• State Resolved
  • Locked Locked
  • Replies 4 replies
  • Subscribers 63 subscribers
  • Views 127 views
  • Users 0 members are here
Support feedback
Options
Options
Similar topics

This thread has been locked.

If you have a related question, please click the "Ask a related question" button in the top right corner. The newly created question will be automatically linked to this question.

UCC28056: ZCD does not switch in valleys and varies duty-cycle

Marius Schmidt
Prodigy 10 points
Part Number: UCC28056
Other Parts Discussed in Thread: TPS92314

Hello,

i am now trying for a long time getting the UCC28056 to work, it's basically operating fine and performing PFC action. But measuring the switch waveforms i notice that valley switching does not really work and the IC operates in discontineous mode. Also the on time and frequency varies with input voltage (different on-time for valleys/peaks).

I took some measurements with the following results:

ON-Time in sine valleys is 3.4µs and period is 15.8µs (63.3kHz) - duty is 21.5%
ON-Time in sine peaks is 2.9µ and period is 17.8µs (56.2kHz) - duty is 16.3%

The variation of duty cycle of course causes current distortion especially in the 3rd harmonic.

Voltage on COMP pin is stable (at 915mV) - the output voltage is stable and in regulation.

I attached some screenshots of drain-source voltage (green / CH2) and ZCD/CS waveform (red / CH4).
ZCD/CS network is (3MOhm||47pF) for drain-source path and (7k5||15nF) for current path - this leads to divider ratio Kzc of 401. The ZCD/CS voltage is a little bit too high - i have already tried to compensate by adding a few nF parallel to the 15nF capacitor - it does not improve the general behaviour.
I am using RC lowpass 10k+10pF to filter out HF noise on ZCD/CS.
Input voltage is 115VAC with 800Hz.

First screenshot shows switch waveform for input valley voltage with long duty cycle and switching occours in a valley (but apparently a bit too early).
Second screenshot shows switch waveform for input peak volage with short duty cycle and switching occours shortly after a valley - current reaches its maximum (VZCOcp1 at 0.5V).
Third screenshot shows switch waveform shortly after input peak voltage with short duty cycle and switching does not occour in a valley - current is not limited at VZCOcp1 anymore.

Question 1: Why does the duty cycle vary over an input voltage half-cycle?
Question 2: Why does valley switching not work? In my opinion the ZCD/CS waveform looks quite ideal.

I hope anyone has some hints for me what I could optimize to make it work - Thank you!

  • 0
    Prodigy 10 points

    I forgot to mention something quite important - the output is isolated and I am (mis)using the controller as a flyback converter - output voltage is 24VDC.
    This is why the drain-source voltage follows input voltage and is not constant.
    This might be a little bit of an issue regarding regulation (?), but i'd expect the valley detection to still work properly.

  • 0 in reply to Marius Schmidt
    TI__Guru 73275 points

    Hello,

    Your inquiry has been received and is under review.

    Regards,

  • 0 in reply to Mike O'
    TI__Guru 73275 points

    Hello,

    Your issue might be that you are using it in a flyback.  This controller was design to be used with boost converters.  From your waveforms it looks like zero current detection is not working in this applicaiton.

    There is a PFC controller that was designed for flyback converters.  It is the TPS92314 and the data sheet can be found at the following link.  I would recommend you use this controller instead.

    https://www.ti.com/lit/ds/symlink/tps92314.pdf

    If you have questions on how to use the TPS92314 please repost with TPS92314 in thread title so the appropriate applications engineer can answer your questions.

    Regards,

  • 0 in reply to Mike O'
    Prodigy 10 points

    I know i was using the regulator off-purpose, but still i can't see / understand why it would not work as a flyback. Well, it was worth trying and i'd really have liked to use the UCC28056 because it packs so many features in a small footprint.

    Thank you for your response, I will try to evaluate the TPS92314 if it can work for our application.