• State Resolved
  • Locked Locked
  • Replies 5 replies
  • Subscribers 63 subscribers
  • Views 223 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.

Original question:

LM27402: LM27402 is damaged after few hours of working

LM27402: Switching frequency not equal to what is set through Rfadj

Jijesh Janardhanan
Prodigy 120 points
Part Number: LM27402

Hi,

Continuation of thread LM27402 is damaged after few hours if working, We have noticed switching frequency of LM27402 is not equal to what we set through Rfadj. As per the design we have set switching frequency to 700KHz for a load of 14A max current. But on this board, max load current on board is around 2A only.

When the PHY devices are in reset (load current is around 800mA), LM27402 switching frequency varying between 200KHz to 700KHz.

When the PHY devices are fully configured (load current is around 2000mA), LM27402 switching frequency is stabilized to 230KHz

 

Is this the expected behavior when load current very low (less than 15% of max current designed for)?

Will there be any stability issue because of this?

Thanks,

Jijesh 

  • 0
    TI__Mastermind 23775 points

    The 3rd waveform you are showing of the SW node, shows a peak voltage 24.9V which is greater than the SW node abs max voltage 22V.     The first image shows 31V on switch node.  The ringing of the SW node implies the CBOOT voltage is exceeded which can damage the device. 

    A RC snubber on the SW node or slowing the turn on of the high side fet will reduce the sw ringing.   Link to Snubber Post

    The first two waveforms look like instability, noise or noise on current limit.    The schematic only has 100uF  + 2x22uF ceramic and the simulation has greater capacitance on the output.

    What capacitors are on your output.     the ceramic will need to be derated for the dc bias voltage.   A 100uF/4V will not be 100uF of capacitance with 1V dc voltage on it. 

    And the derated cout will be needed to determine the compensation components.  

  • 0 in reply to David_Daniels
    Prodigy 120 points

    Hi David,

    Forget to mention, all the waveform posted are probed on CBOOT pin with respect to GND pin. Output capacitor 100uF + 2x22uF are placed close to LM27402. And near to the load (PHY devices) totally around  40x10uF are placed.

    We planning to load RC snubber network or Rboot to reduce peak voltage below 24V at CBOOT pin.

    Why the switching frequency is reduced from the 700KHz to 230KHz?

    Do we need to change compensation network for low load current?

    Thanks,

    Jijesh

  • +1 in reply to Jijesh Janardhanan
    TI__Mastermind 23775 points

    If the max output current is 2A,  I believed you should change the inductor value and adjust the compensation. 

    The control scheme on the LM27402 is voltage mode control and should not need a change in compensation for lighter loads. 

    I suspect the compensation components could be adjusted to improve phase margin, change the 12.4k (R2344)resistor to 10k. 

  • 0 in reply to David_Daniels
    Prodigy 120 points

    Hi David,

    Thanks for your suggestions.

    Ringing voltage on CBOOT is now reduced less than 24V when we add RBoot of 4.7Ohms

    Switching frequency is stabilized once the compensation network resistor R2344 changed with 10K

    For lighter load around 2A, switching frequency comes down to 230KHz

    For a load current of 10A, switching frequency is around 680KHz

    Thanks,

    Jijesh

  • 0 in reply to Jijesh Janardhanan
    TI__Mastermind 23775 points

    The switching frequency should not change at 2A output current.   Can you change the inductor value from 330nH to 680nH and see if the switching frequency changes at 1A output current