I have been playing with my 2018, 2020 and 2022 HD 8's, and made a few interesting discoveries. I had been posting in another thread, but it seems to be pretty dead, and I think this stuff is interesting enough to warrant its own thread.
1. There are UART pins on the USB-C connector of the 2020 (and I think the 2022) HD 8 that appear to be active - but are not readily available due to the general pinout of USB-C cables.
I had a feeling that there had to be some debugging ports somewhere on the device, and given that every other avenue seems to be blocked by the big A - I started testing. I initially started with a USB-C to USB-A OTG adapter, and got some signal that looked UART, but couldn't seem to get the baud right. I didn't really want to wreck my device, but the kids were done with that one - so I sacrificed the 2020. I removed all of the shielding and started probing interesting looking stuff. This lead me to the discoveries that I have made.
After removing all of the shielding on the 2020 board I found an interesting row of pins that I eventually traced to the USB-C pins. I built a cable to see what interesting information they might put out - but fried my USB/TTL adapter on accident before I got the cable done (waiting on a new one to arrive).
2. There are UART pins in the headphone jack that are disabled by the removal of a single component. I believe that this component is a Zener Diode. This disabled interface is present in the 3 models that I have - I am unsure about others.
I was lead to this discovery by comparing the photos of a development device that was posted here on XDA by Lokio27. I noticed the missing component, recognized the component cluster on my other devices logic boards. I feel that if the big A thinks it necessary to remove this functionality from production devices, it is my duty to try to restore it!
There are hardwire connections in the same component cluster that are connected to RX0 and TX0 on the back of the board.
I am working on the assumption that this functions similar to other 3.5mm jack UART interfaces. Google has used them, and I think some MTK devices have used them as well. With the design, and that history I think that 3.3 volts should be applied to the sleeve, then the tip and the ring should be TX0 and RX0.
I am having some trouble getting the interface to turn on - When I power it (either with internal or external power), it shorts the entire board to ground. I don't know if this is a failsafe, or if I bunged up the board with all of my probing.
Also of note - I shorted every test point that I could find and attempted to boot the 2020 device, hoping for BROM. Sometimes it did nothing, booting normally, sometimes it would get stuck in preloader, requiring a battery disconnect to reset, and sometimes it would literally block any attempt to boot the device as long as I held the short( this happened with the CLK pin)
There are photos below with relevant annotations on the photos. I am not a hardware guy - heck I am not a software guy either. I just have interest and the internet. I am sharing in hopes that someone with more knowledge and expertise that I will take what I have learned and run somewhere interesting with it. I have a feeling that if we can get the 3.5mm interface working that there might be some very powerful tools available to us(maybe a root console?)
If not, I will continue to **** around with it until I either destroy all of my devices, or move on to something else that piques my interest. If nothing else I have learned a ton over the last couple of weeks.
My plan as of right now is to wait until I get my TTL/USB adapter. I will test the USB-C interface 1st - If that has priveleged access it will be a much easier method of entry.
I then plan to continue working on the 3.5mm interface. I'd like to get it up and running and see what is hiding there.
At this point - I have a 2018 logic board that has been probed pretty heavily and is no longer functional. A 2020 logic board that has been probed pretty heavily and is no longer functional. Three 2020's that are still in use by the kids, but can be comandeered for less risky testing. A 2022 HD8+ that I really don't want to brick, but given what I paid for it - I can afford to take some risk. If anyone has suggestions on where to go from here, or additional info please let me know.
1. There are UART pins on the USB-C connector of the 2020 (and I think the 2022) HD 8 that appear to be active - but are not readily available due to the general pinout of USB-C cables.
I had a feeling that there had to be some debugging ports somewhere on the device, and given that every other avenue seems to be blocked by the big A - I started testing. I initially started with a USB-C to USB-A OTG adapter, and got some signal that looked UART, but couldn't seem to get the baud right. I didn't really want to wreck my device, but the kids were done with that one - so I sacrificed the 2020. I removed all of the shielding and started probing interesting looking stuff. This lead me to the discoveries that I have made.
After removing all of the shielding on the 2020 board I found an interesting row of pins that I eventually traced to the USB-C pins. I built a cable to see what interesting information they might put out - but fried my USB/TTL adapter on accident before I got the cable done (waiting on a new one to arrive).
2. There are UART pins in the headphone jack that are disabled by the removal of a single component. I believe that this component is a Zener Diode. This disabled interface is present in the 3 models that I have - I am unsure about others.
I was lead to this discovery by comparing the photos of a development device that was posted here on XDA by Lokio27. I noticed the missing component, recognized the component cluster on my other devices logic boards. I feel that if the big A thinks it necessary to remove this functionality from production devices, it is my duty to try to restore it!
There are hardwire connections in the same component cluster that are connected to RX0 and TX0 on the back of the board.
I am working on the assumption that this functions similar to other 3.5mm jack UART interfaces. Google has used them, and I think some MTK devices have used them as well. With the design, and that history I think that 3.3 volts should be applied to the sleeve, then the tip and the ring should be TX0 and RX0.
I am having some trouble getting the interface to turn on - When I power it (either with internal or external power), it shorts the entire board to ground. I don't know if this is a failsafe, or if I bunged up the board with all of my probing.
Also of note - I shorted every test point that I could find and attempted to boot the 2020 device, hoping for BROM. Sometimes it did nothing, booting normally, sometimes it would get stuck in preloader, requiring a battery disconnect to reset, and sometimes it would literally block any attempt to boot the device as long as I held the short( this happened with the CLK pin)
There are photos below with relevant annotations on the photos. I am not a hardware guy - heck I am not a software guy either. I just have interest and the internet. I am sharing in hopes that someone with more knowledge and expertise that I will take what I have learned and run somewhere interesting with it. I have a feeling that if we can get the 3.5mm interface working that there might be some very powerful tools available to us(maybe a root console?)
If not, I will continue to **** around with it until I either destroy all of my devices, or move on to something else that piques my interest. If nothing else I have learned a ton over the last couple of weeks.
My plan as of right now is to wait until I get my TTL/USB adapter. I will test the USB-C interface 1st - If that has priveleged access it will be a much easier method of entry.
I then plan to continue working on the 3.5mm interface. I'd like to get it up and running and see what is hiding there.
At this point - I have a 2018 logic board that has been probed pretty heavily and is no longer functional. A 2020 logic board that has been probed pretty heavily and is no longer functional. Three 2020's that are still in use by the kids, but can be comandeered for less risky testing. A 2022 HD8+ that I really don't want to brick, but given what I paid for it - I can afford to take some risk. If anyone has suggestions on where to go from here, or additional info please let me know.
