Re: [TCML] IGBT paralleling

[Drake Schutt <drake89@xxxxxxxxx>]

that makes sense to me...

On Tue, Jul 13, 2010 at 11:35 AM, DC Cox <resonance@xxxxxxxxxxxx> wrote:

> There still seems to be confusion here.
>
> In summary, I'm saying that operating a 1200 or 2400 Volt IGBT on say a
> line
> doubled 635 VDC is a waste of money and "headroom"  --- the differential
> between 2400 VDC and 635 VDC.
>
> I trust this will clear the qir.
>
> Dr. Resonance
>
>
>
>
>
> On Mon, Jul 12, 2010 at 4:28 PM, Brian <brianv@xxxxxxxx> wrote:
>
> >  Dave,
> >  That is what I meant, and did say stay under the spec max ratings and it
> > should work. Even running them over the max spec on short duty cycles can
> > be
> > ok. I am not sure where this Headroom thing came from and how it got
> > decided
> > that needs to be filled because it don't. But Like I said I might have
> > missed something somewhere. The Handling voltage I was talking about was
> > nothing more than Max rated voltage of the device and nothing more, what
> > ever the absolute MAX is on the spec that is what I was talking about.
> > Sorry
> > for the confusion.
> >
> > Brian
> >
> > I'mve no idea what a 'handling voltage' spec is.
> > In my training there is max voltage, or standoff voltage.
> > or withstand voltage spec.  This is the voltage the
> > device is specced to NOT breakdown under, when 'off',
> > (The actual voltage for any given specimen will be higher,
> >  a bit, but don't plan on it....).  This means that same
> >  spec is more or less the max the device will 'handle' in
> >  ESD as it is 'handled' during assembly, etc.
> >
> > I can think of no reason why the applied voltage
> > 'needs' to 'fill' this rating: the device should be
> > quite happy anywhere under the rating.  BUT: Higher
> > ratings cost more dollars (euro's, whatever...).  In that
> > sense of the full voltage rating is not being 'used'
> > then money is being wasted.  Or.  The resulting design
> > is more rugged/reliable.  Its a design tradeoff...
> >
> > One thing I've not seen mentioned (did I miss it?):
> >
> > Paralleling 'things', especially active devices, can be
> >  tricky.  Most notably, the devices may not QUITE match:
> >   'identical parts aren't': Frumps law of design...
> >
> > Paralleling power devices can result in Not Equal 'sharing'
> > of currents.  Either forced sharing, or conscious
> > allowance is in order...
> >  best
> > New dwp:     dave_p@xxxxxxxxxxx
> >
> > _______________________________________________
> > Tesla mailing list
> > Tesla@xxxxxxxxxx
> > http://www.pupman.com/mailman/listinfo/tesla
> >
> _______________________________________________
> Tesla mailing list
> Tesla@xxxxxxxxxx
> http://www.pupman.com/mailman/listinfo/tesla
>
_______________________________________________
Tesla mailing list
Tesla@xxxxxxxxxx
http://www.pupman.com/mailman/listinfo/tesla