[SOLVED]820-3115 runs slow (Th1h Fin Stack sensor failing)

[ytesfay80]

I got this board sent in from another shop. They said their customer had this fixed by someone in Hong Kong about 18 Months ago. It turns on but runs slow, the fans spin high and the system is being used at 88%. I ran ASD for this and I got an error on the Th1h sensor and this is what the hardware profile page showed.

Th1H Fin stack Error- SMC IO failed
reaing -124C
low-limit: 10C
High: 80C
Fin Stack Temp

Inspecting the board I see some very tiny spots of blue corrosion. I also saw a couple discolored points on the right side of the SMC but nothing much. I read through and followed this post from another similar issue but I still haven't been able to fix this. https://www.rossmanngroup.com/boards...5-running-slow

R5260-R5261 Diode Mode - Pin 1: 0.325v - Pin 2 - 0.52v
R5260-R5261 Voltage Mode - 3.33v on both sides
There is continuity from R5260/1 to U5511
Q5520 looks fine and the traces are good.
R5510/11/12 traces are good.
U5511 is getting the VDD power at 3.3v and 3.33v on Pins 9/10

The only issue I found was R5510/11 both have 3.3v on both sides of the resistor but R5512 has 3.3v on Pin 1 but Pin 2 is unstable. It will show close to 0v and every second or so it will very rapidly pulse and show something like 0.21v, 0.35v, 1.7v. I wasn't sure what could be pulling CPUTHMSNS_ALERT_L down so I replaced U5511 but it still does the same thing.

My last resort was going to be to remove the edge bonding from the SMC and put some heat and flux under there.

 

[2informaticos]

Did you change R5112?
Measured on board, you must get same reading on R5511/12.
Also check continuity between Q5520 and U5511.

 

[ytesfay80]

I replaced R5112 but still no change.
I measured and found out that there was 0v on Pin 1 of C5513 and Pin 1 of U5511 even though there was 3.29v on R5510 and there is continuity all the way to pin 1 of U5511.
So I ran a wire from Pin 2 of R5510 to Pin 1 of C5513 and again there was 0v on VDD of U5511 so I ran another wire from C5513 to Pin 1 of U5511. Now I get 3.3v on the VDD pin but when I tested in ASD it failed with the same reading.

Pin 1 of C5511 is fluctuating rapidly from 2.5v to around 3v and Pin 2 goes from 0.37v to 0.64v. Pin 1 of C5512 goes from 0.78v to 1.1v and Pin 2 is 0v.

I have a working 820-3115 board and it meaasured around 0.85v on Pin 2 so I'm not sure if there is an issue with Q5515 or the donor board that I replaced U5511 with was already bad.

 

[jgraffis]

Have you checked for a short to ground on u5511? was there liquid damage?

 

[ytesfay80]

There is no liquid damage in that area and no short to ground.

 

[jgraffis]

i would assume you've check the resistance of the resistors from R5510 R5511, R5512. what is the voltage on pins 7 and 8 of u5511?

 

[ytesfay80]

The resistance of R5510/11/12 are all fine. Pin 7 is 3.3v and Pin 8 goes from around 0v to around 1.2v and quickly pulses every second or so.

 

[jgraffis]

Change u5511, while u5511 is removed check for short to ground on pin8.

 

[ytesfay80]

I removed U5511 and measured pin 8 in diode mode and it showed OL. I put another U5511 from a good looking donor board and now Pin 1(VDD) is present without me having to run a wire.
Pin 8 is still doing the same thing with the pulsing. I'm not sure what else could be causing this. I already replaced R5512 so should I try replacing C5513 even though it looks fine.

 

[jgraffis]

Does it still fail asd? and when you run asd EFI version check check hardware profile and post what the reading is. could be an issue on the data line.

 

[ytesfay80]

Yes. It fails with the same reading.
Th1H Fin stack Error- SMC IO failed
reaing -124C
low-limit: 10C
High: 80C
Fin Stack Temp

I don't think it is the data line because all the measurements seem fine.
R5260-R5261 Diode Mode - Pin 1: 0.325v - Pin 2 - 0.52v
R5260-R5261 Voltage Mode - 3.33v on both sides
There is continuity from R5260/1 to U5511

 

[jgraffis]

not sure if its the same thing but i just had same issue with a different sensor. I was told there is an issue with something in the line of smbus_smc_1 or the rail coming off pins 9 and 10 of u5511. looks like the go to r5261and r5261. check for short to ground and continuity as well as voltage on those resistors. if all is well i would think this to be an smc issue.

 

[ytesfay80]

R5260-R5261 Diode Mode - Pin 1: 0.325v - Pin 2 - 0.52v
R5260-R5261 Voltage Mode - 3.33v on both sides
There is continuity from R5260/1 to U5511

Pins 9/10 of U5511 are both 3.3v.

dukefawks larossmann Any ideas?

 

[dukefawks]

Crap under SMC

 

[ytesfay80]

Okay so I decided to replace Q5520 which should have been the first thing I tried but since it measured fine in diode mode to my working board I decided not.

I just replaced it and everything is working fine and ASD passes.

Thanks for the help guys!

 

[dukefawks]

Shouldn't affect anything as it is on the "analog" side and would just cause a wrong measurement. There was an SMC IO error so that is on the digital side.

 

[ytesfay80]

Well when I read the datasheet for the U5511 chip it said something like a failing diode would cause the Alert signal to be affected which I assume was the reason why CPUTHMSNS_ALERT_L was measuring so low.