• State TI Thinks Resolved
  • Locked Locked
  • Replies 7 replies
  • Answers 2 answers
  • Subscribers 64 subscribers
  • Views 389 views
  • Users 0 members are here
Support feedback
Options
Options
Related

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.

TPS54308: Keeps burning even with no load

Jayjay_Nexo
Prodigy 90 points
Part Number: TPS54308

Hello dear TI-Team,

we had some PCBs assembled with a TPS54308 and 22uH Inductor as power supply (24V). It keeps bursting into flames when powered. As I go through the datasheet there seems to be a lot of protection circuitry in the chip, such that it shouldn't destroy like this. The voltage comes from a regulated lab-supply. 

What could posslibly cause this chip to get destroyed all the time (some boards work just fine, some start smoking imediately, some work fine for some time but get destroyed after 2nd or 3rd power up.)

Any ideas how I could debug this? Thanks

  • 0
    Prodigy 90 points

    Edit: (didn't know how do edit my original post)

    When the device is powered from the "on"-Button of the supply it works fine. As soon as I just put some wires on the board and connect/disconnect them it gets destroyed. Something with inrushcurrent maybe?

  • 0
    Prodigy 90 points

    Edit2: When I look at the voltages at the input pin and at the output (SW) pin the voltage spikes to about 50V at each pin. This is strange because at least the input should be connected to low-resistance supply. When switching on and off with the power supply button, I measured that the voltage is ramped up such that there are no spikes. These spikes probably instantly kill the chip, but why do they even occur? I have 10uF at the input, ceramic, as recommended by the datasheet.

    Does no one else have this problem??

  • 0 in reply to Jayjay_Nexo
    TI__Genius 9550 points

    hi,

    you mentioned you caught the input voltage up to 50V, so it is the directly reason to damage the device.

    Please upload the schematic/layout and check for you. (including the input voltage circuit)

  • 0 in reply to Jason Wang (BSR-MV)
    Prodigy 90 points

    Hello Jason,

    I have no access to my files at the moment, but the design is straight forward. The schematic only has a 10uF Capacitor at the input, it direcly goes to the supply wires. I found out that there is some kind of overshoot/ringing that destroys the chip.

    However, a 47uF cap solved the problem (no ringing, just a damped, small overshoot).

    I never before experienced an input of a Buck converter causing such a heavy ringing, there must be some kind of parasitic inductance in the chip that makes for a RLC-Circuit with the input cap.

    Do you have any idea what might be the origin of this? Have you ever had a phenomenon like this before? Anyway, the larger cap solved the problem.

    Thanks,

    Jayjay

  • 0 in reply to Jayjay_Nexo
    TI__Genius 10070 points

    Hi jayjay,

    In my experience, 10uF is enough. If the layout of your board is not good, for example, the input capacitor is far away from VIN pin and GND pin of the IC or the trace are too thin, the spike of the SW would be very large, this may result in the damage.

    You can try to add a 0.1uF ceramic capacitor close to VIN pin and GND pin, and check whether this issue can be solved.

    Shawn

  • 0 in reply to Shawn Nie
    Prodigy 90 points

    Ok, on the weekend I will perform some tests. We have 100nF and 10uF close to the Chip (1.5cm maybe).

    You said that spikes occur at the SW Pin, but I was talking about the spikes at the input pin. This might also have to do with the power supply (wires are about 1m long). When I add a capacitor of 47uF at the Supply itself (far away from the device itself) the same problem, spikes at the input. When I put the same 47uF close to the chip (solder it to the PCB) everything is fine. 

    I will perform some further testing, and let you know on monday.

    Thanks for your replies,

    Jo

  • 0 in reply to Jayjay_Nexo
    TI__Genius 10070 points

    Hi,

    The wire is too long, the ringing voltage would be very large, that is to say the input voltage excess the max input voltage which result in the damage.

    Shawn