UCC28220: only one channel works properly

Gabor Futo

Part Number: UCC28220

I built a similar circuit:

http://go.wolfspeed.com/l/101562/2018-07-02/552z75/101562/70670/CPWR_AN24.pdf

With such parameters: Vin=300...400V, Vout=440V, max power 10KW, Cree C3M00651000K.

Slope, SS, CTRL, CHG, DISCHG, CS1 and CS2 circuit is the same as in the document.
I only changed the output and input voltage sensing.

I started testing at ~1KW with a lab power supply (350V input, 440V output, ~200ohm load).

One channel works well, the PWM signal can be measured from the controller normally (from UCC28220 to Cree gate driver module). SiC FET works well on this channel.
But on the other channel I only measure one spike instead of the normal PWM signal. On this channel, SiC FET will only turn on for a moment.

My question:
Is this a normal operation because of the small output load/power?
Or should the two channels work alternately in each case?

Thanks for info!
Regards,
Gabor from Hungary

over 6 years ago

0 Peter Meaney over 6 years ago

TI__Mastermind 31580 points

Hi Gabor,

What is the CTRL pin voltage level at this load level. If you test with a 100R load does the other phase of the UCC28220 start to operate ?

Both phases should be operating but at lighter loads if the CTRL starts to drop to 1V or below it could be possible that due to mismatches between the two channels that one channel starts to provide the largest portion of power.

Regards

Peter

0 Gabor Futo over 6 years ago in reply to Peter Meaney

Prodigy 50 points

Hi Peter!

CTRL pin voltage is 3.56V.
SS, Slope, CHG, DISCHG, REF pins are OK (based on documentation).
"Line good" conditions are also good.
Vin = 12,08V

The enclosed images show the OUT signals of the two channels (same timing and voltage setting).

What can cause this?
Thanks for info!

Regards,
Gabor

0 Gabor Futo over 6 years ago in reply to Peter Meaney

Prodigy 50 points

+Info: Two hardware has been built, same problem on both devices.

0 Peter Meaney over 6 years ago in reply to Gabor Futo

TI__Mastermind 31580 points

Hi Gabor

And how about the current sense signal on the pins of the device?

Regards

Peter

0 Gabor Futo over 6 years ago in reply to Peter Meaney

Prodigy 50 points

CS1 and CS2 are the same, They are around 0V, with little noise at switching.

0 Gabor Futo over 6 years ago in reply to Peter Meaney

Prodigy 50 points

Do you have any idea for this? What could be the reason? Or is it a normal operation at a lower load?
Unfortunately I can't try it with a higher load (lab power supply no more).

0 Peter Meaney over 6 years ago in reply to Gabor Futo

TI__Mastermind 31580 points

Hi Gabor,

Can you swap the part for a new device ? Or alternately, disconnect the gate drive from the good channel, short the gate of the mosfet on the good channel to the source to ensure the mosfet is off and try and load the suspect channel only.

Regards

Peter

0 Peter Meaney over 6 years ago in reply to Peter Meaney

TI__Mastermind 31580 points

Hi Gabor,

I hope you have made some progress, since it has been 10 days since the last update I am going to close the post. If you have more questions please open a new post.

Thanks

Peter

Power management

Power management forum

UCC28220: only one channel works properly