The basic idea is that when the unit is off you have 120v potential - you can pass many milliamps without brightening the lamp. When the load is on there is still a voltage potential across the triac. If that potential is not enough (usually these devices need /very/ little current, probably less than a mA) then you can add a very low resistance wire to add a bit more voltage with lower current lamps. Either way, since the circuit draws only a few milliamps, you can simply use a zener and resister across the dropping element (the triac in this case). When off the zener forces the resister to drop hundreds of volts, but at only a milliamp that's still less than a 1/2 watt resister. When on the resister may only be dropping a few volts. In really cheap circuits the you don't use a bridge, the zener simply passes all the current of one half the wave through the resister. The capacitor keeps the circuit running. The touch element, when there is a wire for people to touch in such a switch, is capacitively coupled to the circuit. Incidently, the X-10 radio receivers have an antenna that is capacitively coupled - the antenna ends in the plastic of the case, and on the other side of the plastic wall is a wire, so the case itself is the capacitive element. Some people have modified them for greater range by using a capacitor soldered between the antenna and the wire. Hope the cap is mains rated, and fail 'safe' (open). -Adam roines reenig wrote: >I agree with your observation about the need for a resistive load. I'm having difficulty getting my head around how the control circuit (some type of CPU) controls the load. so in series: > >live_in --- control --- live_out --- load --- neutral > >i think that control probably has to have a triac that it is phasing on and off in order to dimm/turn on/off the load. at the same time it's got to use that same line to power itself. i can't get my head around that part. i think i'll do some searching for how dimmers power themselves from the live line. hopefully one of them has a CPU in them so then it'll all be clear to me. > >thanks, >roines > >William Chops Westfield wrote: >On Wednesday, May 26, 2004, at 18:39 US/Pacific, roines reenig wrote: > > > >>I think we have drifted a little bit from what I attempted to describe >>initially. >> >> > >I was wondering about that. You were mostly interested in how to get >power at (for instance) a lightswitch, when you only have access to the >"hot" side of the connection, and the load, right? > >Most things that do this (ie X10 dimmer switches, or regular dimmers, >for that matter) rely on the load being essentially a slightly >resistive path to the return (a 100W lightbulb has a resistance of only >130 ohms or so, so it's pretty easy to get 50mA worth of 5V out of >there somehow. >That's why they "only" work on resistive loads. Sneaky, eh? > >BillW > >-- >http://www.piclist.com hint: The list server can filter out subtopics >(like ads or off topics) for you. See http://www.piclist.com/#topics > >--------------------------------- >Do you Yahoo!? >Friends. Fun. Try the all-new Yahoo! Messenger > >-- >http://www.piclist.com hint: The list server can filter out subtopics >(like ads or off topics) for you. See http://www.piclist.com/#topics > > > > > -- http://www.piclist.com hint: The list server can filter out subtopics (like ads or off topics) for you. See http://www.piclist.com/#topics