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.

TPS25982: Cannot turn OFF at low temperatures (~ -30C)

Part Number: TPS25982

I am experiencing an issue where the TPS25982 load switch  would not turn OFF at temperatures around -35 C. My schematic is shown below

The board with the TPS25982 contains data acquisition hardware, so I can monitor the voltage on the output pins through vsns_sw. The output of the load switch (out_sw) is not loaded with anything except for the voltage sense circuitry. vsns_sw is measured directly by an ADC (no pullups or down).

Testing Procedure:

1) Set the EN input (cmd_sw) to toggle ON(3.3V) and OFF(0V) every 5 seconds.

2) Put the board in thermal chamber and set the temperature to -40C

3) Bring the board back to room temperature

Here is what I am observing:

1) When approaching temperatures around -30C, the switch after tuning ON fails to turn OFF.

2) This anomalous behavior still occurs when the board is brought to room temperature.

3) Voltage level of EN pin was confirmed with a multi meter to be correct  (switching between 3.3V and 0V every 5 seconds).

4) Voltage of the output pin was also confirmed with multi meter to be 12V.

5) When the device output is loaded in this faulty state, it is able to power the load. (Power is coming from the input voltage)

Is this a known issue with the device and what is the fix for this?

 

 

  • Hi Aratrik,

    At room temp when you give VIN=12V and EN=0(DC no pulse) is VOUT=12V?

    Regards

    Kunal Goel

  • At room temp when I power on the board and give those inputs, VOUT is 0V as expected. But when I bring the board down to -35C while switching it on and off, it fails to turn off (ie VOUT = 12V). Then I bring the board back to room temperature, it still doesn't turn off (it remains in the fault state). If I power cycle the board at room temperature, the switch works as expected.

    Also note, we have about 8 such switches on one board. On testing multiple such boards, some of the switches fail and some don't. But the ones that fail, fails consistently.

    It has nothing to do with load as the switches fail loaded or not.

  • Hi Aratrik,

    Let me check and get back on this by tomorrow.

    Regards

    Kunal Goel

  • Hi Aratrik,

    1. The unit that recovered from power cycle if it again goes through EN and temp cycle it still fails?

    2. One reason for this could be device undergoing in some latchup state which can happen due to Vout going negative. Can you repeat experiment with 1uf cap at output? If possible please add a schottky diode also at outptut. That will help in preventing VOUT going negative.

    3. Also is it possible to probe dvdt and PG pin?

    Regards

    Kunal Goel

  • 1. The unit that recovers from power cycle faults again on temperature cycle.

    2. The VOUT is not negative, because I have been monitoring it through a multimeter.

    3. The PG pin stays high during the fault. Don't know abut Dvdt

    Also with more testing, it seems that the device faults on temperature cycling. If I power cycle at -35C, it doesn't fault, but if I change the temperature, it faults

  • Hi  Aratrik,

    VOUT going negative is a transient event so might not get captured on multimeter. To get more info probing dvdt pin will help. 

    You mean that if you toggle EN pin or toggle VIN pin in both cases you see that VOUT stays 12V? 

    Regards

    Kunal Goel

  • VOUT stays 12V if I toggle EN pin. Toggling VIN (power cycle) clear the fault. But if I toggle EN again it faults again if temp is below -35C.

    Will a negative VOUT transient event at room temperatures cause such a latched up state at low temperatures? Or does it fault just after a negative voltage transient?

    I have been looking at the output voltage waveforms when I use a normally operating switch to power a relay (our specific application). VOUT does drop but just to -420mV. I am guessing this is not enough to fault the device as it operated fine down to -45C.

  • Hi Aratrik,

    -0.3V is max what we recommend. It could latchup at room temp also. It would be good to retry with high cap at output.

    Regards

    Kunal Goel

  • Okay, I can test it with a schottkey diode and output caps but that still leaves the question of why was it repeatedly failing at low temperatures without any load?

  • Hi Aratrik,

    At low temp you see VOUT going negative,right? 

    Regards

    Kunal Goel