• State
  • Locked Locked
  • Replies 3 replies
  • Subscribers 63 subscribers
  • Views 76 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: TPS61022

Daniel Esperon
Prodigy 20 points
Part Number: TPS61022

Tool/software:

Hi,

I have a design with the TPS61022, the input range is 2.7-4.3V and the output is 5V@3A.

I use the output to power a leds (100ms ON, 900ms OFF). It works fine, but if I touch the DC/DC zone with the finger to check the temperature, the DC/DC is destroyed and is in shortcircuit.

It is the first time that I have this problem with a DC/DC. Could you help me?

I think that the problem is in the FB-VOUT resistors feedback, we probe to change the 732K and 100K to 73'2K and 10K. The result is the same.

Could be possible that a feedforward capacitor was needed?

Best regards,

Daniel.

  • 0
    TI__Mastermind 18830 points

    Hi Daniel,

    What's the failure mode? Is the SW shorted to the GND?

    I don't think a feedforward cap will help. As my checking, the damage should be caused by the bad layout, in which the main power path from the VOUT pin to Cout to Cout GND to chip GND pin is very long. So the parasitic inductance is big, which may cause big SW spike. 

    The max SW spike should be very close to the damage value 9V when your finger doesn't touch the board. You may verify this by measuring the SW spike.

    After your finger touches the board, the parasitic may slightly increased (not quite sure about this, just guess) and the SW spike increased, then damage the device. 

    And also, you can verify above guess by measuring the SW spike with your finger touches the board, and all the other work conditions keep the same.

    Regards

    Lei

  • 0 in reply to Lei Zhong
    Prodigy 20 points

    Hi Lei,

    I put a feedforward cap of 10nF directly to try it and the problem has solved. It was the easy way to probe something.

    I think that the fb loop is the problem, the cap help to stabilize the loop.

    In other hand, I will try to put the mode pin to high level to enter in PWM forced mode. In our application the consumption are not a problem and do it shoul be help to stabilize more the fb loop.

    In the redesign I will optimize the layout. Thak you for your reply.

    Best regards,

    Daniel.

  • 0 in reply to Daniel Esperon
    TI__Mastermind 18830 points

    Hi Daniel,

    I'm surprised that a 10nF feedback cap works! Usually it's too big and may cause instability issues.

    Anyway, I'm glad than you problem solved! Thanks to feedback the good news.

    In your redesign, please optimize the layout and put the footprint for the feedforward cap.

    Regards

    Lei