Re: New SSTC and topology in need of review

["Tesla list" <tesla-at-pupman-dot-com>]

Original poster: David Sharpe <sccr4us-at-erols-dot-com> 

Hi Steven

Tesla list wrote:
>Original poster: "Steven Ward" <srward16-at-hotmail-dot-com>
>
>Hi everyone,
>
>I have been working out my latest SSTC designed in the hopes that higher
>voltage across the primary coil will work better than typical low voltages
>of some 170-340V.  This new design should be able to produce about 1400V
>across the primary in a very unique way:
>
><http://www.hot-streamer-dot-com/srward16/SSSSTC.htm>http://www.hot-streamer-dot-com/srward16/SSSSTC.htm 
>
>
<<<SNIPPO>>>

I found this particular posting on the web and want to investigate with the 
IR2153D
integrated driver chips.  This particular topology was also disclosed 
earlier this
year at the IEEE APEC (Applied Power Eelctronics Conference) 2003 in Miami.

Section 26.6  "A Multilevel Inverter Topology for Inductively-Coupled  Power
                       Transfer", J.I. Rodriguez, S.B. Leeb, Massachusetts 
Institute of
                       Technology, Cambridge, MA.

This basic topology is related to NPC (neutral point clamped) multilevel 
converters,
and "flying capacitor" power converters (both of these phrases can be 
researched
further on Google).  I have not personally built this circuit.  My concerns 
are in the
area of multiple isolated relatively high speed drive signals having to be 
coherently
generated to operate the stages, and each succeeding stage having Vapplied 
between
the isolated gate drivers.  At some point, the gate drive isolation holdoff 
voltage will
have to be greater then  Vin*n stages  (referenced to control ground) for the
control circuits to survive.  The author's opinion is because the power 
stages are
capacitively coupled that voltages are additively rippled through the 
converter, and
voltages will be more or less shared between converter stages.  I'm a 
skeptic...
show the data to substantiate that opinion!  It is interesting to note that 
no photos, jpegs
or drawings of physical construction of this converter was shown, although 
synthesized
output voltage waveforms from a converter were presented (IP, patent 
issues???)

IMO, the only way to get beyond perhaps 1-2kV GDT hold off limitations will 
be with
fiberoptic links and local floating battery power.  This technique has been 
used by
our moderator for the first OLTC control circuits.  These issues also 
explain why
HV POWER circuitry is so darned expensive; CAD modelling, multiple 
interations,
and dumpsters full of smoking, burnt, EXPENSIVE parts.  If you think I jest,
consider the efforts by Marco Denicoli with THOR's very large SMPS...

Regards
Dave Sharpe, TCBOR/HEAS
Chesterfield, VA. USA