SMC Issue on MBP 13 - Hardware Troubleshooting Guide

[Wickeido]

Hello,

first of all, this thread is the most valuable information source for mac logic board repairing. I've learned a lot. Thank you all!.

Excuse my english, I'll try my best to explain myself.

I've bought some days ago a water damaged MBP 13" Mid 2010 model (A1278). First thing I did was to open it and clean all the corrosion, using alcohol andcontact cleaners.

It was not bad, altough the corrosion has broken 3 pins in the screen connector. I should be able to fix it anyway.

So I connected the board to the MagSafe, and nothing happened. Tried SMC reset procedures, and welding the G3Hot points, to force it to start. Nothing, only thing I get is a very dim green light on the MagSafe.

Following this thread, I suppose there are 3 main issues with the logic board:

1. The dual P Channel mosfet.

http://www.vishay.com/docs/68934/si7149dp.pdf

In the one that is near the DC IN socket, I get 17,25v on the drain side, and 16,25v on the 3 pins in the other side. The 4th pin, that is the gate fluctuates (not sure if it's normal, I guess yes). Seems a normal behavior for that component.

On the other side of the logic board, the exact mosfet only outputs 16,25v on the source pins. The drain has 0v.

Seeing that this second mosfet had a bit of corrosion beneath, I decided to lift both of them to check the pad they were welded to.

So... I unwelded both mosfets, and I only get 17,25 v on the DC IN side, not in the other. I'm not sure if this is normal, but I think that as it's the same pad, only mirrored in the other side of the board, should have also 17,25v.

Having the mosfets outside the board, I checked them.

In one, I got connection (open circuit it's called?) between drain and the 3 source pins, and no continuity in the gate. I beliebe it's what is should look like.

In the other one I don't get connection anywhere. I guess is defective.

2. Fuses.

The fuse near DC IN acts normal, have 17,25v in both sides. The fuse in the other side, near battery conector, not getting anything, 0v on both sides. It was the same before unwelding the mosfets. I tried connection between both sides of the fuse, and got nothing.

My next action will be getting a new mosfet, and finding a solution for the pad that doesn't get 17,25v. Also change the battery fuse.

Now... is what I'm saying right? Did I made any mistakes I should take care of?

I would apretiate any guidance or tips on how to follow without making more damage to the board.

Thank you very much! Oh and if there is someone who needs to take a look at the schematics of the 1278 board, below you can find a link to download it.


http://www.4shared.com/office/ynRxJWF4/A1278-820-2879.html

 

[cmdrdata]

So i can tie pin 5-8 together and wire it to the fuse?

Yes, the FIREWIRE fuse, however I don't know the physical location of it. In my late 2009 MBP unibody it is F4260, a 1.1A rated fuse. There two other fuse with higher current rating: these are the Magsafe input fuse for the raw DCIN 16.5-18v source protection, and the another one near the battery connector that protects the 12v PPBUS_G3H bus. Do not connect to any of these two fuses.

 

[cmdrdata]

I've bought some days ago a water damaged MBP 13" Mid 2010 model (A1278). First thing I did was to open it and clean all the corrosion, using alcohol andcontact cleaners.

1. The dual P Channel mosfet.

In the one that is near the DC IN socket, I get 17,25v on the drain side, and 16,25v on the 3 pins in the other side. The 4th pin, that is the gate fluctuates (not sure if it's normal, I guess yes). Seems a normal behavior for that component.

> Yes this is normal MOSFET behavior. The pulsing gate signal is also normal. I believe this is functioning just fine.

On the other side of the logic board, the exact mosfet only outputs 16,25v on the source pins. The drain has 0v.

Seeing that this second mosfet had a bit of corrosion beneath, I decided to lift both of them to check the pad they were welded to.

So... I unwelded both mosfets, and I only get 17,25 v on the DC IN side, not in the other. I'm not sure if this is normal, but I think that as it's the same pad, only mirrored in the other side of the board, should have also 17,25v.

Having the mosfets outside the board, I checked them.

In one, I got connection (open circuit it's called?) between drain and the 3 source pins, and no continuity in the gate. I beliebe it's what is should look like.

In the other one I don't get connection anywhere. I guess is defective.

> MOSFET unpowered or with no gate signal (no pulse or no DC threshold voltage should have very high resistance between DRAIN and SOURCE. If you measure zero or low resistance then I believe the MOSFET is bad. You said you got a connection (off the board) between drain and source so I think it is a bad/shorted chip.

2. Fuses.

The fuse near DC IN acts normal, have 17,25v in both sides. The fuse in the other side, near battery conector, not getting anything, 0v on both sides. It was the same before unwelding the mosfets. I tried connection between both sides of the fuse, and got nothing.

My next action will be getting a new mosfet, and finding a solution for the pad that doesn't get 17,25v. Also change the battery fuse.

Now... is what I'm saying right? Did I made any mistakes I should take care of?

I would apretiate any guidance or tips on how to follow without making more damage to the board.

Thank you very much! Oh and if there is someone who needs to take a look at the schematics of the 1278 board, below you can find a link to download it.

> Thanks for the schematic. Until you get 12v on the "input" side of the fuse by the battery, you will not get 12v needed for charging the battery or PPBUS_G3H to power the system. Somewhere between the DCIN fuse and the PPBUS_G3H fuse you have one or more bad MOSFET or gate signaling that failed to turn on the power MOSFET(s).

 

[longcoo817]

Dadioh do you where exactly where this fuse may be?

 

[cmdrdata]

To posters in this forum

I believe some of you are not from the USA, and I am curious where you are from. When you sign-up for macrumors account or post a question, do you mind updating your user account with a country info? thanks.

 

[Dadioh]

Dadioh do you where exactly where this fuse may be?

Are you referring to the firewire fuse linked a few posts back? I have never needed to work on that so no sure. I imagine it is near the firewire port. With the schematic in hand you should be able to track it down with a multimeter.

 

[longcoo817]

Location updated !
cmdrdata where exactly do you stay maybe we can get together so you can teach me a few here and there

 

[torontomac]

17" Unibody

I have 2 17" mid 2010 unibody units
i have one thats beyond repair and another that has a bad firmware
when you turn it on its gives u 9 beeps 3 short 3 long 3 short
i know its NOT A MEMORY PROBLEM 100%
i swapped memory over 10 times so I know with known good sets
I also looked on apple website and its seems they dont have a
firmware restoration cd with efi's for any unibody models
if anyone knows which chip controls the efi firmware so I can swap the two
Thanks!

 

[cmdrdata]

I have 2 17" mid 2010 unibody units
i have one thats beyond repair and another that has a bad firmware
when you turn it on its gives u 9 beeps 3 short 3 long 3 short
i know its NOT A MEMORY PROBLEM 100%
i swapped memory over 10 times so I know with known good sets
I also looked on apple website and its seems they dont have a
firmware restoration cd with efi's for any unibody models
if anyone knows which chip controls the efi firmware so I can swap the two
Thanks!

I googled the SOS code and it seemed to me that firmware issue was only applicable to "older" Macs. have you tried swapping the HD and see if that help?

 

[torontomac]

yes i have tired with and without a working hard drive with lion installed
some people have left the machines on and beeping and it finally boots for them but I wasnt able to get that on my machine
If i put in no ram i get the standard 3 beeps so i believe the board is good
just the efi firmware, do you which chipset on the boards stores that info?

 

[weluvmacs]

Pull up resistor locations?

SUCCESS

After hours and hours of teadious searching and measuring resistances all over this damn logic board I found 2 resistors together one was 1k ohm and the other was open circuit and Voila ! Once replaced the battery status lights worked and the battery started charging so it was one of the pull up resistors that went open that caused it not to charge the battery. I noticed when trying to trace these suckers from the connectors and charging chip ISL 6258A that the two tracks always ran together so I started to look for 2 1k ohm resistors together anywhere on the board.
The schematic really helps because you can literally see if the tracks and components match what you are finding.

I have attached the original picture (which I cloned from an earlier response in Dadioh's thread) where I asked Dadioh what was the component that I had seen light up like a Christmas tree and he correctly identified it as the overvoltage circuit so I have attached it again here adding Q6915 and U6915 arrows and labels which we also replaced.

I am semi retired and enjoy messing around with these things on the side and got all my components from dead boards I have collected over previous failed "projects".
I will take a picture of the location of the two pull up resistors and post it - probably wont help much as it was probably a one-off issue but it took me so long to find them that if it helps one person then it will be worth it!

Thanks to all for the help and on to the next project.

I've been away from forum for awhile and have been using my MBP2011 with just the magsafe adapter, but the above "success" caught my interest as I am having similar battery issues....was there ever another post showing the location of the 2 1k pull-up resistors?

 

[Dadioh]

Not staying shut down

Wanted to see if anyone has some thoughts on this situation. I picked up a badly liquid damaged unit whose logic board was completely unrecoverable and in fact the liquid had even seeped into the display itself and ruined the LED panel. I replaced the logic board with one that I had already fixed and everything seemed fine except for the fact that as soon as I connected the magsafe the unit would boot up, not waiting for me to press the power button. Also, when selecting "shut down" from the apple menu it would shut down but then immediately restart.

So I put a second logic board in there and the behavior remains suggesting it is not the logic board. I thought maybe the power switch is a little intermittent so I took the cover off the back and inspected it. There was corrosion on the backplate that covers the switch but the switch itself seems like a sealed unit and is part of the keyboard itself so repair doesn't seem like an option.

So watcha think? Keyboard/power switch replacement? The only thing I could think of was intermittent power switch. Or is there some other possibility I am overlooking? Pulling the keyboard is a major PITA given the 4000 tiny screws required

 

[peislander]

Wanted to see if anyone has some thoughts on this situation. I picked up a badly liquid damaged unit whose logic board was completely unrecoverable and in fact the liquid had even seeped into the display itself and ruined the LED panel. I replaced the logic board with one that I had already fixed and everything seemed fine except for the fact that as soon as I connected the magsafe the unit would boot up, not waiting for me to press the power button. Also, when selecting "shut down" from the apple menu it would shut down but then immediately restart.

So I put a second logic board in there and the behavior remains suggesting it is not the logic board. I thought maybe the power switch is a little intermittent so I took the cover off the back and inspected it. There was corrosion on the backplate that covers the switch but the switch itself seems like a sealed unit and is part of the keyboard itself so repair doesn't seem like an option.

So watcha think? Keyboard/power switch replacement? The only thing I could think of was intermittent power switch. Or is there some other possibility I am overlooking? Pulling the keyboard is a major PITA given the 4000 tiny screws required

What happens if you unplug the keyboard & put power to it?Does it fire up?Try power up with g3 hot & see if you can shut down normal.This will tell you if its the keyboard or something else.Keep unplugging things one at a time till you find it.good luck

 

[Dadioh]

What happens if you unplug the keyboard & put power to it?Does it fire up?Try power up with g3 hot & see if you can shut down normal.This will tell you if its the keyboard or something else.Keep unplugging things one at a time till you find it.good luck

Excellent point. I will give that a shot tonight. Thanks!

 

[cmdrdata]

Excellent point. I will give that a shot tonight. Thanks!

Haha, sometimes the obvious in fault isolation don't even come to mind. I've done that many times. Good advice from Peislander.

 

[Dadioh]

Haha, sometimes the obvious in fault isolation don't even come to mind. I've done that many times. Good advice from Peislander.

Yup. Sometimes I let my brain get in the way of common sense

 

[cmdrdata]

yes i have tired with and without a working hard drive with lion installed
some people have left the machines on and beeping and it finally boots for them but I wasnt able to get that on my machine
If i put in no ram i get the standard 3 beeps so i believe the board is good
just the efi firmware, do you which chipset on the boards stores that info?

Looking at a couple of MB/MBP schematics, the boot ROM is called SPI ROM (a 32 Mbit ROM) and in their page 2 of the schematic, the CPU gets its boot code and perhaps POST commands from the chip shown. Both MB and MBP uses identical boot ROM chip. The chip is a 16-pin, MX25L3205, a serial data eFlash memory. Location wise, I haven't got a clue.

 

[torontomac]

Looking at a couple of MB/MBP schematics, the boot ROM is called SPI ROM (a 32 Mbit ROM) and in their page 2 of the schematic, the CPU gets its boot code and perhaps POST commands from the chip shown. Both MB and MBP uses identical boot ROM chip. The chip is a 16-pin, MX25L3205, a serial data eFlash memory. Location wise, I haven't got a clue.

thank you so much
thats the chip!
i found it on my boards
will post after swap
wish me luck
thanks again

 

[mini998]

hi I have a similar issue ,hope someone can help me

!. System was completely dead ,when i connect the charger it turns green and battery was showing full charge on indicator .

2.Left it for couple of weeks , now the battery is totally dead but still charger turns bright green when connected

I measured 3.4 volts at G3Hot and 12 V at the fuse near battery .

There were no known liquid damages but it was bought an year ago from ebay.
Is it worth trying changing the battery?

hope someone can help me
cheers

 

[cmdrdata]

hi I have a similar issue ,hope someone can help me

!. System was completely dead ,when i connect the charger it turns green and battery was showing full charge on indicator .

2.Left it for couple of weeks , now the battery is totally dead but still charger turns bright green when connected

I measured 3.4 volts at G3Hot and 12 V at the fuse near battery .

There were no known liquid damages but it was bought an year ago from ebay.
Is it worth trying changing the battery?

hope someone can help me
cheers

Yes, first thing to try is a KNOWN good battery. When a battery is "old" and the electronic inside the battery enters a shutdown mode (0 volt), it may not be communicating with the SMC/CPU, thus the CPU will think it is not present. The green light tells the SMC/CPU that it is able to communicate with the magsafe and abler to pass power to the system (the 12v on fuse mentioned is the proof). With the battery removed, does the system power on from magsafe? Does it have a successful Power On Self Test (POST) chime? Also try re-seating the SODIMMs.

 

[mini998]

Yes, first thing to try is a KNOWN good battery. When a battery is "old" and the electronic inside the battery enters a shutdown mode (0 volt), it may not be communicating with the SMC/CPU, thus the CPU will think it is not present. The green light tells the SMC/CPU that it is able to communicate with the magsafe and abler to pass power to the system (the 12v on fuse mentioned is the proof). With the battery removed, does the system power on from magsafe? Does it have a successful Power On Self Test (POST) chime? Also try re-seating the SODIMMs.

Thanks

It does not boot up even with the battery disconnected

There are no Power On Self Test chimes ,it is completely dead ,no sounds ,no lights or fans. although I replaced two memory modules about 6 weeks ago ,i tried reseating them but still no luck

I took the logic board out and it does not have any corrosion on it,

 

[cmdrdata]

Thanks
It does not boot up even with the battery disconnected
There are no Power On Self Test chimes ,it is completely dead ,no sounds ,no lights or fans. although I replaced two memory modules about 6 weeks ago ,i tried reseating them but still no luck
I took the logic board out and it does not have any corrosion on it,

OK, I am confused. First pls provide model number and estimated production info if available. Second, I just noticed you said eBay about a year ago. Did it ever worked or you bought already broken?

With battery off and SODIMM of did you get a POST chime or beep indicating no memory? Did you try try the old SODIMM?

 

[mini998]

OK, I am confused. First pls provide model number and estimated production info if available. Second, I just noticed you said eBay about a year ago. Did it ever worked or you bought already broken?

With battery off and SODIMM of did you get a POST chime or beep indicating no memory? Did you try try the old SODIMM?

I'm not in front of computer but from top of my head it was mid 2009 and model number was A1178 , no it doesn't make any noise even with battery and ram removed , I didn't try my old ram . Shouldn't it make some noise without ram? So I guessed ram wasn't the problem

 

[cmdrdata]

I'm not in front of computer but from top of my head it was mid 2009 and model number was A1178 , no it doesn't make any noise even with battery and ram removed , I didn't try my old ram . Shouldn't it make some noise without ram? So I guessed ram wasn't the problem

You didn't answer the system state after eBay purchase.
On the RAM here''s what I expect should happen:
The power on self-test resides in the ROM of the computer. This test runs whenever the computer is turned on after being fully shut down (the power-on self-test does not run if the computer is only restarted).

If a fault is detected during the test, you will not hear a normal startup chime. Instead, the system will beep as explained below. If you experience one of these beeps, you may call your Apple Authorized Service Provider for additional troubleshooting assistance.

1 beep = no RAM installed
2 beeps = incompatible RAM types
3 beeps = no good banks
4 beeps = no good boot images in the boot ROM (and/or bad sys config block)
5 beeps = processor is not usable

In addition to the beeps, on some computers the power LED will flash a corresponding number of times plus one. The LED will repeat the sequence after approximately a 5-second pause. The tones are only played once.

Note: In this case, a flash is considered to be 1/4 second or 250 ms or greater in length.

 

[mini998]

You didn't answer the system state after eBay purchase.
On the RAM here''s what I expect should happen:
The power on self-test resides in the ROM of the computer. This test runs whenever the computer is turned on after being fully shut down (the power-on self-test does not run if the computer is only restarted).

If a fault is detected during the test, you will not hear a normal startup chime. Instead, the system will beep as explained below. If you experience one of these beeps, you may call your Apple Authorized Service Provider for additional troubleshooting assistance.

1 beep = no RAM installed
2 beeps = incompatible RAM types
3 beeps = no good banks
4 beeps = no good boot images in the boot ROM (and/or bad sys config block)
5 beeps = processor is not usable

In addition to the beeps, on some computers the power LED will flash a corresponding number of times plus one. The LED will repeat the sequence after approximately a 5-second pause. The tones are only played once.

Note: In this case, a flash is considered to be 1/4 second or 250 ms or greater in length.

Thanks ,but as I mentioned earlier there are no beeps ,no chimes .its dead as a dodo . no matter what I do it doesn't show any life .

I bought it from ebay about 1 year ago ,it worked perfectly till now . I just checked the model number... its A1278