TPS62143: Part easily damaged on power up

Chris Jones21

Part Number: TPS62143

Hello,

I am using the fixed 5V output TPS62143 switching regulator in my design. It looks like I have a problem with overshoot on my input when power is applied. Is there a filter design TI suggests to suppress the overshoot on power up?

thanks,

Chris

over 7 years ago

0 Chris Glaser over 7 years ago

TI__Guru*** 144230 points

Can you post a waveform of the overshoot you see? Do you know where it comes from in your system?

It is usually not recommended to have an overshooting input supply, so this would be the best thing to fix.

0 Chris Jones21 over 7 years ago in reply to Chris Glaser

Prodigy 170 points

Yes, I will try to capture the overshoot and post it. Unfortunately, I can't change the supply. Our product line uses a common 14V AC adapter to power our products. Our new offering requires 5V and I need to regulate from our existing adapter.

0 Chris Jones21 over 7 years ago in reply to Chris Glaser

Prodigy 170 points

Attached you will see a capture of the overshoot on input. This damaged the regulator and now Vin is shorted to Gnd. I've had other times where all pins end up shorted after the failure.

I have also tried to leave EN pulled to Gnd, and while this seems to help the problem eventually the part failed.

0 Chris Glaser over 7 years ago in reply to Chris Jones21

TI__Guru*** 144230 points

This looks like exactly what I described in a blog a few years ago: e2e.ti.com/.../what-is-that-giant-tantalum-cap-on-the-input-of-the-evm

You can simulate your adaptor cable in TINA and determine how much cap and ESR is needed to snub the hot plug. You should also do bench testing to validate any fix. There are also OV protection devices which can block the voltage until it is done ringing.

0 Chris Jones21 over 7 years ago in reply to Chris Glaser

Prodigy 170 points

Chris,

The Tantalum cap has helped A LOT! Thank you. As a follow up question: In addition to the tantalum cap, if I had an RC on the EN line and increased the SS/TR cap value, would that help with the hot plug at all?

0 Chris Glaser over 7 years ago in reply to Chris Jones21

TI__Guru*** 144230 points

No, this should not help. The overshoot is very likely over by the time the IC turns on.

You can confirm this by adding the SW node to the above waveform. When it goes high, the IC has started switching.

0 Chris Jones21 over 7 years ago in reply to Chris Glaser

Prodigy 170 points

What about delaying the EN pin through an RC?

0 Chris Glaser over 7 years ago in reply to Chris Jones21

TI__Guru*** 144230 points

I don't understand how this help prevent an OV on Vin. Can you explain a little more how this would help?

0 Chris Jones21 over 7 years ago in reply to Chris Glaser

Prodigy 170 points

The idea is that if the regulator isn't switching when the hot plug occurs maybe the overshoot will be minimized. The only current draw will be to charge the caps. Once all the input caps are charged up the EN pin will have charged up and the regulator can start switching from an already stable supply.

0 Chris Glaser over 7 years ago in reply to Chris Jones21

TI__Guru*** 144230 points

Right. That's what I tried to address in my last reply on Wednesday. The overshoot occurs and finishes before the IC is even on. You can confirm this by looking at the SW node when you hot-plug. It should be low until well after the overshoot is complete.

Power management

Power management forum

TPS62143: Part easily damaged on power up