820-00045 Almost booting but not.

[realmacmods]

I have one of the awful 12" MacBooks that came in for repair. Customer puled the bottom off for a battery replacement and damaged the KB/Trackpad connector and cable. Everything besides KB and trackpad worked fine. I used it for a few days through a USB dock with USB KB/Mouse and HDMI monitor.
I removed the connector from the board and put a new one one. Ever since then i get no display internal or external, and no USB power/activity. I removed the connector and put it aside just to be sure its not the conenctor/internalKB/trackpad.

I have measured tons of points and everything i would normally check is coming back fine. CPU VCORE is normal, 3v3 AON is normal. CPU heats up as i would expect, it even responds to the power button pads on the board (grounding will power it off, grounding again will power it on. As evidenced by watching the wattage the charger is pulling from the wall).

The only thing i am finding missing so far is the 5V on l7850. The enable for this is missing. If i force the enable_l to a low state on q4635 that 5V will appear on l7850. With that 5V present the USBc dock still does not work for USB or HDMI. I imagine there is some muxing supposed to be happening that is not.

SDA and SCL on u4650 appear to working at least somewhat. I can watch data and clock happen with my scope on boot up. That said i have no way to know what data if any is being exchanged. It is bi-directional so maybe those lines are sending data but receiving no data back to set that enable low?

As far as i can tell all chips involved in the sequence are powered by 3v3 AON and are getting 3.3V to their VCC.

I know these things tend to have dead CPUs, but this one seems like a lot is actually going on for it to be a dead CPU.

Any advice appreciated, Data is important to owner otherwise i would replace the board and get it out of here.

 

[realmacmods]

P1v05s4SW and it's good are missing. If i force the e85HSMUX_USB_EN high on q4530, those are present.

Randomly e85HSMUX_USB_EN now remains high after multiple power on/off/unplug

 

[2informaticos]

L7850 delivers 5V only if an external USB device is detected; to power it on.
Missing its enable points to bad detection.

Something went wrong when you replaced the connector.
Probablly the heat affected CPU, or other sensitive circuit.

 

[realmacmods]

Thank you for the info. I agree, it must have been the heat. At this point the only real suspect thing ive determined is CPU_VR_READY is missing. u7200 has power via PPVIN_S0_CPUVR_VIN. CPU_VR_EN is present (CPU creates this, i believe) . VR_Ready is the PGood output from U7200 and is not present.
CPUVR_PWM1 and CPUVR_PWM2 seem to outputting something, it kind of looks like noise to me but it is some sort of pattern for sure. It does not look like a typical PWN square wave type of signal, but ive never looked at this signal before so im not sure what it should look like. Those PWMs make me want to believe u7200 is alive and working.

Any thoughts on what else i can check around u7200? It must not be happy about something to not create it's PGOOD.

CPU_VCCST_PWRGD is 1V
PPVC_CPU_S0 is 1.7v for a few seconds then settles to 1.5v and remains.

 

[2informaticos]

If you get correct CPU core voltage (R7310/20), then U7200 must release CPU_VR_READY signal.
Be sure R0850 is good.

 

[realmacmods]

is 1.7v, then 1.5v correct for CPU core voltage (measured at c1032)? I believe it is, but you know better than me. Am i correct that u7200 is responsible for that voltage change after the initial 1.7v?

r0850 measures 10K which matches schematic.

 

[2informaticos]

You can compare the CPU core with known good board.
As per schematic, it is in specs range.
Its voltage is dynamically controlled; BIOS, PCH/CPU and U7200 are involved there.

 

[realmacmods]

So it gets even stranger. It seems like VR_Ready may actually be present, but goes away and shuts down the board when i try to measure it. If i measure with a multimeter i never get anything, if i leave a scope on it while its booting VR_READY will go high for maybe 3 seconds, then go low, then board shuts down (as evidenced by the power meter between the charger and AC). If i try to measure while the board is running, as soon as i touch a probe to r0850, or the VR_ready test point on the board i can see the board switch from pulling around 7-11W to 0.8W and remain that way until i unplug and plug it back in.

So somehow the act of measuring the signal is causing it to go low, i guess?

In normal circumstances that VR_Ready signal should be constantly high, yes?

If i take r0850 out of circuit power stays at 0.8W and he board does not appear to power on. Other side of r0850 is around 1V. Which while would normally seem low to me for an enable, appears to be typical of other measurements i have seen posted.

I do not think this is the case, but lets assume somehow the LCD, LCD connector, or LCD Cable got damaged during disassembly/reassembly. Is there any signal that would be present while the board is fully running/booted/in the OS, but while the LCD is unplugged? To be clear, there is no evidence of damage to any of those parts, im just trying to prove the board is actually booting fully, with the next thought being maybe i can get it into Target Disk mode to get data somehow.

I am even getting the Enables high for BTPWR and WifiPWR, though it does not appear to be pulling a DHCP ip/lease from my server.
Not sure if i mentioned, but i also have 3.3V on ALLSYSPWRGOOD as well.

 

[2informaticos]

1V is normal for CPU_VR_EN; that CPU block is powered from 1V05_S0 rail.
CPU_VR_EN is also used as pull-up level for CPU_VR_READY; through R0850.

Even older boards turn off (sometime) when touch VR_READY with multimeter probe.
I saw this when there is problem with CPU core presence.
Should be stable I think, if you got steady CPU core (R7310/20) voltage.

You have missing internal/external image, nor even USB activity, or chime.
Board probably tries to boot and gets stuck; or doesn't boot at all, just complete power on sequence.
Check if PLT_RESET_L comes high.
Try known good BIOS, just in case; don't forget to make backup first.

 

[realmacmods]

PLT_RESET_L is 3.3v

For what it's worth, it responds to the power button on the keyboard. Holding "t" for target disk mode does not seem to work, as it does not show as connected to another MacBook pro. As well Caps Lock light does not respond. So i think it is safe to say it is not booting into the OS

 

[2informaticos]

If you get PLT_RESET_L, then power on sequence is completed.
No need to check CPU_VR_EN/READY, nor other previous signal.

You have a communication problem on some data bus, or BIOS issue.
Just in case, check the resistors from LPC bus.

 

[realmacmods]

i am waiting on an adapter for my SPI chip programmer for this footprint before i can read and write to it. It does seem like the board is interfacing with the chip at least. If i remove the chip the board will not really even attempt to boot. I replaced the chip with one from a 820-00244 (closest thing i had here right now), and the board will endlessly power cycle very early in the process. I put the original chip back on. Chip is getting power, and I can see clock and data happen on the lines, but of course i have no idea if the data is valid. Next thing will be backing up and over writing the current bios.

Which LPC resistors do you mean?

 

[2informaticos]

R1540-44, page 14.