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

Re: Re: Oil Dielectric Properties



Original poster: "resonance" <resonance@xxxxxxxxxxxx>



If you make this capacitor be advised that "rf tracking" across in insulator (your spacers) will be a major issue to deal with. Typical separation values of 5-6 times the breakdown spacing values will render the 1/d factor so large the capacitance factor will be considerably reduced by 500-600%.

RF tracking in pulse caps is a nightmare problem for the cap engineers. Trichel pulses form off the edge of the foil and produce heating effects on the dielectric.

Dr. Resonance


Original poster: "Anthony R. Mollner" <penny831@xxxxxxxxxxxxx>

Now that's a capacitor! I would like to see a chart on the direct dielectric
properties of oils to find what kind of spacing would be best.

Which properties are you interested in... google and find a Shell Diala AX data sheet
Breakdown strength varies with contaminants, but is quite high (150-200 kV/cm)
Epsilon varies with the kind of oil, but 2 or 3 is typical for "transformer oil" and 5-6 for "castor oil" (which is why castor oil is used in some pulse caps...)

http://home.earthlink.net/~jimlux/hv/insulliq.htm

 Anyone have a
dielectric chat that they can post for me? What did you use for spacers I
wonder since plastics are a better dielectric than oils in the first place?
Tony

-----Original Message-----
From: Tesla list [mailto:tesla@xxxxxxxxxx]
Sent: Saturday, April 14, 2007 7:21 PM
To: tesla@xxxxxxxxxx
Subject: Re: Oil Dielectric Properties


Original poster: gary350@xxxxxxxxxxxxx

I built a cap using metal plates and a fish aquarium.  Cut all the
plates the same size let plates 1,3,5,7,9,etc hang over one side
about 1/4" let plates 2,4,6,8,10 hang over the opposite side connect
the edges together with a soldered wire.  A spacer is needed between
all of the plates.  The spacer is the number one problem because it
becomes part of the dielectric.  I tried all sorts of things
polyurethane 1/8" thick worked the best for me. I tried several
different oils, Caster oil, mineral oil, 100% paraffin oil, 10 weight
hydraulic oil, High Voltage oil, BBQ grill lighter fluid.  All of the
oils seems to work fine.  This cap is 99.9999% bullet proof and 100%
repairable.  The only problem I ever had was getting an arc along one
of the spacers and causing a carbon track on the spacer.  The spacer
was replaced and the cap worked fine again. Cap is quick and easy to
build.  Disadvantage is it is large and bulky.




-----Original Message-----
 >From: Tesla list <tesla@xxxxxxxxxx>
 >Sent: Apr 12, 2007 9:38 PM
 >To: tesla@xxxxxxxxxx
 >Subject: Oil Dielectric Properties
 >
 >Original poster: "Breneman, Chris" <brenemanc@xxxxxxxxxxxxxx>
 >
 >Hello,
 >
 >I was once again considering possible unique methods of homemade tank
 >capacitor construction, and came up with this idea:
 >In a jar, or some other kind of uniform container with an open top,
 >place a plate of metal at the bottom.  Glue the plate of fasten it to
 >the bottom by some means.  Then add a layer of salt to the desired
 >dielectric thickness.  Then add another metal plate cut so that the
 >lead is on the opposite side of the container (of course the leads on
 >alternating plates alternate) and glue or fasten it to the side of
 >the container.  Then repeat with the salt and plates until the
 >container is filled.  Metal plates should have some kind of opening
 >in them, either a drilled hole or a space left for the opposite leads
 >on the sides of the container.  After this is done, the plates should
 >be stationary and fixed to the sides of the container.  The salt
 >could be poured out or dissolved with water, and the container could
 >be filled with a liquid dielectric, such as oil.  Such a capacitor
 >would be self-healing, always a useful trait.
 >Anyway, I was wondering what the dielectric properties of various
 >kinds of oil are at high frequencies to determine whether a capacitor
 >constructed in such a way would be feasible for Tesla coil
 >use.  Also, if anyone sees any general design problems with this,
 >suggestions are of course welcome.
 >
 >Thanks,
 >Chris
 >
 >