ASUS Zephyrus GA401IV 1.4 GA401IU MB._8G_R7-4800S_AS_R1.4fz AC_BAT_SYS ~32Ohms

[abajor]

I have an ASUS Zephyrus board on which the owner shorted the battery clip to a the small filter cap which goes across the "BAT" connections 1 2 and 3 to ground, and one one of the TVS diodes is missing.
Other than the small .1uF cap which was blasted off its pads and the missing TVS diode everything appears fine.

How ever when I get to the AC_BAT_SYS section I read between 28-33ohms to ground (this depends on how warm I've made the area). I'm also reading the same thing across +PVDD_AMP capacitors.
I have tried to inject 1V into AC_BAT_SYS. I draw about .15A, but with the thermal camera I can not see anything at all heating up on this board.

Normally when something draws that sort of current I can easily see the hotspot form even when its on the opposite side of the board, so this is a little strange and confusing.

I'm sort of at a loss of how to proceed. I only have the schematic for a different year of this board I've tested the high side low side mosfets as well as removed them and several capacitors from the board and still I'm reading the ~30ohm to ground.

I tried putting kapton tape over the U4601 USB PD CONTROL SN1905005YBGR, since that appears to be a shiny raw silicon die, but that doesn't appear to be heating up either.

Any ideas where this .657W of heat may be going?

[mcplslg123]

Resistance to GND reading at 28-33 ohms on AC_BAT_SYS very likely indicates a shorted mosfet. It cant be a cpacitor.
If caps is shorted the resistance to GND reading on the same line will be 0 ohms-5 ohms.
So focus on all the mosfets where this line goes and remove them one by one. Could be a pwm controller laso.

[abajor]

Resistance to GND reading at 28-33 ohms on AC_BAT_SYS very likely indicates a shorted mosfet. It cant be a cpacitor.
If caps is shorted the resistance to GND reading on the same line will be 0 ohms-5 ohms.
So focus on all the mosfets where this line goes and remove them one by one. Could be a pwm controller laso.

Thanks I already took out PQH8901, PQL8901 and PQ8903 The three largest N-Channel Mosfets in the area. For what ever reason I can only find a board view for this part number not a schematic so it's extra exciting hunting and pecking for related circuits.

I'm assuming PVDD_AMP and AC_BAT_SYS are a parallel source since they both exhibit the same resistance to ground.

[abajor]

Resistance to GND reading at 28-33 ohms on AC_BAT_SYS very likely indicates a shorted mosfet. It cant be a cpacitor.
If caps is shorted the resistance to GND reading on the same line will be 0 ohms-5 ohms.
So focus on all the mosfets where this line goes and remove them one by one. Could be a pwm controller laso.

Thankfully I was able to find a Schematic on my board schematic service which says it is the GA401QM version of the motherboard. My best guess at the moment is that its one of the APU power IC's since that's the only power IC that has a direct connection to ground which receives AC_BAT_SYS, also I did confirm that PVDD_AMP is just AC_BAT_SYS with a couple of fuses and some filter caps separating the two rail designations, so still could be something there as well.

[abajor]

It looks like +NVVDD might be shorted I'm measuring a DCR of 0.6 Ohm. I don't know for sure I know there are a huge amount of parallel connections in a GPU so perhaps 0.6 Ohm is perfectly normal across the GPU power rail?
+VDDCR_SOC measures 9.6 Ohm.

[abajor]

Found the problem.
It's one of these https://www.digikey.com/en/products/...5018SG/5305044
High Side / Low Side buck converter mosfet packages for powering the V-RAM rails.

The only way I figured out what the components were was watching the resistance drop as I blew hot air in the region of the VRAM. It's pretty wild that the mosfet package itself doesn't heat up when I inject current into the board, but the resistance drops as I pump heat into the traces.

I'm surprised this is what failed, I almost wonder if the buck converter chip failed before the owner messed around with the battery, as far as I can tell he just shorted the battery hot terminal to ground with the metal clip.

I just hope the VRAM is ok :x

[abajor]

Resistance to GND reading at 28-33 ohms on AC_BAT_SYS very likely indicates a shorted mosfet. It cant be a cpacitor.
If caps is shorted the resistance to GND reading on the same line will be 0 ohms-5 ohms.
So focus on all the mosfets where this line goes and remove them one by one. Could be a pwm controller laso.

I'm about to order some mosfets for the VRAM power stage, I'm wondering what a typical DCR of the VRAM "+FBVDDQ" voltage rail should be to ground? I do have concerns that the VRAM is damaged since the buck converter mostfet was shorted, and the thermally varying DCR is coming from the VRAM itself.

[mcplslg123]

Yes, very likely all the VRAM chips are fried.

[abajor]

Yes, very likely all the VRAM chips are fried.

So learning the hard way I removed all of the VRAM chips, not sure if they use some special formulation of solder for the BGA but they were very difficult to desolder, ripped a few pads I need to now fix with micron wire.
The really odd thing is after removing all six of the VRAM chips I found that the circuit still has a resistance of around 56ohm. Very odd for a power rail with no load devices and no buck converter mosfets installed.

I didn't realize I needed 6 VRAM for this board so I'll have to wait to go any further with it. Wondering about this 56ohm resistance I'm measuring where it should just be an open?
Also when I put in a new VRAM module the Resistance went back down to near where it was and also appears to be thermally variable.