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.

TPSM84209: Failure

Part Number: TPSM84209
Other Parts Discussed in Thread: LMZM23601

Following is a supply circuit for an Interface Board. 

Input: 24 Volt DC

Output: 3.3Volt

Minimum Iout : 90 ma

Maximum Iout : 1000 ma

I connected the board to the bench supply set for 13 volts. The current showed 44ma. Output measured 3.3volt.

Connected to 24 volt supply and output measured 3.3 Volt.

The board has the following interfaces:

Typical (ma) Max (ma)
STM32F765ZG 95 258
LAN 50 70
USB 35 40
ESP 32 Wifi 300 500

Programmed the board STM32F7 and powered the board. It worked fine.

Connected USB connector to PC. It was recognized.

Disconnected the power.

Made some changes to code. Reconnected the power and tried to program it and no connection.

Checked the supply voltage. No supply voltage and all power components are burning to touch.

At this stage, only USB and STM32F7 is active.

Connected the bench supply back, and it looks like Vin and Vout of U2 is shorted.

Checked with Ohm meter between Vin and Vout of U2 and it reads zero ohms.

Connected the supply passed Q1 (bypassing Q1) the same.

Connected the supply passed L2 (bypassing filter) the same.

Below is schematic and board layout.

  • Hi Allen,

    It sounds like the failure happened after you made some changes to your code. Vin to Vout short indicates that the part may have been damaged such that the HIGH side MOSFET's body diode is shorted. This may be an EOS issue where the input pin has been over-voltaged pass the absolute maximum voltage rating of 30V. 

    What in the code did you change?

    Perhaps there was a hotplug event that could have spiked the input voltage above the abs max. In this case, I would recommend having a high ESR electrolytic bulk capacitor to keep the input voltage rail 24V during your programming. 

    Regards,

    Jimmy  

  • Hi Jimmy,

    I plugged the power jack. I tried to flash the new program, couldn't flash the new program. 

    I have ordered a thermal imaging camera. I will try to capture and post the image next week.

    Then I am going to Contract Manufacturer and try to X-ray the part. 

    There is no switch on the 24 Volt Supply line. It is just a 5.5 mm DC Jack.

    I will keep you posted.

    Thanks

    Allen

  • Is the failure repeatable (i.e. have you seen same failure and steps to reproduce on multiple boards)? Have you tried replacing the TPSM84209 to see if the circuit works again as expected after replacing it?

    On a working board, can you please take an oscilloscope capture on the VIN pin 2 of the TPSM84209 when you plug in the 24V supply. Maybe that input filter is resonating when the 24V input is applied and the voltage is exceeding the abs max ratings of the device.

  • Allen, have you resolved the issue or do you still need help debugging?

  • I need a couple of more days to finish gathering thermal images and posting them.

  • No, Since I only have one board and it is dead.

    I am going to go to CM and replace U2. Then I capture voltage on VIN pin 2 of TPSM84209. 

    Meanwhile here is a thermal image of the board when I apply 24 Volt power.

  • I replaced TPSM84209.

    Captured Vin Pin on Startup. It shows an overshoot of 3 volts. I am going to use LMZM23601 for future revisions.

    .

  • I noticed there is also ringing below ground in your Vin waveform. Both TPSM84209 and LMZM23601 allow only down to -0.3V negative voltage on VIN pins in their Absolute Maximum Ratings tables. If the waveform is accurate, the ringing goes to about -3V which may be the cause of damage.

  • Do you have any recommendations about fixing the -3V voltage at startup?

  • Is this the TPSM84209 VIN pin 2 waveform when you plug in your 24V into your power jack? Is the power jack J5? Can you please take another waveform of VIN pin 2 with time x-axis zoomed in more, and also capture the waveform at the 24V net before the filter on a 2nd channel?

    I would like to see the ringing in more detail, and also understand if it is there before the filter as well as after.