Screen freeze or black out-- this problem is strange!

[Hook 3,021]

MY Pro1 screen was replaced in Jan 2021 by FxTec and has worked perfectly until about 2 days ago,  This is so weird that I have to give a very thorough description,

Suddenly out of the blue either the screen will stop responding to any touch—however, keys still work and you can, for instance, select apps with arrow keys—or the screen goes totally black.  When the screen goes black, it as if the Pro1 is off (powered off) except the keyboard is backlit  When it first occurred a power-button reset would fix it.  By power-button reset, I mean holding the power button until the Pro1 spontaneously resets.  Since the screen doesn't respond in either case, it is the only way to do a reboot.  Now, however it seems to take several such reboots to get things back.  When things work, everything works, so it's not like there's a problem with the LCD or digitizer.

It seems to only happen when it rides in my front left jeans pocket (nothing else in there). However, once that has happened, it can spontaneously happen after I reboot and it is fine. but will also just work for long periods of time IF I don't carry in in my pocket. Opening and closing the Pro1 doesn't seem to affect it one way or the other.

My Pro1 may be out of warranty or nearly out of warranty depending on what date they re using.  It was sent to me in November 2019, but was blocked by customs (I am in the US), so they sent me another which reached me on December 10, 2019.  Right now, it is working fine, but it is happening more ad more when it spends any time in my pocket. Since I still stay at home a lot, I can keep it out of my pocket most of the time, but clearly I want the problem solved.

Anyone have a clue.  I suppose I will open it up and check the LCD cable seating, but I really don't think that is it.  Why would being in a pocket affect it when josling out of the pocket doesn't? Some kind of static build-up?  I've put it into the FxTec leather pouch and the same thing happens.  I've tried pressing on the case and nothing happens, so it isn't pressure whilr in the pocket.

There seems nothing wrong with either the LCD or the digitizer when they work.  She doesn't hot hot or anything else alarming.

There are too few AICP-R users for me to be sure it isn't some strange AICP-R problem, but I doubt it.  The test would be to flash Lineage, but I can't chance the touchscreen going dead while flashing.

When she's good she's very very good.  When she's bad, there's nothing you can do but reboot.  Maybe several times.  😄

Any clues?

I will likely run it by FxTec support, but I'm still studying what makes it happen when.

[Rob. S. 1,661]

Now that you're reporting this, I've had a few (very few) situations over the last few weeks which were slightly similar, but not as severe. Screen going black while in use, needing a press of the power button to come on again, or screen becoming touch-insensitive until switched off and on again with the power button, and there might have been one occasion or so in which such a situation needed a reboot, too. I'm on LineageOS 18.1, and for me this somehow rather felt more like software rather than hardware, but that's just an impression.

[Hook 3,021]

Well, it has now become a more or less permanent situation (not touch screen, not black out) Many reboots will sometimes get it back, but it seems to spontaneously go back to no touch in little time.

I'll check the screen cable, but if that does nothing, I'll see what FxTec can do for me. If that doesn't work, I'll give my local shop a challenge.  😄

[EskeRahn 5,460]

My guess would be something mechanical too. Since related to the pocket.

Either a cable, or perhaps a superthin crack-line of a print plate.

You could try some (gentle) flexing of the botom and top parts and see if you can trigger it. And perhaps revive when the issue is there.

[Hook 3,021]

3 minutes ago, EskeRahn said:

My guess would be something mechanical too. Since related to the pocket.

Either a cable, or perhaps a superthin crack-line of a print plate.

You could try some (gentle) flexing of the botom and top parts and see if you can trigger it. And perhaps revive when the issue is there.

I tried flexing and also running my fingers over the sides, but nothing seemed to be affected either way.

[Hook 3,021]

Opened the display up (removed the 5 screws on the back and pried open.  Everything looked good as far as I could see.  Cable loked new, connections solid, clean as a whistle in there (why are whistles clean?  I would think they'd be germy 😉 )

One curious thing.  When I went to re-screw, the top left (landscape) screw hole was somehow magnetized.  It kept repelling the screw.  I got out one of my replacement screws from China thinking it was the screw that had become magnetized, but that screw was repelled to.  I did finally fight it in and screwed it down, but I don't remember anything like that when I opened up the display a year ago when I had ghost touches.  WTF?

EDIT: Hahahaa I'm pretty sure I can now explain the magnetized screw hole.  It;s the one closest to the magnetic charging nub when the keyboard is closed. 🤣

I'll contact FxTec just to see what they say and to find out my warranty status.  I guess I'll have to try one of the screen replacements from AlieExpress (I have 2 of them 😄 )

Edited November 13, 2021 by Hook

[VaZso 1,998]

11 hours ago, Rob. S. said:

Now that you're reporting this, I've had a few (very few) situations over the last few weeks which were slightly similar, but not as severe. Screen going black while in use, needing a press of the power button to come on again, or screen becoming touch-insensitive until switched off and on again with the power button, and there might have been one occasion or so in which such a situation needed a reboot, too. I'm on LineageOS 18.1, and for me this somehow rather felt more like software rather than hardware, but that's just an impression.

My original screen became having ghost touches after nearly a year, so I have replaced it but only used double-sided tape for fixing which were too narrow letting the bottom part to start moving a bit by time. At bottom part, there is a ribbon cable coming from the display itself to the back side.

So my second screen became also faulty after a few months but in a bit different way.
There touch function was not affected but sometimes my display has started became brighter and brighter to white colour, but turning the display on and off has helped.
Later, it has occasionally started developing random pixels (in random colours) in my display and in a few seconds it became full of random pixels.
Only a reboot has helped.

...then It has reached a point where it did not work normally, only random pixels have appeared right after turning the phone on, it has became "static".

I have replaced my display the second time and used another glue which can be disassembled when necessary but holds better than double-sided adhesive tape I used earlier.
So far it works well, I think it will exceed the working time of my original display soon if it has not already exceeded it.

Basically when display is powered on, it should has an init sequence to start working (also orientation, color mode and other parameters may be initialized).
It may also has some working modes like enable/disable display refresh and if it interprets an erroneous display off command, it may become blank while the phone sends a display on command on turning the display on (theoretically).
...but init sequence is most likely only applied once at boot time (maybe by boot loader) and if display looses power connection(s), init sequence would be needed to function again.

I have a feeling that the root cause of the symptoms above may be the same as ghost touches which is the damaged bottom cable of display - at least that is how I see it.

5 hours ago, Hook said:

Hahahaa I'm pretty sure I can now explain the magnetized screw hole.  It;s the one closest to the magnetic charging nub when the keyboard is closed. 🤣

I think there is a magnet nearby what Pro1 use to identify if its keyboard is in opened or closed state.

[3zet 100]

Screen is disconnecting from its connector. Which one? I don't know. Unplug, brush both plugs and sockets very gently with fiberglass brush, reassemble. Pins just oxidized, had the same issue and also very bad LTE signal, brushing connectors helped a lot. 

[Hook 3,021]

I emailed support. They are going to send me a new screen assembly under warranty (they honored my two years and accepted my date of reception (Dec 10th, 2019) as the mark on the wall.  Of course, it will be a long while before I get it (probably not until the Pro1x is ready would be my guess), but I have to say my experiences with FxTec CS have been excellent!

[VaZso 1,998]

8 hours ago, 3zet said:

Screen is disconnecting from its connector. Which one? I don't know. Unplug, brush both plugs and sockets very gently with fiberglass brush, reassemble. Pins just oxidized, had the same issue and also very bad LTE signal, brushing connectors helped a lot. 

Be aware brushing connectors using fiberglass brush is not always a good idea.
If connector (or PCB) has golden plating then you will remove this thin layer and the problem will reappear much sooner.
The same applies for connectors of PC RAM for example.

To remove oxidized parts, it is much better to use a piece of paper similarly like a sandpaper which also removes oxids but not as destructive as fiberglass brush.

[zaptastic 3]

I'm seeing log messages like these. There's about 122 of them since I booted the Pro1-X.

08-11 08:54:32.294   594   600 E android.system.suspend@1.0-service: Error opening kernel wakelock stats for: wakeup11: Permission denied
08-11 08:54:47.232   240   240 W kworker/u16:2: type=1400 audit(0.0:2513): avc: denied { kill } for capability=5 scontext=u:r:kernel:s0 tcontext=u:r:kernel:s0 tclass=capability permissive=0
08-11 08:54:47.440   594   594 W Binder:594_1: type=1400 audit(0.0:2514): avc: denied { read } for name="wakeup48" dev="sysfs" ino=57280 scontext=u:r:system_suspend:s0 tcontext=u:object_r:sysfs:s0 tclass=dir permissive=0
08-11 08:54:47.440   594   594 W Binder:594_1: type=1400 audit(0.0:2515): avc: denied { read } for name="wakeup4" dev="sysfs" ino=34968 scontext=u:r:system_suspend:s0 tcontext=u:object_r:sysfs:s0 tclass=dir permissive=0

There is only one app with a waitlock at this time.

$ adb shell dumpsys power | grep -i wake_lock
    no_cached_wake_locks=true
  DOZE_WAKE_LOCK                 'DreamManagerService' ACQ=-9m5s596ms (uid=1000 pid=1547)

I turned off doze. It's still too soon to tell but the "stuck blank screen" issue has not happened yet.

$ adb shell dumpsys deviceidle disable
Deep idle mode disabled
Light idle mode disabled

 

Edited August 11, 2022 by zaptastic

[zaptastic 3]

A day has past and my phone did not get stuck with the screen blank now that doze has been turned off as a work-around.

[dlmarquis 20]

I think my Pro1X just did this a few minutes ago. Suddenly appeared powered-off and unresponsive, had to try holding the power button a couple times before it woke back up. I checked dmesg and saw a huge number of these lines:

 

[ 2836.383934] mas_ioctl: cmd=0x80044d22
[ 2836.383941] mas_ioctl: end.
[ 2836.394128] mas_ioctl: cmd=0x80044d22
[ 2836.394137] mas_ioctl: end.
[ 2836.404314] mas_ioctl: cmd=0x80044d22
[ 2836.404323] mas_ioctl: end.
[ 2836.414717] mas_ioctl: cmd=0x80044d22
[ 2836.414726] mas_ioctl: end.

 

I have no idea what this is referring to.

Edited August 25, 2022 by dlmarquis
Codeblock markdown fail