SMC Issue on MBP 13 - Hardware Troubleshooting Guide

[chigwelldave]

Thanks Chigwelldave for the hint. Did you mean ALL_SYS_PWRGD signal? That will be difficult to catch, because looking from schematics it seems Q7950 produces ALL_SYS_PWRGD which in turn is dependent on PP3V3_S0. In my case, PP3V3_S0 drops to 0.10V within 1 second.

I tried to measure ALL_SYS_PWRGD within that brief duration and it was around 3.36V at pin 4 of Q7950. However, dropped to 0V the next second.

Not sure how to trace it. Has anybody faced this problem here earlier?

View attachment 475644

Hi,

Sounds good. After that, the signal is combined with 5 more power good signals to make "ALL_SYS_PWRGD" (see the Power Block page). Once the SMC sees this, it enables the CPU voltage regulator U7400, if these rails come up then the PCH will be happy and booting can progress. If any of these rails fail to come up within the time allowed, the SMC will abort and switch everything off again. This is why you only see the voltages for a second or so.

Regards,

Chigwelldave.

 

[musicloverdenon]

Hi,

Sounds good. After that, the signal is combined with 5 more power good signals to make "ALL_SYS_PWRGD" (see the Power Block page). Once the SMC sees this, it enables the CPU voltage regulator U7400, if these rails come up then the PCH will be happy and booting can progress. If any of these rails fail to come up within the time allowed, the SMC will abort and switch everything off again. This is why you only see the voltages for a second or so.

Regards,

Chigwelldave.

Okay, so I looked at the power rails on page 73 of K90i schematics and most of the measurements looks fine. Exception is PP3V3_S0 and consequently ALL_SYS_PWRGD drops to around 0V within 1-2 seconds.

Also, S4_PGOOD at pin 4 of U7930 measures 0.22V with power supply and then goes down to ~0.07V. I think it should go down to ~0.01V to enable the boot.

I am yet to take few more measurements.

On another thought, is it possible that this shutdown within 1-2 seconds is being caused because of over-voltage on U3890. As mentioned in my earlier post, I am getting 11.96V or 12.64V on all the components after Q3880 depending on whether board is powered from battery or magsafe. That means my U3890 is overfed. On a working board these components measure 0V and goes upto 0.26V on boot. However, on the non-working board, it's always around 12V.

Pls see the pic below. voltage of around 12V upto Q3880 (blue arrows) is fine and matches with a working board. But all the voltages after Q3880 is 0V on a working board and around 12V (red arrows) on this board.



Earlier, with same measurement, the board was not shutting down itself, fan used to spin continuously and heatsink was getting extremely hot in very short duration.

Now fan stops within 1-2 seconds and board is cool. May be I activated certain censor after cleaning the board which was not working earlier.

Also, with this measurement, I am not sure if Q3880 is bad or U3890. Understanding the schematics properly is bit challenging for a Finance guy like me. One thing is sure, if these components are part of the whole power cycle in any manner, I won't be able to do much troubleshoot without replacing the faulty component.

 

[musicloverdenon]

On further reading of schematics, it seems U3890 regulates voltage supply to display. So with Q3880 or U3890 bad, board may not have display. Which was the original problem when I had received the board.
Not sure if this will impact boot process though.

 

[chigwelldave]

On further reading of schematics, it seems U3890 regulates voltage supply to display. So with Q3880 or U3890 bad, board may not have display. Which was the original problem when I had received the board.
Not sure if this will impact boot process though.

Hi,

The board should boot without any display - I often use a USB flash drive to boot from so I can see activity. You're not getting that far though as the SMC is not happy with something. Try checking your main power rails with an oscilloscope as it will react quicker than a meter. You could have a sensor problem rather than a rail down as there are voltage and current sensors for measuring the main supplies.

Regards,

Chigwelldave.

 

[musicloverdenon]

Yes, I think you are right about booting without display.
I have a hunch feeling that still there is some sort somewhere.
While testing few components, I got continuity on one of the pins of an IC with battery connected but no grounding without battery. Cross-checked with working board and it doesn't ground with or without battery there. Just forgot which IC it was, will have to re-locate.

On a side note, how do you guys keep track of measurements taken. In my case, I realised that I have been measuring quite a few components multiple times. Bit lost on the schematics.
Testing with Oscilloscope is a good idea only if my skills allowed that. With my current skill sets, holding a oscilloscope would be like coconut in monkey's hand
I am going back to read the whole thread. May be some hints somewhere.

 

[Sandy Santra]

I am very sad that this thread is now over 3 years old and closing in on 3000 posts. This thread would have been nipped in the bud early on if Apple had paid attention and/or truly cared about this subgroup of MBP buyers/owners and addressed what is obviously an ongoing issue affecting a significantly consistent percentage of MBP 13" machines. The fact that the thread has been reduced to discussing voltages and circuit boards on what are arguably middle- to top-of-the-line laptops made by a company whose quality control was close to the best until a view years ago underscores the dilemma commenters here face: how to obtain, after repeated exchanges, one of these machines without any significant defects. This should not be part of the equation for any consumer that lays out over 1000 USD for a product.

After closely watching (at times) and consistently returning to check up on this and other similar MBP threads during the past few years, I decided it was still too risky to buy an MBP and chose instead to buy an iMac last month. Even though the iMac may be refreshed to a 4000K retina build later this year, I am glad that I chose an iMac instead of an MBP.

My sincerest condolences to all those who purchased defective MBPs during the last three years who have been unable to yet obtain satisfaction with regard to this issue.

My 2007 MBP lasted like a champ through 3 HD upgrades and only just recently became enfeebled. It will still boot to a vanilla Mountain Lion, however, and even play a DVD.

My new iMac is pretty much perfect. Apple is still in the running for decent computers...I just wish someone (Tim? you there? anyone?) would step up to the plate and address the FACT that MBP build quality has degraded to something like a 3% failure rate as of this last refresh in 2013.

Good luck, MBP owners who are burdened with unsolved/unaddressed MBP defects. Wishing you the best, and here's to hoping Apple returns to a consistently high-quality build in the next MBP refresh and that you have the funds to replace your current (if defective) MBP with a non-defective one.

 

[musicloverdenon]

I am going through the schematics again and again to understand it better. Have a query.

Few components are marked as "CRITICAL" on schematics. What does that mean? Critical for boot sequence or critical for normal device operation?
Also, what does "NOSTUFF" signifies on schematic.

 

[chigwelldave]

I am going through the schematics again and again to understand it better. Have a query.

Few components are marked as "CRITICAL" on schematics. What does that mean? Critical for boot sequence or critical for normal device operation?
Also, what does "NOSTUFF" signifies on schematic.

Hi,

As far as I'm aware, CRITICAL means the part cannot be substituted (from design stage to production).

Do you get a logic high from U7980 (ALL_SYS_PWRGD) into the SMC? If so, do you then get your CPU supplies PPVCORE_S0_CPU and AXG?

Regards,

Chigwelldave.

 

[musicloverdenon]

Hi Chigwelldave,

Thanks for the explanation and fast response.
I did notice that I get ALL_SYS_PWRGD for around half a second. It measured around 3.36V I think and went down to 0V.

I am yet to take more measurements but finding it difficult without anything to compare with. So, I went back to old discussions and trying to gather various measurements for working boards that were discussed earlier.

I don't want to open my working MBP again and again for comparing measurement. Else I'll loose a working laptop if I short something accidentally. Anyways, I am novice and learning my way into this.

It's fun.

I haven't checked the second signal yet. Today I went through the power sequence document I found here and realised, I need to take lot more measurements. Bit held up with month closure accounting and reporting, so I'll take some time to do it.

I really appreciate you taking time to understand the problem and trying to guide a way out of this problem. Thanks a lot.

 

[musicloverdenon]

Hi,

As far as I'm aware, CRITICAL means the part cannot be substituted (from design stage to production).

Do you get a logic high from U7980 (ALL_SYS_PWRGD) into the SMC? If so, do you then get your CPU supplies PPVCORE_S0_CPU and AXG?

Regards,

Chigwelldave.

I thought I'll start documenting the measurements from now onwards. So that I don't end-up measuring same components again and again. It should work as a reference for others as well if somebody gets into similar problem.

So, I started with power rails and will go upto boot sequence measuring PPVCORE etc as suggested by you. Could take very limited measurement today. You were right about late reaction of multimeter. Most of the times I don't get the complete measurement where the voltage remains for a fraction of second. I see something between 1.5V to around 3V on my multimeter on places where I expect voltage to be upwards of 3V and the next moment it comes down to 0V. An Oscilloscope might be needed.

One strange thing. At few places like pin 4 of U7941 I see between 0.7V to 0.11V when only Magsafe is connected without any actual power supply. If I remove magsafe, reading becomes 0V.

Attached below are some of the initial readings where I could not find anything wrong. Measurement in red is with magsafe and in blue color is with battery.

U7200:


S5 RAIL ENABLES& PGOOD:


CPUVCORE ENABLE:

 

[musicloverdenon]

On another thought, so far our assumption has been that PP3V3_S0 is getting dropped from 3.36V to 0.10V because SMC is not happy with something. However, that may not be correct. If SMC is not happy with something, it'll not generate further pwrgood or boot signals which go into booting sequence. However, it should not cause to drop voltages which are already generated before the SMC boot check.

Many users have reported on this forum about no boot with fan still spinning. This corroborates my thinking. My MBP's fan should keep spinning even with problems on further boot sequences.

So it looks like I may have two different issues to take care of. First finding out why PP3V3_S0 is not stable and correct it, and once it's stable, finding out what SMC is not happy with to initiate the boot sequence.

On further digging, PP3V3_S0 gets generated at Q7830, which in turn depends on PP3V3_S5 (stable at 3.36V) and PM_SLP_S3_R_L (drops to 0.10V within a second). If I trace this line back, it depends on PM_SLP_S3_L signal which gets generated at U1800.

I took out the LB today and gave much closure look for any short in PM_SLP_S3_R_L line. It's getting dropped everywhere on the board, so the problem seems to be at source itself. Does it indicate a shorted graphic card?

Also, during this check, I found out that C1756 and connected lines are grounded. So, this may explain the issue with boot sequence.

One more thing, while examining certain components at full 20X zoom under the scope, I found few utensil balls on the board. Some 8-9 balls overall, either stuck to some capacitor or resistor at both the places behind U1000 and U1000.

Does this sound like chances of getting this board back is bleak? Is it possible to reball processor and graphic card if those balls actually belong to them? Any sort of re-balling is certainly beyond my skill level. But I can find somebody if it's at all possible.

 

[pabs13]

SMC troubleshooting

Could somebody explain how I would go about checking for any sort of damage to my SMC on a MBP 13 mid 2009? Any help would be great. Hope this is the right place to post this.

Thanks in advance

 

[Yako76]

Dear Friends,
I have a Problem with a Board of Macbook Air 13 2010.
The Board goes on, chime comes but no external display. It get Signal but stay black. I have no Internal Display, I don't know if the problem is just on external TFT.
Has somebody a Idea wich Problem this is?

I have Schematic and BV.

Thank a lot in Advance for your Answers!

 

[South657]

Fets

Hi all I am trying to obtain Renesas RJK03E1DNS I am in the uk and have looked around and so far no good is there an update part No.to this chip?

Regards

South657

 

[dellxps15]

Hi all I am trying to obtain Renesas RJK03E1DNS I am in the uk and have looked around and so far no good is there an update part No.to this chip?

Regards

South657

e-best on ebay have almost all that little component. express ship is about 18$

i have to make another order with him for some macbook i5 repair..... yes i got another macbook i5 not working. someone tried to repair it but lost many component on the board.... i think is worth repair it right ?

here is the i5 board picture with damage:


for me at least at first looking it need 11 repair:

1° that missing capacimeter
2° fuse
3° diode
4° capacimeter
5° capacimeter
6° missing 0402 capacimeter and resistor
7° check if some component is misisng here.
8° bad ic and need to be replaced tps51019 code
9° bad capacimeter and need replace of both
10° check if something else is bad here

and 11 on the top side i have some component to clean (doesnt look bad )

http://i62.tinypic.com/s1rqk1.jpg
http://i59.tinypic.com/x3aiao.jpg

it's ok to try to repair it right ?

PS: cant understand why all pad are not brillant (without thin) and go like black color.... already passed ipa on them.... maybe some solder paste to clear them properly ?

 

[South657]

Chips

e-best on ebay have almost all that little component. express ship is about 18$

i have to make another order with him for some macbook i5 repair..... yes i got another macbook i5 not working. someone tried to repair it but lost many component on the board.... i think is worth repair it right ?

here is the i5 board picture with damage:


for me at least at first looking it need 11 repair:

1° that missing capacimeter
2° fuse
3° diode
4° capacimeter
5° capacimeter
6° missing 0402 capacimeter and resistor
7° check if some component is misisng here.
8° bad ic and need to be replaced tps51019 code
9° bad capacimeter and need replace of both
10° check if something else is bad here

and 11 on the top side i have some component to clean (doesnt look bad )

http://i62.tinypic.com/s1rqk1.jpg
http://i59.tinypic.com/x3aiao.jpg

it's ok to try to repair it right ?

PS: cant understand why all pad are not brillant (without thin) and go like black color.... already passed ipa on them.... maybe some solder paste to clear them properly ?

Thanks for getting back to me, i have had ago on ebay no luck? I will say ebay is better then most for chips but not in this case I have 4 on boards but not to sure
how to test them after they are lifted to see if they are any good?

Thanks and kind regards

South 657

 

[ZZZAC]

820-3115 baord confusing me badd

Hi guys i have an 820-3115 board here.
Bassicly the 16.5v DCin voltage fluctuates between 16v and 12v and no matter what i have tried nothing sorted it
Ive tried new Q7080 and new Q7085 but this didnt change, New U7000, Tested almost all the resistors around the U7000's area. Tested almost all the capacitors around the area.

I noticed lastnight when i have a charging IC installed i can hear a faint ticking coming from somewhere. My only guess is this is the source of the problem or why the voltage is fluctuating.


any thoughts would be apriciated

 

[ZZZAC]

Hi all I am trying to obtain Renesas RJK03E1DNS I am in the uk and have looked around and so far no good is there an update part No.to this chip?

Regards

South657

there 20 quid/lot on alliexpress.
Only place i have found them

 

[South657]

Fets

there 20 quid/lot on alliexpress.
Only place i have found them

Hi ZZZAC,

Thanks have just orderd.

Is that on pin 2 U7000 Magsafe only? Are you getting 12V at the battery pins?

Regards

South657

 

[South657]

Thanks for getting back to me, i have had ago on ebay no luck? I will say ebay is better then most for chips but not in this case I have 4 on boards but not to sure
how to test them after they are lifted to see if they are any good?

Thanks and kind regards

South 657[/QUOTE

 

[ZZZAC]

Hi ZZZAC,

Thanks have just orderd.

Is that on pin 2 U7000 Magsafe only? Are you getting 12V at the battery pins?

Regards

South657

Yeah Pin two is DCIN.
tried everything in my reach. Almost every capacitor, resistor and IC in the charge discharg circuit. broken the circuit at the fuse to check resistance. NO short to ground, but at certain points the resistance fluctuates as itdoes with the voltage, one moment it will be OL the next itll be 0ohm and it will keep switching in quick succession.

probably going to give up on it soon and sell thee fella another macbook

 

[South657]

Yeah Pin two is DCIN.
tried everything in my reach. Almost every capacitor, resistor and IC in the charge discharg circuit. broken the circuit at the fuse to check resistance. NO short to ground, but at certain points the resistance fluctuates as itdoes with the voltage, one moment it will be OL the next itll be 0ohm and it will keep switching in quick succession.

probably going to give up on it soon and sell thee fella another macbook

My thinking would be D7005 and before? Bad track on magsafe side only?
DC in board? Is green light flickering?

Regards

South657

 

[ZZZAC]

My thinking would be D7005 and before? Bad track on magsafe side only?
DC in board? Is green light flickering?

Regards

South657

got about 10 DC in magsafe connectors. Getting no green light atall, usually whenever ive seen similar its the inrush protection fets or something simple.
I was starting to wander if it could be related to a smc signal being given, so i pulled apart my g3hot rail to ensure the SMC isnt doing anything it shouldn't.

Still no change. going to have another go at the board today i should thing. Re measure resistors, and pull anything i am yet to pull up and test.

Because its a simple charging circuit issues im sure i can find it!!!
plus i have an older core 2 duo board im working on that does something very simalar and if i can get them both working....

 

[musicloverdenon]

got about 10 DC in magsafe connectors. Getting no green light atall, usually whenever ive seen similar its the inrush protection fets or something simple.
I was starting to wander if it could be related to a smc signal being given, so i pulled apart my g3hot rail to ensure the SMC isnt doing anything it shouldn't.

Still no change. going to have another go at the board today i should thing. Re measure resistors, and pull anything i am yet to pull up and test.

Because its a simple charging circuit issues im sure i can find it!!!
plus i have an older core 2 duo board im working on that does something very simalar and if i can get them both working....

If G3HOT line is broken, you'll anyway not get green light. Even if everything else is fine.

 

[ZZZAC]

If G3HOT line is broken, you'll anyway not get green light. Even if everything else is fine.

I know. I broke G3hot because something is pulling my 16.5v down, and although g3hot was running i assumed power to the smc could be whats triggering a protection signal and pulling the line low voltage and cuasing it to fluctuate. Apparently not though.
Have two boards with identical symptoms, Done quite a few baords in the past now and have never seen this. Both boards started in better condition and seemed to have got worse and fluctuated more after changing the isl and the fets.ect
driving me mad ive tested so many components on both boards and havent found a single thing out of place, ive replaced u7000 a few times.ect

im on the brink of using them for parts and cutting losses