepage path is not enough even under oil. Also, many experimenters handle these dielectric materials with their "bare hands" which of course are covered in sodium based (conductive) oil. By the time its in the oil the sodium provides a conductive path for rapid RF tracking of the HV across the semi-conductive surface and failure soon ensues.

If you are making your own you absolutely need to wear rubber gloves and use good engineering --- at RF frequencies leave at least 2 inches on each side of the dielectric material for a min. of 4 inch surface tracking distance. The sharp edges of the plates are usually protected with a small rolled up copper tube forming a less sharp edge.

This is how Maxwell does is --- and without failures. They also use special silicone and other types of oil for best RF tracking resistance.

Dr. Resonance
----- Original Message ----- From: "Tesla list" <tesla@xxxxxxxxxx>
To: <tesla@xxxxxxxxxx>
Sent: Wednesday, March 01, 2006 4:35 PM
Subject: RE: design for new capacitor


Original poster: "Mark Dunn" <mdunn@xxxxxxxxxxxx>


Adriano:

I tried the same thing with 8" X 12" X.032" thick FR-4 sheets double
sided with copper. I stripped off the outer 1/2" of copper along all 4
edges and both sides to keep prevent cross-over.  As you noted, each
sheet is a capacitor and I alternated the voltage on each board so that
"like" voltage copper surfaces would face each other.

Submerged in oil (5) sheets gave me 13.84 nF.  Unit worked well for a
Tesla Coil.  In air, the arc ran across the surface of the FR-4 around
the edge and back across the other surface.

My final copper area was 7" X 11" less an area to pass brass bolts
through to connect the thing together.  This bolting gets difficult to
explain.  I can send you a drawing of both sides of the cap if you are
interested.
Note: Most "List" subscribers hate Parallel Plate Caps!

Mark


>Original poster: "Dr. Adriano Mollica" <adriano.mollica@xxxxxxxxxxx>

>Hello list,

>i just bought on ebay a certain quantity of douple faces copper fiber
glass plate for very >cheap.

>i will like to use them to make a capacitor, and i believe that it is
possible.
>But i'm worring about the design. infact, every single plate is already
a little
>capacitor "as is",  and if  i'm right , the thickness of the fiberglass
will be 1.6 mm,
>that is very big (this will lead to a high punture voltage resistance ,
but low capacity). >So , how should i connect the plates one to another
in parallel cinfiguration trying to
>avoid any "corona" , and external spark? and so on, considering that
the plates will be in >a number of 20-40 and i cannot overlap them due
the double face!!!

Any ideas?

bye bye


----- Original Message -----
From: "Tesla list" <tesla@xxxxxxxxxx>
To: <tesla@xxxxxxxxxx>
Sent: Wednesday, March 01, 2006 5:53 AM
Subject: Re: rolled cap blowew out


 > Original poster: Chris Estes <estescc@xxxxxxxxxxxx>
 >
 >
 >
 >
 > Tesla list wrote:
 >
 >
 >  >Original poster: Adriano.Mollica@xxxxxxxxxxx
 >  >
 >  >
 >  >Hello list,
 >  >
 >  >after the last attempt to start the TC using lithuanian capacitors,
>  >that used to explod very quick and very often, i made a homemade  >
>rolled polyethylene cap.  >  >The dielectric thickness was 0.4 mm,
since 1 mm should resist 60kv, i  >  >thought that 0.4mm should handle
about 25kv.  >  >After the rolling up of the materials, i dip the roll
into a tube  >  >filled with synthetic engine motor oil, and by a vacuum
pump i  >  >removed all the air inside.  >  >  >  >unluckly, after few
successful sparks, the capacitor stopped working  >  >and started to
make a noise (probably an internal arc).  >  >i'm pretty disappointe
'cause i spend around 40 euros in materials,  >  >and many hours of
work!  >  >  >  >how can i built a cheap, reliable, and FAST pulse
discharge capacitor?  >  >  >  >i'm thinking about planar caps in air (
at least is the cheapest  >  >dielectric i know.!  >  >  >  >any
suggestion?  >  >  >  >how much is the dielectric costant of the air,
and its puncture voltage?  >  >37 nF cap wil probably be very large?  >
>thanks!  >  >  >  >  >  > I've just finished an interesting design
thats yeilded about 6.7 nf  > with I'm guessing 50 Kv strength. Its a
stacked plate cap that uses  > corrugated cardboard and mineral oil for
the dielectric. I place 2  > pieces of cardboard between each plate to
give about 250 mil (3/4 cm)  > spacing. All the pieces of cardboard are
oriented so that when looking  > down on the stack you can see through
the corrugation if that makes  > sense. then i place the stack in a
container and fill it with oil. The  > orientation of the cardboard lets
the air bubbles flow up and out quite  > nicely. The most expensive
component is all the oil and, I believe,  > they will have a 'self
healing' quality. If theres a puncture, all I  > need to do is shake it
around to jostle the oil a bit. The dimensions  > are about 20cm by 35
cm by 35 cm for, as I said, about 6.7 nf. If you  > wanted more farads
in less space then you could probably just use one  > layer of cardboard
but I just needed about 7 nf for my NST resonant  > condition and wanted
to be sure about the dielectric strength. I  > haven't tried this out
yet as I'm still building the rest of my coil  > but I can't see any
serious flaw in my design so far. Good luck and  > happy coiling.  >  >