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

RE: drsstc - catastrophic failure!



Original poster: "by way of Terry Fritz <teslalist-at-qwest-dot-net>" <Marco.Denicolai-at-tellabs-dot-com>

Jimmy and Ken,

Sorry for catching on so late (been on holiday).

Don't give up the s.s.! On the contrary, I invite you to investigate
(and document) more in detail the failures. There is much me, you and
all the others too have to learn from these failures. And it's better to
learn from killing a discrete IGBT than from killing an expensive brick.

Ken, you are right: it's a curse that you can't measure "in-operation".
I suggest you to use my approach:
- measure with reduced input voltage
- find the spikes and transients (e.g. on Vcc, Vge, etc.)
- develop solutions to reduce them at a minimum

This way you'll have "things in order" also for full power or, at least,
as good as you possibly can get. Did you read the description of my last
development on Thor's power supply (i.e. the revision B stuff at
www.iki.fi/dncmrc)?

Best Regards

 > -----Original Message-----
 > From: tesla-at-pupman-dot-com [mailto:tesla-at-pupman-dot-com]
 > Sent: 18. huhtikuuta 2003 19:49
 > To: tesla-at-pupman-dot-com
 > Subject: Re: drsstc - catastrophic failure!
 >
 >
 > Original poster: "K. C. Herrick by way of Terry Fritz
 > <teslalist-at-qwest-dot-net>" <kchdlh-at-juno-dot-com>
 >
 > I don't know what that "Hello everyone" paragraph means
 > unless it's curse
 > words.  With "microsoft" in it, I guess it has to be...
 >
 > So welcome to the club, Jimmy.  I just blew out half the IGBTs I'd
 > incorporated into my re-done s.s.t.c.  I'm within a whisker
 > of giving up
 > on s.s. completely.  Primarily because it's impossible to get scope
 > probes near the circuits while the whole thing is running: can't
 > trouble-shoot it with the pedal to the metal.
 >
 > Except for the mains-frequency high voltage, I'd be inclined
 > to give in &
 > go for the tried-&-true; but I hate 60 Hz H.V.
 >
 > Ken Herrick
 >
 > On Thu, 17 Apr 2003 22:10:59 -0600 "Tesla list" <tesla-at-pupman-dot-com>
 > writes:
 >  > Original poster: "jimmy hynes by way of Terry Fritz
 >  > <teslalist-at-qwest-dot-net>" <chunkyboy86-at-yahoo-dot-com>
 >  >
 >  >
 >  > Hello everyone,<?xml:namespace prefix = o ns =
 >  > "urn:schemas-microsoft-com:office:office" />
 >  >
 >  >
 >  >
 >  > A few days ago I had it running at 280 volts, pulsing the primary
 >  > coil and
 >  > capacitors for 2 cycles (400 amps peak) every second. But now it is
 >  > dead.
 >  >
 >  > I was running it with 150 ohms in series with the ac hot,
 > and it was
 >  >
 >  > working fine up to around 280 volts. At around 280 volts, the
 >  > microcontroller started freaking out every few pulses and sometimes
 >  > shut
 >  > off. I am almost certain it picked up noise from the primary coil a
 >  > few
 >  > inches away. I then looked at the power line to the micro. It was
 >  > all over
 >  > the place from 0 to six volts, following the 60khz current in the
 >  > primary.  I had forgotten to add a decoupling capacitor. That was
 >  > stupid,
 >  > as it could have saved my project. I will put some fat ones on
 >  > before
 >  > running it again ;-).
 >  >
 >  > I was just about to turn the thing off, when I heard a
 > loud pop, and
 >  > the
 >  > dog ran into the bedroom to hide under the table  After I turned it
 >  > off, I
 >  > looked over and saw smoke.  If the capacitor bank was at 280 volts,
 >  > there
 >  > were  88 joules in there. The pop killed seven out of the eight
 >  > IGBTs that
 >  > I was using! These things are tough, but not indestructable.  I can
 >  > find
 >  > ways to break anything ;-). The IGBTs are all in one piece
 > but every
 >  > pin is
 >  > shorted together. I think that happens when they are
 > over-volted. My
 >  > guess
 >  > is that the microcontroller got confused and told all of the IGBTs
 >  > to go on
 >  > at the same time (I have no shoot through protection).
 >  >
 >  > I might try some brick IGBTs, because my assembly was pretty much
 >  > destroyed, only the diodes are  left on the heatsink.
 > There are some
 >  > on
 >  > ebay now that look fast enough, so I might snatch them. I am also
 >  > trying to
 >  > contact Powerex about their complete Pow-R-Pak modules. I want to
 >  > finish
 >  > this thing!
 >  >
 >  > I've also gotten advice from Bill Reass, who designed a resonant
 >  > pulse
 >  > transformer for physics research at Livermore Labs.
 >  >
 >  >
 > <http://arxiv-dot-org/ftp/physics/papers/0008/0008125.pdf>http://a
rxiv-dot-org/ft
p/physics/papers/0008/0008125.pdf
  >
  > I want dibs on his power supply when they scrap it out :-). They're
  > pulling
  > 10MWatts peak and 1MWatt average power!
  >
  >
  >
  > Jimmy
  >
  >