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.

LM5176: Unstable behavior & dying chips

Part Number: LM5176

Hello,

I had previously opened a ticket regarding the LM5176, where we had issues with no output voltage, chips heating up, etc.

We redid some of the design, and produced new PCB's. We assembled them, with the same behavior on the boards.

We then proceded to assemble a new PCB, but with all the components from your Evaluation Board LM5176EVM-HP.

Right after assembling this board, during testing, the LM5176 did not heat up, the output voltage was pretty much ok (we were expecting 12V, the voltage was around 12.5V which is ok), and we increased the Load Value up to 1.5A without any issue whatsoever. We had the load at the 1.5A mark for around 5 to 10 minutes, when suddenly, the output voltage dropped to ~9V, and no power could be delivered. Without load, the voltage would stay at 9V, but even with a very small load of 100mA, the output voltage dropped to 0V.

After this, the IC started heating up again, this time less than before, but there is a fixed consumption of about 50m.

In the last response on my last ticket https://e2e.ti.com/support/power-management-group/power-management/f/power-management-forum/1236935/lm5176-output-voltage-too-low-ic-heating-up-fast/4782936, by , the most visible issue was with the VCC line, which is now ok, at around 7V.

My main issues right now are 2 things:

1. Why does this PCB work fine (at least for 10 minutes) with the component values from your Evaluation Boards, but does not work with the component values from your Webench Power Designer? Can the designer provide wrong values?

2. Why does the chip suddently die after 10 minutes, with a fixed & stable power supply, and also a fixed and stable DC electronic load. There were no voltage spikes, no current spikes, and still the chip is not working!

Best regards,

Carlos Mesquita

  • Hi Carlos,

    when operating with the components of the EVM, did you observe any heating up or thermal overstress on any other component (while the load was increased).

    May we first try to identify the root cause with the EVM components.

    Can you check the switch nodes (SW1 and SW2) if there are bigger over or undershoots due to ringing based on parasitic.

    Best regards,

     Stefan

  • Hello Stefan,

    With the load at 1.5A, ~20W, the inductor was heating up a little, and the FET's were also heating up; still, I would say those were "normal" heating conditions, as expected. I wouldn't classify it as "overstress"... but right now i can't really tell you any temperature values since the PCB is non-functional.

    Regarding the SW1 and SW2, same issue, I don't have any functional board right now. I am also assuming all our chips are fried, and that's the root cause of all malfunctions.

    I don't want to be buying and frying chips on a daily basis, it's just unbearable... how can we prevent this from happening?

    Best regards,

    Carlos

  • Hi Carlos,

    assuming that the issue is not due to Thermal overstress which is basically current, the other option is voltage over stress.

    This can happen if there are undershoot or overshoot on signals, most often the switch nodes.
    This happens due to parasitic effects.

    Another issue could be that the power stage is not stable but this you would typically see with a unstable/swinging output voltage.

    To further debug this - measurements would be required.

    Note: also some components could get damaged due to overvoltage but with 9V input and 12V output it think this is not the root cause here)

    Best regards,

     Stefan