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

Terry's DRSSTC - higher power testing

Original poster: Terry Fritz <teslalist@xxxxxxxxxxxxxxxxxxxxxxx>

Hi All,

I set up a test load:


The fake primary system had a Q of about 10 for testing the DRSSTC. I bypassed the buss charging circuit to a variac on the box.


With this I could control the buss voltage into a sort of real load and scope things.

I spent all day yesterday putting the controller guts into the metal box... When it gets warm outside I will paint it up nice and all, but for now, it is finally in a "box"! I really like that controller! With the fiber optics you are not going to get electrocuted and the microcontroller's programing is real nice!! Smooth linear control and when stopped it resets everything to low power so no surprises... Works real smooth and natural... Glad that all worked out since it added a lot of $$$ to the project.

So I fiddled with it.... Spent a lot of time worry about little glitches the Tek P5205 differential probe was seeing: I used the Tek P5100 where I could and my favorite probe continues to please me ;-))


Of course, with a little thought and trying other probes, most of those glitches are just noise being picked up in the P5202's 6 inch unshielded probes... It is a super nice thing, but it has a real "personality" when it comes to noise.... After a while I figured best just to set the scope for 20Mhz BW Limit and be happy ;o)) You just can't get 50V across a 30V TVS ;-)) A common mode choke on the output did seem to help noise so that may be a real improvement...

But after awhile I worked my way up to a 340V buss and 50 amps out:


I had the current limit set to 50 amps and it kicked right in without any problems. I uped the BPS to 300BPS and nicely burnt the dust off the poor little 400+ watt power resistors. The IGBT temperature (I can read that off directly in the protection circuit) went from 22C to 23C ;-)) So the IGBTs seem to run very cool. At 500 amps that is 10^2 so maybe +100C then, but that is a pretty rough situation when the thermal limits probably "should" cut in. I got to where I could "push all the buttons" and play with it at 300V without worries... I can't push anymore into the "fake" system now. So a little more fiddling with dead times and put it all back together...

I think Steve Ward is right about the really fast dead times. Having the IGBTs all off seems to just invite all kinds of weird harmonics and ringing during the dead time... Probably best to really push them as fast as possible.... Have to watch heating and, of course, cross conduction, but really long dead times seem to be a real waste...