• State
  • Locked Locked
  • Replies 5 replies
  • Subscribers 65 subscribers
  • Views 1179 views
  • Users 0 members are here
Support feedback
Options
Options
Related

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.

UCC28950: Abnormal pulse widths issue for OUTC and OUTD occurs only in burst mode

Yoshwa
Expert 4050 points
Part Number: UCC28950

OUTC and OUTD pulse waveforms are abnormal when the customer application circuit operates in the burst mode, as attached waveforms.

Drive pulses in burst.pdf

Yellow is OUTA, blue is OUTB, red is OUTC and green is OUTD.

Customer configured the error amplifier as a gain one negative amplifier with four about 10 kohm resistors without capacitors.

The secondary output voltage is feedbacked through a photocoupler to the about 10 kohm resistor connected to EA-.

He has confirmed that the loop characteristics is stable with reasonable operating margins for his output voltage feedback configuration including the photocoupler for continuous switching conditions for light, normal and full load, not burst.

What kinds of problems are thought?

Can we fix this?

  • 0
    TI__Mastermind 31580 points
    Hi Yoshwa-san,

    I have asked one of our applications engineers to take a look at the waveforms that you provided. You should see an answer later today.

    Regards

    Peter
  • 0 in reply to Peter Meaney
    TI__Mastermind 25570 points
    Hello Yoshwa-san

    This is obviously not the expected behaviour. You can see in the datasheet that we note the first OUTD pulse in a burst may be narrower than usual but the other pulses should be 'full width'. The first OUTC pulse is wider and the last OUTD pulse is narrower than I would expect. This burst also only applies voltage to the transformer in one direction when the OUTC and OUTB pulses 'overlap' - normally a burst always places at least one positive pulse (OUTA and OUTD on at the same time) and one negative pulse (OUTB and OUTC on at the same time) so this is also not correct.

    There are a few things that might cause this
    Is the controller being synchronised to an external source ? if so it might be possible that the controller is sometimes losing synchronisation - this could possible cause erratic switching like you are seeing.
    Noise being coupled onto the RT pin might 'upset' the internal oscillator. If the controller is NOT being synchronised and is configured as a Master (Timing resistor connected between RT and VREF) then the internal clock is brought out on the SYNC pin - it would be useful to look at that pin during normal and burst operation.
    Does this happen with a different UCC28950 device in place
    The PCB layout may be coupling noise into the controller that is disrupting its operation - I can review the layout and schematic if you wish - for confidentiality you can send the gerber files and schematic to me at colingillmor@ti.com

    please let me know how you get on
    Regards
    Colin
  • 0 in reply to Colin Gillmor
    TI__Mastermind 25570 points
    Hello Wada-san

    Thanks for the schematic - it will help me a lot if the customer can send me the Gerber files - of course to me directly at colingillmor@ti.com

    Regards
    Colin
  • 0 in reply to Colin Gillmor
    TI__Mastermind 25570 points
    Hello Wada-san

    Thanks for the added information. Unfortunately I cannot make any real sense out of the layout so I can't really help there. I'd really need to see Gerber files to be able to help because the different layers are in different colours and can be turned on and off so that I can see what is connected where.

    Thanks for the marked up schematic, I see the 10k resistors at the ADEL and ADELEF potential dividers - this should be ok.
    Does the system work correctly when it is loaded above the burst mode threshold and is running continuously?

    What is the input voltage ?
    Can the customer probe the CS pin during this event and also during normal running ?
    Can the customer send a screen shot of the voltage across the main transformer - differential probe across the primary
    I'd be interested in seeing these 'scope plots

    Regards
    Colin
  • 0 in reply to Colin Gillmor
    Prodigy 10 points

    Hi,

    I've similar pulse widths issue than described above. There was lot of noise at output of current transformer. I've disconnected current transformer from CS pin and pulse widths came back to normal.