Richard Prosser wrote:
>
> Yes - I was thinking that the isolation would occur at the flyback
> transformer (e.g a similar size to a 200W computer power supply
> transformer). Pulse steering would be accomplished by using optos or pulse
> transformers to switch the FETs/Bipolars/SCRs in order to retain isolation.
>
> As far as overall efficiency is concened, we manufacture 2 stage recifiers
> (230/110V to 470VDC  to "48V" DC) with peak efficiency of around 93% at >
> 2kW power levels. The output switching frequency is in the 400kHz region
> This is the sort of efficiency figure I was contemplating. By using highish
> frequencies and cunning circuits so called "soft switching" can be utilised
> - at the expense of dearer switches.
>
> Richard P


Hi Richard, I too think this is an intriguing new
market. There is no way I would produce a device like
this without total mains isolation, but that's not a
huge hurdle. :o)

There seems to be two basic ways of going about it:

1. 50Hz toroidal mains transformer. Then just use
PWM to feed current into the secondary which must
be fed back into the mains. Efficiencies >95% for the
transformer, total isolation and very easy to detect
mains phase and absence. Bad point is weight and cost.
But many good points.

2. Hi freq dc step-up variable supplies, generate a high
DC voltage that can be fed into the mains direct as
DC current every +/- phase. Can be much smaller and
lighter (cheaper??) but will require a high speed high
volts switching diode for each +/- step-up circuit,
and also a high volts transistor or fet to isolate
each DC supply when the other phase polarity is active.
It also requires some communicaton from mains back to
low volts circuits, for mains detection and for switching
the isolator semis. This is a lot more high volts semis
to go bad, but would be cheaper in larger quantities
like "real" commercial production.

For smaller production and testing I am leaning back
towards the toroid transformer option 1. Safety is great,
even in the event that one main power semi might short,
there will be no catastrophic load cuurents and it is easy
to fuse. It might take some fine tuning to determine
best phase angles to feel power in as the toroid will
have different effects under different loads I imagine.
-Roman

--
http://www.piclist.com hint: To leave the PICList
mailto:piclist-unsubscribe-request@mitvma.mit.edu