iMac 820-3126 Half Fan Spin


Following the power signals on page 63

I went to the outputs... I guess I measure the right components.

Please check. If not give full detail

C6920 = 3.3V
C6921 = 3.3V

R6946 = 0V BOTH PIN
R6947 = 0V BOTH PIN

R6936 = 0V BOTH PIN


I posted to you before

R6946 = 0V BOTH PIN
R6947 = 0V BOTH PIN

They are the output of U6900

Do you need any specific pin?



1 = 3.3V
2 = 3.3V
3 = 3.3V
4 = 3.3V
5 = 3.3V
6 = 3.3V
7 = 3.3V

8 = 0v
9 = 0V
10 = 3.3V
11 = 3.3V
12 = 3.3V
13 = 3.3V
14 = 3.3V


Remember that to measure at this time I did a jumper from 12V_G3H to 12V_S0 as you requested

1 = 0
2 = 0
3 = 0
4 = 0
5 = 3.3
6 = 3.3

1 = 0v
2 = 8.95v
3 = 0v
4 = 0v
5 = 0v
6 = 1.59
7 = 0
8 = 11.95


Staff member
Try to understand schematic; be sure R7082 is good.

"I hope 12V_S0_PWRCTL is present."
You can choose to reply one question per day, if not in hurry with repair...


12V_S0_PWRCTL is not named this on the boardview file

R7082 was not measuring 2K and only have voltage on pin1 and nothing on pin 2. Replaced with another one and now I do have 12V on both pin and looks like the board is trying to start sometimes. I will assemble to test.

But I still do not have LED 2 and the funny thing is I've got R5281 from a 820-3115 board to replace it and it was measuring 2K on the board.. After replacing it on the iMac board its not giving 2K ohms on the multimeter but I do have 12V on both pin.



Staff member
An onboard measurment for resistor can give you different reading than nominal.
That's normal, more components connected there affect onboard value.

"12V_S0_PWRCTL is not named this on the boardview file"
I'm sure you can locate one component connected to it on schematic in boardview too.
Learn how to use schematic/boardview; such reply is not acceptable for a technician!

Did you test it now with its original iMac PSU?
Remember to remove the short between 12V_G3H and 12V_S0.


Just tested with the original iMac PSU.. Same behavior as of the beginning, half fan spin only LED 1..

For sure I remove the jumper before doing anything else.

What now?


Staff member
Check which voltages try to appear in pages 68-74.
Be sure none is shorted to ground.

Did you test SMC bypass?


Yes I tried SMC bypass and its same behavior I already told you

I will measure the output from the pages you request ok? Cause there is too may things and I am not sure exactly what you want me to measure.

I think would be a way more efficient to give components or lines that could have problem. Giving pages to go for is not that good.

I could go to the whole schematic by myself and take hours doing it but being in this forum I am looking some more specific advice

And some pages that you are asking me I already gave you the info


Its also important to mention that the pages you just requested me to measure the majority of the output are on the side of the board which I am not able to measure with the iMac PSU

Only with LAB PSU injecting voltage from G3H to 12V_S0 WHICH is not the same behavior that I just mentioned above "trying to go high and then goes to 0V"

With LAB PSU Its stable 12V_S0. I think it could make wrong measurements don't you agree?


Also important to mention that before I replaced R7082

When I inject 12V with PSU LAB it was stable at 0.080a after replacing it it tries to go to 0.300a and then back to 0.050a in a loop.



Measuring with PSU LAB

page 68
R7467 which is the VDD tries to go high and then back to 0V

Page 69
C7516 = Stable 0.560V
C7510 which is the output tries to go high and then back to 0V

Page 70
L7750 tries to go high and then back to 0V on both pins
L7710 stable 3.3V

Page 71
L7830 tries to go high and then back to 0V on both pins

Page 72
Mentioned on the start of the thread its 3.40v

Page 73
U8000 PIN6 tries to go high and then back to 0V
U8053 PIN6 tries to go high and then back to 0V
U8050 PIN6 tries to go high and then back to 0V
U8025 PIN6 tries to go high and then back to 0V

Page 74
R8132 stable 11.93 on both pins


Staff member
The problem with lab PSU is that PCH timings are not respected; 12V_S0 being present even in S5 state.
If you can confirm that a good working board boots in this way, then is safe to continue the repair process with lab PSU.
Until then, is better to remove original PSU and use it with the board on the desk.

Mentioned pages contain lot of secondary voltages.
You must be sure first there is no short; on any output rail.
Then check for voltage pulse.
Some of these secondary voltages have a power good signal, which is needed to get correct ALL_SYS_PWRGD.
For example, on page 68, 1V05_S0 is what you need to check (L7420).
On page 69 should check (pulse) voltage for D and S of Q7500.
You forget to check L7850 on page 71.

"Yes I tried SMC bypass and its same behavior I already told you"
Did you try it after changing R7082?

Check for pulse at pin 8 & 11 of U7050; also at L7201.