• State
  • Locked Locked
  • Replies 6 replies
  • Subscribers 62 subscribers
  • Views 887 views
  • Users 0 members are here
Support feedback
Options
Options
Similar topics

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.

LM2841x Fails when the input voltage is turned on sharply

Sanjib Acharya
Prodigy 130 points

Hi,

We have a board that uses LM2841x buck regulator IC to derive 3.3V, 100mA output (IC load) from (4.5V - 40V) input range. We are seeing a lots of failure in the field.

While investigating, we saw that the IC fails when the input power is turned on sharply from 0V to 30V (in ~20-30 usec). Some times the failure occurs after switching the input on multiple times at this rate and sometimes it gets reproduced at low temperature (-25degC and below) at lower input voltage.

Anybody has faced this issue? What might be the reason? Any solutions?

Regards,

Sanjib

  • 0
    TI__Guru 62275 points

    Can you seprate the SHDN input and Vin and determine if the problem occurs when the input or the SHDN is ramped?

    FD

  • 0 in reply to Frank De Stasi
    Prodigy 130 points

    Hi,

    In the circuit, there is a 33.2Kohm resistor between VIN and SHDN pins. When the IC fails, sometimes we see smoke and it gets burned near pin# 5. On some occasions we saw pin# 5 got welded to the pad!

    Reviewing the layout of the design, I found that the GND pin# 2 is connected to the ground plane on the next layer with a single via! Which I think is not good and I connected to wires to the nearby ground points from pin# 2. Which this failures have reduced. Please find the pictures before and after the wire mod on pin# 2.

    With the above mod, things are better. But still we saw some failures. We could capture the waveform when it fails, which is as below:

    I thought the output is getting unstable during the sharp rise of the input. Or due the sharp rise of input, the chip fails to make the output stable. I used a 0.22uF cap from SHDN pin to GND with the 33.2Kohm still connected between the VIN & SHDN pins. This implements the soft start, isn't? It helps a lot. The failure seemed to disappear as we cycled the input power for more than 100 times. But yesterday I saw two out of four modules failing at less than (-25)degC temperature...and the failure mode seems to be the same. It fails during input turn on.

    Any suggestions?

    Regards,

    Sanjib

  • 0 in reply to Sanjib Acharya
    Prodigy 130 points

    Hi FD,

    Could you get a chance to look into my response? Any further suggestions?

    Regards,

    Sanjib

  • 0 in reply to Sanjib Acharya
    TI__Guru 62275 points

    I will pass this along to someone more familiar with this part than I am.

    FD

  • 0 in reply to Frank De Stasi
    TI__Guru 62275 points

    It looks like you input supply is bouncing around at the time that the requlator is trying to settle.  If you can prevent some of this bouncing it may help.  You might try an Al. cap on the input.  Or an external UVLO circuit to keep the regulator off until you input supply has settled.

    FD

  • 0 in reply to Frank De Stasi
    Prodigy 130 points

    Hi Frank,

    After we discussed last time, I incorporated a soft-start using a 33.2K resistor connected between VIN and SHDN# pins and a 0.22uF capacitor between SHDN# pin and ground. Using this circuit, output is delayed by ~500 usec as shown below:

    The failure observed earlier was not seen in our lab after implementing the hot-swap. But recently we have got one issue reported from one of our customers that indicates that LM2841x circuit has failed. Not sure if this is related to a particular customer or not, we are digging into this. I will let you know the details about the failure once I know more about it.

    Regards,

    Sanjib