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

Re: Character Revisited



Hi all,
         I would like to make a few comments on this piece....

> R. Hull recently posted an excellent post comparing the character of TC 
> discharge arcs with different drivers.
 
from RWW......

> As a follow up, I forward a portion of a post from a nonlist member.
>  
>      I had ment to make clear, that if you could drive a TC with a      
>      class A solid state, or vacume tube driver, and a signal with the  
>      same frequancy components as a spark gap, you would see the same   
>      arc lenghts per primary power levels as you would get with a gap.

I think it's important to recognize that the duty cycle of a gap
(a few percent if that) is vastly different from that of an oscillator 
(100%). An oscillator produces a power arc with a continuous high 
degree of channel ionization whereas the gap system produces distinct 
discharges with only some maintenance of ionization and that 
discontinuity allows the discharge terminal to reach a much higher 
peak voltage for each bang. The secondary Q is not constantly low for
a cap discharge system but varies between unloaded (no power through 
between gap fires) and heavily loaded several cycles after gap fire.

>      I think it's not the gap, but the frequancy content that makes     
>      gaps appear to deliver longer secondary arcs for a given power     
>      level.  Unless the frequancy components of the primary current are 
>      the same for the tube or transistor driver, we cannot make diurect 
>      comparisons to gap driven coils.

They never will be identical IMO because of the continuity vs 
discontinuity. Was the arbitrary waveform generator maintaining the
normal gap-type duty cycle or simply putting out the waveform 
repetitiously?

>      I've used some fancy soild state drivers with the same coils that  
>      have been driven with gaps, and while I cannot claim to have       
>      proven this exaustivly, it seems that gap vs. tube or transistor   
>      makes no differance.  A clean primary current (spectrally pure     
>      that is) is less 'efficient' (as measured by secondary arc lenght) 
>      than the mess from a gap.

Same as above. Actually, the gap is not totally bad in this regard 
either. There is a side-frequency generation for a pure unloaded 
ringdown in a cap discharge primary and there should be - its 
amplitude is changing wrt time.

>      To do this, I digitized an actual gap with a HP arbitrary waveform 
>      recorder/generator, and fed this through Phase Linear DC to 100    
>      Khz amplifiers.  A RF wattmeter was used to adjust the gain of the 
>      Phase Linear amps (two bridged together) so that the primary       
>      current was as close to that measured (by the same watt meter)     
>      when the coil was driven with the gap.

Question - was the low duty cycle of the gap discharge maintained? 
This is very important. Note also that at those low energy levels
voltage production in a disruptive system is approaching that of 
comensurate spark length (in my experience).

>      Guess what, same arc lenght.  Alter the frequancy components of
>      the waveform, and the arc lenght appears to change at a faster 
>      rate than when the power levels alone are changed.  So I agree 
    
>      that the arc is important, but why the arc is important may have   
>      more to do with the frequancy profile of the primary current than  
>      it depends on what kind of driver is used.
>      
>      I claim that if you deliver the same primary current waveform with 
>      a tube or transistor as is delivered with a gap, you get identical 
>      performance with each driver.
> 
>      snip . . .
> 
>      Getting back to the gap vs. tube or transistor discussion, in my   
>      mind it all boils down to what is inside the wave, it's internal   
>      structure and frequancy/phase profiles.  If we extend frequancy    
>      and phase to the pure potnetial waves, then scalar and EM really   
>      are the same things.
> 
> 
> Has anyone run a spectral analysis on a spark gap discharge?  It may be 
> informative to alter the spectrum or perhaps delete or add spectral 
> components in order to enhance TC preformance.  Take heart Alan Sharpe.

Be interested to hear other's views on this.
Malcolm