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

Re: Tube TC

Original poster: "Dan" <DUllfig@xxxxxxxxxxxxx>

John:

Ok, I don't know that much about electronics (enough to be dangerous), but I thought that in a colpitts oscillator, the cappacitors and _primary_ winding formed the LC tank circuit. In this schematic, the capacitors are connected in parallel with the _secondary_ ! Unless I don't understand the operation of this oscillator, seems something's fishy...

Dan

----- Original Message -----
From: <mailto:tesla@xxxxxxxxxx>Tesla list
To: <mailto:tesla@xxxxxxxxxx>tesla@xxxxxxxxxx
Sent: Thursday, October 13, 2005 7:06 PM
Subject: RE: Tube TC

Original poster: <mailto:Sparktron01@xxxxxxxxxxx>Sparktron01@xxxxxxxxxxx

Jim

Tubes can absolutely be made to operate as a push-pull power oscillator.
I have built both types (Armstrong, Hartley), and recently (see below) posted a
URL concerning a 50kW (!!!!) Colpits push-pull unit used for a linear
accelerator
HVDC multipler stack.

<http://www.veccal.ernet.in/~vecpage/inpac2005/CD/Contributory%20paper/C-027.pdf>http://www.veccal.ernet.in/~vecpage/inpac2005/CD/Contributory%20paper/C-027.pdf

NOTES
I.  There is a schematic error in the above URL, can anyone catch
it??? (and it is
      NOT minor...  :^D )
II.  The above URL may or may not work, try "push pull colpits" as
key words in GOOGLE
      then look for title:
      _CONSTRUCTION & TESTING OF 50kW/120kHz OSCILLATOR FOR 3MeV, 30kW DC ..._

The difference between the three oscillator types are:
Colpitts uses a series capacitor voltage divider network to establish
grid drive power.
Hartley uses a tapped inductor (dual of Colpitts) to derive grid drive power.
Armstrong uses a grid feedback (tickler) coil to establish 180 deg
phase shift and
     derive grid drive power.  There is a now with Armstrong (3)
degrees of magnetic
     freedom in oscillator (1. Plate Tank to GFB, 2. Plate Tank to
Sec., 3. GFB to Sec.)

I personally like the Hartley (or Colpits) over an Armstrong because...
1.  Grid drive is derived directly from the tank circuit, rather then magnetic
      coupling from a tickelr coil.  It's just one less "fiddle
factor" to deal with.
2.  Push-pull was used in early induction heating due to lack of high
      power tubes, and there is significant history and references (IEEE, IRE)
      concerning use, design, etc.
3.  Double the voltage swing and half the current would equate to ~1/4
      i^2r heating loss in tank circuit, versus a parallel connected
circuit of equal
      power.  Tank caps have to handle 2X the tank voltage, however.
     (TINSTAAFL- There is no such thing as a free lunch).
4.  My experience with PP suggests significantly improved stability, and less
      likelihood of the oscillator to pull off frequency, even with
significant loading.
5.  I've not tried a Colpitts oscillator (as above URL), however,
with HV vacuum
      tuning caps (feedback/drive control) and a MMC array to handle HV and
      high RMS currents, may have considerable design advantage over a Hartley
      PPO at HV, high power.
6.  An excellent power oscillator tube for a VTTC would be a air
cooled 3CX2500H3
      or water cooled Amperex 6960.  Alex Tanjsek and I have culled some
      6960's with socket and water jacket hardware from boneyard high power
      LEPEL IH units.  Fils are intact (12V, 30A), have not built up
a constant current
      test fixture to determine operational parameters.
7.  Staccato mode power control can allow significant pulse power
capability out of
      tubes without overheating; I believe I would now use high power
IGBT's instead
      of SCR's for cathode modulation control.  Much higher voltage,
pulse current
      rating in a cheaper package (IMO).

Regards
Dave Sharpe, TCBOR/HEAS
Chesterfield, VA. USA


> Original poster: "Jim Mora" <<mailto:jmora@xxxxxxxxxxx>jmora@xxxxxxxxxxx>
>
> John or Dan or ?, I have 4 833a's with original Johnson Sockets and have
> wondered if way down the road they could be made to Push Pull. I haven't
> studied Tubes since 74. Can you elaborate on the Culpits vrs. The Armstrong
> oscillators. One thing I was reading stated that 2 Triodes in parallel cause
> Capacitance problems but there were work arounds? Any Ideas?
>
> They are not in my theory or family of curves text books.
>
> I net:
>
> "Colpitts oscillators are somewhat similar to the shunt fed Hartley circuit
> except the Colpitts oscillator, instead of having a tapped inductor,
> utilises two series capacitors in its LC circuit. With the Colpitts
> oscillator the connection between these two capacitors is used as the centre
> tap for the circuit.
>
> Regards,
> Jim Mora
> Ojai Ca.
>
>
> -----Original Message-----
> From: Tesla list [mailto:tesla@xxxxxxxxxx]
> Sent: Wednesday, October 12, 2005 8:16 PM
> To: <mailto:tesla@xxxxxxxxxx>tesla@xxxxxxxxxx
> Subject: Re: Tube TC
>
> Original poster: <mailto:FutureT@xxxxxxx>FutureT@xxxxxxx
>
> In a message dated 10/12/05 8:14:26 PM Eastern Daylight Time,
> <mailto:tesla@xxxxxxxxxx>tesla@xxxxxxxxxx writes:
>
>
> >All:
> >
> >I have been looking at different VTTC's, and noticed that most all of
> >them are variations of the Armstrong tuned grid oscillator. Nowhere
> >is it explained the merits of using such an oscillator. Why isn't the
> >Colpitts, for example, used? what is the advantage? or drawback? has
> >anyone built a VTTC that is wired differently?
> >
> >It would seem to me that a colpitts oscillator would give you a tuned
> >primary; since the DRSSTC coil works so much better on account of
> >having the primary tank tuned, wouldn't a tuned primary benefit a VTTC
> also?
> >
> >By the way, I just bought a 833a tube on ebay, and plan on building a
> >coil around it.
> >
> >Dan
>
>
> Dan,
>
> The traditional modified armstrong VTTC's also use a tuned primary.
> A number of folks have built Hartley VTTC's but I never saw any actual
> efficiency advantage over the armstrong despite theoretical advantages.
> I believe oscillation instabilities of the armstrong may actually be
> an advantage
> but I'm not sure.
>
> The 833A is a good tube for a VTTC and can give 21" sparks
> or so without staccaco (pulsed) operation. Longer sparks can
> be obtained with staccato operation of course with suitable
> decreases in tank impedance.
>
> John
>
>