• State Resolved
  • Locked Locked
  • Replies 6 replies
  • Answers 1 answer
  • Subscribers 64 subscribers
  • Views 619 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.

UCC28780: Issues regarding Only Three PWML Pulses : first pulse is too short and maybe brown-in protection is triggered

Julien MENAGE
Prodigy 200 points
Part Number: UCC28780

Hello,

I am following the application report slua982 regarding debugging ACF converter start-up issues. I have the following waveform at 50% of brown-in voltage.

  • CH1 : voltage of the secondary winding (Vab in §6.2 of the application report)
  • CH2 : voltage of auxiliary winding (Vcd in §6.2 of the application report)
  • CH3 : not connected
  • CH4 : voltage at the switching node

I have identified 2 issues on that waveform that I can not solved:

  1. I have only 3 pulses instead of 4 and I don't get a 1.5s restart. It seems more like a reset every 60ms (which make me think of a brown-out detection of table 5 but since Vin ~ 0.7 * Vbi I don't think so...). It is not a temperature issue (IR picture is fine). HVG is below 13.8V. So it remains the brown-in protection but I don't know where to begin to solve that issue. Note: When I look with a scope probe on the VS pin, I can get the 4 pulses.
  2. The first pulse is too short (200ns-300ns): I have tried several Ropp & Ccs values for 25ns to 100ns current-sense filtering (calculations based on excel calculator SLUC664C). I have also increased the bootstrap resistor, thinking bootstrap current was triggering the 0,28V current threshold, but nothing changed. Is it relevant to put a "big" Ccs (100 pF to 220 pF) with a Ropp of 100 ohms ?

Thank you for your help.

Best regards,

Julien

  • 0
    TI__Genius 9790 points

    Thanks for your interest in TI here. I've contacted the appropriate product group. You should hear from them soon.

  • 0 in reply to Yunsheng Qu
    TI__Expert 7080 points

    Hi Julien,

    3 pulses looks like still ok for me , since you told me that the hiccup time is ~60ms rather than 1.5s , that means the OCP protection happened . the 60ms hiccup time is VDD retry.  And the input voltage still under brown in voltage . the first pulse is CS pin fault detection, even it's narrow . ignore it , you still can see the 2nd /3rd pulses, that means CS pin fault has passed . why there is only 3 pulse rather than 4 pulse . I think maybe on your PCB board , VS pin and its divider resistors layout is not good . make sure the divider resistors should be placed as close as possible to the controller , and  without ground plane under VS pin trace and two resistors .  temporary solutions :

    1. Place a 1~2pF capacitor on upper resistor of VS pin divider ( the resistor which connected from AUX winding to VS pin). this is a temporary solution. if it make sense , please optimize PCB layout finally.

    2. Flying two resistor , Then using wire connect to controller , make sure  VS pin wire is short .

    Sometime, only 3 pulse is still ok , try to increase the input voltage close to Brown in point , to see if you can get 4pulses.

    Thanks.

  • 0 in reply to Jaden Ning
    Prodigy 200 points

    Hi Jaden,

    Thanks for your answer. Indeed I don't have the 1.5s hiccup more like something below 100ms. see image below Vdd1 is in pink (CH2). I have added the current in the primary of the transformer on channel 3, just in case. CH4 is still the voltage at switching node. CH1 is not connected

    Here are some results :

    • Without modifications I tryed to increase the input voltage close to Brown in point => no effect
    • Place a 1~2pF capacitor on upper resistor of VS pin divider ( the resistor, RVS1, which connected from AUX winding to VS pin) => no effect

    But doing both gives me 5 pulses (see below) !

    I will try the flying resistor. I have followed the recommendation of the datasheet regarding layout : no ground plane below VS net and short loop between Rvs2 and UCC28780. So I will make a flying wire to connect AUX to Rvs1 (and disconnect partially Rvs1).

    If you have comments regarding the current waveform, I will be happy to read them. Note : I have chosen a "big" Cclamp (on primary side) and a secondary resonance with Co1.

    Thanks again

  • 0 in reply to Julien MENAGE
    Prodigy 200 points

    update : the flying connection of resistor did not work, we have the same behavior

  • 0 in reply to Julien MENAGE
    TI__Expert 7080 points

    Hi Julien,

    5 PWML pulses means it already close to Brown in voltage , Increase the input voltage you will see controller soft start if there is no other protection.

    Thanks.

  • 0 in reply to Jaden Ning
    Prodigy 200 points

    Hi,

    Thank you for your help. I have other trouble so I will make another post because it is another topic.

    Best regards,