820-00165 possible PP3V42_G3H short in SMC

EsteTron

Member
820-00165 came in with no sign of damage or liquid damage with no power and no light on MagSafe Charger.
PPDCIN_G3H = 16.5V
PP3V42_G3H = 0.3V.
Diode mode reading full short on PP3V42_G3H.
Applied IPA on underside of board, injected 1.8V 1A power on PP3V42_G3H and noticed the SMC (U5000) was heating up first. Specifically heating up in 1 specific spot on the SMC.

Is it safe to say that there is an internal short in the SMC?
 

2informaticos

Administrator
Staff member
You can increase injected voltage untill 3.5V and get higher current.
In fact, if you get 3V/2A (as example) should be good enough to detect if only SMC heats.
 

JohnB8812

New member
Inspect super closely to make sure it?s not a cap by the SMC. But yeah very likely the case. Remove it and see if short clears. Also make SURE PPBUS and PPVRTC are not also short once you remove the SMC to avoid wasting time. Sometimes I?ve seen the SMC shorting cause all of them to short to ground and kill the CPU. Not often, but it can happen. I would assume if the SMC is short this will be an easy one.
 

EsteTron

Member
Thank you for the replies. I injected 3V 2A and only could find SMC heating up. I then removed the SMC and confirmed short is now gone. PPBUS and PPVRTC are not shorted with SMC removed, so it looks like my culprit is the SMC. I have never replaced one before, but just to make sure, all I really need is a stencil and a donor board to pull one from, correct?

Thanks again for all the advice.
 

2informaticos

Administrator
Staff member
Sometime can be soldered even without reballing it; re-using rest of solder on the board pads.
However, I do NOT recommend this method.

I use 0.35mm to reball and sometime I placed them one by one, at hand (aprox 150).
Remember to take it form donor board, same model.
 

JohnB8812

New member
Yep reball as 2informaticos said. Must come from 820-00165 donor board. I?d order 2-3 just to make sure you get a good SMC for the repair
 

EsteTron

Member
It took awhile to get an SMC stencil and 2 donor boards, but I went ahead and replaced the SMC, get green ling on magsafe and all S5 rails are present, but now device only turns in SMC bypass mode. All rails are present in SMC bypass mode. Does this mean the replacement SMC is a dud? If so, I have another one I can reball.
 

2informaticos

Administrator
Staff member
If start in SMC bypass, you must check all SMC sensors first.
Could be bad SMC also, but discard the sensors issue first.
Do you get correct AVREF_SMC?
 

EsteTron

Member
I replaced the SMC again with another one from a donor board before I saw your replay 2informaticos. The board is still doing the same thing, so I am looking into the sensors, now. I have 3.3V on AVREF_SMC.
 

EsteTron

Member
All the SMC sensors appear to be present. I plugged board into LVDS port and performed an SMC bypass, I get 1.8 on CPU VCORE, but no post. I am starting to think this may be a dead CPU.
 

EsteTron

Member
I get fan spin and all power rails present. It originally slipped my mind to hook up board to a LVDS and see if it even posts. When i installed board back into macbook housing, and device has fan spin that restarts every 10 seconds. it appears it doesn't post or chime in SMC bypass mode. I've yet to experience a dead CPU, but from all the research in the forums, this sounds like all the symptoms of a dead CPU.
 

2informaticos

Administrator
Staff member
Very possibly dead CPU, but you can try to discard other options.
Now the board turns on in normal mode, as in SMC bypass too?
Did you install correct SMC (same donor board)?
Try new BIOS, just in case...
 

EsteTron

Member
This is true. The board turns on in both normal and SMC bypass mode. I an smc from the exact same board. Twice just to be sure I didn't get a dud SMC. When installing a new bios, do I have to pair it to the ME first before testing, or can I swap one from the donor board just to see if it even posts?
 
Top