• State
  • Locked Locked
  • Replies 1 reply
  • Subscribers 63 subscribers
  • Views 2745 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.

LM5007: Defect Flyback diode

Sven Langhoff
Prodigy 10 points
Part Number: LM5007

Hello,

since some weeks we are faced with the problem that the flyback diode of our LM5007 dc/dc converter circuit gets broken at our customers system.

But we weren't able to reproduce it at our lab so far.

Our electronics (~100 pcs) works for months 24/7 without any failures.

This is what our 5V voltage regulator looks like:

The only difference is, that the flyback diode D4 is a 60V/1A type (10MQ060NPbF).

The input voltage (HV-IN) is 48V, maybe 52V, and the 5V load appr. 400mA.

We have an almost exact circuit at the same board with 12V output voltage, but until now we received only five defect 5V flyback diodes.

We already noticed that the maximum voltage rating of the diode (60V) could be a problem. But we analyzed the real breakdown voltage and it amounts very reliable 78V.

And since weeks we run the electronic in a climate chamber with 60°C at 58V without any failure.

Does anyone have an idea what might be the reason for the defect diode?

We noticed a negative spike for maybe 2ns of about -5V (load dependent) at the flyback diode (SW node) after control-FET turn-off. Maybe this is a hint.

We are thankful for every idea! Tomorrow we analyze the broken diode in a xray lab. Maybe this shows us the failure cause.

Best regards,

Sven Langhoff.

  • 0
    TI__Mastermind 39980 points
    Hello, I doubt 2ns of Reverse recovery will do anything to the diode? I like you strongly suspect that the issues are Abs max violation of the VRRM. It is possible that the VIN perturbation bac end up on the Switching node. If the unit goes in different applications where the length of input wire is longer then this can create large voltage overshoot on the input on start up? This is just one way I conceive the max VRRM is breaking down. if only a small percentage of units are failing, catching the failure maybe a little difficult, particularly in a lab environment? One suggestion is to increase the VRRM of the diode D4?