Stuck in standby mode: Faulty hall effect sensor?

[abarthch]

My PW2 works fine half of the time, but sometimes it's just stuck in standby mode and I'm unable to wake it up with the power button; which works fine because I can hold it a few seconds longer and the PW reboots.

I made a complete reset and also updated to the latest FW. The only solution (which usually never works) is grabbing the device and hitting it on a hard surface. Works every time. Sometimes I can also reenter the standby mode with the same method.

This led me to believe that the hall effect sensor is somehow "stuck". Because when the problem occurs, even a reboot doesn't fix it. It reboots and just reenters standby mode instantly. As if my PW was in a closed case all the time.

Because I'm a dummy and updated to the latest FW I can't jailbreak it and disable the sensor. But since I don't care for the magnetic lock feature anyway, do you think desoldering the sensor would solve my problem or mess up the device completely?

[knc1]

There is nothing to 'stick' in a Hall Effect sensor - google about the device will show you why.

Best guess - 'completely mess up the device'

Before taking any sort of drastic action -
dump and read the system messages - they might tell you 'why' the device is acting the way it is.

Best chance with the messages would be after it 'stuck' and then was 'unstuck' but before re-booting the device.

Use search bar shortcut:
;dm
Yes, the semi-colon is part of the command.

[abarthch]

Thank you for your help, knc1.

Here's the dump after the routine (I resetted the device before just to make sure).

When I Ctrl-F for 'error' I get 282 results But I also get this a few times:

151031:212957 powerd[3498]: I def:magclose::Mag sensor closed event
151031:212957 powerd[3498]: I def:metrics::prg_name=powerd,prg_source=active_handle_mag_sensor,name=GoingToScreensaverDueToMagSensorClose,value=1

And yes, I know the hall sensor doesn't work mechanically, but I still have a strong intuition that something is wrong with the magnetic lock feature. Is it maybe possible that the sensor is loose on the PCB and hitting the device causes it to make a brief connection?

[knc1]

" I know the hall sensor doesn't work mechanically "
Your the one who wrote that it might be stuck -
I can only answer based on what you write, my crystal ball is off-line and I gave up channeling into other people's minds for Lent.

Control-f for error?
Is this a PW-1?
How are you pressing a ctrl-f?
When, what, why . . .
https://www.mobileread.com/forums/sho...d.php?t=272087

[abarthch]

I'm sorry, forgot to mention I searched for 'error' on my PC, not on the PW.

Oops, just updated my profile details. But it's in fact a PW2, the title is correct.

I just removed the outer frame to uncover the sensor. It looks fine but when I put pressure on it, my PW instantly wakes up from standby. When I release the pressure, standby comes back on. I don't think that's supposed to happen, unless my small finger is magnetic (No, I don't have a prostetic).

[knc1]

Originally Posted by abarthch

I just removed the outer frame to uncover the sensor. It looks fine but when I put pressure on it, my PW instantly wakes up from standby. When I release the pressure, standby comes back on. I don't think that's supposed to happen, unless my small finger is magnetic (No, I don't have a prostetic).

You where correct, a loose connection.

If you have, or have access to, a hot air soldering system, first try to just 're-flow' the existing solder without re-moving the part.

Careful where the hot air goes - you don't want to un-solder anything nearby.

[abarthch]

Hey knc1, just an update to let you know how it worked out.

I finally got my hands on a hot air soldering system with a small opening (about Ø4mm) but as much as I tried to be careful I still melted the surroundings a little bit (the transparent LED stripe and the black part of the housing). Maybe it would work when the screen and the LED stripe are separated from the housing but I didn't want to take the risk in doing so.

But what worked was placing a tiny amount of tin next to the loose pin of the sensor (in my case the bottom right pin) and holding the tiniest SMD soldering iron I could find next to it. It didn't work right away, since there's a white plastic sheet that covers the pins. But after 3-4 attempts the tin made it's way around the sheet and now my Kindle finally behaves normal.

Thank you once again for helping me with one of my devices. If it weren't for you, i probably would have just desoldered the sensor and ruined my PW

[geekmaster]

Originally Posted by abarthch

I finally got my hands on a hot air soldering system with a small opening (about Ø4mm) but as much as I tried to be careful I still melted the surroundings a little bit (the transparent LED stripe and the black part of the housing). ...

It is customary to place a heat mask (aluminum foil with a one or more holes cut into it) over the circuit board, taped in place to prevent airflow migration, for DIY hot air repairs.

Here is an example using aluminum tape instead of foil (but for a small device like a kindle, I would cover everything including all plastic, except for the component to heat/reflow):

[knc1]

Originally Posted by abarthch

- - - - -
Thank you once again for helping me with one of my devices. If it weren't for you, i probably would have just desoldered the sensor and ruined my PW

Any of this sort of repair work is tricky.
One of the reasons SMD re-work shops can charge such high dollars.

Just hope you never have to replace that SoC chip -
your probably talking a second mortgage on the house.