- The power on/off button on the Slug generates an interrupt which, broadly speaking, calls the shutdown procedure.
But if one has hacked the slug to always boot upon power-on (http://www.nslu2-linux.org/wiki/HowTo/ForcePowerAlwaysOn Alternative 11) the power button is redundant, and its circuitry can be used to cause interrupts for other nefarious purposes - you simply need to rewrite the shutdown procedure to do whatever you want.
But just arranging for your circuitry to pull down what the button used to pull down is not good enough - for a start the slug will no longer boot and moreover the slug seems not to want to obey code when that signal is Down (and the chip I am using to generate the interrupt insists on maintaining that Down until it is reset by the program).
So I need to insert a capacitor to do the pulling down so that it comes up (with time constant derermined by the 10K R13 (see http://www.nslu2-linux.org/wiki/Info/PowerSwitchingSchematic). And indeed that works. I make a contact that triggers the interrupt, my "shutdown" procedure notes the interrupt, resets it, and does other useful stuff. And when I release that contact I get another interrupt which is duly noted.
The problem is the size of the capacitor needed to do the job. I find I need 1uF, which is Huge considering what I am trying to do, and it results in a long recovery time before I can repeat the process. I had originally calculated that 0.1uF would be more than ample.
And so I deduce that the capacitor C3, which sits across the old power button, and which is effectively in series with my 1uF, must be comparable with the 0.1uF I had intended to use. And to design a proper circuit, I really need to know the value of C3.
So my question is: "What is the value of C3?"
It is not given in http://www.nslu2-linux.org/wiki/Info/PowerSwitchingSchematic, and I can see wyt because, if you peer at it through a large magnifying glass, it appears to be totally devoid of markings.