• State
  • Locked Locked
  • Replies 5 replies
  • Subscribers 63 subscribers
  • Views 210 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.

LM3478: LM3478MMX-NOPB+TPS23861PWR

Jiang Leven
Prodigy 60 points
Part Number: LM3478
Other Parts Discussed in Thread: TPS23861

Tool/software:

We use LM3478MMX-NOPB to boost DC12V to DC58V, supply power to TPS23861PWR, and realize POE PSE function. Now the circuit cannot work normally.
1, during the boot process, DC12V will be severely pulled down, resulting in a restart of the machine;
2, Using external power supply to VDD58V_PSE to TPS23861PWR power supply, POE PD end can not work normally.
May I ask what causes it? Is there anything wrong with the schemtic design?

POE_PSE-SCH.pdf

  • 0
    TI__Mastermind 30220 points

    Hi Jiang,

    Thanks for using the e2e forum.

    Unfortunately, I am only familiar with the LM3478 device questions and cannot comment on the TPS23861 device. (This would be another product line, so a new e2e thread would need to be created)

    The first step would be to find out if the problem comes from the LM3478 or the TPS23861 device.
    You mentioned in 2.), that you tried to power up the TPS23861 device with a external 58V supply and it did not work normally. This would indicate that something is wrong with this device circuit.

    Do you know the load requirements of your applications?
    If I know the output power, I can review the LM3478 section of the schematic and check for problems here.
    The structure itself looks good to me, but if the 12V input is dropping, it might be caused by too high output current requirements at the 58V rail.

    Best regards,
    Niklas

  • 0 in reply to Niklas Schwarz
    Prodigy 60 points

    Hi Niklas：

    We disconnected LM3478 and TPS23861, and separately tested the LM3478 booster circuit and TPS23861 POE circuit. The test results are as follows
    1. LM3478 test NG;
    2. TPS23861 POE circuit test OK: The input power of TPS23861 adopts the regulated DC power output 58V test, and the function and index test meet the expectations;
    According to the above tests, the LM3478 DCDC Boost output is abnormal.

    Best regards,

    Jiang

  • 0 in reply to Jiang Leven
    TI__Mastermind 30220 points

    Hi Jiang,

    Thanks for providing these test results.
    You mentioned TPS23861 input power adopts to regulated DC power output. Can you give me the expected peak power requirements? This will help me in reviewing the LM3478 circuit.

    Thanks and best regards,
    Niklas

  • 0 in reply to Niklas Schwarz
    Prodigy 60 points

     Hi Niklas：

    Thank you for replying so quickly.

    Can you give me the expected peak power requirements?

    The IEEE 802.3at average output power is a maximum of 30W,

    that is, LM3478 Input and output parameters are as follows
    1.Input voltage 12V
    2.The output voltage is 58V and the average current is 517.2mA

    Best regards,

    Jiang

  • 0 in reply to Jiang Leven
    TI__Mastermind 30220 points

    Hi Jiang,

    Thanks for the info.
    This might be related to a stability issue.
    I checked the compensation and it shows rather low phase margin.
    I would recommend to reduce C806 down to ~200pF to increase stability of the loop compensation.

    A more important point is the slope compensation.
    As this design switches with more than 50% duty cycle, subharmonic oscillation is possible.
    I would recommend to place a slope compensation at the ISEN pin to avoid oscillations.
    There is no footprint for a resistor in your schematic, but for testing it might be possible to use the filter resistor R910 for this.
    A good starting value for the slope compensation would be 1kOhm.

    Is it possible for you to perform waveform measurement of the switch node voltage? Here you would switch if the switching is stable or oscillates.

    Best regards,
    Niklas