submerged Macbook A2141 - no power on, no voltages

[Sanevent]

I tried to inject voltage to PPBUS_G3H Yesterday. As recommended i started with 1v and set a limit of 0.5a at first.

With 1v injected nothing could be observerd. My lab bench power supply was showing that no power was consumed at all. At best i get a reading of 0.001a on my PSU but most of the times it stayed 0.

I soldered 2 wires to one of the big electrolytic capacitors that connect to PPBUS_G3H as well as to GND.

My next step would be to step up the voltage until i can see something, but the small/absent current draw made me unsure if i should.

Because if we had a short, or at least a partial one, at least a bit of current should be drawn.

I tested my solder joints in continuity mode, and they are good. I tested from the end of wire to a different exposed point on the board.

 

[2informaticos]

"set a limit of 0.5a at first"
Where I told you to set such limit???
The board will take the current dictated by the low resistance to ground.
It will be low anyway, but don't set a limit!
If lab PSU will detect an overcurrent, wil get blocked and doesn't inject anything; unless knows to work in constant current mode.

Arriving to 5V it should take 0.5A more or less.
That may be insufficient to find a hot spot with the fingers; but easy detected by IR camera.
With 10V the current should be around 1A, or more; 10W should be found even with the fingers.

I hope you didn't forget to check continuity between PPBUS_G3H and any coil.
More than 2V can kill PCH, CPU, or GPU.

 

[Sanevent]

Sorry for the confusion, let me clarify. Of course I'm using my PSU in CV mode - at first set fixed to 1v.
I just need to set an upper limit for the amps, because my PSU can only detect overcurrent when I set that limit, because mine does not do it by itself.
I do not use a CC PSU - I just set the ampere limit to something that felt reasonable for me.

Is there general advice on how to set the current limit correctly?

Before proceeding today, I will check every coil again and then crank up the voltage step by step.
I do have a Thermal camera, so maybe I can find something.

 

[2informaticos]

Set always amp limit to max, when injecting voltage.
If you adjust the voltage, can't adjust the current at same time.
The current depends by the value of the resistance to ground.

We don't have any interest to reach the PSU amp limit.
If resistance is low, high current can be achieved with low voltage.
If current limit is too low, PSU will get blocked and doesn't inject the desired voltage.

BTW, a good lab PSU for voltage injection, should support 20A.

 

[Sanevent]

Ok. I tried to inject some voltage today and was met with a very stange behaviour of the board.
I tried to raise the voltage - at first to 2, then 3 and finally 5 v but I got the same result.
The first few moments ( not even a second) the board draws some current, but then it falls back to 0. Flat 0.

As I tried with 5 v I got 0.077a but only for a brief moment before going back to 0.000a.

I did not try to go further, as I was not sure if this behaviour is expected or not.
I thought this was rather strange, because I would expect the current draw to be constant. However this is not the case.

 

[2informaticos]

What is actual resistance there (to ground)?

 

[Sanevent]

Strangely, the resistance starts around 200 ohms and raises towards 3 mohms.

 

[2informaticos]

Wasn't 12 ohm???

 

[Sanevent]

Im probably as confused as you are. Before injecting voltage it was 15 ohms. Now it starts with 200 ohms increasing to 3mohms.

The only thing i changed is that I soldered 2 wires to the board to inject the voltage and i removed the F8400 fuse before doing so.

After removing the fuse (and before i injected voltage the first time) i checked both pads with the multimeter - because i was curious if that changed something.
The PPBUS_G3H side was 15 ohms and the other side was n.c..

Then i injected voltage as recommended and now the situation is as I described in my last post.

By the way, the new Chip has arrived. I can replace it as soon as im back at the lab.

 

[2informaticos]

No reason to inject voltage at all, if no low resistance to ground.

Always allow 10-15s before checking in diode mode (or ohm scale), since power was removed.
Remanent voltage changes real behavior.

 

[Sanevent]

When I came into the lab this morning, the first thing I did was to measure again. The board was without power since Friday.
I checked in diode mode, and when I put the red lead on PPBUS and the black to GND, I get a value of 0.00 and rising well above 1.1 and when I reverse the probes (after waiting a few minutes), I get 0.400 as a reading.

My best guess is, that the short burned itself free somehow and is no longer present - at least this is what I could imagine. I don't know how to verify my theory though.

Is there anything I could check before bringing the fuses and the IC back to make sure that the short, which magically disappeared, will cause no problem if I restore the state of the board?

 

[2informaticos]

"I put the red lead on PPBUS and the black to GND"
Almost not used this configuration.

0.400 is a good reading.
Put back the components and test the board.

 

[Sanevent]

I put everything back, including all fuses and the IC. I measured again in diode mode from PPBUS to GND, and the value has not changed. It still is 0.400, which I suppose is good. Can I now safely plug in the charger and test for 20v, or is there anything else that I need to perform first?

 

[2informaticos]

Is good for test now.