820-4924 - Did I fry the PCH?

sbergin

New member
820-4924 with corrosion around U3220 and U3320, no Magsafe LED or boot. I replaced those ICs and got the unit to boot fine, but without keyboard or trackpad. While testing, I did not have the fan attached. Before I could start troubleshooting the KB/TP issue, I had to go take what I thought was going to be a short phone call. I forgot to disconnect from power and the board ran fanless for over 90 minutes. When I came back, it was pulling about 10ma and missing all S4 and lower rails including PM_SLP_S4_L. The behavior was mimicking the symptoms in this other thread, so I followed the same troubleshooting steps... https://boards.rossmanngroup.com/threads/820-4924-missing-pm_slp_s4_l.57094/

Bottom line is that it appears to be a dead PCH. I haven't tried heating the PCH with flux applied, not sure how to do that without affecting the CPU in a negative way.

Did I fry this thing by running it without a fan for 90+ minutes? I would have assumed that the CPU would have throttled down according to heat sensor input.

Don't feel too badly for me, this is a board from an eBay machine I bought to practice on, so I'll just consider it a learning experience if it ends up being a dead PCH.
 
Last edited:

2informaticos

Administrator
Staff member
"When I came back, it was pulling about 10ma and missing all S4 and lower rails including PM_SLP_S4_L."
Isn't this the actual state of your machine???
Post basic signals of power on sequence.
Do you get 3V3_S5/SUS at least?

Any reason to quote an entire post visible few centimeters above?
Please, maintain forum aesthetic...
 

sbergin

New member
Apologies about the re-quoting, I just left the default text in place when I hit the Reply button.

Yes my current state is 10ma and no S4 or lower rails. Your initial reply seemed to only be addressing my initial problem with MagSafe, included for context, and you made the jump to that not being a PCH issue. I'm not trying to be a smart ass here, I'm just trying to make sure I'm clear on everything.

My question is simple: Could I have fried the PCH by leaving the board powered on with no fan for 90+ minutes? I don't want to troubleshoot further if that is the case.
 

2informaticos

Administrator
Staff member
Is possible, but bord should tunr off if CPU overtemp is triggered.
You should now check the signals for power on sequence, to make an idea about actual PCH state.

You can also check resistance to ground on CPU rails and compare with good board.
That can help too.
 

piernov

Moderator
Staff member
No, PCH would not die in this case. It's not even cooled by the heatsink.

CPU could overheat but it'll throttle to maintain temperature below 105°C. If the heatsink is attached properly it, throttling should be enough to keep the CPU below 105°C. If the heatsink is not attached, CPU can still run for several minutes, but it may end up overheating and shutting down to prevent damage. After letting it cool down it should turn on again without issue.
It's not impossible that the CPU died as a result of overheating, but still very unlikely, MacBooks are designed to operate close to max temperature. It's not unusual to see the CPU at 100°C under load, and it'll easily stay at 105°C or a couple of degrees above until fans ramp up. But it's not that common to have a plain dead CPU (except on A1534).

That said, PCH could still have died randomly, but it's more likely to be something else related to the initial damage.
 
Top