# TRIAC cuts peak voltage in one half-wave

I am building a dimmer for resistive loads/lightbulbs using the T2650 TRIAC. I use a phototriac to drive this TRIAC normally. For testing I have removed the phototriac and used a 220 Ω resistor instead to show and easily reproduce the issue I am having. The circuit looks like this:

simulate this circuit – Schematic created using CircuitLab

The lamp is a 40 W lightbulb for testing. What I am seeing is, that at the light bulb on one half-wave the peaks are capped. So if I measure at the point where TRIAC, lightbulb and resistor are connected to each other I see a waveform that looks similar to this:

_____/\__________/\________


The "/\" look like the tips of a normal sine wave, they have an amplitude of about 100 V. As the TRIAC is in my neutral line it should be a flat line if it's constantly conducting as it should, but instead I am seeing that it's cutting of the voltage partially.

Reducing the resistor to 33 Ω fixes it, however my driver has a maximum peak current of 1 A, so that would probably kill the driver. Looking at the parameters from the datasheet I don't really understand why this is happening, I don't see anything that indicates that this behaviour is "okay" or wanted. So does anybody have a clue what's going on here?

• Have you confirmed that the '220Ω' is actually 220Ω? You are operating in QI and QIII which is not hard for the triac. OTOH, if the resistor is actually 2.2kΩ that would be expected behavior. Commented Mar 5, 2023 at 4:38

## 2 Answers

The resistance of a fully lit 230 V 40 W incandescent bulb is ~1320 ohms. Combined with 220 ohms in series you have ~1540 ohms between the Phase line and TRIAC Gate. The T2650 TRIAC has a maximum required Gate trigger current of 80 mA. 1540 ohms x 80 mA = 123 V. So the TRIAC may not turn on until the line voltage reaches ~123 V.

If your Gate driver always supplies at least 80 mA of triggering current it should be fine.

• Well I am seeing the opposite: The TRIAC turns on and then somehow turns off again/starts dropping voltage, until it then turns on again. But only on one half-wave, on the other one it works perfectly fine. It doesn't even fully turn off, it only causes voltage drop. Commented Mar 4, 2023 at 23:32
• The TRIAC will turn off when load current drops below the holding current, which in this case is 75 mA max. With a 40W bulb this could occur at as much as 99 V, so you may see it turn off early too. The answer to this is to make the Gate driver supply current continuously throughout the cycle. The half cycle effect is caused by asymmetry in the TRIAC (trigger and holding current may be much higher in one direction). Commented Mar 5, 2023 at 0:49

Triacs are somewhat asymmetric so more current will be needed to trigger in one direction than the other, which could explain what you’re seeing. Using a resistor to trigger is rather crude as it requires a substantial voltage to trigger, so you have to trade off how small you want to make the resistance. Hopefully the real driver circuit will deliver a pulse that’s enough to trigger the trial reliably once there’s enough AC voltage for the trial to latch.

• But why is the peak of the voltage cut away? I would understand if it would switch on later, but it switches on, then acts like a resistor when the peak is reached and causes voltage drop there, and then switches on again. Commented Mar 4, 2023 at 23:12
• My knowledge of triacs is rather rusty. I see from Wikipedia that not all triacs can be triggered by a negative voltage, and the datasheet for this part doesn’t explicitly say that it can so maybe that’s the problem
– Frog
Commented Mar 5, 2023 at 0:04