Re: IR - SSTC Half-Bridge Driver

["Tesla list" <tesla-at-pupman-dot-com>]

Original poster: "David Sharpe by way of Terry Fritz <twftesla-at-qwest-dot-net>" <sccr4us-at-erols-dot-com>

Hi Jan!

Tesla list wrote:

 >
 > Well, rather like 10-100kHz limited because the output dead time is,
 > according to specs, 1.2us typical and 0.75us minimum.

You are correct, I've set my circuit up for an Fo from ~25kHz to 200Khz
just to give adequate tune range for unit.  Higher frequency is possible of
course with FET's, but I think IGBT's are much more robust, since I shouldn't
need the series schottky + additional HV fast recovery diode around the
"simple application" FET (yea... right  :^)  ).

 >  > 3.  Chip will operate from ~10V to 600V input (with proper
 >  >       Vcc 5mA current limiting resistor or 5mA constant current
 >  >       source)
 >
 > I wonder what the peak output current of those gate drivers might be...
 > The specs don't say it. Around 250mA maybe? (but it seems to be "enough"
 > for IRF820 in the PFC/ballast circuit, anyway :)

I don't know what the drive capability of the chip is, will converse with 
IR App
Engineer next week.  Real issue is Qsw drive capability to FET or IGBT.  I'm
going to get some Fairchild Induction Heating IGBT's samples on order,
part number SGH40N60UFD.  600V, 40A, 120A pulse rated. Total switching
loss ~ 0.6mJ per cycle, device rated to 160W.  At 100Khz, switching loss
is ~60W each, so will need a good size heating and force air cooling, but this
is at a current of 40A and rated voltage.  That's a whopping hel.. of a bridge
(600V * 40A / 2 = 12kW!!!)  I'd be happy with 2-5kW, with OVP and OCP
integral to the design.

 >
 > If I read the functional block diagram correctly, you could just put some
 > feedback signal from the TC secondary directly into the C_t pin, with
 > DC level shifted up to 5V. And leave out R_t, and C_t.

Could you elaborate concerning the auto-tune possibility?  If you use a
high frequency current transformer (like LEM), detect, amplify, and run
into a schmidt trigger, you might be able to "fake" the unit into self
oscillating at resonator Fo.  Another cool thought I had was if you
could invert and signal isolate the Ct signal from the first chip and
feed into Ct input of a second chip (external time synchronization),
you might be able to create a full bridge system by adding one additional
chip and some minor interface electronics.

 >
 > As long as the internal prop delay of this chip isn't too much, this'd be
 > the ultimate single-chip-two-IGBT SSTC solution! 8-)
 > Thanks for sharing the info about this IC! :)
 >

It really is beginning to appear that way!  Hope to find out soon!

 >
 > cheers,
 >
 >   - Jan
 >
 > --
 > *************************************************
 >   high voltage at http://www.hut.fi/~jwagner/tesla
 >   Jan OH2GHR

Best Regards
Dave Sharpe, TCBOR
Chesterfield, VA. USA