Macbook air 820-00165-A fixing

[bargo]

Hello folks,

I hope you are doing well.

I have had a macbook Air, from 2013 with 8GB 1.7GH one day (almost 1 year ago) it had failed, so you know as a human being, I had to try to fix it, but then i caused a lot of more trouble by shorting PPDCIN_G3H with PP5V_S4RS3 by mistake, which I then fixed to a good extent (macboook was powering up three times then normal boot, into internet recovery (which was not working) but booting into ubuntu via a usb stick went halfway, etc... so after so much time spent on this logic board replacing stuff to get it to that working state, i was hit by the issue that there is no SSD visible when its there, so I was investigating that but while doing so, I shorted the PCH via the PCIE SSD resistor R1884 with again the PPDCIN_G3H through the body of the Q7180.

Now to the problem. I decided to buy a bunch of cheap macbook air logic boards, I got two of them which may have a good chance to work, I started with one that had obviously PCH replacement as well as SMC replacement, this board actually has all voltages until ALL_SYS_PWRGD, the only thing that I found a little bit strange, iS SMBUS_SMC_0_S0_SDA and SMBUS_SMC_0_S0_SCL, which are not present, what makes SMC generates these lines, or is it generated by default? I looked for anything that could pull to ground but nothing also is there..

Any idea on what to look further, or SMC needs to be replaced again (which I dont think so)?

All my best,

[mon2]

Re: Macbook air 820-00165-A fixing

SMBUS is an open collector design. Locate the pull-up resistors on each line. Confirm the pull-up voltage to ground for each resistor. If the pull-up voltage is missing, SMBUS will not function correctly.

Also, remove all power, meter in resistance mode.

Check the resistance to ground of:

SMBUS_SMC_0_S0_SDA
SMBUS_SMC_0_S0_SCL

Post each measurement.

[bargo]

Re: Macbook air 820-00165-A fixing

Thanks a lot mon2, I appreciate your help..

SMBUS_SMC_0_S0_SDA: is at 70k ohm with 2.6v on diode mode
SMBUS_SMC_0_S0_SCL: is at 73k ohm with 2.65v on diode mode

I believe both of them are being pulled up to PP3V3_S0SW_LCD which I can't find where are they being pulled up from, but I may be wrong, I thought these are being generated from the SMC internal circuit. I may have seen designs where the I2C is being generated internally

The only logical explanation would be a missing touch point between the SMC and the PCB, but both balls are being missed is quite strange thus I dont think its an SMC issue, but still it could be.. On a working logic board it would be great to know if the SMBUS_SMC_0_S0_SCL gets pulled up at the same time as SMBUS_SMC_1_S0_SCL? because my SMBUS_SMC_1_S0_SCL is being pulled up to 3.3v just fine, I think there is some activity also on the SDA line SMBUS_SMC_1_S0_SDA.


Again my measurements, apparently the capacitors were not fully discharged on my first readings:

SMBUS_SMC_0_S0_SDA: is at 27k ohm with 2.6v on diode mode
SMBUS_SMC_0_S0_SCL: is at 27k ohm with 2.6v on diode mode

SMBUS_SMC_1_S0_SDA: is at 8k ohm with 0.96v on diode mode
SMBUS_SMC_1_S0_SCL: is at 8k ohm with 0.96v on diode mode

I can see difference here, its high resistance on SMBUS_SMC_0_S0_SDA/SCL which means something is broken, and the only place it goes to is the SMC, right?

[mon2]

Re: Macbook air 820-00165-A fixing

Be sure that the measurements for the resistance are without any power to the logic board.

Can you measure again the voltage to ground but use a low DC volts scale on your meter? That is, since we know the voltage should not be higher than 5 volts (max); ok to use 10v or higher scale on your meter.

In addition to the DC volts measurement, to use DIODE mode scale -> remove all power.

Red meter probe, yes RED to GROUND.
Black meter probe to the point to test.

The DIODE mode of testing is great to check if the pin is damaged. A low DIODE reading means the internals of the IC are damaged - perhaps from an over voltage or ESD static event.

Please update your measurements.

[bargo]

Re: Macbook air 820-00165-A fixing

Thanks again mon2, I actually measured the voltage with a handheld oscilloscope there is a around 6mv rise on SMBUS_SMC_0_S0_SDA/SCL.

For diode readings red to ground I read 0.57v.

And of course all readings for ohm and diode mode, are done without any source of power..

[mon2]

Re: Macbook air 820-00165-A fixing

The SMC is the bus master so it initiates the smbus traffic.

What is the voltage to ground of PP3V3_S0SW_LCD?

This is the pull-up voltage source for this bus pair of lines. See R8363 and R8364.

Also since your power good status is low means that at least one power rail is not valid.

[bargo]

Re: Macbook air 820-00165-A fixing

The SMC is the bus master so it initiates the smbus traffic.
Also since your power good status is low means that at least one power rail is not valid.

Which power good status you are referring to? ALL_SYS_PWRGD that one is also high, maybe i did not describe it correctly.


PP3V3_S0SW_LCD is at 0v same as SMBUS_SMC_0_S0_SDA/SCL

I actually asked earlier

I believe both of them are being pulled up to PP3V3_S0SW_LCD which I can't find where are they being pulled up from, but I may be wrong, I thought these are being generated from the SMC internal circuit. I may have seen designs where the I2C is being generated internally

PP3V3_S0SW_LCD do you know what is the source of voltage for this rail? I can't find it, its connected to the L8304 but going to PP3V3_S0SW_LCD_UF which goes to J8300, so does it take power from the display!! or what is the source of it?

if it takes it from display, then that explains it, as the EDP_BKLT_EN is low at 0v so the main screen power is not present to the LCD so it can not give power to the SMBUs anyway, but if the source of power is not the display, what is it ?

[bargo]

Re: Macbook air 820-00165-A fixing

I noticed this statement..

it could be I have a bad q7707,I think I will just remove it.. OK it is not q7707. I read on various places that EDP_BKLT_EN will not go high unless CPU know there is a connected screen, so I'm trying to actually validate this assumptions, and if its correct, which pin is the sensing pin?

I think the sensing pin is DP_INT_HPD, I will test this network.. I could not find anything wrong, its unconnected at 75k ohm, and connected at 20k, which seems logical, DP_INT_HDP also stays low at 0v as well

I feel like I'm on windows 95 and refreshing the screen every few seconds, to see how fast is my machine

I thank the Lord of earth and skies, for the knowledge he gives

[bargo]

Re: Macbook air 820-00165-A fixing

Just to make sure, I'm looking in the right places, I found it interesting that PM_THIRMTRIP_L is high at 1v , i can also see that SMC_THRMTRIP is low which I suspect that it should be high!

Why SMC_THRMTRIP is low is a long story I guess

on the other hand SMC_PROCHOT is high after few seconds, and thus CPU_PROCHOT_L is correctly low.

Ok I think I need to know a little bit more information, the SPI communication from CPU to BIOS is happening, and its reading some stuff, at first, then after few ms it acknowledges or say something else, now, that means bios is being read and CPU should know what to do as well as SMC, now the SMC part is a bit tricky, I have not yet understood correctly, but I think there is a separate rom that resides in SMC, which basically explains why it also has these lines connected to it.. if SMC has its own ROM could it be that it has the wrong ROM in my case? I'm no expert.

attaching SMC picture, I will try to plug my self into it, and see what I can get it.

[mon2]

Re: Macbook air 820-00165-A fixing

Can you confirm you are using the original and official power adapter? What is the color of the led inside the magsafe head?

What is the voltage to ground of ppbus_g3h?

[bargo]

Re: Macbook air 820-00165-A fixing

Yes, I can confirm original macbook adapter, the 85w variant. (so not the official, the official is 65w) but that should not matter I assume

The magSafe color is green at the first second then turns to orange

Voltage at ppbus_g3h is static at 8.55v

[mon2]

Re: Macbook air 820-00165-A fixing

More power is ok. Except when the 7 million dollar man was a guest on the six million dollar man show. 'The power Steve...'

Is the ppbus_g3h the same on both sides of the fuse? Do not have the schematic open but may be F7000 or F7040.

[bargo]

Re: Macbook air 820-00165-A fixing

Ok i actually think there is a problem with the charger, its says output is 20v which i assume is not actually the correct input voltage for this poor logic board, I think the correct output should be 14.5v or I'm maybe wrong. I already check voltage on the fuse, its equal 0 ohm between end points. Do you know what should be the official charger output for this variant?

[mon2]

Re: Macbook air 820-00165-A fixing

To me, the adapter voltage seems high. Would you have another magsafe adapter for testing ?

Measure the voltage to ground of each pin on D7005.

Also, review the 11+1 voltage sequence here:

https://www.it-tech.com.au/mac-q-a/pm-slp-s4-l-timing/

[bargo]

Re: Macbook air 820-00165-A fixing

Thanks mon2, I checked D7005 and pin 1 is at 8.6v pin 2 and 3 are at ~17v , I'm still using the same charger, I currently don't have another charger, but soon I will.

regarding the PM_SLP_S4_L 11+1 I already checked that too, as I say, I have PM_SLP_S4_L as I'm correctly on S0.. but maybe not, I did also check the prerequisites for PM_SLP_S4_L anyhow and they are good

[bargo]

Re: Macbook air 820-00165-A fixing

I'm a 0 million man, but that was a good one

What is the start voltage for SMC_SENSOR_ALERT_L on a working 1466 board? with an oscilloscope I see it at 3.3v at start and then down 0v, then after couple of seconds again at 3.3v

[bargo]

Re: Macbook air 820-00165-A fixing

Just to summarize, at this state, I currently have no power rails missing, what I have is digital signals missing, these are as far as I know:

- SMC_S4_WAKESRC_EN (A must check)
- SMC_THRMTRIP
- SMC_SYS_KBDLED (when the keyboard led lights jack is connected tested with two keyboards)
- EDP_BKLT_EN (when the LCD jack is connected tested with two screens)
SMC seems to be doing most of its job but not the complete job, digitally speaking. So I don't want to put it a suspect while it is, the EDP_BKLT_EN comes from CPU, I dont want to put that as a suspect too because its doing what it should be doing as far as I can see.. But of course I don't see it all.

What is missing from forums and other places or maybe google did not yet index, is how can I know if the memories are being trained successfully or are being communicating to the CPU correctly?

is there something like RAM_DEBUG led?

additionally I think i am missing one more signal to be fully in an S0 state, SMC_S4_WAKESRC_EN, what is this signal? what makes SMC thinks its time to generate, what if I provide this with a nice 3.3v connection, could that be what is missing for the other signals to come in?

btw i could not find the SMC_S4_WAKESRC_EN listed for any boot sequence requirements so far, I think this signal should be one of the signals to check if you are in S0 or not.. it also says SMC, its becoming more of a suspect at this stage

[bargo]

Re: Macbook air 820-00165-A fixing

̶O̶k̶ ̶t̶h̶i̶s̶ ̶i̶s̶ ̶s̶t̶r̶a̶n̶g̶e̶,̶ ̶I̶ ̶h̶a̶v̶e̶ ̶P̶1̶V̶5̶S̶0̶S̶W̶_̶A̶U̶D̶I̶O̶_̶E̶N̶ ̶o̶n̶ ̶2̶.̶9̶v̶ ̶i̶t̶ ̶s̶h̶o̶u̶l̶d̶ ̶b̶e̶ ̶1̶.̶5̶v̶. No that was correct its the _EN signal, P1V5S0SW_AUDIO is at 1,.5v

[bargo]

Re: Macbook air 820-00165-A fixing

btw i could not find the SMC_S4_WAKESRC_EN listed for any boot sequence requirements so far, I think this signal should be one of the signals to check if you are in S0 or not.. it also says SMC, its becoming more of a suspect at this stage

I think SMC_S4_WAKESRC_EN is not needed, It actually has no reason as the u8170 is no stuff, thus I am on S0 state at the moment, with all power rails apparently, as well a valid 3.3v all_sys_pgood

I think the SMC that was planted on this board is not the correct version, I would be looking for a LM4FSXAH5BB and what I have attached in the previous picture is LM4FSXEH5BB so the letter A and E differs, could that be flashed or I need to get LM4FSXAH5BB? or it should not matter, I think it should matter a lot!!

[bargo]

Re: Macbook air 820-00165-A fixing

Just to make sure, I'm looking in the right places, I found it interesting that PM_THIRMTRIP_L is high at 1v , i can also see that SMC_THRMTRIP is low which I suspect that it should be high!

Is it actually correct that SMC_THRMTRIP is low? Should that indicate no thrmtrip, or is it thermal tripping when low (I think its not thermal tripping, because when its high the machine shuts down)?

According to this logic in the last chart here, https://logi.wiki/index.php/Bootup_Signal_Order (Thanks to Piernov) in the orange blocks, I have activity on CS pin for SPI reading bios and then confirming something after some second, then I also have SMC_LRESET_L, i dont see any activity on LPC_FRAME_L or R and I think thats because SMC did not yet initialize, does it mean bad SMC or wrong flash on SMC, not compatible with BIOS or any ideas? I think I will need some tools