• State
  • Locked Locked
  • Replies 5 replies
  • Subscribers 63 subscribers
  • Views 736 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.

LM27403 shuts down or won't start

Simon Brown
Intellectual 340 points
Other Parts Discussed in Thread: LM27403

I've got a power supply design, designed almost entirely using Webbench.  The DC input voltage is between 15 and 20V, and I need to generate a variety of voltages for my circuit.  Webbench proposed stepping the incoming voltage down to 6V using an LM27403 and then stepping that down further to generate the required voltages for my board.  This all works, except for a few issues, and it's slightly board dependent, which is worrying me.

1) We need at least 18V to make the system start reliably.

2) In some cases, the 6V works to start with, then shuts down after 5-30s.  In a few cases the board always starts up reliably, even with the same firmware loaded.

3) I've now changed the main inductor to 6.8uH (from 4.7) the main reason being that the 6.8uH inductor has 23mOhm DCR, whereas the 4.7 only had 13mOhm.  Since I've read a few other threads on here suggesting that you need at least 20mOhm on the inductor I've changed to that (though webbench disagrees with this).  Now the 6V doesn't start at all - it produces 1.8V and nothing more.

4) I've tried using the LM27403 Regulator Design Tool, and can make sense of most of that, but I don't understand the compensation design.  What sort of values do I want for Pole and Zero frequencies?  What do I want to get out of the Bode Plot?

I've attached my Excel spreadsheet with various values plugged in to make things match.  I need some way of getting the 6V to start up and then STAY running.LM27403 PWM Regulator Design Tool.xlsx

Many thanks,

Simon

  • 0
    TI__Expert 7935 points
    Hi Simon,

    Could you please share the schematic and layout of your design?

    When I download your Excel file, all graphics are stripped out and Excel informs me that the file was repaired. So, I'm not 100% sure if all the info in the Excel file is what you entered.

    From what I could see, the only item that stood out is the output capacitance. I see 2uF as the selected value. Knowing that the ceramic capacitors de-rate significantly with the bias voltage, I would suggest using a 10uF capacitor instead.

    Regards,
    Davor
  • 0 in reply to Davor Glisic
    Intellectual 340 points

    Hi Davor,

    Many thanks for your reply.  I've attached a schematic and layout file to show you what is where.  Sorry if it's not very clear.

    As for the output capacitor, it was reading some other threads on this forum that warned me about voltage derating on ceramic capacitors, so I added a 10uF capacitor in parallel to the 2*1uF capacitors I already had - and that stopped it starting at all.

    CAM output.pdfpsu_6v.pdf

  • 0 in reply to Simon Brown
    TI__Expert 7935 points
    Hi Simon,

    -How are you driving the EN pin? The schematic does not show a connection to a source of the EN signal.
    -Suggest tying the SYNC pin to GND
    -Suggest setting C389 to 1000pF, C388 to 100pF, R145 to 5.36kohms, and R144 to 11ohms
    -Suggest setting R139 to 1.47kohms

    I assumed 6.8uH, 23mohm inductor.

    Regards,
    Davor
  • 0 in reply to Davor Glisic
    Intellectual 340 points

    Hi Davor,

    Ok, following your suggested changes (thank you), the 6V once again starts up, but again turns off as some other components come online.  It isn't directly related to current (I can't see the current go up significantly on the bench supply providing the 18V in), but might have some bearing.  I will admit that due to limited resistor values available to me, the 11ohms for R144 is actually 10, and the 5k36 for R145 is actually 5k0.  I'm assuming that these are close enough.

    Shorting SYNC to GND didn't help either.  I've not got EN connected to anything - Webbench doesn't suggest connecting these to anything.  I'll try connecting it to a bench supply of 3V and see if that helps.

    Yes, the inductor is 6.8uH, 23mohm.

    Regards,

    Simon

  • 0 in reply to Simon Brown
    TI__Expert 7935 points
    Hi Simon,

    While there is a weak pull-up current source on the EN pin, contamination on the board may create a low enough impedance from the pin to GND to bring the EN down. Have you tried biasing the EN pin with an external supply? Cleaning the board with isopropyl alcohol may do the trick as well.

    Regards,
    Davor