Re: [TCML] SSTC half bridge

[Steve Ward <steve.ward@xxxxxxxxx>]

Sure Dan, but we are talking about also having a DC blocking cap, which as
you know will not allow the Volt-seconds to become un-balanced (in the long
run - after a few LC time constants), thus flux walking is not an issue.

Herwig,

I almost exclusively prefer using GDTs.  Unless the frequency is really high
(say more than 500-600khz) then GDTs are quite good.  Im not sure where your
ringing was, if it was on the gate, then you must question whether or not
the ringing matters.  If the gate voltage stays sufficiently high to keep
the MOSFETs on, then the ringing isnt a big deal.  Ive developed high-side
gate drivers in the past for special jobs, and they are fine, but the
robustness and simplicity of a GDT is really hard to beat when you can take
advantage of it.

If you would like to elaborate on your troubles, there are at least a few of
us who can offer up suggestions on how to reduce them.  Its important to
keep in mind that many papers on power electronics may often idealize
waveforms, which are typically not practical, so it takes some experience
and judgment to say what is acceptable or not.

Steve



On Thu, Jun 10, 2010 at 7:03 AM, McCauley, Daniel H <
daniel.h.mccauley@xxxxxxxx> wrote:

> Herwig,
>
> The issue with non-exact 50/50 duty is that you can develop flux imbalance
> in the core, and if the core doesn't reset fast enough, you can end up with
> whats called flux walking where the flux densitiy on the core starts
> trending towards either positive or negative quadrant of the BH curve where
> it will saturate.
>
> Dan
>
>
> Steve:
>
> >  I cannot think of any severe/negative consequences of non-exact 50/50
> > duty cycle
> - Thank you very much for sharing your knowledge.
>
> Up to now my half bridge consists of two IRFP 460s with Schottkys and
> external freewheeling diodes.
>
> I have tested two drivers:
>
> a.  GDT driven by TC4421. Problem was the ringing.
> b.  IR2113. Problem was VS undershoot. After many killed semiconductors I
> eventually solved this problem.
>
> Now I am planning on testing a more robust driver with IXDD414s, the high
> side driver using an isolated power supply and a fast opto coupler for
> signal transfer.
>
> What kind of driver do you favorite ?
>
> Regards,
>
> 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
>
_______________________________________________
Tesla mailing list
Tesla@xxxxxxxxxx
http://www.pupman.com/mailman/listinfo/tesla