What is on the sillicon in the upper right corner of the nRF52832?

[CurtisHx]

I have a QFN48 nRF52832, and it's getting hot. I'm looking at it with a thermal camera and there's a hot spot directly under the "N" of the "N52832" that's on the package. I think that would put the hotspot on extreme upper right hand corner of the die itself. If I can figure out what's in that general area, that will tell me where to start looking for problems.

[Sigurd]

Hi,

Could you explain a bit more about what do mean by hot? What temperatures are you measuring?

Please make sure that you are not using the nRF52832 outside the Recommended operating conditions, and that you are within the electrical specification on the maximum current you can sink/source on the GPIO pins.

[CurtisHx]

I'm seeing a hot spot of about 115F. Everything I've checked (voltage / temperature) is within the operating conditions. All of the GPIOs are set as disconnected inputs, so they shouldn't be sourcing / sinking inputs.

[Sigurd]

Is this a custom board? Are you seeing this on several devices, or only one device? Could you measure the current consumption on this device?

[CurtisHx]

It's a custom board. I have 2, but the other one is misbehaving in other ways. Current consumption is ~80mA into the nRF52832 with the LDO enabled. It's 100mA with the DC-DC converter. The inductors and cap for the DC-DC are on the board, just like the reference design. I loaded the same firmware onto a completely different (but still custom) board, and it draws 3mA. The firmware starts up the DC-DC converter, sets all of the GPIOs as disconnected inputs, and then falls into an infinite loop. It's just running the CPU. I've looked at the bad board under a microscope and under an xray machine. I cannot find any explanation for why the bad board is drawing so much current and why the nRF52 is getting hot. I'm looking at it with a thermal camera and there is a clear hot spot in the upper left corner of the nRF52

[CurtisHx]

I'm getting closer. On the good board, I can see the DC-DC converter switching nicely on pin 47 of the nRF52. On the bad board, pin 47 is stuck at a constant 2.35V. It's not switching even though the DC-DC converter has been enabled in firmware. Pin 47 snaps to 2.35V on power up and stays there. I can't see any attempt at switching happening.