• State Resolved
  • Locked Locked
  • Replies 5 replies
  • Answers 1 answer
  • Subscribers 63 subscribers
  • Views 253 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.

TPS61022: Dead short on input of TPS61022

Sean O'Leary
Prodigy 70 points
Part Number: TPS61022

We have 5 boards we have built with this design. Two of them are showing signs of failure. One is working fine.

One of the boards that failed shows a short on the VCAP net to ground.  The VBAK net show 14K to ground.  After removing U10, the short is gone. This indicates to me that U10 was damaged.

C53 (68uF, 1206, 10V) is as close to pin 3 as I could get it, about 170 mils.

C52 (10uF, 0402, 10V) is the input cap.

I am not sure what event caused the failures or why some boards work fine while others are seeing failures.

  • 0
    TI__Genius 12245 points

    Hi Sean,

    It looks like the low side FET is damaged, so VCAP is shorted to ground.

    It should be caused by the poor layout, especially the placement of the output capacitor, C53. Please place the output capacitors as close as to the VOUT GND pin.

    You can refer to the layout guidelines in d/s, or refer to the app note below.

    www.ti.com/.../slvaes4.pdf

    Regards,

    Nathan

  • 0 in reply to Nathan Ding
    Prodigy 70 points

    Hi Nathan,

    I will fix the layout on the next rev.  I am surprised that C53 would cause this issue since it is only 170 mils away from the pin. Could that really cause the issue?

  • 0 in reply to Sean O'Leary
    TI__Genius 12245 points

    Hi Sean, 

    Yes, that is what we usually found in some other cases.

    The longer trace, larger package size capacitor (1206 larger than 0805) would introduce larger parasitic inductance, that would cause higher voltage spike at SW node. 

    You can test SW waveforms to check if the SW voltage spike is so high that may have a risk of damage. Please refer to this app note as a guidelines about how to test SW by a correct way.

    www.ti.com/.../slva494a.pdf

    And more, may I know the operation condition in your test?

    For example, the VIN range, IOUT range, if your input voltage is very low and IOUT is very large, maybe a feedforward capacitor is also needed to avoid the potential risk of instability. 

    You could also share your new schematic and layout to us after you finish it, so that we can help to review in advance.

    Regards,

    Nathan

  • 0 in reply to Nathan Ding
    Prodigy 70 points

    Thanks Nathan,

    Most of the time the output current is very low uAs.  When the main voltage fails, the supercaps feeding the 61022 become the preferred source and the output is switched into the main current path. It must provide about 1.2A for about 5 seconds while the processor shuts down. The input voltage starts at 5.2V and continues until the MCU shuts down at about  3V.

    Sean

  • +1 in reply to Sean O'Leary
    TI__Genius 12245 points

    Hi Sean,

    With this condition, the feed forward is not necessary. 

    If possible, at least one 0805 package size output capacitor is also recommended, because of the lower ESL I mentioned above. I mean place at least one 0805 package size capacitor at COUT as close as possible.

    And more, when input voltage starts at 5.2V, is output voltage equals to about ~0V? If that, we recommend to add a pre-bias diode as below. Otherwise they will also a risk of damage.

     

    Regards,

    Nathan