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

Tube coil ideas



Tesla List wrote:
> Original Poster: "David Trimmell" <davidt-at-pond-dot-net>
> Problem is the ARRL Handbook (although, invaluable), and all "radio" theory
> goes out the door when designing a "tube coil"... At least that the way it
> seems to me, a "novice".

   Edd.B writes;

    Let's see if I can start a "new" thread...........

      Just about all of the tube coil designs use the same very old
    close coupled / grid feedback oscillator with a large triode
    in class B service. The efficiency of a typical class B circuit
    is about 60% to 65% ( power in / power out ).

      An 833-A triode with cooling ( fan or blower ) has a plate
    dissipation of 450 watts. With 3000 Vdc at 400 Ma ( .4 Amp )
    power input is 1200 Watts, at 60% eff the power output is
    720 Watts. 1200 - 720 = 480 Watts plate dissipation, this
    over the max dissipation and the tube will run red hot with
    a shorter life expectancy.

      An NST design with a 15Kv/30Ma transformer has a 450 watt
    output and will have a much better spark discharge, much longer
    and thicker sparks from less power.

      I think we all know that, But ..... WHY.....

      The reason is ( I think ??? ) the NST circuit is a pulse
    discharge circuit, with a very powerful pulse and oscillatory
    ringing in a classic tuned circuit.

      The tube circuit has a sign wave output not a pulse. After
    all it's nothing more than an old fashion "ticker coil" design.
    
      So what would happen if we pulsed our 833-A example ???.
    
      A variable oscillator with a saw/tooth output from ... say
    50Kc to 200Kc driving a voltage amp ( 6L6 ) biased AB1 to a
    power amp ( 2 / 6L6 ) biased Class C .... this is how we change
    from a saw/tooth to a pulse. Next is a driver stage having
    about 25 or 30 watts output, an 807 in class C should do it.  
      
      It's not that complicated, it would be all tubes, nothing
    small and modern to burn up, no tuned circuits in the amp
    or driver and with the 50% duty cycle that we obtained by
    pulsing the 833-A ( which is now biased class C ) a higher
    input power can be used without exceeding the max plate
    dissipation. The last benefit but certainly not the least
    is the efficiency of the pulsed 833-A , it will be or should
    be 90% or better.

     Comments anyone ??????

    Edd.B