• State TI Thinks Resolved
  • Locked Locked
  • Replies 5 replies
  • Answers 2 answers
  • Subscribers 63 subscribers
  • Views 532 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.

LM5010 - No SW node waveform

Hithesh
Genius 3290 points
Other Parts Discussed in Thread: LM5010

I am using LM5010 as step down converter. Vin is 54V, Vout is 6.6V

Max current draw is 350mA

When the load is enabled, sometimes the LM5010 does not turn on. Around 1 out of 10 times. 

When it doesn't turn on, there is no waveform at the SW node. 

Output is at 4.5V, SW node is also at 4.5V. 

Vin is 54V, Enable is ON(connected to Vin via 180K resistor). FB node voltage is 2.25V (when the output is good, it is 2.5V)

It's not thermal shutdown. The chip temp was hardly above room temp. 

Would it reach this stage, if the part hit current limit somehow?

Strange thing is, when I short the output, it starts working again.

  • 0
    Genius 3290 points
    Only way I can get it work again (other than shorting) is power cycling.
  • 0 in reply to Hithesh
    TI__Intellectual 2320 points

    Hello Hithesh,

    Sorry for the late reply.
    Do you see the failure only when the load is turned off? I believe there can be two reasons:

    1. The part may be going below BST-SW UVLO especially at light load (due to the lack of enough pre-load current). You can start by reducing the feedback resistors such that the pre-load current is at least as high as 1mA. That is why you may seeing the lacking of SW node pulses. Please section 8.3 in the LM5010 DS for more details.

    2. The ripple injection circuit may just not be enough for this application. The ripple injected at the FB may be less than 25mV for normal operation.

    I would suggest you check out the feedback resistors first and try out what I have suggested in the first Bulletin. Also, it would be great if you can share the schematic with us.

    Regards,

    Sourav

  • 0 in reply to Sourav.Sen
    Genius 3290 points
    The feedback resistors are 2K total. This should be enough to keep enough current flowing.

    The no load ripple at output is 200mVp-p.

    What else can I check when the part is in failed state.
  • 0 in reply to Hithesh
    TI__Intellectual 2320 points
    Hello Hithesh,
    Can you please send me a snapshot of the failure (like the SW NODE, VOUT (DC) etc) when you get a chance. Also, if you could share the schematic it would really help.
    Are you using a ripple injection circuit or an external ESR for your ripple injection in your circuit. If you have parallel output capacitors, I would suggest you use a ripple injection circuit as the ESR resistor may not be enough, in that case, to add enough ripple at the FB pin and there can be resultant regulation issue there as well?

    When you mention 'the load is enabled', are you actually toggling the load during this operation? Would it possible for you to probe the inductor current in your application during this operation. That could be another place to check so that we can confirm that no current limit is being triggered, especially when you toggle the load from no load to full load.

    Regards,
    Sourav
  • 0 in reply to Sourav.Sen
    Genius 3290 points

    The issue was under voltage lockout because of voltage from the load. The load has internal LDO, which generates a voltage of 4.4V.

    The voltage difference between SW pin and BST is <4V. The regulator hits UVLO.  

    Question: is there a path from the FB pin to the output pin ?