Re: [TCML] DRSSTC ring-up

[Steve Ward <steve.ward@xxxxxxxxx>]

Herwig,

As Antonio said, there is nothing "wrong" with your setup.

What most coilers do is tune the primary lower than the secondary frequency
such that the primary is in tune with the secondary once it has a big
streamer on top of it.  If you try to tune the primary too low
(anticipating a gigantic streamer) you will find that the driver just
cannot put out the current required to ring up to such a high voltage, and
either the IGBTs fail or your circuit is smart enough to limit the current.
 Or its possible your circuit is so lossy that it cannot develop a high
enough voltage to impress you.

It turns out another efficient tuning is to leave the coils more or less
"in tune" with eachother, but have a driver thats smart enough to drive it
at the upper pole frequency (which might be significantly higher).  This
can excite just the one resonance (upper one) after enough cycles, and i
find in my latest experiments that driving at the upper pole while
"detuned" for streamer loading is much more efficient than driving at the
lower pole with streamer detuning.  The only reason people havent done this
more is because most people use a simple self oscillating controller which
will happily excite whatever frequencies naturally come about when starting
up the TC.

Steve


On Thu, May 23, 2013 at 12:41 PM, Herwig Roscher <herwig.roscher@xxxxxx>wrote:

> Hi Daniel,
>
> > Do you have any pictures of your setup?
> - Unfortunately there are no pictures of my setup. Primary is 5 turns AWG
> 8 on 3.55" dia
> form. Secondary is about 1000 turns AWG 32 on 3" dia form in a "standard
> arrangement".Topload is made of 2 small toroids.
> The half bridge is "standard" as well:
> GDT, bridged gate resistors and protection suppressors at the inputs. PP
> caps and 400 V
> TVSs at the output. The only "special" feature is the connection of the
> MMC to GND
> instead to the usual midpoint of 2 serialized capacitors across the rails.
> I've tested this setup with a resistive load of 1 kW and rail voltages up
> to 400 V
> already.
>
> >  with only 75A peak and 57V on the bus, you're simply not developing
> enough voltage at
> > the topload to break out properly.
> - So obviously I need to be more courageous.
>
> > What IGBTs are you using
> - IXGH48N60C3D1 (diode integrated), V ces = 600 V, t fi = 38 ns, I C25 =
> 75 A, I C for 1
> ms = 250 A
>
> > absolute maximum pulse pulsed Ic-e as 180A, but are completely reliable
> up to 300A
> > peak current in a DRSSTC.
> - That's an important info!
>
> > A better way to test your coil might be to set your OCD less
> conservatively
> - Thank you very much for your advice. I'll give higher currents a chance.
>
> Herwig
>
>
> _______________________________________________
> Tesla mailing list
> Tesla@xxxxxxxxxx
> http://www.pupman.com/mailman/listinfo/tesla
>
_______________________________________________
Tesla mailing list
Tesla@xxxxxxxxxx
http://www.pupman.com/mailman/listinfo/tesla