# Triac Circuit

1. Mar 10, 2017

### pierce15

I am making a very simple circuit, shown below (from the MOC3041 datasheet). The circuit consists of a MOC3041 optoisolator driving a BTA24 triac (click either for datasheet). I have a number of short questions and don't think it makes sense to make a separate thread, so I'll just put them all here.

1. First, I don't really understand what the optoisolator is supposed to do. When a current flows through 1 and 2, is there supposed to be a constant voltage between 4 and 6?

2. How the values of the resistors and capacitor are chosen. (Also I am using 120 Vac). Should these values depend on the properties of my triac/the output of the MOC3041? I understand the reason why the 39 ohm resistor and the cap are there, since they form an RLC circuit if the load is inductive and hence can dissipate power if the load is switched off. Are these just standard values?

3. Can someone give me a simple explanation of how to operate a triac? I have tried reading a number of explanations and am unable to understand them. My current understanding is this: if there is a voltage at the gate above a certain threshold (called $V _\text{GT}$ in the datasheet) and a sufficient current flows through either other terminal ($I_\text{GT})$, then the triac will conduct across the main terminals in either direction. Is that right?

2. Mar 10, 2017

### Baluncore

When current flows through the LED (pins 1 and 2) the resistance between pins 4 and 6 becomes very low which then causes a current in the TRIAC gate circuit, so the TRIAC is triggered and turns on. When the LED turns off, the TRIAC remains latched in the conducting state until the current falls to zero. Then it stays off until the LED is turned on again.

There is no inductive kick when the TRIAC turns off because that can only happen as the current reverses through zero.

The 39R and 0u01F limit the voltage slew rate across the TRIAC so as to prevent triggering by noise on the supply.

The 330R makes sure that the TRIAC will turn off when there is no LED current, and prevents false triggering.

The 360R protects the optoisolator from a high current spike during the turn on transient.

3. Mar 10, 2017

### Jony130

The long time ago I drew this diagram. And it fits to the description given by Baluncore.

4. Mar 10, 2017

### pierce15

Thanks to both of you for the help. So just to clarify: there should be no voltage drop across the detector pins when there is a current through the MOC3041 LED pins, right? And it should be able to conduct either way?

I tested the below circuit with just the BTA24 (so no optoisolator) and it worked with a 100 ohm resistor, since 100 ohms $\approx$ (120 V - 1.3 V) / (.035 A). But when I use the same resistor value in the place of the 360 ohm resistor in the first diagram, it doesn't work. My understanding is that the circuit should be exactly the same except the switch SW1 becomes pins 4 and 6 of he MOC3041. Let me know if this is wrong. If it is right, can you think of a simple test to see if the MOC3041 is broken? I measured the voltage/current through the MOC3041 LED and it is above the threshold current at 35 mA. However, the forward voltage is at only .7 V whereas the datasheet says it should be at 1.3 V. Does that sound strange?

Last edited: Mar 10, 2017
5. Mar 10, 2017

### Baluncore

There will be some voltage across the MOC3041 output pins when conducting. It will only be +/– 5 volts.
Yes, it should conduct either way. See; Figure 1. “On–State Characteristics” on page 2 of the data sheet.

The MOC3041 is drawn as having a low power TRIAC output stage. There may be a DIAC in the zero-crossing turn on circuit.

In your test circuit, post #4 you do not have an RC series snubber across the triac. You also do not show a resistor between triac G and MT1 to mop up any charge at the triac gate. That may make reliable testing difficult.

The IR LED voltage should be between 1 and 2 volts when conducting, with pin 1 more positive than pin 2. If you reverse bias the LED you may see a protection diode voltage drop of close to 0.7 volt, but I doubt that is provided since the usual LED Vr = 6V reverse breakdown voltage is included in the data sheet.

Take care how you measure voltages in the AC power circuit. You should have an isolation transformer for the testing you are doing. A digital multimeter will not be a good solution because it will give silly answers in switching AC circuits. Use a moving coil meter with a range switch.

Know someone interested in this topic? Share this thread via Reddit, Google+, Twitter, or Facebook

Have something to add?
Draft saved Draft deleted