[Prev][Next][Index][Thread]
Re: Spice simulation pictures
Tesla List wrote:
>
> > > Subject: Re: Spice simulation pictures
> Re: Spice simulation pictures
>
> >From rwstephens-at-ptbo.igs-dot-netFri Oct 18 22:58:41 1996
> Date: Sat, 19 Oct 1996 00:28:21 -0500
> From: "Robert W. Stephens" <rwstephens-at-ptbo.igs-dot-net>
> To: tesla-at-pupman-dot-com
> Subject: Re: Spice simulation pictures
>
> Date: Thu, 17 Oct 1996 22:25:01 -0600
> From: Tesla List <tesla-at-poodle.pupman-dot-com>
> To: Tesla-list-subscribers-at-poodle.pupman-dot-com
> Subject: Re: Spice simulation pictures
> Reply-to: tesla-at-pupman-dot-com
>
> > > Subject: Re: Spice simulation pictures
>
> >From hullr-at-whitlock-dot-comThu Oct 17 22:20:40 1996
> Date: Thu, 17 Oct 1996 10:41:39 -0700
> From: Richard Hull <hullr-at-whitlock-dot-com>
> To: tesla-at-pupman-dot-com
> Subject: Re: Spice simulation pictures
>
> Tesla List wrote:
> <mega snip>
> >
> Bert wrote:
>
> >... Another thought - are any tube coilers using heavy top-loading (other
> > than Robert Stephens' 15 KW oil immersed beastie)? Sounds like it would
> > make these systems less sensitive to detuning from operator movement.
> > You could still control breakout threshold via a "bump" on the toroid.
> >
> > -- Bert --
>
> Richard Hull wrote:
> >Bert,
> >
> >In theory a tube coil could be top loaded. I have never seen one with
> >even a 1" ball on it!!!! These systems are very sensitive to capacitive
> > top loading. Remember a tube coil is just an oscillator in which the
> >primary and associated coupled grid coil are merely part of the delecate
> >oscillator tank. Any loading on the top of the coupled secondary would
> >require a major re-think back in the oscillator circuit which is rarely
> >equipped with the broad tune range devices needed.
> >
> >I eagerly await the first guy to have a 18"X4" secondary on his tube coil
> >with a 20"X5" toroid on top to step forward!!!!
> >
> >Richard Hull, TCBOR
>
> Richard,
>
> Would a 24 inch by 3 incher please you? Been there, done that. Oh
> BTW, no streamers.
>
> One important aspect characteristic of vacuum tube Tesla coils which
> is not often considered because it is not an issue with disruptive
> coils is the duty cycle of the circulating current in the secondary.
> Didn't someone (Bert?) recently post that the disruptive system has a duty
> cycle of some 0.03%? In a vacuum tube coil operating from a full
> wave filtered DC power supply the secondary circulating current is
> 100% CW RMS toaster watts power. At 100 kilovolts the resonator
> circulating current starts climbing rapidly as you add a little topload
> capacitance to your vacuum tube Tesla coil.
>
> I made a 100pF cap which sustained 150 KV before
> breakdown from a 50 US gal steel drum and a smaller metal cylinder
> with rounded ends suspended inside. I wound a secondary out of #14 AWG
> enamelled copper wire on a 6 inch PVC form that resonated with this 'topload'
> at 150 kHz and drove it with my 1000 watt vacuum tube oscillator
> (3 x 810 triodes).
>
> I was able to swing the 100 pF load to about 85 kilovolts RMS (120 kV
> peak). The secondary winding got so hot in just 120 seconds of operation that
> I shut the system down for fear of melting the PVC tube on which it was
> wound! My circulating current was 8 amps RMS according to the
> measured output voltage across a capacitive reactance of some 10.6 K
> ohms, but the heat produced in the wire felt more the equivalent of
> maybe 30 amps RMS. Can anyone explain where my figuring leads to
> such an error? The interconnect between the top of the secondary and
> the drum capacitor was #14 PVC covered solid wire. It got bloody hot
> too. If I had had one then, that's where I would have liked to place an RF
> thermocouple ammeter.
>
> The commercial client who I am working on the large vacuum tube 100
> kHz system for with the oil filled resonator wanted me to consider
> also building them a system whereby I would be able to swing a 100 pF
> load to 200 KV RMS at 1 MHz. This is a formidable power challenge.
> Using the same figuring based just on voltage divided by capacitive
> reactance the continuous secondary resonator circulating current this time
> would be 125.63 amps RMS! This is a job only for large copper ribbon or
> tubing with silver plating. I declined on the big 1 MHz project BTW, at least
> until I get more high power CW experience. How about these guys with the
> vacuum tubes in the linear accelerators developing nearly three
> quarters of a megavolt at 50 MHz across a ten turn helix with a 50 kilowatt
> tube oscillator. At least they don't have to swing a big toplaoad capacitance.
>
> Food for thought, and actual projects for the well funded mentally deranged.
>
> rwstephens
Robert,
Wow! I never thought about the circulating current problem in a
capacitively-loaded CW system. It sounds like "skin effect" losses are
the root cause of the heating. Doing a little behind-the-scenes math: At
DC, #14 AWG has about 2.5 Ohms/1000 feet. At 150 kHz, the current flow
will be constrained to just an outer layer of the wire (called the skin
depth). The skin depth decreases dramatically as the frequency
increases. Because of "skin-effect", the same #14 AWG wire now has an
equivalent resistance of about 60 Ohms/1000 feet, or about 24 times the
DC resistance.
Since skin effect losses create I^2R heating just like DC resistance,
running 8A RMS at 150 kHz causes the same degree of heating as if you
were running 39 Amps of DC through it! Or, looking at it from a
conductivity standpoint, #14 AWG behaves as though it was between #27
and #28 AWG at 150 kHz. Even going to silver-plated or Litz wire reduces
these losses only slightly (5-10% max). If you truly need CW operation
with significant (100 pF) top loading at high voltages, and cannot live
with some type of pulsed operation, you'll probably need to circulate
oil to cool the secondary as well as insulate it.
Sounds like your clients are into some wild and crazy stuff! :^)
-- Bert --