RE: AC vs DC charging circuit question

["Lau, Gary" <Gary.Lau-at-compaq-dot-com> (by way of Terry Fritz <twftesla-at-uswest-dot-net>)]

This is something that is commonly misunderstood, so you are in good
company.  The short answer is that there is no efficiency advantage to DC
over AC operation, and in fact the conversion to DC probably results is
additional (though probably not too significant) losses.

The thing that is not obvious with AC charging is that energy is never
wasted.  Let's say that if a cap starts out with zero charge on it and the
AC waveform starts charging it.  Let's also assume that the cap is "matched"
to the NST secondary to be mains-resonant.  The peak voltage on the cap will
occur at the end of the first mains half-cycle as the input voltage to the
NST is passing through zero, or at 180 degrees.  Let's also say that the
spark gap was set too widely and that this peak cap voltage is not high
enough to trigger the gap.  Now the NST will be charging the cap in the
opposite polarity, and it might seem as if the charge that's there now will
be somehow wasted.  This isn't true. 

Instead the voltage will swing in the opposite polarity but even faster than
before, and it will reach a peak voltage at 360 degrees that is much higher
than the previous peak after the first half-cycle, presumably high enough to
now trigger the gap.   If there is no gap or it is set much too widely, the
cap voltage will build higher and higher on successive half-cycles until
either losses limit the peak voltage, or your NST or cap go Pffft! from too
much voltage.

The same principal also applies when a gap is set narrowly and the cap
reaches the gap trigger voltage and discharges the cap before the end of a
half-cycle.  It will try to recharge to Vgap following the first bang, but
won't be able to reach it before the half-cycle ends.  But again, the
partial charge will just be applied to the next half-cycle's charge in the
opposite polarity.

Hope this helps,
Gary Lau
Waltham, MA USA
Original poster: ghub005-at-xtra.co.nz 

>I've been wondering about something for the last few days. 
>
>In an AC charging circuit it is common for the SG to fire near the 
>peak charging voltage (I'm disregarding LTR circuits). What 
>happens to the energy carried in the latter half of the charging half-
>cycle? I guess that this energy is just dissipated through the 
>'shorted' spark gap - but wouldn't that make a fullwave DC charging 
>circuit twice as efficient? (i.e. since the charging circuit's energy is 
>stored right across the half-cycle)
>
>However, I understand that equivalently powered DC and AC coils 
>don't display a large amount of difference in their spark output(?).
>
>Can anyone tell why this is? Of course I'm assuming that the DC 
>filter cap is isolated from the tank cap (otherwise it would also just 
>discharge through the SG). 
>
>Of course that begs the question of how to actually isolate a DC filter 
>cap(s) from the resonance in the tank circuit. I'd be interested to 
>know what methods are in use.
>
>Best regards,
>
>Gavin Hubbard