ASRock B450 Steel Legend - no reaction to power button

[karlp028]

I've got an ASRock B450 Steel Legend that I'm trying to diagnose for no reaction to the power button.

The SIO receives the PWRBTN# signal and sends PSOUT#, but the CPU doesn't return SLP_S3# so nothing further happens. I have been looking around the board trying to figure out why and I happened to touch the STEP-DOWN CONVERTER NB685GQ-Z (UP10 in photo) which was VERY hot (like burn your finger). This seems to provide VDDP_S5, and indirectly, VDDCR_SOC_S5, which the CPU needs in order to handle the startup signals.

The VCC voltage powering the converter does not look right. It comes from the +3VSB rail and goes through resistor RP196 (4.7 ohm) before connecting to pin 3 on UP10. I measured 3.32V for +3VSB, but on the other side of the resistor it is only 0.12V and I'm not sure why. Pin 3 of the converter seems to be grounded because I get 0.6 ohms between that and GND (my meter shows 0.4 when you touch the probes together). Does that mean the chip has failed, or could it be something else?

[FriedFred]

With a resistance of just 4,7R it's quite impossible to measure such a high voltage difference. So I think, this resistor has a much higher resistance now. Worth to desolder it and check. Probably it's burned.

[karlp028]

I think it's more likely that the chip is shorted and is drawing high current. That would explain the voltage drop and the heating of the chip.

[Herbs n spices]

I'd wager the hot chip is shorted like you said, but make sure the downstream things it powers aren't what actually shorted, and double check the pin descriptions in its datasheet. The 4.7 ohm resistor probably turned into a pseudo-fuse too like FriedFred said.

P.S.: If you're doing a lot of motherboard repairs, I'd get a cheap thermal imaging camera. It'll pay for itself in time and burnt fingers saved! I've seen parts hit >250C with shorts exactly like you've got here.

[karlp028]

I got a new microscope a couple of weeks ago and was looking at this board again tonight. I noticed some damage to the capacitor (CP132) that was next to the resistor in question. I removed it and now the short is gone.