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.

TPS25924: TPS25924 malfunction

Expert 2385 points
Part Number: TPS25924

Hi Ti,

The part fusing up after power up for around 10 seconds and found that output voltages is shorted to ground. The max input current was at 3.2A only which it’s still within the limits. After latched the part have to be replaced for it to function again. After replacing the part the board works well again.

Please advise what is the possibilities correction needed on the schematic below. Thanks.

  • Hi Incognito,

    Since you are driving >470uF at the output, I recommend to use 4.7nF CdVdT capacitor at DVDT pin to limit inrush current.
    Do you have any captured waveforms of VIN, VOUT, IIN to share?

    I also suggest to use TVS at input to protect eFuse during POWER-JACK Hot-plug event and Schottky diode at output to absorb negative voltage spikes. please refer Section 10 and figure 47 in the datasheet.

    Best Regards,
    Rakesh
  • Hi Rakesh,

    In the data sheet it says if we use 4.7nF If CdVdT = 4.7 nF was used, the device must have tried to charge the 4700-μF output cap with inrush current of 986 mA in 57.24 ms, dissipating power of 5.94 W. This is outside the safe starting condition of the device, and must have led the device to enter thermal shutdown during start-up.

    Can I choose 22nF instead?

    Even if I am using this and blocking diode, the problem is the output pins are shorted to ground when the part is not working and it takes off fuse as well and it happen when the board is running with stable voltage. It happened on 4 boards same way. There are others that are running for more than 2 days and few I'm putting it loop to power up and down to see the behavior.

    Another concern to check that if I am using a chassis ground and the part ground is digital ground, is that ok? and should I place the IC ground completely to chassis ground, but I don't want to do that. I am not sure how stable would that be.
  • Hi Incognito,

    Large output capacitance requires higher CdVdT value. For 470uF+3*22uF, 4.7nF CdVdT value would be good enough. Please use the design calculator tool available at www.ti.com/.../toolssoftware for CdVdT calculations.

    Can you help with below information to debug the issue.
    What kind of load the device is driving at V_12_VIN? Since it is repeating in the similar way on multiple boards - are you able to capture some test waveforms? If yes, please share.
    At the instant of failure - Is there any sudden load or device power up & down happened?
    Have you seen the failure even when all the ferrite beads are shorted?

    Best Regards,
    Rakesh
  • Thanks Rakesh,

    Here are the graphs for input and output of of the IC. Seems like there is  surge on the input for 100ns and can go up to 40V, This didn't take off the IC, but there might be surge that is more than 40 and longer period. So I think the capacitor and TVS will take care of that. On the output i have more than 470uF, so its pretty clean.

    How much surge it can sustain this IC?

    Yellow is INPUT and Green is OUTPUT

  • Hi Incognito,

    Please use 1uF capacitor and a TVS diode (part no: SMBJ15A) at the input of eFuse device.

    Does the board need to pass surge test? what voltage level?
    TPS25924 device can withstand absolute maximum voltage of 20V. Enough transient protection should be provided at the input to limit the voltage seen by the eFuse to under 20V for surge protection.

    If it is just POWER-JACK Hot-plug event, capacitor and TVS will take care


    Best Regards,
    Rakesh