# Re: Racing sparks, streamer growth, coupling, etc.

```Original poster: "Malcolm Watts by way of Terry Fritz <twftesla-at-uswest-dot-net>" <m.j.watts-at-massey.ac.nz>

On 5 Mar 01, at 21:51, Tesla list wrote:

> Original poster: "Jim Lux by way of Terry Fritz <twftesla-at-uswest-dot-net>"
>
>
> > Original poster: "Antonio Carlos M. de Queiroz by way of Terry Fritz
> <twftesla-at-uswest-dot-net>" <acmq-at-compuland-dot-com.br>
> >
> > > Original poster: "Jim Lux by way of Terry Fritz
> > > <twftesla-at-uswest-dot-net>"
> > >
> > > I've been thinking about what's really going on in a TC during the
> > > first few half cycles, in connection with the spark growth.
> > >...
> >
> >
> > > 2) If the primary and secondary are very tightly coupled, the
> > > secondary voltage will rise faster, fast enough that the charge
> > > that is flowing
> into
> > > the top load can't be "absorbed" by the growing spark, so the
> > > voltage
> will
> > > rise higher.  If the voltage gets high enough, it will exceed the
> breakdown
> > > voltage along the secondary, causing "racing sparks".
> >
> > I don't think that "racing sparks" are related directly to coupling.
> > When the coupling is increased, there is no significant change in
> > the maximum dv/dt of the voltage waveforms, compared to what would
> > occur close to the end of the energy transfer proccess with lower
> > coupling.
>
> Why wouldn't the dv/dt be higher with more coupling? (having not done
> the calcualtions myself, I am lazy)... I would think more energy in
> the same time would imply more voltage at the Vpeak, so the dv/dt must
> be higher at any given time.
>
> I don't think the dv/dt is much higher with better coupling, just that
> the max voltage the top load can reach is higher.  If the spark forms
> earlier, it loads the secondary, reducing the amount of voltage rise
> as the charge moves into the secondary.  The spark can only grow so
> fast...  I'm working out a rough simulation, where the capacitance of
> the top load grows after breakout (assumed to occur at some
> predetermined voltage).

dV/dt can be significantly higher with increased coupling but
obviously depends on the degree of coupling. It is in fact the
changing slope of the secondary ring which gives rise to the
sidebands as viewed in the frequency domain and these move further
apart as k increases. dV/dt increases with k on both a per cycle
basis and on an average (rectified if you like) basis. Note to all
those non-believers out there - the sidebands are present if you view
the secondary e-field only - waveforms looking like a ringup/ringdown
generated by whatever means can be mathematically shown to have
dual/multiple frequency content. Although a tuned primary is part and
parcel of a TC it is not a precondition on producing what is
essentially a DSBSC envelope.
But output voltage can also reach a higher value with the same
Ep and increased k as the primary gap is given less time to lose it.

It stands to reason that if the secondary attains some voltage more
quickly, dV/dt must have increased. Scope observations suggest that
the secondary pretty well rings up before a spark lets go in most
systems whether it is the initial discharge or part of a repetitive
train.

Regards,
malcolm

> > The secondary maximum voltage (and maximum dv/dt) just occurs in
> > less cycles when the coupling is higher. Just observe that close to
> > the end of the energy transfer, the energy is flowing between the
> > secondary capacitances and the secondary inductance. The maximum
> > current at the secondary is fixed by the energy in the system
> > (0.5*L2*I2max^2), and this also fixes the maximum dv/dt, as
> > IL2=C2*dv/dt, independent of the coupling coefficient. I think that
> > the problem is related to the proximity between the coils (but this
> > is -another- consequence of high coupling).
> >
> > >...
> > > In a disruptive coil, the amount of energy that can be stored in
> > > the top load at a reasonable voltage (based on radius of
> > > curvature) is much,
> much
> > > less than is stored in the primary C, so there has to be a
> > > mechanism
> where
> > > the energy is transferred from primary C to secondary C, bit by
> > > bit,
> while
> > > keeping the voltage on the secondary C low enough so that you
> > > don't get racing sparks, but fast enough to keep the spark
> > > growing.
> >
> > This is not true. The energy that can be stored in the terminal and
> > in the self-capacitance of the secondary coil is easily all the
> > primary energy. This happens in all those coils where a point must
> > be added to the terminal to have breakout.
>
> But.. for a big coil, where the top load is perhaps 100 pF, and the
> max voltage is, say, 500 kV.  The stored energy in the topload is 12.5
> Joules. At 120 breaks per second, this is only 1500 W.  There is
> substantially more energy stored in the primary C, and if you presume
> that the energy stored in the topload is discharged into the spike on
> each half cycle, and, where you get, perhaps 5 or 6 half cycles before
> "quench", you could get rid of perhaps 50 Joules on each "gap firing",
> for a total power of 6000W...
>
> Clearly, it's somewhere in between.  Much to contemplate here...  I
> suspect that a quick simulation will tell if this is even close to
> being a reasonable theory...
>
> >
> >
>
>
>
>

```