• State
  • Locked Locked
  • Replies 4 replies
  • Subscribers 63 subscribers
  • Views 338 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.

LMZ10505: Part failures: short to GND

Jon Mason
Prodigy 190 points
Part Number: LMZ10505

Hi - We're evaluating this part but we've had two failures where the input and output are both shorted to GND. Can you explain what could cause that to happen?

Vin = 5V, from a bench PSU

Vout = 3.3V

Iout = 2A resistive load.

The output waveform looks well-controlled. Blue is the enable input. Red is the output. 1mS/div, 1V/div

  • 0
    Prodigy 190 points

    C47,49,9,10,11,66 = 47uF ceramic
    C12,64 = 1uf ceramic

    C13,65 = 100nF ceramic

    C7 = 2.7nF

    C8 = 82pF

    R7 = 100K

    R8 = 32K0

    R9 = 665R


    I notice other people on this forum have had similar problems but no resolution has been posted that I can find. 

  • 0 in reply to Jon Mason
    TI__Guru**** 150900 points

    Hi Jon,

    The 2 waveforms in your first post did not come through.  Can you please repost them?

    Many power supply failures are related to the PCB layout.  Can you post yours for review?

    Is there anything between the 5Vin and the external power supply?  Have you already checked for hot-plug overshoot of the 5Vin?

  • 0 in reply to Chris Glaser
    Prodigy 190 points

    There is nothing between 5V-in and the bench supply. Would you recommend a MOV across the input of the LMZ10505?

    Can you explain the likely failure mechanism when a LMZ10505 input and output are shorted to GND?

    I have attached some waveforms and the PCB layout for review. I notice that for some reason, the GND plane for the EP of the LMZ10505 does not extend to the input and output caps We will certainly fix that in the next PCB spin. Could that potentially be the problem?

    LMZ10505 part failures.pdf

  • 0 in reply to Jon Mason
    TI__Guru**** 150900 points

    Thanks for sending Jon.

    Yes, both the Cin and Cout should connect directly, on the top copper layer, to the exposed pad of the device.  Vias in this path add inductance and are not recommended.  For a quick test, you could scrape away some soldermask to create places to solder the caps, where shown in figure 30 in the D/S.

    When Vin and Vout and GND are all shorted like that, it is most likely EOS on one/both power FETs inside the device.  This is usually the result of a too-high voltage occurring on the IC pins--either from excessive ringing during switching or Vin hot-plug.

    This technical article explains Vin hot-plug: 

    What is that giant tantalum cap on the input of the EVM?
    I often receive questions about why we place a physically large tantalum capacitor on the input of most low-power, step-down converter TPS62xxx evaluation modules (EVMs). Tantalum capacitors are generally…
    By in Archives > Fully Charged
     Do you get the same Vin waveform when you turn on the external power supply and then plug it in, compared to plugging in the power supply and then turning it on?