Primary Frequency too High

[alan sailer]

I am trying to troubleshoot a DRSSTC. It's a small coil using SKM100, full bridge and it designed
to run at about 120KHz. I'm using the UD2.7C thru-hole board.

The problem I am trying to figure out is that the coil is oscillating at about twice the
expected frequency. The SKM100s are not happy and the drain current (such as it is) looks like
a triangle wave. I am running the coil at a buss voltage of about 60 volts to troubleshoot the thing.
The buss voltage is generated through a Variac driving a doubler.

I've built a fair number of coils and they have all started at near the design frequency. I have
measured the tank cap and the primary inductance using a DE-5000 LCR meter and they are all within
5% of the expected value. Even a basic LC resonance calculation says that I should be getting 120KHz.

EDIT:

I decided to disconnect the FB transformer and drive the UD2.7C directly with a signal generator. Set it at 120KHZ and 500mVpk-pk.
The bridge drives the primary just dandy with a textbook current and voltage waveform. I'm even getting
little bitty sparks from the secondary/toroid. So the bridge works, the UD2.7C works. And since I got oscillations when I attch the FB transformer
I think the transformer is functioning.

Cheers.

[davekni]

Have you tried reversing CT phase?  That would be my guess.

[alan sailer]

Dave,

Crapping crud. It works.

I know my knowledge of coils is shallow but I though it was binary. Get the phase wrong, nothing. Get the phase right, bingo.

So can you, for a me being a dunderhead, explain why?

If not, much thanks anyhow. You made my afternoon. I was ready to go to an astrologer.

Cheers.

[davekni]

With reversed CT connection you are starting with 180 degrees phase shift, exactly out of phase.  At 240kHz your driver and IGBTs have enough delay (additional phase lag) to total close enough to 360 degrees to oscillate.

One possible contributing factor is if you don't have phase-lead enabled on your UD2.7 driver.  Adjusting phase lead for ideal switching at 120kHz (H-Bridge outputs switching just before current zero-crossing) would likely preclude enough lag at 240kHz for reverse-CT oscillation.

[alan sailer]

Dave,

Thanks for the explanation. It makes sense to me and is something a lot more subtle than my basic SSTC
knowledge covers. It also explains why I have not seen this problem in any of my previous SSTC builds
(two QCW and one conventional, all full bridge). I was just lucky.

The phase shift inductor was in place but at this point I have no idea if it is the right value.

Once again thanks. I am not sure that I would have ever figured it out given my basic understanding
of FB.

Cheers.

[davekni]

Alan,

You are certainly welcome.  Great to hear that it made sense.