Re: flashing vs. whatever
From: Bert Hickman[SMTP:bert.hickman-at-aquila-dot-com]
Reply To: bert.hickman-at-aquila-dot-com
Sent: Thursday, November 27, 1997 3:23 AM
To: Tesla List
Subject: Re: flashing vs. whatever
Tesla List wrote:
> From: Barton B. Anderson[SMTP:mopar-at-mn.uswest-dot-net]
> Sent: Tuesday, November 25, 1997 10:24 AM
> To: Tesla List
> Subject: Re: flashing vs. whatever
> Tesla List wrote:
<SNIP of most previous text from Bert Hickman>
> > How big were you planning on making your secondary and toroid, and what
> > voltage/current is your power source??
> Bert,Yes, it's a large value cap! Here's my project: (100KHz, 17KW)
> Items completed:
> Secondary: 12.5"OD, 33.2" H. 783 turns. #18 polythermalese magnet wire (1/4
> Wave of 2460ft).
The physical wire length is about 2460 feet (at 100% winding density),
but your resonator's natural frequency (without a top-load) will be
about 150-155 kHz. This corresponds to a quarter-wavelength of about
1600 feet. A coil will always have a self-resonant frequency that's
higher than the equivalent length of straight wire because of lower
propagation velocity (much less than the free-space velocity of light).
> Primary: Flat pancake 3/8" soft copper tubing. Currently 8 turns (may be
> expanded), 3/8 spaced, 23" OD. Adjustable up/down first winding of sec. by 9
> inches each direction.
You'll need to provide at least 1.5" - 2" of clearance between the
innernmost turn and the secondary to avoid secondary:primary flashovers.
With a ctr-ctr spacing of 0.75" (3/8" spacing between turns), this will
bring your outside primary diameter to about 27". A pancake coil with a
15.5" inner diameter and 3/4" ctr-ctr spacing will give you about 49 uH
at turn 8.
> X-Former: Pole Pig 14.4K, 240, 10KVA, 1 Ph
> Toroid: 8 x 30".
This is a good size for this coil - you might even want to increase the
diameter a bit to 36" - 40". A 30" diameter toroid should add about 25
pF of top capacitance (discounted for the effects of secondary
shielding), which should drop your secondary's loaded operating
frequency down to around 100 kHz.
> Not completed or attained:
> Cap: 0.256uF (Just now starting to seriously think about this design)
I suspect that because of the problems you had in estimating too low an
operating frequency, this also caused you to need an abnormally high
tank cap value. To bring your system into tune, you'll only need a bit
over 0.05 uF. I'd also recommend adding a few more turns to your primary
to give you a bit more tuning range.
> Sparkgap: now at 8 electrodes (should finish this weekend)
> Variac (not attained yet but found 3, 30 amp'ers. Have yet to inspect and
> Ballast (not built or attained yet. Either variac modified or welder;
> Stand-alone breaker box to be put in garage (off incoming).
> And a CRAP LOAD of safety!!! (including labels).
> Your exactly right about parallel caps NOT increasing their voltage handling.
> I think I said "series" in my post (brain must have been turned off last
> night) Yes Parallel! What I'm doing is deciding on purchasing vs. building a
> cap for this big dog. I have a contact at CSI. He offered to build the cap
> (all polypropylene) for $300 (might be the way to go). He said this particular
> process takes about 6 weeks to build one of these caps (unlike the paper and
> poly type used for wicking the oil).
Hmm.. sounds too cheap for this size cap. Are you sure he understood
that it's for disruptive Tesla Coil operation off a 14.4 KV RMS source??
> In my equation I used 2.3 as the dielectric of x-former oil which is near
> polyethylene film. However, I see what you mean. 32 mils (total) isn't even
> close. I did this calculation pretty fast and with not a lot of digging into
> the archives. It looks like it would take more along the lines of 4, 30 mil
> sheets on each plate side. I would end up with a distance over a 1/4inch
> between plates. This seems extreme and is based on 240V/mil at breakdown. Is
> this value correct? If so, I could end up with a huge and maybe impossible to
> build cap capable of about 60KV. However, this assumes a linear build and
> doesn't account for layering of dielectric. Is there an equation (or good
> assumption) for finding voltage handling capability of layered film plate
You can probably get away with a total of 90 mils of LDPE by using three
30 mil sheets between plates. Also, use a dielectric constant of 2 in
your design calculations to account for the lower dielectric constant of
the oil and the "looseness" of the capacitor construction. Another
approach would be to build three identical caps, each made from at least
two thicknesses of poly, and connect them in series to reduce the
voltage stress seen by a single cap. There's really no design equation
you can use, just the practical "real world" experiences of coilers. 60
mils just isn't enough for reliable pole-pig operation at 14.4 KV.
Hope this helped!
Safe coilin' to you, Bart!
-- Bert --