PX5 stuck in bootloop

[dottorniù]

Hi,
Confirming ctl_wakeup (logic 0) should be pulled to ground and remain pulled low.
Ctl_recovery pulled 3.3v via resistor (logic 1) and stay 3.3v.

I measured all SOM pins today and can't see any reason why the suggested combination doesn't seem to work for you.
Ensure that the SOM is also grounded via 3 mounts.

Also load pX5 USB 'rockchip USB' driver. Even if USB DM and DP are transposed, you should see something happen in the windows device list (code 21 from memory.)

Good morning, after a night to study on this thread are even more confused ....

first question:

in the first post we see the connection scheme between mcu and psu / usb
pins 9,15,20,28 are all 3.3 volts but no one with a logic resistance of 1 or 0 is chosen

In the post instead quoted (if I misunderstood) you say that the pin 20 must be put to 3.3 through a pullup resistor. Quite right? since 10K
and 9.15.28 always pullup?

NOTE: SINK [CTL_WAKEUP) to GROUND (0v) if unit does not power ON

and if 15 goes to direct gnd or through pulldown?

Could you post a picture where you see how you connected the som?

Probably for you and the other people who read the thread are things taken for granted, but before I give tension I prefer clear any doubt .... thanks for your patience.

Second question:
I can no longer find where I had read it but I seemed to have seen among the procedures that a pin was fed only until the som was not recognized by the pc. Can you tell me what it is or if I'm wrong?

Thank you all.

 

[Malle355]

Good morning, after a night to study on this thread are even more confused ....

first question:

in the first post we see the connection scheme between mcu and psu / usb
pins 9,15,20,28 are all 3.3 volts but no one with a logic resistance of 1 or 0 is chosen

In the post instead quoted (if I misunderstood) you say that the pin 20 must be put to 3.3 through a pullup resistor. Quite right? since 10K
and 9.15.28 always pullup?

and if 15 goes to direct gnd or through pulldown?

Could you post a picture where you see how you connected the som?

Probably for you and the other people who read the thread are things taken for granted, but before I give tension I prefer clear any doubt .... thanks for your patience.

Second question:
I can no longer find where I had read it but I seemed to have seen among the procedures that a pin was fed only until the som was not recognized by the pc. Can you tell me what it is or if I'm wrong?

Thank you all.

Hi,

thanks for your help. No it isn't a "pin" that you need to connect. You can try to ground the first four emmc lines with a tweezer or something else like on this Picture.

 

[scorillo_ro]

I attach a picture of my som that has only 72 pins, in the picture I indicated what my test is pin 1 and 72.
Looking for some feedback with the multimeter I have identified the destination of the following pins
71, 72 -> usb rtl8723BU
70, 66, 56, 12, 10, 6, 5, 4 -> GND
17 -> touch int
18 -> touch scl
19 -> touch sda
29, 30 I attach image
So despite my pins are less it seems that the numbering matches with the one posted by Marchnz. Quite right?

For the usb connection 30 is d + and 29 d- right?

Thanks for any answers

What you are referring as pins 71 and 72 are in fact pins 74 and 75. Look at the attached image provided by @marchnz which is cropped from the original schematics and has the supposed required pins highlighted in colors. The attached pdf contains two tables made by me witch may help you identify the pins more quickly. The colors are the same as in marchnz's schematics.

I don't know what 71,72 and 73 from the schematics are, but because they goes to GND, they might be the three mounting holes of the SOM...

Remember that nobody actually succeeded to do the OTG using just the SOM...

Marchnz used an old spare GS motherboard which he modified to make a JIG.
I managed to enter loader mode OTG but not on my bricked SOM but on an functional SOM and using my KLD unit which has an external connector used for OTG. I have not been able to enter maskrom mode by exposing and shorting some paths between emmc and the px5 processor.

Marchnz is waiting for some headers in order to try to simulate his JIG without using his old motherboard, so this is still a work in progress.
But in the end I'm sure that things will become clear and everybody will be able to do OTG on PX5.

Be patient and don't give up, your tests and experience may help all of us.

 

[scorillo_ro]

deleted

 

[scorillo_ro]

I can not insert images, please open them with the links below .... sorry

SOM-->h**p://i66.tinypic.com/oky9oh.jpg

Interesting SOM.
My bricked SOM is version 4 and looks different. I didn't took off the heat sink of my new SOM to see if it is similar to yours, mine is blue, but I'll do it next days.

I wonder what are those pads near the tracks between the emmc and px5 processor... Might be for entering maskrom mode?

 

[marchnz]

Interesting SOM.
My bricked SOM is version 4 and looks different. I didn't took off the heat sink of my new SOM to see if it is similar to yours, mine is blue, but I'll do it next days.

I wonder what are those pads near the tracks between the emmc and px5 processor... Might be for entering maskrom mode?

Might be, keen to try?

 

[dottorniù]

What you are referring as pins 71 and 72 are in fact pins 74 and 75. Look at the attached image provided by @marchnz which is cropped from the original schematics and has the supposed required pins highlighted in colors. The attached pdf contains two tables made by me witch may help you identify the pins more quickly. The colors are the same as in marchnz's schematics.

I don't know what 71,72 and 73 from the schematics are, but because they goes to GND, they might be the three mounting holes of the SOM...

Remember that nobody actually succeeded to do the OTG using just the SOM...

Marchnz used an old spare GS motherboard which he modified to make a JIG.
I managed to enter loader mode OTG but not on my bricked SOM but on an functional SOM and using my KLD unit which has an external connector used for OTG. I have not been able to enter maskrom mode by exposing and shorting some paths between emmc and the px5 processor.

Marchnz is waiting for some headers in order to try to simulate his JIG without using his old motherboard, so this is still a work in progress.
But in the end I'm sure that things will become clear and everybody will be able to do OTG on PX5.

Be patient and don't give up, your tests and experience may help all of us.

Thank you for the exhaustive answer,
now I understand the situation better and I believe that I will proceed to the motherboard mod. I thought marchnz had already tried ...

Interesting SOM.
My bricked SOM is version 4 and looks different. I didn't took off the heat sink of my new SOM to see if it is similar to yours, mine is blue, but I'll do it next days.

I wonder what are those pads near the tracks between the emmc and px5 processor... Might be for entering maskrom mode?

Might be, keen to try?

I had also noticed those two pitches, one goes to GND and the other is connected to the first track of the emmc. I tried to close them with tweezers but nothing happens.

I put other photos so if you see errors maybe give me a suggestion ...
wire colors:
orange 3.3V
red 5V
Black GND
Green DM
White DP
To feed the som I used a psu atx

Meanwhile, the motherboard mod.

jig1 h**p://i68.tinypic.com/23h63yc.jpg

jig2 h**p://i68.tinypic.com/124bqe9.jpg

jig3 h**p://i64.tinypic.com/34qpslu.jpg

jig4 h**p://i63.tinypic.com/15g2t8n.jpg

pad som h**p://i64.tinypic.com/2w4ijqs.jpg

full album link h**p://it.tinypic.com/a/abpe8k/4

 

[scorillo_ro]

PX5 SOM rev8.0 MaskROM

I have the confirmation that those round pads near the tracks between PX5 processor and the eMMC are for entering MaskROM!

 

[scorillo_ro]

PX5 SOM rev4.0 MaskROM

Tracking the MaskROM pads on the rev8.0 SOM I found that exposing the tracks between the PX5 processor and the eMMC is not necessarily needed for the rev4.0 SOM. Take a look at the attached pictures.

 

[scorillo_ro]

full album link h**p://it.tinypic.com/a/abpe8k/4

Your JIG looks very good even if it is not working yet.

Why don't you try to track the OTG pins of the GL850G to see if you already have a hidden OTG connector like my unit have?

PS. If I remember correctly, after 10 posts on the forum you will be able to directly attach pictures. Please upload them here when you'll be allowed.

 

[dottorniù]

update:

raise the pins of the USB hub removed the pin 28 of the som from the motherboard and brought the 3.3 to the pin 28 of the som.

I soldered a USB cable under the motherboard, connected the USB to the PC and fed the motherboard ... but still nothing is detected by PC.

Also in this case I have to try to connect pin 28 to GND?

Is PC detection immediate or does it take a few minutes?
@scorillo_ro

fantastic discovery!

to activate maskrom join pins before power or after?

Thank you very much!

 

[scorillo_ro]

update:

raise the pins of the USB hub removed the pin 28 of the som from the motherboard and brought the 3.3 to the pin 28 of the som.

I soldered a USB cable under the motherboard, connected the USB to the PC and fed the motherboard ... but still nothing is detected by PC.

Also in this case I have to try to connect pin 28 to GND?

I first tried to lift pins 27 and 28 of the GL850 USB hub and directly connect the white and green USB wires to the SOM. Also wired CTL_RECOVERY pin on the SOM in order to connect it to 3.3v. Now I think that this approach was a bad idea!!!
I ended soldering back both pins (27 and 28) because I have found the OTG connector on the back side of my unit.

Is PC detection immediate or does it take a few minutes?
@scorillo_ro

fantastic discovery.

to activate maskrom join pins before power or after?

Tank you very much!

The maskrom mode is activated almost instantly after the unit is powered on with the round pads already shorted. It will take one second or two at a maximum to hear the windows chime for founding new device.

Remember that the first time the device will be discovered, windows will start installing the rockchip drivers for the corresponding USB port. This might take a long time and the SOM will eventually give up waiting for the connection and will exit the maskrom mode, trying to boot from eMMC. Don't worry, windows will finish installing the drivers even if it will report that the device was disconnected. Power off your unit, short the pads, power it on again. This time the device will be properly discovered and you will be in maskrom mode for a little bit more time.

As @marchnz already said, the maskrom mode will be available for only about a minute. My stopwatch showed a minute and about 10 seconds when the device got disconnected from windows. So you don't have enough time to flash a full update.img firmware. You can use one of the sdupdate.img files kindly provided by @SkyBeam2048 on his google drive here. I will suggest to use the one for android 6, but android 8 will also work.

I think @marchnz can provide you more details on how to use the Rockchip Batch Tools software.

Later edit: you should release the maskrom pads after the windows device connected sound

 

[dottorniù]

I first tried to lift pins 27 and 28 of the GL850 USB hub and directly connect the white and green USB wires to the SOM. Also wired CTL_RECOVERY pin on the SOM in order to connect it to 3.3v. Now I think that this approach was a bad idea!!!
I ended soldering back both pins (27 and 28) because I have found the OTG connector on the back side of my unit.



The maskrom mode is activated almost instantly after the unit is powered on with the round pads already shorted. It will take one second or two at a maximum to hear the windows chime for founding new device.

Remember that the first time the device will be discovered, windows will start installing the rockchip drivers for the corresponding USB port. This might take a long time and the SOM will eventually give up waiting for the connection and will exit the maskrom mode, trying to boot from eMMC. Don't worry, windows will finish installing the drivers even if it will report that the device was disconnected. Power off your unit, short the pads, power it on again. This time the device will be properly discovered and you will be in maskrom mode for a little bit more time.

As @marchnz already said, the maskrom mode will be available for only about a minute. My stopwatch showed a minute and about 10 seconds when the device got disconnected from windows. So you don't have enough time to flash a full update.img firmware. You can use one of the sdupdate.img files kindly provided by @SkyBeam2048 on his google drive here. I will suggest to use the one for android 6, but android 8 will also work.

I think @marchnz can provide you more details on how to use the Rockchip Batch Tools software.

Later edit: you should release the maskrom pads after the windows device connected sound

maybe I found the usb-otg pins ... are set to those for the usb internet key.
In the cable provided with the two pins are empty, checking with the multimeter those pins connect to the GL850G pin 8 (usb-DM3) and 9 (usb-DP3)

Also in yours are they in the same place ??

Update:

update:
* I brought the motherboard back as originally
I put a button connected you have pad of the SOM to start masckrom.

I connected the usb to the pc and powered while holding the button pressed but still nothing.

I turned off, unplugged the usb and restarted to check the voltages on the SOM and I saw that pin 15 (CTL_WEKEUP) stands at zero. it's normal?

 

[scorillo_ro]

maybe I found the usb-otg pins ... are set to those for the usb internet key.
In the cable provided with the two pins are empty, checking with the multimeter those pins connect to the GL850G pin 8 (usb-DM3) and 9 (usb-DP3)

Also in yours are they in the same place ??

Update:

update:
* I brought the motherboard back as originally
I put a button connected you have pad of the SOM to start masckrom.

I connected the usb to the pc and powered while holding the button pressed but still nothing.

I turned off, unplugged the usb and restarted to check the voltages on the SOM and I saw that pin 15 (CTL_WEKEUP) stands at zero. it's normal?

My unit is different. It has all 4 pins used by USB: GND, DATA+, DATA- and VCC
https://forum.xda-developers.com/showpost.php?p=75777649&postcount=90

Have you also connected the ACC to 12 volts? I have both 12v wires connected at the same time (red and yellow) when powering on with the mask pads shorted.

LE. DM3 and DP3 are normal USB port. You should look for 27 and 28 (OTG-DM and OTG-DP).

 

[dottorniù]

My unit is different. It has all 4 pins used by USB: GND, DATA+, DATA- and VCC
https://forum.xda-developers.com/showpost.php?p=75777649&postcount=90

Have you also connected the ACC to 12 volts? I have both 12v wires connected at the same time (red and yellow) when powering on with the mask pads shorted.

LE. DM3 and DP3 are normal USB port. You should look for 27 and 28 (OTG-DM and OTG-DP).

Yes, +12 yellow and red, on the other pins tested (1,2,3,4,5,6,9,20) the voltage is correct.

For the first test I used usb 1 that measuring with the multimeter is connected to pins 27 and 28 of GL850 and nothing happened.

Checking the wiring diagrams posted by marchnz I found that the transil diodes D5 and D8 are connected to otg-dp and otg-dm, on the motherboard I found the position of the diodes and from that they went back to the pins of the GL850 and the output ones on the back of the stereo.

Later or tomorrow I will try to connect the USB cable directly on the pads and see if I'm lucky but I start to suspect that the problem is just the motherboard ....

I add the link to a photo that simplifies my explanation, to better understand it should be downloaded and enlarged.

h**p://i66.tinypic.com/2rw0rr6.jpg

 

[marchnz]

I have the confirmation that those round pads near the tracks between PX5 processor and the eMMC are for entering MaskROM!

Brilliant! Nice work.

 

[marchnz]

As @marchnz already said, the maskrom mode will be available for only about a minute. My stopwatch showed a minute and about 10 seconds when the device got disconnected from windows. So you don't have enough time to flash a full update.img firmware. You can use one of the sdupdate.img files kindly provided by @SkyBeam2048 on his google drive here. I will suggest to use the one for android 6, but android 8 will also work.

Hi,
What happens is that the MCU watchdog tries to reset the SOM by powering it off/on, when it doesnt receive a response from the SOM. I found a workaround, to keep the SOM powered up during the process.

The workaround: U22 [MP584] add a temporary 470Ohm resistor between Pin 2 of U22 (enable) and +3.3v.

This will keep the SOM powered ON.

-M

 

[marchnz]

I have the confirmation that those round pads near the tracks between PX5 processor and the eMMC are for entering MaskROM!

Hi,
See this thread for making a simple OTG cable from a USB cable and two resistors.....

https://forum.xda-developers.com/an...px3-5-headunit-mod-recover-flash-som-t3766892

 

[scorillo_ro]

Hi,
See this thread for making a simple OTG cable from a USB cable and two resistors.....

https://forum.xda-developers.com/an...px3-5-headunit-mod-recover-flash-som-t3766892

Very nice!

 

[adioperator]

Hello friends,
Today I got into trouble.
I wanted to root to my Android 8.0 drive using the method:

ROOT rights for Oreo based on GS firmware from 20180118

Receiving the root only through the loading card. The load map contains a modified recovery, and the update.zip update is modified by the boot.

Download archive
Yandex - https://yadi.sk/d/pFSNkK4_3SwNx5
Google - https://drive.google.com/file/d/1xpzsUg1EAI6Am4lfw1rUNDcGmaEPfsjd/view
The archive contains the following folders:
Androud-8-sdupdateROOT
CreateCD
SD Formatter
UpdateZIP-patch-root
- In the CreateCD folder, create a bootable card using SD_Firmware_Tool (the program is configured, no checkboxes are set)
- Image to create a map from the folder Androud-8-sdupdateROOT
- After creating the boot card, delete sdupdate.img from the card
- Copy from the UpdateZIP-patch-root folder to the root of the card's SD update.zip
We insert the card into the recorder, reboot, install the update from the SD Card
Remove the card and reboot the radio.
Map formatted with the program from the folder SD Formatter
The root system / app script writes to MagiskManager-v5.6.1.apk

After install the update from the SD Card, remove the card and reboot the radio, does not load Android and sometimes enters Recovery Mode like in video
https://www.youtube.com/watch?v=llz_H-6Vw2M

My head unit model:
https://photos.app.goo.gl/6Mj3SYc1vJD5KRY72

What should I do now?
Thanks!