TPS82150: Components fail during operation

reto hermann

Part Number: TPS82150

Tool/software:

Dear support team

we are using the TPS82150 in one of our designs and have the issue that two parts (different productions lots of the parts and the boards) failed during operation after initially working properly. I have concerns regarding a potential design issue. Can you please review our schematics?

0160.pmic.pdf

In both cases the IC102 (FPGA1V2) failed. After replacing the defective parts the boards are working properly. The defective parts both have a short circuit (2..7 Ohm) between VOUT and GND.

Here are some measurement from a working reference board:

If you need more information, please let me know.
Thank you for your support!

Reto

5 months ago

0 Colin Chen 5 months ago

TI__Mastermind 21141 points

Hi Reto,

The sch looks good to me.

Could you check the VIN, Vout and ss waveforms to me?

Thanks, Colin

0 Andy Chen 5 months ago

TI__Mastermind 34395 points

Hi Reto,

Did you saw similar failure on 3V6 socket? can you test SW node waveform at inductor terminal?

Can you upload the screen shot of layout?

0 reto hermann 5 months ago

Intellectual 720 points

Hi, regarding your questions:

Vin and Vout waveforms see in my post above.

SW node waveform

SS waveform

Did you saw similar failure on 3V6 socket? No, this part did never fail so far.

Reto

0 Colin Chen 5 months ago in reply to reto hermann

TI__Mastermind 21141 points

Hi reto,

These waveforms looks normal.

Could you show me the layout?

Thanks,

Colin

0 reto hermann 5 months ago in reply to Colin Chen

Intellectual 720 points

Hi Colin

sure, here is the layout:

All components are located on the top layer. The next layer below (mid-layer1) is a GND plane

BR
Reto

0 Colin Chen 5 months ago in reply to reto hermann

TI__Mastermind 21141 points

Hi Reto,

This layout is not good. According to your description, Vout short to GND means low side mos is damaged. This is probably caused by SW rise ring which is the voltage stress for low side mos. To reduce SW rising ring, please put the 0.1uF cap at closest to VIN pin like below shown.

Thanks,

Colin

0 reto hermann 5 months ago in reply to Colin Chen

Intellectual 720 points

I have just realized a difference between the 3V3 and the 1V2 supply when looking at the inductor voltage ringing:

With the 1V2 supply the SW node voltage drops below 0V at it appears as if it is limited by a diode to approx. -700mV whereas with the 3V3 supply the voltage is always > 0V. Could this be a problem? Could this be the reason why we never have seen failing the 3V3 supply.

0 Colin Chen 5 months ago in reply to reto hermann

TI__Mastermind 21141 points

Hi Reto,

This is caused by dead time of internal mos driver and inductor current would flow through body diode of low mos. The -0.7V v is body diode voltage drop. This wouldn't cause any problem.

Thanks,

COlin

0 reto hermann 5 months ago in reply to Colin Chen

Intellectual 720 points

Hi Colin
thank you for your answer / suggestion. I have placed an additional 100n cap as close as possible to the input as you have suggested, but I cant see any difference of the ringing at the SW node.

I'm not sure if I understand your explanation. You are talking about the "SW rise ring" but when I look at the waveform I can't see ringing at the rising edge of the SW node voltage:

Power management

Power management forum

TPS82150: Components fail during operation