Asus G512 no power (BQ24780S problems)

[Dannyx]

Good day folks. It's been a while since I repaired a laptop, so here's a challenge: a friend of mine asked me to take a look at an Asus G512LI killed by the liquid metal that got smeared all over the board when someone tried to clean it (I'm not blaming it on him - could have been someone before him). I was able to clean it enough to get it going on my own, without any help and thought I was all done, but then I ended up breaking it again

See, the battery wasn't charging and the laptop wouldn't start on battery either. If you were to start it on AC and THEN unplug the charger, it would still run on battery until it ran out, but it wouldn't charge the battery back. I figured the problem lies with the IC responsible for charging, the BQ24780S, on page 77 of the schematic, so I ordered one, replaced it.....and now the laptop doesn't turn on at all

Worth mentioning the chip gave me a lot of trouble to replace - either my technique or my tools suck - I HATE soldering QFNs, as it never seems to work the way others make it look. The solder didn't seem to want to stick to the pads of the chip - I could still see some gaps between the board and the pads when inspecting it sideways as close as I could with a regular magnifier, as I don't have a microscope. I pushed on for, I'd say, a few hours (!), even knocking some components around it off the board, until I got it looking "good enough". I replaced the supporting components as best I could and tried plugging in the adapter. Nothing happens now: I no longer get AC_BAT_SYS and subsequently, no other power rails. The only light at the end of the tunnel is that nothing dramatic happened either, like sparks or shorted components.

What happens is the first and second FETs (PQ8901 and 2) do not open to let 20v through to AC_BAT_SYS.

On the IC itself (adapter, no battery):

- VCC is 20v (good)
- ACDET is 2.7v (good)
- REGN is 0v (first bad sign right there, as it should be 6v)
- anything else: don't care at this stage or don't remember

With JUST the battery (11v at pack's output):

- VCC is 11v (good)
- BATSRC is 10.8v (good)
- BATDRV is 11v (bad) - PQ8903 is not open (Interesting that I get 3v at keyboard connector, but only because power from the battery goes through body diode of PQ8903. Pressing the power button, nothing happens)
- REGN is 0v (?) The datasheet says it should be 6v even in battery mode - paragraph 7.3.1.1 on page 17 says this. In my case, I was getting 6v on the OLD IC but only in AC mode, which might explain why it wasn't starting on the battery alone.

No shorts on REGN as far as I know. All resistors are in spec, even those that I accidentally knocked off. The 5 FETs have high resistance on all pins. I think I somehow killed the IC or didn't solder it properly. I now ordered a new one and will try lower heat and better flux. I wasn't pleased with how the joints ended up looking from the beginning...maybe this was it.

Link to schematic - https://www.badcaps.net/forum/showthread.php?t=119245

Reminder: rules and organization of this section - https://www.badcaps.net/forum/showthread.php?t=117483

.

[mcplslg123]

Re: Asus G512 no power (BQ24780S problems)

If VCC is present,the charging IC should output REGN unless there is short to GND on REGN or else the charging IC itself is bad.

Soldering a QFN is not at all tough. Just pretin the IC pad and pins before soldering it on PCB. Use a good quality BGA flux.Thats it.

[Dannyx]

Re: Asus G512 no power (BQ24780S problems)

As an experiment, till a new chip arrives, I'll try soldering two thin wires directly on the IC itself on VCC and REGN and power it up on the bench and completely off the board, to see if it gives me REGN. I might not be able to do this, since the pins are VERY small...

[Dannyx]

Re: Asus G512 no power (BQ24780S problems)

Ok, I got a new BQ chip and after a LOT of struggling, I finally got it to stick. The laptop works now, but I'm back where I started, which is a relief at least: battery doesn't charge, won't start up on battery and only runs on battery for like 10s....

First problem is that the battery voltage immediately tanks when the adapter is removed, to like 8-9v, which no doubt triggers UVLO, either via SMbus or via PU8900 directly after a few seconds. I can confirm the laptop does at least switch to battery mode, because if I add my bench power supply set to 11v on top of the battery connector (with the battery still attached), the laptop stays on but seems to draw a lot of current: 3.5a, which, to me at least, seems a bit much at idle, although it IS a ROG, so it may be normal. This battery is "new" and shows 10.5v when idle. It won't go above this when on the laptop. I guess I could push it with the bench power supply, but it won't make any difference. The original battery is "locked out" (0v at the connector) and won't accept being charged with the bench supply, either on or off the laptop. Tried it with a third battery, same problem...

I tried digging into the power sequence to try and figure out why it won't fire on battery (page 89 of schematic). I get stuck at step 7: +3VA_DSW doesn't start when I hit the power button. I can see a flicker on the multimeter on +3VA_DSW, so it TRIES to fire up but cuts out immediately.

It's generated by PU8701 on page 75. What happens is: when you hit the power button, the EN2 pin goes high for a brief moment but immediately stops, so the output also stops. We know PU8701 is fine, because it does work in AC mode.

I think there's a problem with the EC chip which runs the whole show, because 3VADSW_EN (the enable pin for 3VA_DSW) is connected to the EC on page 29, so if for some reason, the EC decides it doesn't want to run, it pulls this pin low...not worth the risk probably...

[alex26]

Hello. I have a problem that is beyond me with an Asus Rog laptop model G512LI. If I power it, the board consumes 30-35mA and does not start. The 19V rail is ok. I only have 5V voltage on the coil and LDO voltages of 3.3V and 5V, and if I press the power button, the 5V voltage drops to 0V, and the voltage on the 3.3V coil from 0V increases to 1.15V. I measured the coil values and I have 2.1 ohms on the CPU, 5.8 ohms on Vcore, 0.9 ohms on the GPU, 143 ohms on VRAM, 309 ohms on RAM, 10.5K ohms on the 3.3V coil, 40 ohms on the PCH.

Under the thermal camera, the TPS51285A integrated circuit, which is the 3V and 5V chip, but also the EC IT5125VG, heats up very little.

I replaced the TPS51285A, but nothing changed. Then I removed the EC chip and the 5V voltage and the 3.3V voltages, the 1.8V voltage and the 1.05V voltage appear on the circuit. So, with the EC chip unsoldered I have voltage on 4 coils, and with it soldered I only have 5V voltage and when I press the power button, the 5V voltage disappears. I put a new EC chip and the problem is exactly the same as with the original integrated. By the way, with the EC chip unsoldered, if I powered the laptop, it would heat up to about 30 degrees on the maximum circuit, but no more than that even after 10 minutes. I don't know what else to measure, I measured pretty much everything I knew. There are 2 EC chips on the board, but I understand that one is for the backlit keyboard and such. I also have a 3.3v bios that has power and another 1.8v that has no power except probably at startup. I need some advice if you can help me.

[mon2]

Since you lose the 5v, suggest to test the resistance of every usb port, vbus line. Perhaps a defective usb load switch or esd TVs diode causing this shutdown.

[alex26]

I have mega ohms on every usb port. The esd diode is ok. If i inject 19v the board draws 30mA, but if i inject only in 5V coil ,the board draws 140 mA. Under the thermal camera I only see the pch and a network chip. I removed the network chip and the board now draws 110mA. I'll have to check the USB C and the audio card too. But i don t have any solutions.

[mon2]

Test the resistance to ground of each and every inductor on the board. Perhaps a downstream buck regulator is shorted.

[alex26]

PL9401 = 131 OHM
PL9201= 1.3 OHM - GPU
PL8105= 19.8 OHM
PL8103= 2.4 OHM - CPU
PL8106= 5.1 OHM
PL8401= 11.5 KOHM 1.8V
PL8302= 49.5 OHM - 1.05V
PL8601= 340 OHM - RAM
PL8901= 33KOHM - BATTERY
PL8702= 8.9 KOHM -3.3V
PL8701= 6.6 KOHM -5V
The values and voltages are measured with the EC unsoldered from the board.

[alex26]

I find the problem. It was short on the +5VS .