Asus FX517ZC charges but power button does nothing

[TimeRabbit07]

Hi all!

I've got an Asus FX517ZC on the bench at the moment and I just can't figure out what's going on with it.


Here are its various numbers and identifiers:
Model short: FX517Z
Model long: FX517ZC-HN114W
SN: NANRCX01S57842E
CN: 3SDB 12M
MFD: 2022-10
Motherboard model: DA0NJLMBAE0 REV E


Symptoms:
The laptop's charge light comes on, but when the power on button is pressed nothing happens. No fan spin, no caps lock, no screen or keyboard backlight, nothing.


History so far:
The customer believed the laptop's BIOS was corrupt and wanted to flash it themselves, but decided against this after taking the laptop apart and seeing they couldn't just clip on to the BIOS chip. They then bought it to me, and I continued to desolder and reflash the BIOS. After reflashing the BIOS the symptoms remain the same, so either I haven't flashed it properly or something else is causing the issue.

Overall, the laptop has no clear signs of physical or liquid damage and looks to be in near-mint condition.


What I've tested:
I've removed, re-verified and re-soldered the BIOS chip (displacing a couple of VERY small resistors in the process, getting them back into place wasn't fun) and everything regarding the BIOS is fine as far as I can tell, though BIOS flashing is an area I have little experience in so I may be missing something.

I've also checked the laptop's lid sensor which seems to be working fine and isn't damaged.

I've checked that the power button is working, and I can measure it where it connects to the mainboard pulling the voltage down to something like 47mV when pressed. When the button is pressed I can see 3.3V going to the BIOS chip, and also the 5V, 1.8V, and 1.05V rails come to life so I know the button is being registered. While checking power rails I also found that the main power rail was showing 20V to the high side/input of the main CPU MOSFETs, but nothing on their output. I assume this is because they're missing their gate signal, as there was no voltage on the gate pin, but I haven't followed that line of thinking any further yet.

If I remember correctly, the laptop pulls about 0.1A after the power button is pushed (and with the battery disconnected, so no charge current) - which is an increase from before the button is pushed, so it is trying to start. It also doesn't appear to loop at all. I haven't got access to a thermal camera so figuring out where this power is going would presumably be impractical.

If anyone has faced similar faults in the past and feels like weighing in it would be much appreciated, otherwise, I'll have to mark this as a no fix-no-fee and give the customer a number for a place with more board repair experience than me 😅
Thanks for reading!

[reformatt]

FYI, Asus/Acer boards that have power buttons on the keyboard usually have a unpopulated switch on the motherboard. I solder on a microswitch to that as this makes debugging the board on the bench much easier. Just get a bag of different switches off Aliexpress, usually one will fit or be close to it. There is a boardview on the forum for a DA0NJLMBAD0, which is pretty close and the switch is KPSW1. As usual with Asus boards, no schematic available.

First do a thorough examination of the board for corrosion or corroded components (or testpads). Lots of Asus boards have corrosion somewhere near the fan outlet. Like Apple, blowing or sucking air across a board is really bad for electronics and is a common failure point for corrosion on these boards. Don't take too much stock in what a customer tells you. If you aren't confident in the BIOS, get the original dump and post it in the BIOS section and someone will fix it for you in order to rule it out.

My biggest indicator as to what a machine is doing is by observing input current (without battery). See if it is static or dynamic. Check with or without RAM. What voltages are present on each coil when the power button is pressed. This can often localise an issue relatively quickly.

[TimeRabbit07]

Hi Reformatt, thanks for dropping in - and for the advice on the power button! I've been using that boardview too, but hadn't realised the pads for a button were there; being able to test without the keyboard is so much easier.

I'm fairly confident in the BIOS as I was the one who sourced it, though I'm new to flashing so it's far from impossible something is wrong with it - I've just decided to trust that it's okay rather than get fixated on that one possible issue and miss something else!

As for missing something else, in complete contradiction to my original post, I double-checked voltages on the coils again and am getting nothing on any of the coils EXCEPT the 5v and 3.3v rails which are fine, and the 1.8v (1.8v_S5 on the boardview) rail, which, after the power button is pushed has an odd constant jumping between 200 and 600mv - but never reaching 1.8v as it should. The chip (PU2001) appears to be getting 20V fine, so could this chip be dead, is it missing an enable signal, or is it normal for the S5 rails not to be on yet? thoughts are appreciated!

Input current wise, the board starts out jumping between 0.012A and 0.023A, but then when the power button is pressed goes up to 0.040A and remains solid. Then after some time (a few minutes) drops down to 0.029A solid. Which, to me seems like a very very low current.

As you suggested I have also taken a more thorough look around the board but didn't see any indicators of corrosion - though knowing that's a common issue will come in handy at some point, thanks!

[reformatt]

I don't have the correct boardview for this model so can only give you general advice. Each rail will come up in accordance with the power sequence. When a buck converter indicates power good on a specific rail, then the next one in the sequence will be enabled and so on.

So it would seem your 1.8V S5 rail has an issue. Is this rail shorted or very low impedance? According to the DA0NJLMBAD0 boardview, 1.8V_S5 is generated by PU501, boost network is PR505 and PC504 (these are first things to check for a low output if the rail resistance is ok). But this rail leads off to the CPU, which I'm assuming is the PCH part of it as it's a S5 rail. So could be a bad PCH in this instance.

[TimeRabbit07]

I fully appreciate you're working with very limited information, even general advise and rough theories are of colossal help!

You're quite right, I meant to say PU501 but got the 1.8v and 3.3v mixed up, dough! I'm seeing 4.75v on PR505's and PC504's inputs (I assume input?), then the same 4.75V on PR505s output, but only 004mV to 030mV on PC504's output (its constantly changing then spikes up to 030mv and back down again), which is shared with the coil that itself is now reading the same 004 to 030mV and not jumping between 200 and 600mv as it was before, which is strange.

I also double checked the resistance on the 1.8v rail's coil and its reading 1.7kOhm, which seems plenty high enough(?).

My somewhat limited understanding of these symptoms brings me back to thinking either PU501 needs replacing as it seems to be receiving proper input and have no shorted output but isn't generating the 1.8v rail, however, am I jumping the gun a bit in saying that or could your suggestion of a bad PCH also be able to cause the 1.8v rail to not activate? (In theory, obviously, the 9443 mile gap between you and the board makes saying anything about it for certain a bit tricky).

[TimeRabbit07]

You're quite right, I meant to say PU501 but got the 1.8v and 3.3v mixed up, dough! I'm seeing 4.75v on PR505's and PC504's inputs (I assume input?), then the same 4.75V on PR505s output, but only 004mV to 030mV on PC504's output (its constantly changing then spikes up to 030mv and back down again), which is shared with the coil that itself is now reading the same 004 to 030mV and not jumping between 200 and 600mv as it was before, which is strange.

Update: the 1.8V rail's coil/PC504's output has gone back to its old behaviour of jumping between a few hundred Millie volts if that matters - perhaps I had a bad connection while measuring earlier, or some being from the warp was interfering - who knows!

[reformatt]

Is the enable on pin 2 staying high when power button is pressed or fluctuating? Until you get power good high on pin 1, the board won't go further in the power sequence. Check all passive components around the IC and if ok, replace the IC. Impedance on the output is ok. Is this an Intel or AMD CPU BTW?

[TimeRabbit07]

Argh, the plot thickens! After being left overnight I started the board up this morning to check for the pin 2 activation signal - it was there! And I had 1.8v on the 1.8v coil, and the board was drawing slightly more power than it had been!! However, after restarting the board it's back to how it was before with no 1.8v and no activation signal on pin 2 either! I also noticed that PR505's input raised to 5.5v rather than 4.7v, which means whatever caused this jolt of life also affected other power rails too... Which leaves me with a great many questions I'm starting to think I won't be getting answers to.

PR508 does have 3.3v on one of its pads, however there's no component in that position on this board. PC511 is empty too. PR509 also isn't there, but there is a trace in that area which I'm assuming serves the same purpose and must be where the activation signal comes from? I've attached a picture to clarify.

The sudden appearance of the 1.8V power rail and its activation signal assumably means that something before PU501 in the power sequence has partially failed and is still able to occasionally do its job, hence the inconsistent 1.8v rail. The only clue I can think to follow is that PU501 isn't getting its activation signal and that following the signal back through the board may reveal our culprit. I've tried to figure out where this signal comes from, but after looking over the boardview for 20 minutes I've still gotten nowhere. Would you happen to have any thoughts/pointers?

Oh! And I believe this board is home to an i5-12450H and a RTX3050 though I'm not 100% certain.

[reformatt]

VCCIO_EN-1 is fed from SLP_SUS#, which comes from the EC (KU1). It's likely that a pre-requisite is not right check the 3V/5V power good signals back to the EC and RSMRST#. I did have an Asus board recently where the power good signal for the 3V was being randomly pulled down. I found this to be a faulty diode.

[TimeRabbit07]

Thanks! I followed it to U6008 but then couldn't figure out which pins were meant for what 🤦

I'm getting 3.3v on one side of KD1, 2, 3 and 6 which is 3V_5VPGD. I'm guessing as it's called "3V_5V" that this signal is responsible for communicating both voltage's power good states?

On the OTHER side of KD1, 2 and 3 however, I'm reading nothing. In contrast, KD6 is reading 3.3v on DP_PWROK. Should all of these diodes have 3.3v on both sides, or in your experience is their current state to be expected?