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

Re: Pulsed Tube Coils




On Mon, 6 Jan 1997 22:43:07 -0700 Tesla List <tesla-at-poodle.pupman-dot-com>
writes:
>Subscriber: FutureT-at-aol-dot-com Mon Jan  6 22:35:09 1997
>Date: Tue, 7 Jan 1997 00:03:16 -0500
>From: FutureT-at-aol-dot-com
>To: tesla-at-pupman-dot-com
>Subject: Re: Pulsed Tube Coils
>
>In a message dated 97-01-06 00:55:41 EST, you write:
>
><<    
> >>snip
> >>  >>  I was thinking of building a tube coil using a pair of
> >  >572-B's (basically heavy duty 811's) with a filtered DC
> > >> >plate supply of about 2500 volts,  the power supply would
> >  >have a output filter capacitance of about 15 UFD and the
> > > >>circuit would be the typical link coupled feedback setup.
> >  >>  But I was going to use a square or triangle wave signal
> > > >to drive a power mosfet (IRF-511) switch in the filament
> >  >center tap (cathode return) to gate the tubes conduction
> > > >> much the same as what a variable rate/dwell rotary gap
> >  >would do in a disruptive coil, or even use a audio 
> > > >source to modulate a fixed bias reference in the cathode
> >  >>return. Any thoughts on this approach?
> >   
> > > >>			Mark Graalman
> >   >>
>>>>snip
>
> >> pulsed tube coil using this system will still require a tetrode 
>and 
> >>AC (or
> >> >other up-ramping voltage) on the screen grid for long spark 
> >>production.   I
> >> definitely recommend pursuing this cathode-type control technique. 
> 
> >If you
> >> >try this technique Mark, I'd definitely be interested in your 
> >>results.
> > 
> >>    John Freau
> >  
> >>Mark,  All,      (update)
> >
>> >snip
> >got a nice 18" staccato spark most of the time, but since I was 
>still 
>> >using
> >the non-sync, pulsed AC system, sparks sometimes "broke into 
>multiple 
> >>spikes"
> >and lost spark length, due to incorrect pulse timing.   
> >
> > Next, I hooked up the DC power supply, and ran the system.  I only 
> >>turned up
> >the power part way so far (I don't like seeing my components burn 
> >up!), but
> >>the sparks seemed to be maybe a little better than using by grid 
> >controlled
> >DC staccato (neither case here using the "AC on the screen-grid" 
> >voltage
> >>technique).  What I mean by "better" is the sparks seemed less 
> >"plasma-like".
>> > But it's hard to judge at low power.
> 
> >>>>snip
>
>> >John Freau
> >
> >
> >    Hi John,
> 
>>   How do you think it may work out if the screen is 
>> supplied by a 500 -600 volt DC source rather than AC? Or
>> tying the cathode to ground through a 2 or 3 volt zener
> for a slight amount of bias and switching the screen volt
>> age on and off with the transistor switch? Of course I'm
> still talking about using DC on the plate circuit, I guess my ordinal
>> thoughts on this were to have a "silent"
> output discharge then work on different modulation
>> techniques. It will be some time before I can do much in
> the way of any experiments, my shop is loaded up, torn up, with
>> construction stuff from working on the house etc. 
>>  But the ideas still flow.......................
> 
>> 			Mark Graalman (WB8JKR) TCBA#1399
> 
>> BTW, what type of waveform are you using to control the
>> switching transistor?
> 
>  >>
>   Mark,
>
>I knew I forgot something in my my posts---I used a squarewave for all
>tests--it's all my generator can provide.  And Yes, I suppose that if 
>the
>proper waveform is fed to the "cathode lifting"  control transistor, 
>this may
>make it unnecessary to use the AC on the screen grid.  Then DC on 
>screen grid
>would be OK, or use a triode.  But if duty cycle is high, a lot of 
>power will
>be drawn.
>
>BTW, I did try connecting my negative control pulse to the screen, 
>instead of
>the control grid in one of my earlier experiments--and the technique 
>worked
>well, and in fact eliminated the potential problem of interaction 
>between the
>neg. control voltage and the grid-leak network.
>
>I also like your idea of using cathode bias to eliminate the need for 
>a
>negative control voltage or screen voltage to cut off the tube.  I'm 
>not sure
>what zener voltage rating would be needed here, may need to be more 
>than a
>few volts--if this is  what you generally had in mind--and I suppose 
>you'll
>lose power due to cathode degeneration.   There are certainly plenty 
>of
> possiblilites here.
>
>In general, when pulsed, a DC coil and an AC coil will sound about the 
>same,
>this is because if only one half cycle is passed, the ear will hear 
>this as a
>"pop" rather than a tone.  depending on the pulse rate, the sound can 
>be like
>a chain saw revving up, or like a  rat-tat-tat, or staccato sound.  If 
>you
>use voice or music modulation, that's another story, and I can see why 
>you
>would want the DC for a quiet coil starting point.  I think the DC 
>with audio
>modulation should work very well, but sparks will probably be short, 
>and the
>coil will probably draw a lot of power, you should hear the audio 
>coming out
>of the spark.  I never tried this but it should be pretty cool.  
>Perhaps
>someone else has tried this?   The coil could be voice actuated, so 
>the tube
>can cool between words, music will probably draw a lot of power, you 
>may have
>to keep the power low, or use a giant tube to keep it cool, etc.   
>
>It sounds like your onto something here with these ideas, I'll be 
>interested
>to hear what you come up with once you get your lab back on line.   I 
>don't
>know just how much I'll be able to do along these lines, always glad 
>to hear
>of any new ideas that you'd like to share.
>
>Happy (modulated) coiling,
>
>John Freau
>


   Ok John,

  Thanks for your input. I hope to get back into the swing of things this
summer.

 As far as the cathode bias, with a High Mu tube like a
811A or 572 or 3-400/500 etc I would think that 5 to 8 volts would put it
at just about cutoff with a 2-3000 volt plate supply, a 50 watt zener and
a 100 ohm 20 watt
resistor would work well. As far as the bias voltage being degenerative,
bypass the bias diode/resistor with a
.1 ufd 1KV cap, this will be the RF path around the bias
network and it will cut way down on the RF degeneration.


			Mark Graalman

TTYL!