On 2008-04-09, ~V~ <V.37mmtt DeleteThis @no.email.invalid> wrote:
>
> solder an Optocoupler to a mobile phones vibrator leads and connect it
> to the Computers power switch wires
>
> [image: http://img172.imageshack.us/img172/3347/optovc6.jpg]
>
> Pin 1 to the positive lead of the vibrator
> Pin 2 to the negative lead of the vibrator
> Pin 3 and 4 to Power Switch wires
>
> when the Phone's Vibrator is activated, the Opto is turned On, and Pins
> 3 and 4 are Bridged - same as pressing the Power Button
Do you realise how old this thread is? It's basically dead now.
In any case it isn't quite as simple as simply wiring up an
optoisolator in the manner you describe. For a start you need a
current limiting resistor in series with the LED to prevent thermal
runaway, and the output is polarised as well as the input. All
the machines I've examined in this regard have one side of the
switch linked to ground - this is probably specified in some standard
somewhere. The phototransistor emitter goes to that pin, the
collector to the other pin in an open collector arrangement.
In addition, any mobile phone I've seen doesn't constantly vibrate
- it pulses the vibrate unit to better get your attention. Those
pulses need to be bridged so that you are not rapidly turning the
system on and off. In the real world you are also going to need
to leave the phone ringing for several seconds to be sure that the
handset has actually started ringing - in modern phone systems,
and especially cellular, the ring is generated locally and the fact
that it is ringing at your end is no guarantee that it is ringing
at the remote end.
However, the power on switch is intended for momentary activation
and not being held for sevaral seconds - try this with a desktop
machine, press the power button in and hold it. On many machines
the system will start up for a few seconds and then power off again
because you are still holding the button. What this means is that
you need to convert the level into a pulse with a maximum length
of maybe a second.
If you look at the schematic I posted in response to this problem
at
http://andrews.freeshell.org/news/20080125.achh/pwr_ctrl.pdf
you will see that the majority of the circuitry is performing this
kind of signal conditioning - effectively it is everything to the
left of R4. Sure, you could drop an optoisolator into the circuit
in place of the transistor but you aren't saving anything in terms
of complexity and the additional isolation is not needed in this
context.
Incidentally, the central portion of that schematic, with slightly
different values, would have neatly solved that problem posed by
another poster more recently, enquiring about schematics for a
computer on/off button, and with much less complexity than he
eventually adopted. I would have pointed this out at the time
until I realised the reason that I recognised his email address
was that he had been such a jerk over on comp.arch.embedded.
--
Andrew Smallshaw
andrews DeleteThis @sdf.lonestar.org
>> Stay informed about: turn on remote pc through mboile 
FAQ
Profile
Private Messages
Log in
