zero crossing detector circuit - IC

[fuseshut]

I have two three zero crossing detector circuits for my triac light dimmer. With only a small PCB footprint I don't enough room for the FW rectifier diode bridge, optocoupler resistors. Is there a IC available that's optically isolated that I could AC line voltage directly (115VAC) and the chip detects the zero crossing, hence switching the opto on and off?

 

[fuseshut]

I found a AC input optocoupler, phototransistor output with bidirectional led's. AFter looking at the datasheet it seems to be that I can just hook up AC voltage with a big wattage resistors to drive the LED current. Does this make sense?

 

[Noway2]

Yes, you can use the optocoupler with 115Vac input. You need to make sure that the value of the resistor is sufficient to limit the current below the diode maximum and that the power rating of the resistor is adequate. In one of our products we use this technique to monitor the 120 VAC 'digital' inputs, using a 33K 1W resitor.

During the zero crossing period, the optocoupler shuts off momentarilly. In our application we counter this by placing a capacitor on the transistor output (5V logic side), which is the opposite of application of finding the zero crossings. Of course wether or not this will work for you depends on wether or not the time resolution provided by the opto shutoff is sufficient for your needs.

 

[cbarn24050]

Hi, an ac input coupler will work fine as you describe but you also need a resistor in parallel with the input diodes.

 

[itsmoked]

So leakage doesn't turn them on.

 

[BobM3]

Why the parallel resistor? Leakage? You mean the diodes leak a significant current when they are off? If it is leakage wouldn't it be better to not give the current a path (through the parallel resistor)?

 

[itsmoked]

No. You want the leakage to rather go thru the || resistor instead of thru the LEDs. [we're tlaking parallel to the LEDs (which are also diodes)]

 

[fuseshut]

So I need a resistor in parallel to the diodes, referring to part number Agilent HCPL 3760, and two resistors connected to the each AC leg of the optocoupler. If I am using 60K ohms, 1W, connected to both AC line pins of the optocoupler, what's common practice for resistance value across the diodes (to prevent this leakage effect)?

 

[cbarn24050]

It's not leakage as such, you dont get a reliable output pulse without it. 1k is a good starting point.