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.

LM5143-Q1EVM-2100: What fault conditions do the slaves cover and monitor via power good?

Part Number: LM5143-Q1EVM-2100

Hi Team,

my customer would like to know what fault conditions are covered by the slaves and monitored via power good in the dual-channel automotive synchronous buck controller evaluation module LM5143-Q1EVM-2100.

Thank you,

Franz

  • Hi Franz,

    Each channel has its own power good (PG) indicator that monitors output UV, OV and OC events. It also pulls low during an IC overtemperature event.

    Regards,
    Tim

  • Hi Tim,

    I'm the customer, thanks Franz  for posting it.

    Some further details: I have a multiphase 4ch single output interleaved design, realized as current source to drive laser diodes.  Function is prooved, everything is fine, now I have to look at reliability and safety.

    If one Controller (1 of 2) or channel (1 of 4) have a failure I have to power down the whole system. Thats why some questions,  may be a yes/no answer….

     1. Is "Power good" PG  controlled only by the regulation voltage?

    2. Is PG "hard wired" to the PWM Outputs (power bad --> Outputs off)?

    3. What happens with the other channels if one channel is switched off?

    4. Should I AND-wire all PG Outputs to shut down all if needed?

    5. UV means UV of bias voltages and output voltage, UV of suppy voltage must be done via EN Input?

    6. OV is OV of Output voltage

    7. OC is clear. 

    8. every channel (slaves & master) monitors these failures, except output voltage?

    9. finally, do you have an idea how to identify the source of the possible failure?

    Thanks in advance

    Frank

  • Hi Frank,
    1. Yes power good is controlled by the regulated output voltage.
    2. Power Good is internally connected (hard wired) to the regulated feedback voltage. PG will change states if the output goes out of regulation.
    3. If you are running 4 phases, you can disable one or more phases. If you disable a phase the controllers will continue to operate, sharing the load current. As long as the load current does not exceed the over current set point, in that case the controllers will go into hiccup mode over load protection.
    4. ANDing all the PG signals is not necessary. All the outputs or connected together, only the Master controller is monitoring the output voltage. You can leave the other PG’s open.
    5/6 In the case of Tim’s answer about, it means the output under voltage (UV) over voltage (OV) OV.
    7. Thanks
    8. When operating in 4-phase mode, only the Master is monitoring the output.
    9. With all the outputs connect together it will be difficult to tell which one failed.
  • Hi Terry,
    thanks..
    One Thing according 3.&4.
    I have to vary the Output current over a wide range (2...50A) and I need a good current sharing also at lower currents. Thats why I need current sense resistors as much as possible (66mV at 50A + ripple/2 + some margin)
    The probability of OC issues is the highest. If I understand you right, OC does not effect the PG Output of the slave al long as the master keeps the whole system in control? What happens inside the overload slave controller: it goes to Hiccup without any Information to outside?
  • Frank,

    The outputs are all connected together, each phase Master and Slave measure there individual phase currents.  If there is an overcurrent each  phase will detect it, and the controllers will go into hiccup model overload protection.

    Terry

  • Thanks Tim,

    I'll check it out...

    Frank