switch OLED with 100ma power draw

[jjloki222]

Hi to all! I have an OLED switch with 100ma power draw and doesn't turn on. The board has one of the four inductors around MAX77812 with 0.6v output and the rest three don't have any voltage present (I expect all should have 0.6v present on a normal board). MAX77620 1.3v power rail is not present. (others are present). I have changed EMMC for this board before.

Is this more like an emmc problem or MAX77812 issue? I'm running out of idea and any input is really appreciated!

Thanks!

 

[jkyoho]

Hi to all! I have an OLED switch with 100ma power draw and doesn't turn on. The board has one of the four inductors around MAX77812 with 0.6v output and the rest three don't have any voltage present (I expect all should have 0.6v present on a normal board). MAX77620 1.3v power rail is not present. (others are present). I have changed EMMC for this board before.

Is this more like an emmc problem or MAX77812 issue? I'm running out of idea and any input is really appreciated!

Thanks!

very likely emmc issue, What did you do on Emmc before?

 

[jjloki222]

very likely emmc issue, What did you do on Emmc before?

It was a failed picofly installation attempt. I rebuilt the emmc with previous backup data by a programmer and after I reballed/reinstalled the emmc, it stuck in this status. (I have no access to the date0 point so I can not verify the installation is correct or not with an oscilloscope. However, after I remove the emmc and measure the date0 pin on the board, it's in stable 1.8v)

I am curious to know how to verify if the picofly installation killed the ARM boot processor within the SoC or the emmc installation is not correct？

just think aloud out of my mind and thank you for response!

 

[jkyoho]

It was a failed picofly installation attempt. I rebuilt the emmc with previous backup data by a programmer and after I reballed/reinstalled the emmc, it stuck in this status. (I have no access to the date0 point so I can not verify the installation is correct or not with an oscilloscope. However, after I remove the emmc and measure the date0 pin on the board, it's in stable 1.8v)

I am curious to know how to verify if the picofly installation killed the ARM boot processor within the SoC or the emmc installation is not correct？

just think aloud out of my mind and thank you for response!

Was the "previous backup" from same console? Was the console working before picofly installation failed?
It's hard to verify SOC failed without CFW or L4T Ubuntu boot check since picofly won't kill hardware but just data corruption(rebuild would fix data) yet bad soldering work kills stuff

 

[jjloki222]

Was the "previous backup" from same console? Was the console working before picofly installation failed?
It's hard to verify SOC failed without CFW or L4T Ubuntu boot check since picofly won't kill hardware but just data corruption(rebuild would fix data) yet bad soldering work kills stuff

yes, it was from the previous backup from the same OLED switch. actually, the picofly installation worked for several weeks and stopped working after a child dropped the console on the ground. so I got rid of the picofly and want to do a clean boot to do more diagnose. unfortunately, it's not booting up.

actually the 1.3v power rail is not present and I suspect this rail is only opened by boot processor after it checks emmc encrypted data on boot0/1? (Please correct me if my understanding is wrong. if my theory is wrong, I am more inclined to do a MAX77620 replacement instead of emmc)

 

[jkyoho]

I suspect this rail is only opened by boot processor after it checks emmc encrypted data on boot0/1?

Thats correct.
Do you remember what fuse/HOS version with your backup and what's in the fuse count in the emmc before you restore?
I wonder if OFW/HOS been upgrade and your backup mismatch the burned fuse

 

[jjloki222]

Thats correct.
Do you remember what fuse/HOS version with your backup and what's in the fuse count in the emmc before you restore?
I wonder if OFW/HOS been upgrade and your backup mismatch the burned fuse

Thanks for your confirmation! I was planning to do a power rails and critical signals dependency analysis by a logic analyzer just for learning the boot procedure.

I don't really remember the versions unfortunately. The backup was a raw backup including boot0/1 done by a bga programmer so it should be ok.

Actually, I didn't see any current fluctuation and it goes straight to around 100ma. Shouldn't the boot processor cause some current fluctuation if it really runs? Looks like my best bet is still to reprogram the emmc and do a reinstallation? （ I hate doing emmc reinstallation :'( )

 

[jkyoho]

THE fuse burn in SOC internal storage so even you have emmc entire dump wont stop mismatch.
I hate emmc reball&reflow too but since hekate worked before, GPP raw dump should be better/useful choice than full dump-in-1File

Post automatically merged: Oct 23, 2023

Actually, I didn't see any current fluctuation and it goes straight to around 100ma. Shouldn't the boot processor cause some current fluctuation if it really runs?

Those checking process finishes in less than 1ms, I dont think you could catch the process without Oscilloscope

 

[jjloki222]

THE fuse burn in SOC internal storage so even you have emmc entire dump wont stop mismatch.
I hate emmc reball&reflow too but since hekate worked before, GPP raw dump should be better/useful choice than full dump-in-1File

sorry for my ignorance and thanks a lot for pointing this out! I didn't even notice there is such a mechanism there! (I should read more about this. ) so even I install the Emmc correctly, there is a chance that the boot processor would refuse to boot if the fuse internal to the SoC is mismatch to the fuse in the emmc? (I rememered that I did do several firmware update on sysnand)

In this case, I have to install the picofly and boot to hekate to fix the fuse problem? Otherwise, it won't boot up right?

Post automatically merged: Oct 23, 2023

Those checking process finishes in less than 1ms, I dont think you could catch the process without Oscilloscope

oh, that makes perfect sense.

 

[jkyoho]

In this case, I have to install the picofly and boot to hekate to fix the fuse problem? Otherwise, it won't boot up right?

picofly wont fix fuse mismatch but You can use homebrew upgrade sysnand or run other thing to ensure all hardware works properly.

 

[jjloki222]

picofly wont fix fuse mismatch but You can use homebrew upgrade sysnand or run other thing to ensure all hardware works properly.

Thanks a lot and I will do more research on this!

The fuse is burnt in the boot0/1 or normal storage area? if it's in boot0/1 and the checks is done by boot processor at the very early stage of the boot procedure, it's messed up quite seriously... ...

 

[jkyoho]

Thanks a lot and I will do more research on this!

The fuse is burnt in the boot0/1 or normal storage area? if it's in boot0/1 and the checks is done by boot processor at the very early stage of the boot procedure, it's messed up quite seriously... ...

EMMC fuse info I believe stores in boot0/1 but SOC has its fuse count as well

 

[jjloki222]

EMMC fuse info I believe stores in boot0/1 but SOC has its fuse count as well

got it. Thanks a lot for all your wonderful knowledge sharing. really appreciate your help!