>> don't you think you need to provide a DC path for the LEDs in the opto coupler to conduct?
No. Think outside the box. Look at what C1 is doing at the supply frequency.
>> the resistor would be across the two back to back LEDs inside the opto coupler
Again no. This particular opto only has one LED.
>> even use a series resistor (1Meg or more) to any input as the internal clamping diodes will limit this to Gnd and Vdd.
You really want to rely on a Resistor and the PICs clamping diodes to stand between you and oblivion?
In my circuit the LIVE AC is on the primary (input) side of the opto. No LIVE circuitry (nope not even 240uA of it) ever touches the PIC. If your Resistor failed, the PIC would become LIVE at supply potential. In my circuit, if either C1, D1 or the opto failed, you would still be insulated by the ISOLATION characteristics of the opto (usually 5kV breakdown or more).
This is not a 'Theoretical' circuit plucked out of the air... it actually works and is in use in thousands of products every day. In the snapshot appended, here is the very circuit where the PIC is monitoring the state of two totally separate but critical AC Supply fuses.
Bookmarks