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

SSTC and the Current Drive Ring (IDR) Concept



Original poster: "David Sharpe by way of Terry Fritz <twftesla-at-qwest-dot-net>" <sccr4us-at-erols-dot-com>

Ken, All

I've been working on a concept and floating several trial
balloons privately concerning this post.  I believe
what Ken is saying is correct, and this conceptual driver
concept may dramatically exploit these physical concepts.

The concept is simple, the implementation may not be...
but at least it is theoretically scalable to any power level,
and may even work for a SG TC system.  This has not be
built by anyone at anytime as far as I can ascertain... And
may lead Tesla Coil construction into a significant departure
from past construction techniques.

The Current Drive Ring (IDR) for SSTC
------------------------------------------------------
I'm the "power guy" at work and trying to stay in touch
with latest power electronics techniques.  I've down loaded
much info concerning SSTC, I'm getting ready in fact
to build a small (1kVA) unit.  I was driving to work one
morning and contemplating the previous nights postings
concerning SSTC.  I had seen similar effects of resonator
loading variation causing streamer cut off ("blow-back"
is a term I like to use), because when the resonator streamers
disappear, the impedance shift causes a reactive power
blow-back ("kickback"), which usually results in power
electronics (FET, IGBT) failure.  This is a real problem
for a SSTC driving a reactive resonator load, while a VT
just deals with it...  :^).  In recent discussions with John Freau,
he in fact has had VT's expire (actually high power
arc over and die) from a positive transient riding in on
tank voltage, because of reactive power swings when the
resonator load abruptly shifted.

While thinking about this, I remembered the stability and
performance of similar drivers at work when driving a
low value resistive load while inductively heating this load..
The performance is predictable, there are no reactive
power swings, and in fact you can overload the driver and
as long as you keep the switches cool (by duty cycle or
cooling media, fans, etc.) the system will continue to run.  I
have had very few failures with my units (4 out
of 50 I was responsible), and all were too high a DC supply
voltage, never a power "blow-back" or kickback condition.

I always felt that inductive heating and Tesla Coils had a
linkage, and that maybe significant advancement of the
state-of-the-art was possible if somehow the two separate
technologies could be combined.

Then the "light bulb" went off :
1.  The consensus in the Tesla List (and most of us serious
     experimenters and builders) is that a resonator at its
     fundamental frequency is a lumped series resonant circuit.
2.  A lumped series resonant circuit is basically a current
     driven application.  In fact to get more power (real
     or reactive) into a series resonant circuit (resonator)
     requires the application of more current.  The
     resonator base current also equates into more
     charging current for streamer formation, regardless
     of the capacitive top load or resonator geometry.
3.  More streamer charging current under the right
     circumstances should equate into longer streamers
     and sparks, dependent of top load geometry, hold off
     voltage, and streamer/spark propagation physics.
4.  The base current in the resonator as a first approximation
     is equal to the primary coil current times the coupling
     coefficient.  I have seen this relationship on my magnifier,
     and Greg Leyh also saw it on Electrum.
5.  So based on this logic, one would want to maximize current
     in the primary of a driver (spark gap, VT, or SS driven), to
     maximize base current in the resonator.  Obviously there are
     physical limits to this; for example, if the resonator is
     over coupled to the primary coil, "racing sparks" will
     frequently occur, and there are power input limitations
     and implications.
6.  SSTC, with their inherently low impedance would be
     most capable of performing as a low impedance source
     to a resonator.

Suppose you used a SSTC driver to drive an impedance
matching transformer.  The Primary coil could consist of
50-250 turns, and wound on a ferrite toroid former.
The output of the transformer would be a copper one
turn ring (imagine a Weller pistol soldering gun operating
at 100Khz).  The ring could have water passing
through it to keep it cool.  The idea is reduce the source
impedance seen by the resonator to << 1.0 ohm, stabilize
the transformed load impedance seen by the SSTC driver
and force it to be resistive, and maximize the coupled
current into the resonator. The SSTC driver
is now running as a crude induction heater, and I believe
operational constraints are significantly more relaxed
then with a driver directly feeding a resonator.  You have
built a RF current transformer, running backwards into
a current drive ring, the IDR, and the IDR drives
the base of the resonator.  The overarching goal is to
swamp any reactive power that is reflected back toward
the driver, and make the load seen by the SSTC generator
appear to be resistive (ok, it will have an inductive
component, but hopefully small and predictable...  :^)  )

Multiple drivers could be arranged in series (proper phasing
please) around the IDR.  The resonator can then be mounted
right on top of the current ring, and tight coupling there (even
0.3-0.4) won't bother the driver a bit when there will be currents
approaching 1kA running in the ring.  A cone shaped resonator
could be used very effectively here, allowing a large diameter
copper drive ring to be used.

The ring would be grounded, therefore the voltage
differential between the bottom of the resonator and
the drive ring would be limited, and I would bet
<<100V.  The base resonator drive voltage should
scale with additional in phase driver primaries (due to
increased current and nearly constant current drive
ring resistance).  Heating losses will escalate
exponentially due to current (2 drivers 4X, 3 drivers 9X,
6 drivers 36X).  I discussed this at length with the
TCBOR/High Energy Physics Group at a recent
meeting and we couldn't blow a hole in basic theory.

CT's work, their physics are mature and well understood,
and this application is a impedance transformation CT
running backward into a CT that is unabashedly open
circuited.  A CT open circuited is definitely a no-no,
but a yes-yes in the base of a Tesla Resonator.  The
unknowns as I see it are:
1.  Heating of the IDR at these current levels may
     require supplemental cooling.
2.  Will the <<1 ohm impedance transformation of the
     drive ring really translate into reduction of reactive
     power flows into solid-state drivers, and will the ring
     really look like a resistive load to the driver(s) at power
     levels that are scalable.
I am planning on building a small (1kVA) SSTC, and this
will be one of the first experiments.  Several design issues
I've run into is toroid sizing and proper design of IDR
drive transformer.  A U-U structure would be better
with bobbins and such, but you are really limited in
sizes and material compositions.  A large ferrite like
used for fly back transformers might work, but the DC
biasing for discontinuous current flow requires different
materials then a forward converter... so you're maybe
stuck with a toroid design.  It for sure is the design used
for CT's, maximum volumetric efficiency at what not.

I gotten in touch with Magnetics, Inc. in Pennsylvania, USA.
They are a major supplier of high power high permeability
magnetic cores for SMPS industry.  A
2.25" OD x 1.5"ID x .75" thk. toroid (epoxy coated, 'P' material)
is capable of handling up to 12.5kW -at- 250kHz, and
~7.4kW -at- 100kHz. Heating limited to 25 deg C above ambient.
According to app. engineer, several kA in drive ring NO
PROBLEM with off-line 1/2 bridge topology.  Cost, ~$40.00
US ea. in qty of 10.  So it's just a matter of building something
and trying it...

Regards
Dave Sharpe, TCBOR
Chesterfield, VA. USA


David Sharpe wrote:

> Ken
> Are you suggesting that a 170:1 CT feeding a 1T ring as a drive primary
may be
> any optimum way to feed a resonator?  I have an idea that I've been
working on...
>
> Regards
> Dave Sharpe, TCBOR
> Chesterfield, VA. USA
>

<<SNIP>>