On Wed, 20 May 1998, Morgan Olsson wrote:

> At 10:26 1998-05-20 +0000, you wrote:
> >> At 04:39 PM 19/05/98 +0200, you wrote:
> >> >...
> >> >>  Note that if you do NOT use a transformer, then you will have a
> >> >>triangle-type output and will not be able to ground the center point
> >> >...
> >> >I have distant plans of a small wind power plant.
> >> >I have been thinking about a 400V 3-phase generator like this:
> >> >DC input is +340VDC, GND/Neutral, and -330VDC
> >> >Switching stages are one pair IGBT(or something) for each phase, connectin
g
> >> >to +, -, or both off.
> >   ^^^^^^^^^^^^^^^^^^^^ Gulp. What is going to filter that ? Do you know
> >what happens if you put an inductive load on such a 'phase' ? Seriously,
...
>
> A little exaggeration, i presume... It is a question of design.

Yes, but 1kW is something you might work with soon, if you own a computer
and OSes go the way they go... I use a 400 W one ;)

> Actually I have been studying a 48VDC to 230VAC 250W converter I bought
> before, that uses the above design.  Works very well.

If it does not use some 'sinus magic' then you had better use it ONLY for
what it is designed, unless there is some filter on the output (LC).

'sinus magic' is the key, and it usually consists of a chopper DC/DC
converter disguised as the switching IGBT pair, with the transformer coil
being the other key ingredient: its inductance integrates the HF switched
DC voltage into reasonable sinus, aided by a few large capacitors that
tune the circuit properly to remove the chopper frequency from the output.

  This works from 5 W, up to about 10 kW. After that you start having
problems with components and stray radiation. SERIOUS problems. The
mentioned 5% distortion becomes 500 Watts, mostly HF, and keeping those
out of the consumer's power grids is awful hard to put it mildly. Remember
that pesky light dimmer that hummed into all the radios in the house no
matter what you did, in despite of its miserly 50 Watt bulb load ? Well,
think of the same effect, only about 200 *times* worse.

> Im am interested if anyone can describe how theese class 600MW converters
> for long distance power lines work.  (Made by ABB (Asea) to transfer power
> to other side of seas, from distant power plants etc.  How do they do the
> 600MW HVDC to AC conversion??

Why, you must have seen them, only they were so big you did not see them
as what they were. Remember when you drive past something called 'an
electrical substation' ? And drive, and drive, and drive, and drive, and
then there is an end to it and you can listen to the radio again ? Well,
that's the place. Ah, and, did you notice that I said 'listen to the radio
again' ? Hehehe. You were saying something about my exaggerating with my 5%
above, yes ?

As to the components, well, I had to do with some low power ones <G>. They
were thyristors for 8 kW loads at 380 V AC, and each was about the size of
an onion, with wires the size of my thumb, and a screw terminal that makes
a 1/2" pipe thread look tiny, for the water-circulated heatsink block.
They were used in groups of 6 for stage lighting control... This was 10
years ago. So figure...

Peter