(347) 552-2258 [email protected]

820-4924-A No Trackpad or Keyboard

A

Alex Thayer

Member
I've got a 820-4924-A with trackpad click, but no trackpad movement, keyboard response, or caps lock light. The power button functions and can trigger SMC bypass mode, which does not help the situation, but the keyboard is otherwise unresponsive to startup combinations and the device will not enter recovery mode. The logic board shows no signs of liquid damage, but is a bit "crummy and dusty", I'm currently wondering about crap under the SMC but unsure if the lack of improvement in bypass mode rules this out. The keyboard connector looks good, and no pins are loose. An external mouse / keyboard functions just fine.

I am aware of the frequent trackpad flex failures on this model, A new top case, keyboard, trackpad, and trackpad flex assembly did not change anything.

I've seen several instances of this exact board and problem set over the years and often come up empty handed despite testing and replacing everything I could think of, and read some threads here with the exact same symptoms but no resolution. Any ideas as to what might be behind this issue?
 
2informaticos

2informaticos

Administrator
Staff member
I bet you have trackpad related problem; this also explains non-responsive keyboard.

Check J4802 power rails first; 1, 25, 27.
Also check pins 20, 28, 38 level.
 
A

Alex Thayer

Member
Everything is disconnected from the board except the keyboard and magsafe, the power rails look good, I'm not totally sure what to expect from the later three pins.

1 12.5V
25 5V
27 3.3V
20 0V
28 0V
38 3.3V
 
2informaticos

2informaticos

Administrator
Staff member
Keyboard will NOT work, if trackpad is not good.
I was thinking you understood that.
You need to fix the trackpad problem.

Looks like bad SMC_LID.
Check R5171 and R5250.
Also inspect Hall sensor (J5250) for corrosion.
 
A

Alex Thayer

Member
Yes, I'm aware of that, just wanted to provide you with the current configuration at the time of taking measurements.

I didn't catch that pin 28 was SMC_LID and put two and two together yesterday, the board was in the device with the screen closed, checking again today with the board out of the top case gave 3.3V for pin 28, sorry about that.

Pin 20, TPAD_VBUS_EN, remains unchanged at 0V.
 
2informaticos

2informaticos

Administrator
Staff member
Pin 20 must have 3V before enter to macOS; pin 38 0V at same time.
Once logged in, values must chnage; you've probably measured with SSD inserted.

Of course, nonsense to measure without trackpad connected; keyboard not needed yet.
 
A

Alex Thayer

Member
Okay, I provided the system configuration I was using because I wasn't sure what was appropriate for taking these measurements. It sounds like you'd like the SSD removed and the trackpad to be the only peripheral connected to the board, so I've repeated the measurements with that configuration, they are unchanged from my last message.

1 12.5V
25 5V
27 3.3V
20 0V
28 3.3V
38 3.3V

I attempted to troubleshoot Pin 20 TPAD_VBUS_EN by tracing it back through the U4910 AND gate, and found PM_SLP_S4_L at 3.3V and TPAD_USB_IF_EN at 0V, also checking all of the resistors on those lines. TPAD_USB_IF_EN appears to come right from the PCH, where should I look next? It seems like we want the PCH to be providing TPAD_USB_IF_EN and not TPAD_SPI_IF_EN in this configuration if I understand correctly, but are getting the opposite.
 
2informaticos

2informaticos

Administrator
Staff member
Trackpad works in USB mode before macOS loads its SPI driver.
Change U4910, just in case.
Also check all its solders, traces and adjacent resistors.
Be sure R1618 is good.

Try to login to macOS using USB TP/KB and check if pins 20/38 maintain the same voltages.
Remember to keep internal trackpad connected, for measurments.
 
A

Alex Thayer

Member
No dice on U4910, and the IC is functioning properly if I force both of the inputs high.
Zero visual issues around U4910 or R1618, checked for shorts, resistor values, and continuity all around both but came up empty handed.
The voltages on pins 20/38 do not change.
 
2informaticos

2informaticos

Administrator
Staff member
I found only one time bad PCH in similar case.
Seems to be your turn now.

I know Apple has an update for trackpad issue for some machine model.
Check if your model is listed.
Just in case, try known good BIOS.
 
A

Alex Thayer

Member
Thanks for your guidance on this one, as the board has an overall "crummy" look to it, after testing with a new BIOS, is there any point is trying to heat up the pch a bit to burn out crap under it? Or more likely to cause other issues than help?
 
You must log in or register to reply here.
Top