• State TI Thinks Resolved
  • Locked Locked
  • Replies 5 replies
  • Answers 1 answer
  • Subscribers 65 subscribers
  • Views 1249 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.

LMZ14203: Lost output regulation once, and I can't make it happen again

Damien Ramrattan
Prodigy 50 points
Part Number: LMZ14203

Hello, 

I have a prototype board based on the layout and bill of materials of the eval board.  The output is set to 13.2v and the input is from a full wave rectifier coming from a transformer. I was doing load step testing and I was stepping from very light loads, less than 30mA or so to full load, about 2amps.  I had been testing for many hours and everything was behaving as expected, then for some reason, when I stepped from light load to full load, the output dropped to about 3v and the current dropped to about 400mA.  When I took the 2amp load off, the voltage went back to 13.2v, but when I put the full load on again it went back to about 3v with about 400mA.  I did this several times taking the load on and off but the condition would not clear.  I then removed power and re applied power and the condition cleared itself.  

I have since been testing for many hours trying to reproduce this problem and I can't get it to happen again.  I've looked at all the signals on the pins and compared them to the eval board and they all seem to look the same except for a slight difference on the Ron pin.  I'm not sure if this is of concern or not.  I am using a 301k 1% Ron resistor.  Any help would be appreciated.

Thanks.

the Ron pin.

  • 0
    TI__Intellectual 1535 points
    Hello Damien,

    first I can recommend for your application the LMZ14203H for your desired output voltage of 13.2V. The LMZ14203 is only recommended for output voltages up to 5V.

    Second, some more information could help to solve the issue:

    - Please provide your schematic and layout of your application.
    - Can you also take some scope shots of the input voltage while loading and the error occurs?
    - Try to catch the switch node with the scope. To do that approach the top of the package with the probe tip. The tip should catch up the switch node without having an electrical connection. But it can help to get an idea of the switching behavior of the device.

    Please do not hesitate to contact us again.

    Best regards,
    Michael
  • 0 in reply to Michael Daimer
    Prodigy 50 points

    Hello Michael,

    that is my mistake, I am using the LMZ14203H.  Here is the power supply part of the schematic, layout and some shots of the switch node under low and high load. I am still unable to create the error unfortunately so I cannot take a scope shot of the input under the error.  I did take a shot of the input voltage under full load.  The ripple there is about 2v p-p.

    This is a 2 layer board with 2 ounce copper, .062" thick with a ground plane on the bottom of the board.

    This is the high load switch node envelope.

    Here is the high load switch node switching at 374KHz

    This is the switch node at light load this waveform repeats at about a 17HZ interval

    Please let me know if you need any further information.

    Thanks,

    Damien

  • 0 in reply to Damien Ramrattan
    TI__Expert 4365 points

    Because this is a 12V output and the FB input regulates at 0.8V there is 11.2V across CFF (C71 in your design).

    In the case where you managed to get it to latch, let's suppose you created an overcurrent (or maybe a shor)t on Vout.

    A rapid fall on Vout can then allow CFF to push the FB input lower than it's absmax of -0.3V. This probably caused a nondestructive latch condition that eventually cleared once input power was cycled.

    See if tapping a short across the output with a test lead can recreate the fault mode you experienced.

    For your application, the addition of a reverse biased Schottky (BAT54 in SOD123 or SOD323) on the feedback input to GND may prevent the latch up from occurring in the future

  • 0 in reply to Alan Martin
    Prodigy 50 points
    Good morning Alan,

    I tried shorting the output and although it does cause the CFF cap to push the FB pin lower than -.3 (it goes down to about -.7v) it still does not cause the latch up. I have extensively tried this and the output recovers flawlessly every time. I am going to continue load testing today and try to get this problem to occur again. By all means the part is working flawlessly and has been through all my testing over the last couple of days except for that one event that I can't seem to trigger again.

    Thanks,
    Damien.
  • 0 in reply to Damien Ramrattan
    TI__Mastermind 19891 points
    Hello Damien,
    Thank you for the update. Let us know if you see an issue again.
    Cheers,
    Denislav