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.

TPSM84205: Catastrophic failure, fails short to Vin

Part Number: TPSM84205

Tool/software:

Testing a board assembly with this regulator, I needed to be able to turn the 24VDC input power on briefly while probing, to avoid having to align a scope probe on very tiny contacts with live power. Align probe, watch scope, press foot switch to connect the 24VDC supply. Everything's working great for a few minutes....and then on one power-up, there's smoke everywhere. Time interval between cycles was on the order of 30 seconds; long enough to locate the next node to test, get the scope probe tip lined up right, verify no shorts, then look at the scope screen. I was checking low voltage 3.3V connections on the processor, where no 24V is present.

The TPSM84205EAB regulator module failed short-to-Vin, dumped 24V onto the 3.3V linear (5.5Vmax in) which failed short-to-Vin, and thereby dumped 24V onto both the 5V and 3.3V rails. And every IC on the board went up in smoke. :-( 

Is this a behavior that might reasonably be expected? I don't see anything in the datasheet that would suggest so. There's much talk of soft-start, over-current, over-temp and other protections, but nothing warning of random failures on startup that would blow up an entire board. 

I'm adding TVS diodes all over the place to try to intercept a fault and pop a polyfuse before every IC on the board dies. I'm also looking at other regulator alternatives, which I'll be power cycling to verify they don't have a similar failure mode. And I'll be testing some other examples of this same part to see if I can replicate the problem in a less-lethal environment. Is repeated power cycling something that could induce a failure?

  • Hi Mark,

    Add some 150uF or 220uF electrolytic capacitor at the VIN pin of U1.

    If length of the wires which connects from power supply to U1 VIN is more then we will see higher voltage at VIN during High Side FET turn OFF.

    I we see input is appearing at output of a buck means mostly due to higher VIN more than 30V AMS max HS FET is seen and it might have got damaged.

    My suggestion is add some 150uF or 220uF electrolytic capacitor at the VIN pin of U1 and test it.

    Reagrds,

    Haseen.

  • Interesting suggestion. I don't have room to put a big electrolytic on the board, at least not on this pass, but I can look at raising the input capacitance within the space I have available. I think I can get a transzorb in there too, to protect the input from overvoltage spikes. As you say, longer wires might be an issue, and that could be a consequence of my bench test setup rather than the final configuration as installed in the instrument we're building.

  • Hi Mark,

    The idea is due to this series inductance of wires which are connecting from power supply to buck which will have stored the energy and whenever HS FET opens the energy stored in the series inductance will be absorbed by input 220uF capacitor.

    Regards,

    Haseen.