MacBook Pro 15" A1286 (820-2532-A) no chime no video

[JackM]

I am trying to repair an old MacBook Pro which I acquired that does not boot. I know that most people on this forum either don't use Apple laptops or likely haven't tried repairing one before, but I figured I'd try asking here since most other Apple repair communities online require a paid subscription to get help or post on their forum boards... Even just trying to find a schematic and boardview (component locations) files for Apple products is a real pain.

Specifications:
MacBook Pro 15" late 2008
Model A1286
Logic board part number 820-2532-A
https://support.apple.com/kb/SP499?viewlocale=en_US&locale=en_US

Symptoms: no chime, no video, white front power LED lit, green magsafe LED lit, DVD drive spins up, battery power indicator working.

I did a visual inspection of the logic board but I could not find any evidence of liquid damage or any obvious corrosion or damage. To be honest the logic board looks quite pristine for its age.

To power on the board without the battery installed, I do the following (not sure what this procedure is called):

• Disconnect magsafe power cable (no battery either)
• Hold down the power button
• Attach magsafe power cable, continuing the hold down power button
• Continue holding power button for another 5-8 seconds or so
• Release power button then immediately after press the power button once
• Both fans spin up to full speed

I measured the following voltage rails and important nets. Battery and HDD were removed before any measurements were made.

Magsafe power connected, not powered on:
 PPBUS_G3H at 12.25V
 PPBUS_CPU_IMVP_ISNS at 12.25V
 PPDCIN_G3H at 17.12V
 PP3V42_G3H at 3.43V
 PP3V3_S5 at 3.324V
 PP1V2R1V05_S5 at 1.104V

When powered on (S0 state):
 PPDCIN_G3H at 16.76V
 PPVCORE_S0_CPU at 0.988V
 PPVCORE_GPU_REG at 1.005V
 ALL_SYS_PWRGD at 3.30V
 S0_PWR_PGOOD at 3.283V
 CPU_PWRGD at 1.059V
 PP1V2R1V05_ENET at 1.098V
 SMC_RESET_L at 3.429V
 PP5V_S3 at 5.020V
 PP5V_S0 at 5.120V
 PP3V3_S3 at 3.287V
 PP3V3_S0 at 3.278V
 PP2V5_S0 at 2.514V
 PP1V2_S0 at 1.209V
 PP1V8_S0 at 1.815V
 PP5V_S0_CPUVTTS0 at 5.11V
 PPCPUVTT_S0 at 1.059V
 PM_SLP_S4_L at 3.287V


From what I can tell every single one of the voltage rails and other signal nets that I measured are all good. I don't know much more about the Apple laptop system architecture to debug much further (never even used one before). I believe that the SMC and CPU are getting power based on what the voltage rails show, but with no chime after power on means something must still be wrong. I think that no video is a consequence there being something wrong with some other part of the system, mainly because I still don't hear a chime. Connecting an external monitor to the micro DisplayPort also doesn't show any video output either.

Is anyone familiar with MacBook Pro repairs and could tell me what the next things to check for are? I don't think that power delivery is the problem. Also if you know of any other online communities for MacBook repairs that doesn't require a paid subscription please let me know!

[kripton2035]

try to change the ram modules for some known good one. you need 1067 MHz ram modules.
no chime means : cannot check ram.
3 chimes means : bad ram
if you remove the rams, you should hear the 3 chimes.
also remove the HDD which can be bad too.

[KhronX]

Late 2008, you say? Odds are the GPU crapped out, if everything else seems in order.

https://semiaccurate.com/2010/07/11/why-nvidias-chips-are-defective/

[james_s]

My friend had one with a similar symptom, he had someone resolder one of the BGAs and that fixed it. Apparently it's a fairly common issue.

[KhronX]

Plain reflowing is, at best, a band-aid (and nowhere near a cure). Just FYI 

[james_s]

I think the guy who fixed his removed and re-balled the chip, I don't remember though as it was several years ago. Heating it can provide a clue as to whether that's actually the problem.

[JackM]

If this really is an issue with the GPU being dead, then it's not even worth repairing and I might as well try to sell it for parts or something. How can I confirm that the GPU is dead? Is there a procedure for that?

try to change the ram modules for some known good one. you need 1067 MHz ram modules.
no chime means : cannot check ram.
3 chimes means : bad ram
if you remove the rams, you should hear the 3 chimes.
also remove the HDD which can be bad too.

With no RAM installed, I still get no chimes when I power on the board.

[texaspyro]

Have you tried booting it up with the PRAM reset keys pressed?

[JackM]

Have you tried booting it up with the PRAM reset keys pressed?

One of the first few things I tried was doing an NVRAM reset. Before it didn't seem to do anything; no response chimes and nothing on screen.

I just tried it again and this time after holding down Command+Option+P+R for about 25 seconds straight and then releasing, I got an all white screen! This is the first time I've ever seen anything other than a black screen. I can't seem to replicate it again though...

[ar0311]

Check out Louis Rossmann's Youtube channel.

[kripton2035]

If this really is an issue with the GPU being dead, then it's not even worth repairing and I might as well try to sell it for parts or something. How can I confirm that the GPU is dead? Is there a procedure for that?

try to change the ram modules for some known good one. you need 1067 MHz ram modules.
no chime means : cannot check ram.
3 chimes means : bad ram
if you remove the rams, you should hear the 3 chimes.
also remove the HDD which can be bad too.

With no RAM installed, I still get no chimes when I power on the board.

then the cpu seems unable to test the rams. you should search for ram power rails ?

for the gpu, when it fails on these 2008 models, most of the times you can't even power on the machine. at best you switch on and it switches off by itself 5 to 10 seconds later. black screens with bad gpu happens on later models 2011+

[KhronX]

It's not the balls between the GPU substrate and the mainboard that fail - it's the even tinier ones (called "bumps") between the small silicon die and the green substrate that fail (due to them being high-lead, and the crap underfill nVidia chose). No reballing's gonna cure that.

I think the guy who fixed his removed and re-balled the chip, I don't remember though as it was several years ago. Heating it can provide a clue as to whether that's actually the problem.

[kripton2035]

well performing a light reflow of the gpu is something simple to try.
even if it is something that won't last more than some months.
just pretty sure it won't cure the problem on these 2008 models.

[Rasz]

NVIDIA GeForce 9600M = sell it "for parts" making it some other idiots problem

[wraper]

You should not "reflow" it right away. Heat GPU to around 120-150oC and see if it starts working. If not, the issue, most likely, is somewhere else. If it starts working, you can be sure that GPU is faulty and there is not much to loose. Also I don't recall about which model it was but Louis Rossmann was talking about some faulty ceramic cap under the GPU causing the same issue. BTW google about that cap first before heating anything, heating shorted MLCC together with GPU may fix it for some short time as well. Therefore you may have wrong impression that GPU is faulty.

[Rasz]

C7771, but he already measured PPVCORE_GPU_REG and its 1V

[wraper]

C7771, but he already measured PPVCORE_GPU_REG and its 1V

I was talking about ceramic capacitor, not tantalum. Likely it was different motherboard. Although it's still useful to check C7771 first.

[jc101]

I have a Modal A1286 sitting in a cupboard, the machine itself was fine but the display had de-laminated, caused by someone constantly picking it up by the corner of the screen.  Worked OK hooked up to an external display last time I powered it on.

Time to dig it out and see what state it's in, then possibly sell it on for parts...

[breadbox]

I don't think this is related to the original poster's problem, but someone mentioned a tantalum capacitor and someone else mentioned Louis Rossmann, which reminded me that Rossmann describes a GPU problem related to a failing tantalum capacitor (C9560) in a 2010 Macbook Pro here:
https://www.youtube.com/watch?v=DzcgT_fiVTA&feature=youtu.be
My 2010 MBP (with NVIDIA GeForce GT 330M) had off-and-on periods of GPU panics for about nine months. Otherwise the machine booted and operated fine. After replacing the cap in Sep 2017 I have not had a single GPU panic.

[mbless]

PPBUS_G3H at 12.25V

From what I remember PPBUS is supposed to be 12.6V and anything else indicates a problem. You'd have to dig through Rossmann's videos and website to find more info.

Edit:
It appears I remembered correctly. Here are some threads discussing low PPBUS.
8203115-board-no-green
820-3115-b-no-green-light-ppbus_g3h-12-3v
solved-820-2915-no-backlight

[JackM]

Hmm I'll have to check PPBUS_G3H again and some of those resistors mentioned in the threads that mbless linked above.

I really hope it's not a bad/dead SMC. I certainly won't be trying to fix it if that's is indeed the case. Although it would explain why I get no chimes (even with no RAM), no video, no backlight or much else.

[JackM]

Super late update, but I just wanted to say that I ended up taking the Macbook to a local electronics repair guy who specializes in Apple products. He identified it as a problem with the pin contacts on the GPU BGA chip, which he said was a common problem with this model of Macbook. He had his own BGA reflow station which he used to reflow the GPU and after that it was able to boot up again.

[frozenfrogz]

Thanks for the update. I hope, that it actually was "just" a case of broken solder balls.
Not wanting to be a grinch, but from personal experience I can tell you, those NVidia GPUs are a pain in the butt. Reflowing seems to somehow realign parts of the internal structure, but odds are it is going to die for good anyhow.

There is a guy, that found a way to completely disable the NVidia crap, this involves flashing custom firmware to a mux-chip with on the board though.
-> In case your GPU is going to die again.

[wraper]

Super late update, but I just wanted to say that I ended up taking the Macbook to a local electronics repair guy who specializes in Apple products. He identified it as a problem with the pin contacts on the GPU BGA chip, which he said was a common problem with this model of Macbook. He had his own BGA reflow station which he used to reflow the GPU and after that it was able to boot up again.

Will fail again soon, wasted money. That guy ripped you off by frying dead ship. There are maybe a few percent maximum of those GPUs which actually have soldering problem, the rest have problems with solder bumps inside the chip itself.

He identified it as a problem with the pin contacts on the GPU BGA chip, which he said was a common problem with this model of Macbook.

Common problem is dead GPU, not soldering. The only real repair is replacing with the chip with date code not earlier than 2009.

[Rasz]

He had his own BGA reflow station which he used to reflow the GPU

fraud, unless you got 12 month warranty with the "repair"