TPS54240 Failure, appears to be thermal

Clifford Pontbriand

Other Parts Discussed in Thread: TPS54240

I'm using a TPS54240 device in a DC/DC converter application. In 2 out of 50 units I have seen a shutdown of the TPS54240 after operating for "a while". The converter design is not near the performance limitations of the part:

24 volts input

5 volts output at ~400 mA output current

One of the parts that failed will operate properly when initially powered, and will shut down after prolonged operation (~2 hours). The part will not function upon power cycling. It will operate after resting for > 1 hour.

The other failed part functioned intermittently when initially powered on. After running for ~4 hours, the part shut down and will no longer operate. Neither part draws excess current when in "shutdown". Die temperature does not exceed 40 C. At my load, I would not expect a thermal shutdown until ambient temperature in excess of 140 C, based on the calculation in section 8.2.1.2.12 of the TPS54240 data sheet (page 37). Is there another mode of failure I should be looking at other than a thermal one?

Thanks,

Cliff

over 9 years ago

0 JohnTucker over 9 years ago

TI__Guru**** 191445 points

The thermal performance of the TPS54240 depends greatly on the PCB design. The hysteresis on the OTP is only about 15 deg. C. and it is non-latching, so the device should re-start on its own after some seconds or possibly around a minute. Can you post your schematic and PCB layout?

0 Clifford Pontbriand over 9 years ago in reply to JohnTucker

Prodigy 35 points

TPS54240_Pontbriand_15JAN2016_excerpt.pdf TOP_BOTTOM_U6_15JAN2015.pdf

Tried to insert a schematic and pdf of top and bottom copper layers. Ground layer is contiguous underneath part and surrounding layout. I have 4 vias from the power pad into the ground layer. There is no copper plane area on the bottom layer underneath the part.

Thanks for your quick reply!

0 Clifford Pontbriand over 9 years ago in reply to JohnTucker

Prodigy 35 points

An update-- I've been running the 2 failed units for 2 days now and I cannot get them to fail! TPS54240 temperature is not elevated over board temperature (~40-50 C). If the power supplies have not shut down by next week, my next step will be to run them in an environmental chamber at increasing ambient temperatures. The long duration of the outages (when they happen!), combined with the relatively light load, causes me to doubt the thermal shutdown hypothesis. I am looking for possible connector problems with the input power to the converters, but so far everything looks good.

Cliff

0 JohnTucker over 9 years ago in reply to Clifford Pontbriand

TI__Guru**** 191445 points

OK, keep us posted.

Power management

Power management forum

TPS54240 Failure, appears to be thermal