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

Re: [TCML] Re: Position of baffle inside secondary



The sparks start to diminish in length when the systems are too tightly
coupled because overcoupling causes the main resonant freq to begin
splitting into to separate component frequencies.  Increase values of k and
the split becomes more pronounced.

Wavelength is tied to freq via the velocity of wave propagation, usually
around 0.96 x C (speed of light) for a classic coil.  In classic fashion the
large inductance of the sec coil tries to delay the ability of current flow
to produce a magnetic field, hence the slightly lower than theorectical
velocity.

Each of these two separate component frequencies causes the potential
distribution along the sec coil to occur at different 1/4 wavelengths in
accordance with their specific
frequencies.

If you have the top of the sec hitting best resonance at one freq, and
another point down from the top hitting another resonant peak, then the
potential distribution is hitting
peaks at various points, usually 3/4 to the top of the sec coil.

The localized peak potentials may begin cycles of constructive and
destructive interference which leads to local ionization along the top areas
of the sec coil.  When this occurs the path to the lower (near ground
potential) areas of the coil shortens and massive ionization begins
occuring.  This collective process is called "racing sparks".  The air is
now overstressed and can cause breakdown (as example) from top 3/4 winding
to lower 1/4 (near ground) winding.

This process really isn't anything new as it was well documented in many of
the old wireless textbooks circa 1915-1925, Morecroft, et. al.

Most of the 1980 thru 1990 ARRL Radio Amateur's Handbooks also contained a
section on the undesireable effects of overcoupling between two magnetically
linked inductors.

Dr. Resonance




On Tue, Jul 14, 2009 at 12:46 PM, Paul Nicholson <tcml88@xxxxxxxxxxx> wrote:

> Dex Dexter wrote:
>
> > I was surprised to learn of the existance of the optimum
> > coupling with respect to the best performance of SG coils.
>
> Theoretical considerations must be guided by the hand
> of practical experience here.   It is regularly but not
> always observed that onset of racing arcs limits coupling.
> Nobody knows why this is.   High k leads to fewer cycles for the
> energy transfer, but all the voltages and currents remain about
> the same - at least at the fundamental operating frequency.
> So why should racing arcs and flashover appear?  Unknown.
>
> > But even if the conditions are such that secondary itself
> > is free from undesired discharges ,beyond certain coupling
> > point spark lenghts might start to diminish!
>
> Here too, the reasons are unknown.  Coupling affects the
> envelope of the RF, and therefore perhaps the ability to build
> long streamers.  Little is known about how the coil/topload
> combination interacts with its irregular and complicated spark
> loading.  So many variables involved, instrumentation is a
> challenge, and realistic modelling of the TC/spark interaction
> is, well, a pipe dream.
>
> Nobody has yet had the audacity to square up to these problems.
>
> I think that practical experience has led to 'typical' TC
> design rules that (re)produce coils which are probably close
> to optimum.  I don't think there's any yet-to-be-discovered
> recipe that will go way beyond the present state of the art.
> Our inability to explain why a certain range of k is desirable,
> and so on, is merely an embarrassment to theorists, rather than
> a hidden key to great new performance.  Perhaps that thought
> is why nobody is pushing these boundaries of our understanding.
>
> (Not to mention the hideous cost of a digital scope with enough
> RAM to capture the entire bang waveform at high time resolution!)
> --
> Paul Nicholson
> --
> _______________________________________________
> Tesla mailing list
> Tesla@xxxxxxxxxxxxxx
> http://www.pupman.com/mailman/listinfo/tesla
>
_______________________________________________
Tesla mailing list
Tesla@xxxxxxxxxxxxxx
http://www.pupman.com/mailman/listinfo/tesla