Hp g6 r23 motherboard fault

[Ndagger]

Hi guys.
I've got an hp g6 1325 with the r23 motherboard.
It won't boot and the caps lock key is flashing once which all relevant faqs say is a cpu issue.
I doubt the cpu is dead as I've got another amd e2 3000 cpu which when installed doesn't cure the issue.

Any tips on where to look first?
I have the motherboard schematic and am going to take this on but thought I'd ask the pro's if you have any advice.
Also if anyone has a boardview for this board is very much appreciate a link if you could.
Regards
Neil

[piernov]

Re: Hp g6 r23 motherboard fault

First check that you have CPU VCore, if you do, check that you've other power rails as well (the S0 ones especially, the other ones are probably present), and the APU_RST# signal. If you have all of this, you should try to flash the BIOS.

[Ndagger]

Re: Hp g6 r23 motherboard fault

Hi piernov thanks for the reply.
I'll check the power rails this evening and report back.
This is my first foray into motherboard repair, most of the repairs I do at work are on heating equipment so I don't have the equipment necessary for bios flashing. Since I'm looking at expanding my skills and doing more work with motherboards I'm open to getting myself a programmer. Would the ezp2010 be a good shout? Looking at the images I'm assuming I would desolder the bios ic, solder it onto the small board for the programmer that has the headers underneath and insert into the unit for programming?

[Ndagger]

Re: Hp g6 r23 motherboard fault

Ok.
I've attached the schematic I'm working from and have measured a few points on the board.
U46 pin8 3.2v
Pq47 pin1 5.08v
Pr279 19.4v one side 17.37v other side
U30 pin1 1.5v
Pu13 pin3 3.36v pin6 0.7v
U23 pin2 5.02v
Pr281 3.3v

The fan spins and the cpu gets warm

Anyone see anything amiss here?

[piernov]

Re: Hp g6 r23 motherboard fault

No idea about this programmer. I personally use a RT809F because it was cheap and I needed to flash computer displays too. Otherwise for BIOS, anything should be able to flash an SPI ROM (a Raspberry Pi can do it too).
0,7V on PU13 pin 3 doesn't seem right, schematics say it should be 1,1V, check again. If it's still low, check the resistance to ground.
Also check voltage on PL15, PL24, PU12 pin 6, PL19, PL23, PU8 pin 6, PU5 pin 6, PL22, PQ1 pin 4, PQ22 pin 1, PQ41 pin 1, PQ56 pin 1, PQ38 pin 4, PQ4 pin 4, PQ13 pin 1. Those are the other power rails you did not check I think. (not all of these measurements will be relevant but it doesn't hurt to check them anyway)

[Ndagger]

Re: Hp g6 r23 motherboard fault

I rechecked pu13 pin 6 and I measured 1.1v this time so I guess I didn't have good contact before.

Of the components you listed
Pl15 1.1v
Pl24 1.25v
Pu12 pin 6 2.52v
Pl19 0.9v
Pl23 1.13v
Pq41 pin1 5.07v
Pq56 pin1 1.53v
Pq38 pin4 3.3v
Pq4 pin4 3.36v
Pq13 pin1 3.36v

And
Pu8, pu5, pl22, pq1, pq22 were all bare solder pads with no components fitted, picture attached.
I also have attached the model of the motherboard incase the schematic is for a different board or is it normal to not have some components fitted? Ie is the schematic for several variants of board?

[piernov]

Re: Hp g6 r23 motherboard fault

The missing components are for the discrete GPU only. This is probably because you don't have a discrete GPU on this board.

Before anything, check with known good parts (CPU, RAM, HDD with an OS and good display), and make sure that the system is indeed not booting (HDD not being read for example). Also, with good parts, check that the caps lock key doesn't trigger the caps lock LED.

Anyway all the voltages are ok and the APU reset signal is deasserted, so if it still does not work, this points to either a bad BIOS or a dead FCH. You should try to flash the BIOS first. To check for a dead FCH, you can try to heat it up to 200°C for a couple of minutes (no more). If the board works afterwards, the FCH is dying and will need to be replaced. The heat may not change anything but this doesn't mean the FCH is ok.

[Ndagger]

Re: Hp g6 r23 motherboard fault

piernov you've been really helpful, thanks very much!
I was kind of thinking the FCH might be the culprit, nearly all google results on the g6 with this error sends you to YouTube videos of reflowing the chip, even though this board might be done for I've enjoyed tracing out the power rails and cross referencing with the schematic.
If you wouldn't mind one more question, you originally told.me to check APU_RST and then mentioned that the apu reset signal is deasserted, can you expand on that, what does this mean / how did you come to this conclusion?

[piernov]

Re: Hp g6 r23 motherboard fault

Reflowing the chip won't fix it, but you can at least try to heat it up with a heat gun. If you don't have one, you can cook the board in an oven if you've got nothing to loose. 6 minutes at 200°C max, with aluminium foil all over the board except on the chip itself, all components removed (especially the lithium cell battery). If it's not your board obviously don't do it. And remember that this is not a fix. But at least the computer may boot again for a short time.
It would be wise to try to flash the BIOS first if you can.

APU_RST# de-asserted means that the signal is at a high level (1.2V or something like that here) so this signal is not active anymore. When this signal is active (asserted, at a low level, 0V) the CPU is in reset mode, ie. it's not doing anything. When the signal is de-asserted (goes high), the CPU starts up at the reset vector and will begin processing BIOS' instructions.
APU_RST# is controlled by the FCH and is de-asserted when all the power rails are present and the FCH has finished its initialization.

On Intel platform, this is working in the same way, the CPU_RST# line is controlled by the PCH. On newer platforms when the PCH is integrated onto the CPU package, the CPU_RST# is burried into the package and can't be measured. In this case you can measure PLT_RST# which should come just before or at the same time as CPU_RST#