# Re: [TCML] Spark Gaps and Sine Waves

Hi Dan,

The waveform graphic - Where is that measurement being taken?  If that is
supposed to show the voltage across the tank capacitor, I've got some
explaining to do.

The waveform across the capacitor in a TC looks nothing like a sine wave.
It charges towards the gap firing voltage in a sinusoidal manner, but as
soon as the voltage across the cap (and also the gap) reach the threshold
voltage of the gap, the gap conducts and the cap voltage is reduced to zero
as the cap's energy is transferred to the secondary.  Briefly.  If you were
looking at this on a scope and one 60 Hz cycle filled the horizontal
screen, the duration of the gap conduction would be too narrow to even see
on that screen, I'm guessing in the 10's of microseconds.  While your
graphic shows the gap to be conducting for something like 50% of the time,
the actual gap conduction is only a tiny fraction of one percent. And it's
only during this very narrow time interval when the gap is firing that the
RF oscillations occur and the sparks flow from the secondary.  The other
99.99% of the time it's just slow 60 Hz charging of the cap and nothing at
all going on with the high voltage.

I posted a web page showing the simulated charging waveforms for a variety
of capacitor value and static gap breakdown voltages - see
http://www.laushaus.com/tesla/gapsim.htm  The waveforms between a static
gap and a rotary gap aren't too different - just that with a static gap,
the voltage at which the gap fires is more or less a constant, whereas with
a rotary gap, the time interval between firings is the constant.  But the
waveforms look essentially the same.

Note that the time scale of these simulations shows only the 60 Hz
behavior, it's too slow to show the RF oscillations that occur during gap
conduction.

1) For a NON rotary gap, what happens after the gap stops firing but the
voltage has not yet crossed the zero line? I am guessing the tank capacitor
charges initially and then drains as the supply transformer voltage drops.
Is this primarily hard on the transformer's secondary windings (heating due
to current) or hard on the caps (heating due to current) [No] ? Do the caps
discharge through the primary coil in the tank (since it has less impedance
compared the supply transformer's secondary)?
[Yes, the fully charged cap is placed across the primary coil when the gap
closes to create a parallel resonant circuit, the it resonates at the
chosen frequency and transfers this energy to the secondary]

2) For a NON rotary gap, once the voltage has crossed the zero crossing
point is the charge on the capacitor (coulombs) preserved and carried
forward, thus helping to charge the caps on the next cycle (or below the
zero line cross in this graph)?
[If there is energy in the cap and it starts charging back towards the
other polarity before the gap fires, this energy is not lost!  It is stored
in the inductance of the NST as the cap voltage goes towards zero, then as
it climbs in the opposite polarity, that energy is transferred back into
the cap.  But it doesn't matter whether this is a static or rotary gap.]

3) FOR A SYNCHRONOUS ROTARY GAP, to my knowledge, it has always been
suggested to break on the peaks and valleys of the sine wave (for max
power) AND the zero line crossings to discharge the caps. Which makes me
3a) How is there enough voltage on the caps to be able to break down the
gap if there is zero volts?  [The gap won't fire if there is near-zero
volts.  For a sync RSG, you want to fire somewhat AFTER the voltage peak -
the reasons are rather complex.  I've not herd it suggested to fire at zero
crossings]

3b) It has also been suggested that a NON-SYNCHRONOUS gap when used with
NST's will damage the NST's secondary windings...so if my above assumptions
in questions 1 and 2 are correct...then why don't static gaps destroy NST's
(ignoring random voltage spikes and dirty RF noise)?
[The suggestion that NST's should not be used with async RSG's is just
nubbie advice.  The only hazard to an NST with an async RSG is if the speed
is too low and there is no properly set safety gap to fire if the RSG speed
falls too low.  If care is taken to keep the speed above ~100BPS and if a
properly set safety gap is employed in parallel with the RSG, the NST
couldn't care less if it was synchronous or not.  No gap type is dirtier
than any other.  Did I mention that one should always have a properly set
safety gap ;-) ]

Regards, Gary Lau
MA, USA

Virus-free.
www.avg.com

On Thu, Mar 15, 2018 at 2:32 PM, Daniel Kunkel <dankunkel@xxxxxxxxx> wrote:

> Hello all,
> I have been thinking a lot about spark gaps lately. Specifically: what
> happens on the backside of the sine wave (of the supply transformer
> voltage)?
>
> Refer to this crude image (note the green numbers correspond to each of the
> three questions below). The values and limits are arbitrary just to
> illustrate the idea and questions below.
> https://i.imgur.com/z7fGKZB.png
>
> Questions:
> 1) For a NON rotary gap, what happens after the gap stops firing but the
> voltage has not yet crossed the zero line? I am guessing the tank capacitor
> charges initially and then drains as the supply transformer voltage drops.
> Is this primarily hard on the transformer's secondary windings (heating due
> to current) or hard on the caps (heating due to current)? Do the caps
> discharge through the primary coil in the tank (since it has less impedance
> compared the supply transformer's secondary)?
>
> 2) For a NON rotary gap, once the voltage has crossed the zero crossing
> point is the charge on the capacitor (coulombs) preserved and carried
> forward, thus helping to charge the caps on the next cycle (or below the
> zero line cross in this graph)?
>
> 3) FOR A SYNCHRONOUS ROTARY GAP, to my knowledge, it has always been
> suggested to break on the peaks and valleys of the sine wave (for max
> power) AND the zero line crossings to discharge the caps. Which makes me
> 3a) How is there enough voltage on the caps to be able to break down the
> gap if there is zero volts?
> 3b) It has also been suggested that a NON-SYNCHRONOUS gap when used with
> NST's will damage the NST's secondary windings...so if my above assumptions
> in questions 1 and 2 are correct...then why don't static gaps destroy NST's
> (ignoring random voltage spikes and dirty RF noise)?
>
> If there is already some whitepaper or documentation written on this please
> direct me to it! You can also attach and email me directly as well if there
> is not a tidy URL to point me to.
>
> Thanks,
> ~Dan
> Kansas City area
> _______________________________________________
> Tesla mailing list
> Tesla@xxxxxxxxxxxxxxxxxx
> https://www.pupman.com/mailman/listinfo/tesla
>
_______________________________________________
Tesla mailing list
Tesla@xxxxxxxxxxxxxxxxxx
https://www.pupman.com/mailman/listinfo/tesla