820-00165 Wont turn on, unless you perform a manual smc reset.

[Melamanti]

Customer donated this A1466 to our store after he had spilled something on it.
After some inspection, the machine had liquid damage on the lcd connector, the backlight fuse was burnt, and there was some liquid dmg on the keyboard as well.
Replaced keyboard, backlight fuse, led drive, lcd connector and lcd cable as it was burnt as well, and the machine booted up and worked perfectly.
It charges, it works with battery, in fact it's the one i'm using to write this post.
However, every time you shut it down, if you leave it aside for a couple of minutes, it wont turn on in any way unless you do a manual smc reset.
Any ideas?

 

[2informaticos]

Check if always starts on charger only (no battery).

 

[JohnB8812]

Inspect trackpad flex and trackpad for corrosion especially side of cable that plugs into the trackpad

 

[Melamanti]

It actually starts every time with no battery connected.
I have ordered a new trackpad and trackpad cable. will post results when i test it.

 

[2informaticos]

Disconnect trackpad flex and connect battery.
Solder an wire at SMC_ONOFF_L signal (R5115/16 pads) and use it as power button.
Use USB mouse and keyboard and see if it works correctly this way.

 

[Melamanti]

Have done that already, not with actual wires, with a pair of tweezers and it works normally.

 

[2informaticos]

Then do what John said before.
You may need to change keyboard, or trackpad flex.
Check SMC_ONOFF_L value when TP/KB is connected.

 

[Melamanti]

As i said in the description i have already replaced the keyboard.
And have ordered a new trackpad and trackpad cable. will test whey they arrive and will post results.
Still wonder how that trackpad might be damaged tho. Nothing broken bent or corroded under the microscope.

 

[2informaticos]

"Check SMC_ONOFF_L value when TP/KB is connected."
No comment yet...

 

[Melamanti]

Oh, sorry! My bad!
SMC_ONOFF_L is at 3.42V.

 

[Melamanti]

What i also noticed is that, now (after removing the trackpad so that i could take a close look) it does turn on, but i got no keyboard and trackpad..
Checked R4830 and it is indeed 0?. Is it that the trackpad is totally doomed now?

 

[2informaticos]

R4830 must be 0 ohms, can be replace by just a short.

Did you try to acces boot menu?
Confirm if TP/KB not working on OSX and before login too.
Check SMC_LID level at J4800.
Also post U4810.

 

[Melamanti]

Ok, the boot menu is not accessible by holding alt on power on so i got no keyboard on osx or before that.
SMC_LID on J4800 goes from 2.57v to 2.65.

Measurements on u4810:

Trackpad Connected : Trackpad not connected:
1. 3.34 1. 3.34
2. 0.002 2. 3.3
3. 0.494 3. 3.35
4. GND 4. GND
5. 3.34 5. 3.34
6. 1.84 - 1.93 6. 3.35
7. 0.437 7. 0.002
8. 3.34 8. 3.34

 

[2informaticos]

Disconnect trackpad flex and check again SMC_LID.
If still strange (it must be steady 3.3V), disconnect LIO flex and check it again.

Need U4810 voltage with TP flex connected; specify if measured logged in OSX please.

 

[Melamanti]

Measurements were done with the board extracted from the case due to the position of u4810, but with TP and TP cable connected (no keyboard).
i let it boot so the measurements were done in osx. Or at least at the point that it asks for a password.
I ll measure smc_lid with no tp cable connected.

 

[Melamanti]

SMC_LID with TP disconnected still at the same voltage.
With IO Flex disconnected 3.3 steady.

 

[2informaticos]

OK, seems to have a problem with LIO cable, or flex; maybe just hall sensor issue.
Test the machine now with TP/KB connected, but no LIO flex.
You must get 3V at pin 3/U4810 in OSX.
If file vault is enabled, on password screen should be 0V; still works on USB mode (3V at pin 7 then).

 

[Melamanti]

Well, this is awkward. I tested the machine with keyboard and trackpad connected, with no IO cable and still got nothing. Nothing on boot menu and nothing on password screen. With no IO cable and with TP/KB connected on pin 3 of u4810 i get 0.493 again.

 

[2informaticos]

Pin 3 should go high on OSX; and password screen (if no file vault enabled).
Pin 7 should stay low then, but must be high in boot menu; also high on password screen, when file vault is enabled.

Voltage of pins 3 & 7 is irrelevant if you don't post voltage for pins 1, 2, 5, 6 at the same moment.
Only then we see if U4810 receives correct signals and process them acordingly.

Be sure SMC_LID is 3.3V at J4800 and check continuity from MLB to TP board.

 

[Melamanti]

I got an update, replaced the charging/audio dock thing and the machine works perfectly.
It had some corroded stuff on it, thought it was better to just replace it.
So thank you so very much for the help.
You can mark this as solved.
Hopefully it will help others.