Ok, so my screen printing dryer (basically a large pizza oven, not really)... has a PID that controls voltage to the solid state relays and has an "alarm out" which sends 105v (120v) to a light tree at the top to say that it's up to temp.
The bulb is dimmly lit and no one pays attention to it. Further more we sometimes bump the off button which puts it in cool down mode with a box. We don't notice this until it's too late.
I was thinking I could monitor this "Alarm 1" output and light up a bigger LED as well as give out an audible alert. Then when the Alarm 1 goes off I could have it trigger another condition of a momentary more obnoxious alarm sound to make sure we intended to shut it down.
As I looked into doing this, vs a full on bridge rectifire and such. I came across the idea of using a H11AA1 with some resistors to knock the voltage down to ~1.7v where the optocoupler likes it. Ordered some, but as I sit here impatient and thinking about how simple of a design the H11AA1 is with just two led's firing on each swing of the AC line... why couldn't I just do the same thing with a couple of 817's in parallel.
Now, the LED's are a bit wimpy when it comes to reverse voltage since they aren't meant to see reverse voltage, so I decided to add some 1N5399's in series to help out with that. This does come with a voltage drop that I wasn't wanting, but in practice, I was getting 15v to the 817 without it and 1.6v with it. So it worked out.
My problem is on my output side, I'm still only getting about 50% duty cycle due to the nature of the AC swing.
I don't see this as a problem for the arduino but when I put an LED on there as a test output it drops to 1.5v and the LED is very dim.
Am I just over thinking this or am I doing something wrong?
Here is a crude representation (simulator doesn't have a proper optocoupler so I made this up)... I think you get the idea.
http://tinyurl.com/y3e57z5t