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.

TPS2419: Keeps on frying

Part Number: TPS2419

Hi All,

We've recently integrated a TPS2419D ideal diode controller along with it's N-channel MOSFET into a DC/DC schematic

While it does work a treat at first start (powering the DC/DC via it's 48V HV input), it's the second time I get a weird behaviour (gate = 0V all the time) on some edge cases.

The thing is : 

  • The DC/DC is being shut-down by opening the negative side (via a photoMOS relay)

  • The DC/DC is NOT isolated

  • We need to apply 12V first (from an external source) onto the +12V OUT in order to start and then control the SSPC by putting it's pin 2 (Cathode of the emitting LED) to the ground.

  • The TPS2419D seems to be working correctly at first but then we apply 12V at the input (or we open the 48V GND) and it's then down


What I suspect : when the +48V GND is left floating, we get some strong common-mode that is enough to fry the chip.

Hereafter is a screenshot of the schematic, please bear in mind the Schottky Diode has been added as a safety measure to be able to get the card working (at the expense of a big thermal constraint).

  • A precision also : 

    Even when in "non-operating/fried" mode, the pin 8 (BYP) still has 23V on it meaning the internal bootstrap still works, right? Only it doesn't get to the gate...

  • Hi Marc,

    The TPS2419D device is rated only for 18V on the positive side and -0.3V on the negative side at its input and output. 

    Any transient voltages at the input beyond these voltages can damage the device. You will have to use a TVS or Schottky to clamp the transient voltages across the device to to within -0.3V to 18V. 

  • Hi Praveen,

    Thank you for the answer.
    As you can see on the schematic, I've used a 14V TVS on the input of the TPS2419D so it should be ok on that front.

  • Hi Marc,

    The Max Break down voltage of SMCJ14A is 17.2 V which is close to the Absolute Maximum Rating (18V) of the IC.

    You can consider using lower voltage TVS like SMCJ13A  or even lower - just make sure that the minimum breakdown voltage is greater than the maximum DC voltage and the clamping voltage during transients is less than 18V. 

  • Hi Praveen,

    I've ordered some SMCJ13CA which also happens to be bidirectional, let's hope it makes it better.
    However, I'm not completely convinced it's ok since the DC/DC is outputting a little over 13V which means there might be some leakage current in the TVS...

    Also, it doesn't really cover the negative voltage.

    Last thing : I might put another TVS at the output of the MOSFET / Ideal diode, in case that comes from there.

    In any case, I cannot find a valid reason as to why the chip keeps frying (with no exterior trace of fault by the way).

  • Hi Marc,

    Although the rating on the negative side is only -0.3V, it is only -0.3V DC. Even with a Schottky diode across the input, you will not be able to clamp the input during negative transients to -0.3V. 

    For short transient voltages (in the order of few us), there will be some margin on the negative voltage spec - You can target to clamp the voltage on the negative side to within -0.8V. 

    To be 100% sure, you need to consistently replicate the failure condition of the IC and verify if any of the voltage across the IC has violated the Absolute Maximum rating of the IC. Then you will have to add a TVS (for positive transients) or Schottly diode (for the negative transients) accordingly.