Reply
Thread Tools
Posts: 8 | Thanked: 0 times | Joined on Nov 2009 @ Los Angeles, CA, USA
#11
I vote for Hall effect sensor, usually you can hear a tiny ping when a reed switch operates. A compass brought near the N900 shows strong fields near the ends of the device, particularly the end near the headphone jack. But this could also be the motor for the vibrator.

The camera door is sensed by a optical sensor that triggers on the reflective white patch on the cover.
 
Posts: 21 | Thanked: 13 times | Joined on Dec 2009
#12
Originally Posted by MrGrim View Post
You realize it's impossible for software to distinguish parasitic magnetic fields from the useful ones, right? Don't expect your bug to be fixed too soon
If you actually read the bug, you'd see it was a request for an option that opening the keyboard tray not wake up the device, which it does currently.

If the software is changed to not wake up the device when the tray is opened, it presumably will not wake up when a parasitic magnetic field passes through it, either.

But as for parasitic vs. non-parasitic fields -- it would actually be possible to tell them apart if they were modulated by the hardware, e.g., frequency detection. Instead of a fixed magnet, an electromagnet would emit a small signal which the Hall Effect device could detect and demodulate to distinguish it from background noise.Of course emitting the modulated electromagnet would use up more power.

(OT: I once submitted a bug against a paper shredder company whose shredder thought that the bin was full whenever any light passed through the bin window, and refused to shred. I had to tape up the window to keep it from picking up parasitic light. Had they simply used a modulator-demodulator for the LED transmitter-receiver pair, they could distinguish parasitic light from their own LED's.)

Last edited by leek; 2009-12-08 at 01:23.
 

The Following User Says Thank You to leek For This Useful Post:
YoDude's Avatar
Posts: 2,869 | Thanked: 1,784 times | Joined on Feb 2007 @ Po' Bo'. PA
#13
Originally Posted by leek View Post
If you actually read the bug, you'd see it was a request for an option that opening the keyboard tray not wake up the device, which it does currently.

If the software is changed to not wake up the device when the tray is opened, it presumably will not wake up when a parasitic magnetic field passes through it, either.

But as for parasitic vs. non-parasitic fields -- it would actually be possible to tell them apart if they were modulated by the hardware, e.g., frequency detection. Instead of a fixed magnet, an electromagnet would emit a small signal which the Hall Effect device could detect and demodulate to distinguish it from background noise.Of course emitting the modulated electromagnet would use up more power.

(OT: I once submitted a bug against a paper shredder company whose shredder thought that the bin was full whenever any light passed through the bin window, and refused to shred. I had to tape up the window to keep it from picking up parasitic light. Had they simply used a modulator-demodulator for the LED transmitter-receiver pair, they could distinguish parasitic light from their own LED's.)
I believe it does have something to do with how Nokia determines if the devices keyboard opens. The same thing occurs with the N810 but not the N800.
The N800 had an optional leather case/cover that used magnets. When I tried this thing on the N810 it would wake it up as soon as the cover was lifted. Not so with the N800.
At first I thought the dang light sensor had something to do with it.
Perhaps someone can find the code in an N810 and a work around might be the same for the N900.
__________________

SLN member # 009
 

The Following User Says Thank You to YoDude For This Useful Post:
Posts: 21 | Thanked: 13 times | Joined on Dec 2009
#14
@YoDude: I tested it, and it's definitely not the proximity or light sensor, but is magnetic. Like you, I hope someone came up with a workaround for the N810 which can be transferred to the N900, but I wouldn't hold my breath

I was shopping in the store tonight, replacing my nice leather magnetic-clip case with a cheaper-but-okay slip-in case. 95% of the cases in the store were magnetic, so I had to keep searching

I'll probably eventually get a Piel Frama or similar.

Correction: Piel Frama is also magnetic. S**T!!!

Last edited by leek; 2009-12-08 at 08:02.
 

The Following User Says Thank You to leek For This Useful Post:
meyma's Avatar
Posts: 127 | Thanked: 30 times | Joined on Oct 2009
#15
It may help to insert a piece of iron plate as a shield. Not even sophisticated but better than always empty battery. Care that you don't shield the GSM antenna area (see the manual)!
 
Posts: 329 | Thanked: 142 times | Joined on Oct 2009
#16
A hall effect sensor needs a constant current flowing through to work. This would surely kill the battery quickly, and be a useless drain. I'm betting it's a small coil, which generates an impulse every time a magnet is moved around it. This makes it impossible to modulate the impulses, to distinguish between internal and external fields.
Even if modulation were feasible, it would surely require some hardware changes. So i'm afraid no fix from nokia on existing devices.
A metal plate should reduce the device's sensitivity, but possibly not enough to block it out. The device will also only lose gsm signal if it's completely encased in metal
__________________
I don't mean to crush your hopes, I just can't help it
 
Posts: 21 | Thanked: 13 times | Joined on Dec 2009
#17
Even a coil requires a circuit to detect and amplify its signal, which requires no more current than a Hall effect circuit.

Nokia may not fix the hardware, but the firmware/software can be fixed to avoid turning on the device when the sensor thinks that the keyboard tray opens.

That is, for all intents and purposes, a fix.

The other possibility is hacking your device by opening it up and cutting a wire or trace, thus stopping the sensor from working.
 
Posts: 21 | Thanked: 13 times | Joined on Dec 2009
#18
Apparently Nokia agrees

Please add your votes!!!
 
Posts: 39 | Thanked: 9 times | Joined on Oct 2009 @ USA
#19
I voted for this bug on bugzilla. My nice new case will be useless unless a fix is made available in software. I really hope Nokia adds an option.

Last edited by kwurk; 2009-12-27 at 05:21.
 
YoDude's Avatar
Posts: 2,869 | Thanked: 1,784 times | Joined on Feb 2007 @ Po' Bo'. PA
#20
The software solution to this hardware bug is to have an option to "Not wake on keyboard" or "only wake from switch" or somesuch.
__________________

SLN member # 009
 
Reply

Tags
case, cases, hardware, keyboard switch, magnetic closure, magnetic latches, magnetic monopoles attack, magnetic switch, magnetic switches, magnets, magnets wake device, magnets wake n900, n900, n900 needs hard case, switch, switches

Thread Tools

 
Forum Jump


All times are GMT. The time now is 19:22.