820-3115B running slow ASD error NDC001 but camera seems to be ok

[Ictc]

I got this previous liquid damaged machine back after 6 months of normal usage. The machine runs slow at boot and after logon as an effect of a sensor failure perhaps. The camera seems to work, the only strange thing I noticed in the beginning were some spikes on video using photo boot. I also checked with the oscilloscope the signal directed to the camera line smc and i2c bus and also the power line which has no spikes at all. I couldn't find a replacement camera in stock just to give a try but I'm afraid the problem arise somewhere else. Anybody has an hint to share wit me in order to find the right spot to look at?
Ps. what also sounds strange is that fan spins normallyin every conditions. No matter if other periferrals are disconnected, the fan spins normally and never boost up. Maybe some older issue with smc I had to replace after the previous liquid damage?

 

[Ictc]

forgot to mention that SMC reset actually made the fan spins at maximum speed as expected. As far as I know, sensor issues cause fan spins faster and in my case this doesn't happen. May be the issue is not related to a sensor but only the smc? Is there a way to know it for sure?

 

[2informaticos]

If you doubt about camera, just disconnect it.
Use HWMonitor and check temps and I/V sensors in real time.
Also try clean ME BIOS, even full new BIOS (clean too).

 

[Ictc]

Swapped EFI ROM with a clean one and losted video. Repalced with the original ROM and video still not showing up anymore. Now I get video only after smc bypass, what a hack !???
Couldn't test any sensor with HWMonitor because of the SMC bypass...

 

[Ictc]

Could the ROM change be the cause of the SMC failure at this stage? What could have cause the video to fail even with the original ROM chip? Should I consider an SMC replacement or is there something else am I missing?

 

[2informaticos]

Try first reballing SMC.
Also test the board disconnecting TP/KB and BIL connectors.
Monitor SMC_LID signal.

 

[Ictc]

SMC_LID was low, I had this problem before and I fix it with a jumper between pin 2 of R5071, pin1 of C4906 and PP3V3_S5_SMC. Now the backlight is back but still have problem with slow "beachballed" performance... Takes forever to boot as well. I'll try to test sensor one more time.

 

[Ictc]

Last time the problem with Low SMC_LID didn't involve C4906 where on pin 1 I found PP3V3, today that voltage was missing on that cap. Isn't strange? Something related with the inner traces on the pcb failing or something else like smc putting to ground that signal making the traces soffering somehow? That's also strange because there's 100k resistor between PP3V3 and SMC_LID...

 

[2informaticos]

Did you disconnect BIL connector???
It can kill SMC_LID.

Don't force that signal to 3V3_S5; it has corresponding pull-up resistor.
Run ASD and HWMonitor.

 

[Ictc]

Of course I did, I also measured before and after BIL connection. I suppose the SMC on pin 1 of the pullup resistor is not pulling down the SMC_LID itself. I also removed the pullup resistor and measured for PP3V3 but I got 0 volt that's why I decided to reroute a jumper to it from PP3V42_G3H as it should be as showed in the schematic.

 

[Ictc]

What is also strange is that on the BV the voltage on pin 2 of the pullup resistor is showed as PP3V42_G3H but on the schematic is PP3V3 coming from L7220.... May be I should re-route the jumper to the appropriate voltage rail... What's your suggestions?

 

[2informaticos]

3V3_S5_SMC comes from 3V42_G3H, is correct.
So is OK to run the wire from pin 2 of R5071 to 3V42.
You should always have 3V at SMC_LID; it comes through R5071.
If R5071 is OK and voltage low, remove R6961 and check again.
Less probable to be bad Q5701.
Last to check is changing SMC...

 

[Ictc]

Yesterday I took another board previously repaired from a liquid spill damage, put it in the same chassis and guess what? Same identical issue of the other board. Never tested that board before but that sounds really weird. Same ASD error with the camera and slow performance on boot and runtime with many beach balls. Should I better start believing in ghosts or am I missing something environmental that is causing these issues? Sound also like a sata cable issue that?s why I also exchanged some flat with no luck. May those batch of cable be bad? I don?t have any other fully working a1278 in stock to tests the sata cable right now.

 

[2informaticos]

Why don't disconnect internal HDD and test with USB OSX?
Also, did you try disconnecting camera cable?

 

[Ictc]

Yes with camera cable disconnected nothing change... I decided to reball the smc on the new board but I had a problem with two pads on the chip itself that decided to stick on the PCB ... both pads were g3h, don?t know if that smc once rebelled will work again... this morning I?ll solder it back and will see. I need to find a good working smc which I do not have in stock anymore.

 

[Ictc]

SMC reballed and resoldered but now I get 3 beeps when ram inserted. only one beep with ram slot empty. Can this be the SMC as is the only thing I've messed up with...I'm triyng to find a reason why those 3 beeps, I also checked SMC_MEM_ISENSE wich is between 0,12 with only one module and 0,3 with both modules insertend in ram slots.looks fine, don't know what else to check for.

 

[2informaticos]

Heat applied to desolder/resolder SMC affected other board areas.
You could now have problems with RAM slots, or CPU...

 

[Ictc]

never happened before but I know this board is proneto fail due ram slots issue... I ordered the ddr3 test board andI'll wait for it before consider the board gone... I'll pm you by the way..