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

OLTC update - good news :-))



Original poster: "Terry Fritz" <twftesla-at-qwest-dot-net>

Hi All,

Some good news.

I got some spare IGBTs from DigiKey and tested them at high current using
this fast setup that is sort of close to the real OLTC application:

http://hot-streamer-dot-com/temp/OLTC08-30-01.jpg

I managed to run one at over 700 amps!!!

http://hot-streamer-dot-com/temp/OLTC08-30-02.gif

What is cool is that I seem to have gone higher if I had more gate voltage.
 Everything was rock stable and you could easily see the current limit as
the IGBT starved for gate drive for more even more current.  It just mildly
started to dissipate more heat and all as it lost full conduction.  But it
still didn't explode or anything (just a 'tiny' heat sink on it).  So they
'can' be run well over their rating (204 amps) and survive.  This means I
should be able to easily run my coil of 240 VAC with the ten IGBTs I have
without having to add more or making major changes :-)))

BTW - Very interesting post from Greg Leyh tonight!!!

I will need to modify things a little bit:

1.  I should change the gate drive resistors to 10 ohms instead of 100.
These IGBTs don't seem to have any problem with low gate drive resistance
even at very high currents.  The transorbs can take care of the shoot
through effects, if any. 

2.  I will see if I can go to 27 volt gate drive instead of 18 volts (3
batteries instead of 2).  I am a bit worried as the 9 volt batteries wear
down that they may go too low in voltage which would be "bad".  With three
batteries and 30V transorbs, I would have plenty of margin.  27V is well
over the gate "spec" voltage but that does not worry me at all (me and
Marco will have to talk on that ;-)) 

3.  I have to remember to change the gate transorbes to 30 volts and the
rail ones to 600 volts...  Watch me forget that one :o))

BTW - Marco wanted to see the gate voltages:

http://hot-streamer-dot-com/temp/OLTC08-30-03.gif

It's the red line.  I hate these differential measurements especially with
cheap 100:1 voltage probes like I used here.  But there it is...


The scope picture at:

http://hot-streamer-dot-com/temp/OLTC08-30-02.gif

Is really very interesting.  You can see where the IGBT turned off just
after the forth top hump.  The reverse diode conducted what was left of the
negative side current.  Then, the waveform stopped like a rock.  An amazing
feat considering the current and the one poor little device that was doing
it ;-))  I imagine the transorbes across the "coil" were seeing some spikes
but they seemed to handle it just fine.


Jimmy's post and Paul's secondary virtual data base seem to suggest that my
secondary Q should be more like 200 rather than only 36.  So I probably
screwed something up there.  I bet it was the SonoTube.  I was in a hurry
and just picked that up instead of a nice PVC tube...  


Oh oh!!!........... Paul just wrote me back with a tcma Q of 45 :-P
So,....  It may get messy...