David, The thing that makes me question if the gap really fires
so far ahead of alignment is that I know that in many rotary gap
systems the gaps have to be set very close (much closer than
one would expect for a given voltage) or they fail to fire reliably.
the whirling air around the electrodes may be blowing away
ions and delaying the air breakdown. Also it takes a certain
amount of time for ionization to occur (although this time might
be rather short... I forget). A lot of medium powered coil systems
quench on the 2nd or 3rd notch, and this makes the firing time
much longer than it would be compared to if it fired on the
1st notch. This makes me think that the the firing event is
likely not complete before the gaps at least begin to overlap.
(I don't have proof of this however). As I mentioned, in viewing
a video of a firing rotary gap, I see the gaps aligned (by strobe-
effect) while firing. It would seem to me that if the firing event
is complete before the gaps align, then I'd see that in the
video (I'd see the gaps not yet aligned as they are firing)...
unless I'm not thinking this through correctly?
The electrodes tend to ablate on the leading edges which does suggest
some firing before full alignment, then again, the power is greatest
at the beginning of the firing event, so therefore would cause
the greatest erosion at the start of firing.
I've seen cases using an oscilloscope in which the voltage
waits at a steady voltage for some time until the gaps get
close enough then it fires in NST systems, as you describe.
I agree with your comment that;
"one would set the initial setting early so the
firing voltage would be low and turning up the
control would delay the arrival of the spinning
electrode so as to increase the firing voltage,
and then continue to delay more to go past the
point of max firing voltage, so one could
comfortably find the optimum point."
I found in my setups that when I went past the optimal point,
the coil would suddenly stop firing as the NST went into a new
non-firing charging mode. I would then note this point on the
phase variac, and the next time I'd keep the setting just below
this point.
Regards,
John
-----Original Message-----
From: David <zipo@xxxxxxxxxxx>
To: tesla <tesla@xxxxxxxxxx>
Sent: Fri, Feb 6, 2015 5:50 pm
Subject: Re: [TCML] Synchronous Gap understanding
John and All - Thank you very much for the explanation. With all of the
modeling, calculations and guessing I've done on this I can't verify any
of it, hence these "Burning" questions.
A quick over view of my charging system:
xformer - 15kv, 120ma NST --- (yes I'm the guy that bought the 15/120
nst from this list a couple of months ago)
Cap - 62.5nF mmc 12x5 - the largest cap that this xforme can charge to
25kv (max safe voltage for this xformer)
Sync Gap - "Propeller type" 3600 rpm, 8.5"dia rotor, with a tip speed of
1.6"/mSec.
Fo of system ~90khz
According to modeling and theory the firing point for max voltage
happens some time after max voltage on the cap. Modling shows that the
cap is at or above 25kv for ~3msec, sitting and waiting for the gap to
get close enough. This says to me that the gap must fire some distance
before alignment. Now just to consider 3 notches would be about .18 -
.20mSec. It looks to me like the party is done by the time the
electrodes actually align. If this is correct ??? then let's consider
the initial setup of the gap timing and what direction does the Freac
controler move the spinning electrode. According to theory one would set
the initial setting early so the firing voltage would be low and turning
up the control would delay the arrival of the spinning electrode so as
to increase the firing voltage, and then continue to delay more to go
past the point of max firing voltage, so one could comfortably find the
optimum point.
Wheew this is tough, sorry guys but I gotta know.
Thanks again
David
On 2/6/2015 11:07 AM, Futuret via Tesla wrote:
> Dave,
>
> Another thing to consider is the spacing of the electrodes (at closest
> approach), and the maximum available voltage. I've seen various cases
> where the spark could barely jump the gap, and even missed some
> firings. In these cases I had to set the electrodes so close that they
> almost touched as they whirred by the fixed electrodes. In general,
> when viewing the sparking gap with a video camera, The brightest glow
> at the gap is seen while the electrodes are aligned. I was not able
> to visually see evidence of firing before the gaps aligned. I would
> imagine that in these cases, the
> spark is not able to jump very much before the electrodes align,
> perhaps not at all. In almost all cases, the gaps quench (stop
> firing) while the electrodes are still aligned (before they pull apart
> from each other). Under certain conditions this can cause a bad
> problem of "re-firing" of the gaps while they are still aligned. Factors
> that contribute to this are; slow rotary rpm, small diameter disc,
> wide diameter electrodes, and small main capacitor. This problem
> is unlikely to occur with 120bps sync rotaries however. It's mostly
> likely to occur in higher break-rate systems (whether sync or async).
> In any case rotary gaps do not quench the spark by stretching it.
> The sparks quench on their own while the gaps are still aligned.
> The only case where the sparks could possibly be quenched by
> stretching is in a huge coil running at super low resonant frequency
> of maybe 10kHz or whatever. The math shows that the mechanical
> dwell time of the rotary gap tends to be large compared to the
> typical quench times for a typical Tesla coil. This is what makes
> the bad problem of rotary "re-firing" possible. But again it's usually
> not a problem as explained above.
>
> In any case when setting up a sync rotary 120bps system, you just
> adjust the phase controller for maximum output spark length at full power.
> Adjustment has to be done at full power because the best adjustment
> variac position varies as the power level is varied (at least in NST powered
> setups). I forget offhand if this occurs in PT/Pig powered setups (it
probably
> does since it has to do with how fast the capacitor charges and how much
> voltage is available to jump the gap.
>
> In some cases with NST powered coils, best spark length was obtained
> when the spark gap fired below the maximum voltage that was reached.
> The gap fired as the voltage was decending from maximum.
> It was not possible to make the gap fire at maximum voltage. I think
> the energy storage within the NST was upset when this was attempted,
> and the gaps refused to fire and kicked the resonant charging system
> into a different mode. This seems to have a lot to do with the
> setting of the ballast (in ballasted setups), and the manner and rate of
> voltage rise on the capacitor which depends on the relative charging
> current of the transformer and the capacitance value. This may also
> be a part of the reason that certain sized LTR cap values are best
> for certain NST's.
>
> Regards,
> John Freau
>
>
_______________________________________________
Tesla mailing list
Tesla@xxxxxxxxxx
http://www.pupman.com/mailman/listinfo/tesla