SMC Issue on MBP 13 - Hardware Troubleshooting Guide

[U-234]

NO, just unplug the BIL flex cable. No unsoldering of anything.

All my measurements were taken with the logic board taken out of the case, all cables/connectors are disconnected. The board is sitting on the table by itself only with the battery connected to provide power, nothing else is connected not even megsafe.

Both mesurements were taken using two identical boards with just battery connected nothing else. The only difference is i'm getting 1.9V on SDA lines instead of 3.4V, so there is nothing else to disconnect to eliminate the faulty load. In this case what is my next step? Use bruteforce method and start sucking components out to see which one effected the 1.9V voltage or you have other components in mind to check? Thanks.

 

[mac-n-sauce]

What an interesting thread! I've been searching for a solution for my A1261 17'' MBP and came across this thread. I think the LIO board or the logic board fried after a power outage. Actually, I don't have a battery and my dog pulled the Magsafe a couple of times in the past months...

Now, when I plug in the Magsafe, I get a green light (is this a sign that the LIO board is ok?), but when I turn on the MBP, I only see the fans spinning for a sec and ear the HD and the DVD start and shutdown after a sec (but black screen). I tried the SMC reset procedure, but same thing happens. I also tried holding the power button while plugging the Magsafe and it does something different when I press the button again: the fans runs at full power for a sec and then shut down (but the power-on led doesn't turn on anymore). It keeps doing so after each button press until I remove the Magsafe. Then, its back to the previous symptoms. I've also tried to unplug/replug the PRAM battery, but same result.

Now, I removed the LIO board and I wanted to check the voltages (G3HOT and so on), but it seems the architecture of the 820-2274-A is a bit different than the one of the MBP 13''. I don't see any LT3470. The fuses seem ok, I measure 0.2 ohm. Can anyone please tell me which pins to check? Or if I need to check the LIO board at all if my Magsafe led is green?

What you are describing is the notorious Nvidia failure issue. Graphic card chip has come unsoldered and would need to be reballed/reflowed. If the I/O board is giving you a green magsafe light then it is more than likely fine. Because your board is going through the boot process. Apple warranted them for up to 4 years from the original purchase date.

 

[xbert]

What you are describing is the notorious Nvidia failure issue. Graphic card chip has come unsoldered and would need to be reballed/reflowed. If the I/O board is giving you a green magsafe light then it is more than likely fine. Because your board is going through the boot process. Apple warranted them for up to 4 years from the original purchase date.

I can't say it's really booting properly as I can definitively hear the HD and the DVD shutdown after only 1 sec (there is no boot sound, unlike the videos from youtube I've seen about the Nvidia failure issue). If I hold the power button longer, the HD, DVD and fans spin longer until about 5 sec, where the power on led flashes and then they all shutdown, but still no boot sound.

 

[mac-n-sauce]

I can't say it's really booting properly as I can definitively hear the HD and the DVD shutdown after only 1 sec (there is no boot sound, unlike the videos from youtube I've seen about the Nvidia failure issue). If I hold the power button longer, the HD, DVD and fans spin longer until about 5 sec, where the power on led flashes and then they all shutdown, but still no boot sound.

The problem starts as chime with graphics anomalies, then chime but no video and then progresses to power on/off. I can not say for sure that it is the case with yours, but i bought and sold many of those units and had many of them repaired. That was typically the progression. It could also be complete failure of the chip all together. Has it experienced any liquid? You can probably rule out the I/O board though.

 

[xbert]

The problem starts as chime with graphics anomalies, then chime but no video and then progresses to power on/off. I can not say for sure that it is the case with yours, but i bought and sold many of those units and had many of them repaired. That was typically the progression. It could also be complete failure of the chip all together. Has it experienced any liquid? You can probably rule out the I/O board though.

From the 8 pin connector of the LIO board I can measure the following voltages (with reference to the pin closest to the Magsafe): 0V, -16.9V, -16.9V, -16.9V, -16.9V, -4.265V, -4.265V. So I'm guessing that this is alright.

Is there a way to verify that the Nvidia is the cause? Should the Nvidia chip be detached or something? The MBP hasn't experienced any liquid but my battery bulged a year ago. I kept operating the MBP with the Magsafe without battery since. Should've got a new battery long ago, but the reviews are bad for this battery and the price is high.
Is Ebay the cheapest place to get a new logic board?

 

[cmdrdata]

From the 8 pin connector of the LIO board I can measure the following voltages (with reference to the pin closest to the Magsafe): 0V, -16.9V, -16.9V, -16.9V, -16.9V, -4.265V, -4.265V. So I'm guessing that this is alright.

Is there a way to verify that the Nvidia is the cause? Should the Nvidia chip be detached or something? The MBP hasn't experienced any liquid but my battery bulged a year ago. I kept operating the MBP with the Magsafe without battery since. Should've got a new battery long ago, but the reviews are bad for this battery and the price is high.
Is Ebay the cheapest place to get a new logic board?

If it is the NVIDIA logic board problem, google "baking macbook". Your Mac of that generation do have NVIDIA BGA solder problem and it would be worth your while before spending money on buying a used logic board that may develop the same problem. I've had success baking a 15" MBP of that vintage (400°F for x minutes, forgot the exact time detail though).

 

[mac-n-sauce]

If it is the NVIDIA logic board problem, google "baking macbook". Your Mac of that generation do have NVIDIA BGA solder problem and it would be worth your while before spending money on buying a used logic board that may develop the same problem. I've had success baking a 15" MBP of that vintage (400°F for x minutes, forgot the exact time detail though).

Just remember if you "bake" the LB it is only a temp fix. It takes 7 - 10 minutes depending on your oven. I send my unit to a company in California. They fix it for about $125. You just ship your board to them. eBay is definitely the cheapest for the LB. Not sure how to verify graphics card, but you will not see any physical evidence. It is on the bottom side of LB under the heat sink.

 

[cmdrdata]

All my measurements were taken with the logic board taken out of the case, all cables/connectors are disconnected. The board is sitting on the table by itself only with the battery connected to provide power, nothing else is connected not even megsafe.

Both mesurements were taken using two identical boards with just battery connected nothing else. The only difference is i'm getting 1.9V on SDA lines instead of 3.4V, so there is nothing else to disconnect to eliminate the faulty load. In this case what is my next step? Use bruteforce method and start sucking components out to see which one effected the 1.9V voltage or you have other components in mind to check? Thanks.

1. With battery disconnected and magsafe plugged it, do you see battery SDA line going to 3.4v? If yes, then maybe you have a bad battery. If 1.9v then skip step 2.

2. Do you have another good battery to try? If it is still have 1.9v with a good battery then try the next step.

3. The partial schematic you posted showed the SDA lines going to pages 6,45,48,..., and 84. I don't know what p6 and p84 shows, but I think one of the component hanging on this line is bad. If it was me, since G3HOT is normal (the fact that SCL line is at 3.4v), I'd start by removing D6950 chip. This chip is strictly a "transient protection" so without it the system should still power-up. Perhaps one of that zener diode has broken down.

 

[xbert]

Wow, it did work. Thanks!
The GPU temp is a bit high I think, not sure. I get 58C at idle. For the GPU heatsink and CPU, I get 48C. Ambient temp is 27C. The fans are spinning at 2k. Is this normal for the MBP 2008 (I get much lower temps on a newer MBP on idle)? I cleaned all the old thermal paste and applied new Artic MX-2. The SMC was also reset.

 

[mac-n-sauce]

Wow, it did work. Thanks!
The GPU temp is a bit high I think, not sure. I get 58C at idle. For the GPU heatsink and CPU, I get 48C. Ambient temp is 27C. The fans are spinning at 2k. Is this normal for the MBP 2008 (I get much lower temps on a newer MBP on idle)? I cleaned all the old thermal paste and applied new Artic MX-2. The SMC was also reset.

Awesome, just remember... this is usually a short term fix. So watch the temps, you might be able to adjust the fans to come on at lower temperatures to help from overheating.

 

[mac-n-sauce]

MacBook Pro 15" 2009 Shuts off immediately

Ok, I have another MBP 15" 2009 2.66 GHz. When plugged in with ac adapter or battery and powered on, white light flashes three times. Then I get boot chime. Apple logo, loads almost into OS and then shuts off.

Any thoughts?

 

[Dadioh]

Ok, I have another MBP 15" 2009 2.66 GHz. When plugged in with ac adapter or battery and powered on, white light flashes three times. Then I get boot chime. Apple logo, loads almost into OS and then shuts off.

Any thoughts?

Have you tried verbose mode to see what messages are displayed just before it resets?

 

[mac-n-sauce]

Have you tried verbose mode to see what messages are displayed just before it resets?

Yes, thank you. I said mac hd malfunction.
So I replaced with known good HD. Then it boots completely and the moment it finishes loading, shuts off.

 

[cmdrdata]

Yes, thank you. I said mac hd malfunction.
So I replaced with known good HD. Then it boots completely and the moment it finishes loading, shuts off.

I am pretty certain that this is not a hardware problem. My thought is that after the OS has loaded it "reads" a config file that tells the MBP to shutdown/hibernate and thus it immediately shuts down. I don't recall how to change this stored configuration status, but in a normally working system as an example, if the battery has gone past a critical stage it would shutdown. Yours is powered by magsafe, so that's not it, but somehow the OS thinks it is time to do so. Can you boot from CD/DVD and keep it up?

 

[mac-n-sauce]

I am pretty certain that this is not a hardware problem. My thought is that after the OS has loaded it "reads" a config file that tells the MBP to shutdown/hibernate and thus it immediately shuts down. I don't recall how to change this stored configuration status, but in a normally working system as an example, if the battery has gone past a critical stage it would shutdown. Yours is powered by magsafe, so that's not it, but somehow the OS thinks it is time to do so. Can you boot from CD/DVD and keep it up?

It did have a bad battery in it when it happened. Then I took the battery out and it still did it. i wonder if there is a way to reset it. I will try to boot from external drive.

 

[cmdrdata]

It did have a bad battery in it when it happened. Then I took the battery out and it still did it. i wonder if there is a way to reset it. I will try to boot from external drive.

Just a thought: if you haven't already done so: SMC reset and PRAM reset. These are two separate activities and the PRAM reset is probably the one that will clear the shutdown/sleep status.

 

[mac-n-sauce]

Just a thought: if you haven't already done so: SMC reset and PRAM reset. These are two separate activities and the PRAM reset is probably the one that will clear the shutdown/sleep status.

Tried it, it stayed booted for maybe two minutes and then shut off.

 

[jugger18]

Have you tried a safe boot?

 

[U-234]

1. With battery disconnected and magsafe plugged it, do you see battery SDA line going to 3.4v? If yes, then maybe you have a bad battery. If 1.9v then skip step 2.

2. Do you have another good battery to try? If it is still have 1.9v with a good battery then try the next step.

3. The partial schematic you posted showed the SDA lines going to pages 6,45,48,..., and 84. I don't know what p6 and p84 shows, but I think one of the component hanging on this line is bad. If it was me, since G3HOT is normal (the fact that SCL line is at 3.4v), I'd start by removing D6950 chip. This chip is strictly a "transient protection" so without it the system should still power-up. Perhaps one of that zener diode has broken down.

With megsafe or without I'm still getting 1.9V also i'm sure the battery is good because I'm using the same battery during the test on both logic boards. On one I'm getting 3.4V and the other one with the same battery I'm getting 1.9V, so all i'm changing here is the board.

Now, I thought the same like you about D6950 diode because it hits SDA line I thought it might be the culprit so i removed it, took measurements and still the reading 1.9V didn't change so removing D6950 didn't make any effect.

Next on my list was - C6952 50V capacitor which after my measurements looked suspicious, the reading was "LO" in the capacitance mode. The same capacitor on the other board had higher volume but not "LO".

So I removed it also and took measurements which didn't effect 1.9V. I'm thinking that I need to get a new one solder it and hopefully 1.9V will change.

If after replacing a new Cap won't change the value then it might be issue with SMC which I don't intent replacing.

Please let me know your thoughts on this.

Thanks

 

[cmdrdata]

If after replacing a new Cap won't change the value then it might be issue with SMC which I don't intent replacing.
Please let me know your thoughts on this. Thanks

Replacing the cap will not bring that line back to 3.4v. I'd replace the 2k Ohm pull up resistor for that SDA line and see if that makes a difference. I think papabear did this a while back.

If that didn't do it, then I'd relegate this board as a parts source. Only two other components are connected to it, i.e., the CHARGER IC and the SMC. Both will not be easy to replace and you may not be able to buy the part anyway.

REVISED: I just had a thought: In some MBP logic boards, the pull-up resistor is 1K Ohm. Your's is 2K Ohm. So using basic electronic calculation, the 1.9v is a result of SDA line having a DC impedance of slightly above 2K Ohm. Say you replace the pull-up from 2K Ohm to 1K Ohm, then your SDA static voltage should go up to about 2/3 x 3.4 v = 2.2v. I don't know if this is enough voltage to allow SDA line to pass address/data pulses.

 

[mac-n-sauce]

With megsafe or without I'm still getting 1.9V also i'm sure the battery is good because I'm using the same battery during the test on both logic boards. On one I'm getting 3.4V and the other one with the same battery I'm getting 1.9V, so all i'm changing here is the board.

Now, I thought the same like you about D6950 diode because it hits SDA line I thought it might be the culprit so i removed it, took measurements and still the reading 1.9V didn't change so removing D6950 didn't make any effect.

Next on my list was - C6952 50V capacitor which after my measurements looked suspicious, the reading was "LO" in the capacitance mode. The same capacitor on the other board had higher volume but not "LO".

So I removed it also and took measurements which didn't effect 1.9V. I'm thinking that I need to get a new one solder it and hopefully 1.9V will change.

If after replacing a new Cap won't change the value then it might be issue with SMC which I don't intent replacing.

Please let me know your thoughts on this.

Thanks

I have a similar problem to yours. Can you tell me what the parts look like or where they are located? Any pictures? "D6950","C6952" etc.

 

[U-234]

Replacing the cap will not bring that line back to 3.4v. I'd replace the 2k Ohm pull up resistor for that SDA line and see if that makes a difference. I think papabear did this a while back.

If that didn't do it, then I'd relegate this board as a parts source. Only two other components are connected to it, i.e., the CHARGER IC and the SMC. Both will not be easy to replace and you may not be able to buy the part anyway.

REVISED: I just had a thought: In some MBP logic boards, the pull-up resistor is 1K Ohm. Your's is 2K Ohm. So using basic electronic calculation, the 1.9v is a result of SDA line having a DC impedance of slightly above 2K Ohm. Say you replace the pull-up from 2K Ohm to 1K Ohm, then your SDA static voltage should go up to about 2/3 x 3.4 v = 2.2v. I don't know if this is enough voltage to allow SDA line to pass address/data pulses.

Hm...very interesting, I never thought about that. Yes, if we replace it with lower resistance it should definately increase our 1.9V line.

Now, here is the decision, I already ordered replacement for my cap that gives "LO" which raises red flag compared to 9646 on the other board.

It appears that the pull-up resistor when measured gives 2K Ohm like it supposed to, so should I still go ahead and get replacement? (going with 2K will have possibility if the original is ok that it won't affect 1.9V line)

Note: Also on the battery charger (pin 10) I'm getting 1.9V as well. So, If replacing resistor won't cut it than re-soldering SMC and Battery Charger would be a pain in the butt something I wouldn't try.

----------

I have a similar problem to yours. Can you tell me what the parts look like or where they are located? Any pictures? "D6950","C6952" etc.

D6950 is a small component right in front the battery connector. You have to flip the board bottom up to see it. Are you also getting 1.9V and solid green on the megsafe? What year is your Macbook Pro?

 

[cmdrdata]

Hm...very interesting, I never thought about that. Yes, if we replace it with lower resistance it should definately increase our 1.9V line.

Now, here is the decision, I already ordered replacement for my cap that gives "LO" which raises red flag compared to 9646 on the other board.

It appears that the pull-up resistor when measured gives 2K Ohm like it supposed to, so should I still go ahead and get replacement? (going with 2K will have possibility if the original is ok that it won't affect 1.9V line)

Note: Also on the battery charger (pin 10) I'm getting 1.9V as well. So, If replacing resistor won't cut it than re-soldering SMC and Battery Charger would be a pain in the butt something I wouldn't try.

C6952 is only a 47pf cap and you've already measured the line with it removed and still got 1.9v, so IT IS NOT the culprit. Insteadof taking out the resistor, IF you tack on another 2K Ohm on top of it, in effect you're half-ing the value, thus you'll get 1K Ohm for that pull up as temporary test to see if the voltage will go up. at this point my guess is that the SMC or CHARGER IC has an INTERNAL "open-collector transistor" that is bad thus unable to pulse the SDA line for signaling.

Attached is a schematic of the SMC (H8S2117) chip SDA OUTPUT showing how it drives the SDA signal line.

 

[mac-n-sauce]

With power off check the impedance across Q7030 and Q7035. Check D-S, G-S, D-G on both FET's and see if there is a low impedance indicating a blown FET.

Another quick check, if the impedance Q7035 D-S is low is to lift off the fuse and see if the low impedance is after the fuse or before the fuse. That wilt ell you if it is the charger circuit or if the problem is further down stream on one of the board's power supplies.

DADIOH: HAVING THE SAME PROBLEM WITH MINE. Checked both mosfets see photo for values. I do not know where the fuse is at. Do you know? I am guessing that the bottom mosfet is no good since it has the low resistance value. Should I just replace it.
On a side note: Since I am a novice at all this. What is the best way to remove one of the small resistors. I have tried with my soldering tip and solder wick, but could not get it off. Should I be using a hot air tool instead. Used my Hakko at about 390 c.

 

[U-234]

C6952 is only a 47pf cap and you've already measured the line with it removed and still got 1.9v, so IT IS NOT the culprit. Insteadof taking out the resistor, IF you tack on another 2K Ohm on top of it, in effect you're half-ing the value, thus you'll get 1K Ohm for that pull up as temporary test to see if the voltage will go up. at this point my guess is that the SMC or CHARGER IC has an INTERNAL "open-collector transistor" that is bad thus unable to pulse the SDA line for signaling.

Attached is a schematic of the SMC (H8S2117) chip SDA OUTPUT showing how it drives the SDA signal line.

I just received the cap 47pf, replaced it but I'm still getting 1.9V. Now I'm wating for 2k Ohm resitor, I will tack it on the top of the other one to see any voltage changes. Let's say if that won't be enough to get the SDA line going do you suggest maybe a quick reflow of the charger IC or SMC chip. Or you think that wont help?