Ok, to clarify/address a few things....

The circuit as currently designed uses an 240VAC primary transformer and=20
a lightly loaded secondary.   The transformer and associated parts=20
(diodes, caps, etc on the secondary) cost about $2-$3, but is biggish=20
and heavyish.   I.E. about 1.5 inches square by 1 inch tall, and=20
probably doubles the weight of the product.  Which is a lot of space to=20
consume - since I really have two of these on board.

The advantage of the transformer is that any inefficiencies will=20
dissipate out that big heavy device pretty efficiently - no heat sink=20
required.   Plus, due to the lead arrangement, I can generally keep the=20
AC on the bottom of the board - which is important since this is an=20
industrial type product designed to be in a cabinet where people will be=20
doing work from time to time and I'd like to limit finger-contact options.

As I mentioned in my original email - the problem I'm running into with=20
*any* circuit which uses an optocoupler is that the LED in the opto,=20
plus a bit for the rest of the circuit usually adds up to a few mA,=20
which is in the watts category when you use a resistive power supply off=20
of the line.   Especially when you want it to operate over a wide range=20
of voltages - I need to think internationally here to some extent (I=20
have a few places where 230V is the norm).    If I put the circuitry on=20
the bottom of the board, I run into a heat entrapment issue.   On the=20
top of the board, I run into the 'dangerous voltages for fingers'=20
problem.   Or I run into having to somehow insulate that circuitry.  =20
All with a limited budget of say $5 or less for this portion of the=20
circuit, including any insulation.

I was looking at the capacitive power supplies, but it looks like the=20
required capacitors for those are rather large, and probably not surface=20
mount.   Of course, I may have missed some schematic somewhere which=20
uses smaller sized/valued caps for this... but the general one needs big=20
ugly capacitors.  Plus, it seems these are very line voltage sensitive,=20
which means I get to burn off some more heat in a voltage regulator (aka=20
resistor and/or zener, or a linear regulator).   Or I can spend money on=20
a switcher which isn't in the budget.

In all, I'm back to the $3 transformer - big, ugly, but apparently=20
elegant solution.

I'm intrigued by the idea of using some sort of hall effect sensor.

I'm also intrigued by the idea of using something more like an audio=20
isolation transformer...  with some circuitry to prevent overloading/flames=
..

Or I'm interested in someone pointing out something I missed.

-forrest
--=20
http://www.piclist.com PIC/SX FAQ & list archive
View/change your membership options at
http://mailman.mit.edu/mailman/listinfo/piclist
.