I was working on a TRIAC based dimming circuit. Initially I plaaned to use a PCB mountable SSR (G3MB-202P), though this was a cheap but the device didnt allow me to do dimming with the same. I can only turn ON or Turn OFF the load. and the other SSR cost \$(lots of them). So i decided to take a fairly cheap path of using an OPTO-DIAC and a TRIAC.

I have actually copied the RCR circuit from an online blog. Can someone please help me with the design of this. I know that this circuit is a phase compensation circuit for the current and voltage(lags) due to the inductive load.

Thanks and regards. Siddharth

It's not entirely clear if you're planning on dimming with the aid of a microcontroller or if you're hoping for a purely analog solution, so I'll go through both. If you aren't currently using a microcontroller and aren't hoping to do anything further with this circuit then I suggest you use the analog design.

Microcontroller: If you have a microntroller already in your design then all that is needed is to have a zero crossing circuit input wired into a digital I/O port. When the zero crossing circuit is triggered you'll use a timer internal to the microcontroller to drive the OK1 opto-TRIAC high after a certain amount of time into the mains half cycle. The output waveform will look something like this.

Analog: This one is relatively simple, you will use some resistors and capacitors along with a DIAC as the time delay rather than the microcontroller to trigger the TRIAC at different times in the half cycle. Here is an excellent application note from littlefuse detailing this circuit design. The output will look the same as above. Below is probably the simplest example of an analog dimmer circuit, with R1 indicating a 250K potentiometer.

Additional Notes: (R8, C6) and (R6, C2) make up snubber circuits for both the OK1 opto-TRIAC and the T1 TRIAC in the schematic you provided, these are meant to prevent the TRIACs from triggering spurriously from inductive loads. An alternative is to use a snubberless TRIAC which are designed to not trigger from inductive loads (Example).

• Thank you for your reply. But I am actually looking for the design of the RCR circuit which you see (R7,R8,C6). I am actually Using a flip flop to toggle the state of the input(i.e. input is controlled with a flip flop. – siddharth taunk Jun 14 '17 at 16:41
• The datasheet for your optotriac shows an example that uses a 360 ohm and 470 ohm resistor for R7 and R8 respectively, and a .05uF capacitor for C6. Snubber values in general are not all that important so those values should work well. – nreath Jun 14 '17 at 19:12
• I(GT) is mentioned at 15 ma and the current if i calculate the current which comes around 200 ma which does not match to that of the I(GT) mentioned. I am asking as I have to use a higher wattage resistor of around 1W for each resistor? – siddharth taunk Jun 18 '17 at 13:56
• No, you don't need to use a higher wattage resistor because after the gate current of the triac exceeds the 15 mA required for turn on the triac will essentially be shorted and there won't be any current going through those components. Also, having a value greater than the 15 mA listed in the datasheet is desirable, generally it is best to have the gate current limited to around 10x the I(GT) listed in the datasheet, 200 mA shouldn't be a problem. – nreath Jun 19 '17 at 14:52
• Hi, I didnt wanted to open a new thread for this question. Can I stack 5 of my above circuits side by side with heavy inductive loads? Can you please explain with the consequence and the remedies - of course with the modifications mentioned by @nreath – siddharth taunk Dec 26 '17 at 5:16