820-00165 Pulsing. No Fan spin. Liquid damage near U7600 and the Bios chip

[phuketmymac]

Hello,

I have that 820-00165 which won't start however it is getting 2 pulses of the CPU_Vcore and ALL_sys_pwrgd.

Liquid damaged near the Bios and U7600.
I have reflowed the area and replaced the dirty looking resistors. U7600 was also replaced.
I have followed the SPI traces and everything seems to connect. I haven't checked the resistors values yet around the Bios (Note to self)
I was able to make a backup of the Bios 3 times with no errors.

PPBUS_G3H, PPVRTC_G3H and PP3V42 have a nice stable voltage.

SMC area looks fine, no liquid damage there. I haven't reflowed it yet.

I have tried plugging a battery but no cigar.

Any idea?

 

[aprendiz]

Corrosion under J6100??, corrosion around U1900?? PPVRTC_G3H present???

 

[dukefawks]

Termination resistors and/or traces to U6100.

 

[phuketmymac]

Hello guys,

Well I did some more work and ended up replacing both U7600 and its transistor.
Now I can see the fan shaking when the charger is plugged.

I did spend hours trying to measure each individual resistors surrounding the Bios + checking the traces/pads but everything seem to connect.

Only the ones going under the SMC or the CPU can't be trusted. Shall I start removing the SMC to test?

To you, that issue has to be related with the Bios not being seen by the CPU?

 

[dukefawks]

Is CPU Vcore pulsing now? Can you post a HiRes picture of BIOS area?

 

[phuketmymac]

Hello Duke,

Yes it is pulsing. I even have the time to see it showing up on the multimeter.
It is so frustrating, you feel so close to have it work.

I will post a picture later on. Thanks Duke!

 

[phuketmymac]

No Title

Hello Duke,

Here are some shots. Pretty tough to tell. Shitty camera.

 

[smiba]

The bios and clock chip area look abused

Can you try and see if the balls on U6101 look alright? If they look merged or messed up reball or replace it.

If you dumped the BIOS by using J6100 you can ignore this, but if you didn't make sure the BIOS (U6100) has proper connection to the board by cleaning up the loads of flux surrounding it and manually inspecting the solder joints. If its messed up desolder it, put some solder on the pads and solder it back on again.
You said you already checked the bios traces and termination resistors so I assume that part of the circuit is alright. Make sure the traces that do no longer have their coating on them are not shorting to other traces though

Inspect the clock chip if it looks questionable or the solder joints aren't that great you can either touch them up or just replace the chip to be sure (Its just about $2).

Maybe duke has some other tips as well

 

[dukefawks]

That just looks like everything was messed with. Can't trust any of those traces/resistors without 100% verifying all of them have the correct value and continuity to the next part, what a mess!

 

[phuketmymac]

Thanks for answering both!

Well after digging again on that one, I realized that it may not be a problem near the Bios chip but probably on the PP1V05_S0 line.
I guess a possible short could make the the Fan quarter spin right?
Since you can see on the pictures that line also had liquid coming onto it (near U7600), I've checked the voltage drop to ground and it seems low compared to a known good working board.

Working board has a 1V drop to ground whereas this damaged board has .4 drop to ground.

I can assume it is not a cap since it is not totally shorted right?!
Most likely another line connected with a resistor to that line is shorted, don't you think?
Or maybe the CPU...

I've removed the inductor L7630 and saw it was on the system side, not the creation side, even though I have replaced U7600 and Q7630.

I do hope the CPU is not shorted...

Looking forward hearing some of your thoughts.

 

[smiba]

Diode mode is a bad way for comparing the resistance to ground on low resistance lines like PP1V05_S0. (Its ok for finding power lines or shorts to inspect though, but when comparing boards do it in ohm).

A semi-short can cause a quarter fan spin in rare situations.

Anyways what is the resistance to ground in Ohms on PP1V05_S0? Keep in mind this is going into the CPU so it will probably be just a couple of Ohm.

 

[phuketmymac]

Hello Smiba,

Thanks for your precious help.
Bad board has a 186ohms resistance to ground whereas the known good board a 300ohms resistance to ground.

EDIT: By the way, I have replaced the clock chip with a new one. same behavior.
I have removed the bios chip and U6101 which I have both replaced. Same behavior.

I have followed every traces from the bios area.
The only traces I can't confirm are the ones going to the CPU and SMC.

I also use the diode mode near the bios chip and U6101 and the only difference I get is on SPI_MLBROM_CS_L which is different between the known good board and the bad one.

 

[smiba]

186ohms to ground is alright, its probably not the issue.
(We're looking at 0.003675W vs 0.005927W)

You replaced the BIOS chip? Well now we have a whole different issue on our hands because you can't just swap them over like that! Please place back the original one

When you made this topic, did you had the bios replaced already? Or did you only do that later on?

 

[phuketmymac]

I did backup the Bios first before messing with it.

Now that I have replaced the chip, I have also upload the image on the new chip.

At the time I've started the thread, Bios chip was the original one. Only the U6101 has been replaced with a new one but I assume they are not programmable, am I right?

 

[smiba]

The replacement bios, did it come from another Macbook or was it a new one? If it's new, did you set the quad enable bit?

See if replacing the clock chip U1900 resolves your issues.

 

[dukefawks]

What are diode measurement on SPI_MLBROM_CS_L? You can only compare it to a working board that has the exact same type of BIOS chips as diode readings will vary between different brands.

 

[phuketmymac]

Hello Duke,

Known good board: (MXIC chip)

- Diode mode: .625V
- Resistance to ground: 7M ohms

Faulty board: (MXIC chip)

- Diode mode: .54V
- Resistance to ground: 7.3M ohms

 

[dukefawks]

There are only 2 components on this line so should not be that hard to figure out. But really they are kind of close and probably it is fine. I have no more ideas for this one besides still an issue with the U6100 data lines.

 

[phuketmymac]

Ok thanks for your help Duke!

 

[phuketmymac]

The replacement bios, did it come from another Macbook or was it a new one? If it's new, did you set the quad enable bit?

See if replacing the clock chip U1900 resolves your issues.

Hi Smiba,

Can you explain how to set up that quad enable bit please?