New to AC

**schu4647** · - 28th June 2006, 22:33

I am new to AC electronics, and I am realizing how much of the world I have missed. There is one thing I am really confused on. I have noticed several threads on here that explain a dimmer circuit. I see that the pic control and optocoupler which in turn controls a triac. What is the reasoning behind the zero voltage detection aspect of it. I am guessing it is timing. Maybe you need to start the PWM signal when the AC line is at zero volts. I don't really see why that is though. I have searched a lot on google and all I can find is that it need to be done, but not why. Hope somebody knows. Thanks.

**dhouston** · - 29th June 2006, 00:15

Triacs automatically turn off at zero crossing. You control how long they are on during each half-cycle by controlling the point where the triac is turned on by a gate signal. Powerline control signals (e.g. X-10) are usually sent just before or after ZC as that has the lowest noise.

**dhouston** · - 29th June 2006, 12:20

There's a Microchip App Note that gives a very concise explanation of triacs.

http://ww1.microchip.com/downloads/e...tes/00958A.pdf

**schu4647** · - 29th June 2006, 13:14

Thank you for all the help. You said something that really cleared it up for me when you said the triac turns off at zero voltage. That made me realize the need for detecting the zero crossing is to get an accurate reference every time. If you didn't do it, you may turn on the triac at the beginning of the cycle the first time, and the next time you would turn it on in the middle which would cause the light to flicker. I didn't realize that I was not using the PWM command, but using the 60Hz to creat my own PWM. Let me know if this is correct thinking. At 60Hz the period is 16.6 ms. This means that it is positive for 8.3ms and negative for 8.3 ms. I would then detect the zero crossing and turn on the triac for a certain time. If i turn it on for 4.15 ms I would have a 50% duty cycle or about 60 volts at my bulb. Without the zero detection Part of my on time could be the zero voltage, which would significantly drop the 60 volts and cause a flicker.

**schu4647** · - 29th June 2006, 14:22

I think I finally get it. That documentation was awsome along with TB049 from microchips application notes. A Triac is a latching device. You only need a small pulse from the micro to turn it on, and then it latches until the zero voltage event. I was thinking about it oppisite. I thought I would turn it on at the beginning for the cycle, but that would keep it on the entire time. I need to turn it on at some point mid cycle. Going back to my previoius post if I want a 50% duty cycle I detect the zero crossing, wait 4.15ms and then turn on the triac. It would then stay on for the remainder of the cycle. I think this all makes sense now. I will have to experiment. Hopefully I won't electrocute myself.

**dhouston** · - 29th June 2006, 15:19

It is possible to reverse the turn-on/turn-off sequence by using an IGBT instead of a triac. Search Google on "IGBT dimmer".

**mister_e** · - 29th June 2006, 23:45

IGBT have good advantage.. reducing lamp buzz, reducing the noise the AC line, and and, and... but if my memory is good they need to drive in DC... another Bridge needed.

What's best? It's up to you to decide.