820-3437 Fan Spins, No Boot

[awbower]

The fan on this MBA does not go through the normal start, stop, start, stop, start sequence. Instead, it starts for about 10 seconds, stops briefly and then spins continuously.

All S_0 rails are present, but PPVVC_S0_CPU is a bit low (1.7V, should be 1.8V). PM_PCH_SYS_PWROK out of U1950 is present. The following have been replaced: SMC, clock chip (U1900), and BIOS chip. Still no boot after replacement. The following signals are all present: PM_DSW_PWRGD, PM_RSMRST_L, PM_SYSRST_L, PM_PWRBTN_L. From reading through previous threads, it seems like the next logical thing to do would be to upload a new good BIOS to the computer (I believe I will have to do this anyway since I replaced the BIOS chip). Could someone please outline the steps I need to take in order to do this as well as provide the file?

One other thing of note: U3900 has been mentioned in past posts. This chip does heat up noticeably when the board is powered on, but not to the point that I can't keep my finger on it.

 

[2informaticos]

First of all, welcome to the forum!

Do you get PLT_RESET_L?
Any history available, liquid sign?

Check all SPI Bus Series Termination resistors and their traces.
Be aware, there are 16 resistors (page 46).

Use this dump for test, as is.

820-3437clean me- tested ok.rar

drive.google.com

 

[awbower]

Many years ago I joined the Advanced Reworks forum. Louis helped me fix a few boards there. Glad to see you guys are still fighting the good fight. Every Macbook brought back to life is one new sale taken away from Apple.

This board had some liquid damage through the back near the BIOS chip. U1950 was destroyed. After replacing that I got PM_PCH_SYS_PWROK back.

PLT_RESET_L is not present.
I checked all 16 SPI Bus Series Termination resistors and the associated traces. Everything is ok there.

I extracted the .rar file and got a binary file. How do I upload that into the BIOS chip?

 

[awbower]

I read through some past threads and found that I need a CH341a programmer. I have ordered one.

 

[2informaticos]

If you deal with Apple repairs, get Medusa programmer for the future.
In case like this, don't need to remove the SPI chip; Medusa does the job on the board.
As you do not use it now, remove J6100 from the board; corrosion between its pins can block BIOS communication.

 

[awbower]

I got my CH341a programmer up and running. I just uploaded a new bios with clean ME region and the computer still will not boot. The computer has exactly the same symptoms as before... fan spins, stops and then just keeps spinning. No boot.

PLT_RESET_L is still missing. At the top of pg. 13 of the schematic (820-3437), it shows the signals going into the PCH that are required to obtain PLT_RESET_L. I have all of the required inputs. I decided to apply some flux to the PCH and heat. I kept heating for about 15 seconds after the flux started to bubble. After doing this the computer still would not boot.

The last thing I tried was wiring 3.3V to PLT_RESET_L to force it to be present (I first checked for a short to ground on the line... there wasn't one). After forcing 3.3V into PLT_RESET_L, the computer still will not boot.

Suggestions?

 

[2informaticos]

Bad idea...
PCH pulls down PLT_RESET_L because some reason.
Injecting 3V there could fried it.

Do you get VCCST_PWRGD?

 

[awbower]

VCCST_PWRGD is present

 

[2informaticos]

Then should be PCH/CPU issue...

 

[awbower]

That's what I was thinking...

After replacing the PCH/CPU does the bios need to be updated? Or should it just work automatically with the new chip?

 

[2informaticos]

It should work, but ME region must be cleaned.

 

[piernov]

CPU_VR_READY and LPC_PWRDWN_L present? PLT_RESET_L can be pulled low by a few other things, see top left corner of "Project Chipset Support" page.

 

[awbower]

Whenever I try to measure CPU_VR_READY the computer shuts off. This happens anywhere that I try to read that signal on the board... just putting my probe on the pad causes the computer to power down. I replaced U7200 to see if it was the problem, but nothing changed. LPC_PWRDWN_L is not present. Everything associated with U2071 on the "Project Chipset Support" page seems fine.