820-00426 Shuts down during boot

[SMMRepair]

I'm working on an 820-00426 board, and it's giving me tons of headaches, especially since there's no schematic for it yet. I'm doing my best with the 3787/00138 schematics, but it's tough.

The symptoms of this machine are that it restarts itself during boot with certain components in place. I'll put things in bullet format to make it easier to follow:

- Machine had liquid near U7501, and U7100. Replaced ~4 ugly caps and R7151/7152.
- Machine will start and boot fully 100% of the time without the WiFI card, SSD, or keyboard in place. Once either of those components is connected, it restarts during boot about 20 seconds after the Apple/loading bar appears.
- Thinking something was up with U7501 (since it had some liquid nearby, and that the WiFi card, SSD and keyboard all run off of 3v3 lines), I replaced it. Same result.

Not sure what else I can do to check things without a proper schematic and board view. They don't appear to be released yet, so I'm stuck here. Again, the board powers on and loads OSX fully (via USB) if the SSD, WiFi card and keyboard are all unplugged. If I plug in any of those 3 devices, it will cut out during boot. Should I try replacing Q7520/Q7560 (or rather, their equivalent on the 00426 board)?

Any ideas? All help appreciated.

 

[jason_microsoldering]

Plug in a charged battery and use an 85w charger.

 

[SMMRepair]

Hey Jason, yes, we tried that. Tried it in a known-good test assembly (with good battery, that charges normally, etc), and we've been using an 85W charger. The battery isn't 100% full, but it's in the 70s. If I let it charge a while it would be at 100% I'm sure.

 

[SMMRepair]

I wanted to correct some information in the OP--the board actually restarts (cuts off then powers on again). I wanted to distinguish that from it cutting off and not turning on again, etc.

 

[SMMRepair]

Right before leaving for the day, I was booting the computer up without SSD, WiFi or keyboard attached, and it cut off once again. It may not be related to the SSD/WiFi/Keyboard after all. I'm going to do some more testing this weekend and will update/revise the thread if needed. Spent too much time on this one today already.

Edit: Looks like it may be the trackpad after all.

 

[peterfixit]

is there ripple on the 3v3 line, maybe look at the 3v3 s0 power supply and caps

 

[jason_microsoldering]

Probably dead cpu.

 

[SMMRepair]

No Title

OK, I have an update on this thing. It's weird.

I decided to connect everything and then run verbose mode to see where the machine is cutting off. It looked to be cutting off with the trackpad for some reason (see attached image). I removed the trackpad and booted again in verbose mode, and it shut off immediately after some text about the HD not being verifiable completely, and then "syncing disks...Killing all processes" and it cut off. So I remove the SSD and replace with a known-good SSD, and low and behold it booted to OSX (both to external USB and to SSD).

It's weird, because the unit boots with JUST the trackpad connected, but as soon as anything else is connected, it shuts off/restarts. I thought this indicated an issue with something other than the trackpad, but didn't make the connection that it was the trackpad causing conflictions until this morning. Pair that with the SSD and battery both turning out to be bad, and it caused a confusing hell storm.

I've come to conclude this machine had a bad SSD, trackpad and eventually found the battery to be bad as well ("replace now" with only 18 cycles). However, before I confidently tell the customer they need to shell out a few hundred bucks + labor costs to have the assembly replaced (bad battery + trackpad = might as well replace the entire assembly), can someone please confirm that the attached verbose text is saying the trackpad (or cable--I don't have a cable to test, but the connectors both look flawless, as well as the cable) is causing the shut-down issues? I'm not very familiar with the new forcetouch trackpads, but it looks like it's killing the boot for whatever reason.

Thanks in advance.

 

[jason_microsoldering]

Something something troubleshooting with known good parts before assuming its a board issue... You need to test it with a known good battery and trackpad still.

 

[SMMRepair]

Yeah, definitely. We didn't have a 2015 top-case assembly. so we have a 2015 LCD paired with a 2014 assembly, and just didn't plug in the trackpad when testing it. Since originally it seemed like the trackpad was fine, we didn't think that was an issue. Now that we know otherwise, we definitely dropped the ball on that point. We've ordered a test unit for 2015 now (so damn expensive, even for one in C grade), and will get it all worked out next week.