230V AC dimmer, mains isolated.

[DanyR]

230V AC dimmer, mains isolated, using P12F629

This is a very simple dimmer with a triac output (phase control). It uses timer0 to do the timing.

The code is written in mikroPascal.

The control part (here a P12F629) is competely isolated from the 230V mains supply:
- the zero crossing detector contains an optocoupler (4N25 or similar),
- the drive of the triac also contains an opto coupler (MOC3025 or similar). Be aware: the latter optocoupler should NOT have a zero cross detection!

The 5V power supply is not drawn in the schematic.

The ignition angle can be set in the example by setting the value of Tmr0 at the end of the zero crossing detection. The example uses only these 3 values, see the comments below:

Case Step of
0: Tmr0 := 0; // never ignite triac (delay > 10 ms)
1: Tmr0 := 140; // low value = long delay = late ignition = low power
2: Tmr0 := 175; // average power
3: Tmr0 := 250; // high value = short delay = high power
end;

Schematic:


Code:

 

[Kurenai_ryu]

could you post the correct circuit? isn't isolated, as pin 2 from the 220-lamp connector (which goes to the triac) is connected to VSS and thus to the 5V power supply...

 

[swapan]

As far as the load is resistive, this type of dimmer is ok. What is your opinion about driving inductive load like transformer?

 

[DanyR]

could you post the correct circuit? isn't isolated, as pin 2 from the 220-lamp connector (which goes to the triac) is connected to VSS and thus to the 5V power supply...

You are so right, the circuit diagram was wrong. Thanks for the reply.

This is the corrected diagram:
.

The corrected zipfile (including the code):

So sorry for the inconvenience! :shock:

 

[DanyR]

As far as the load is resistive, this type of dimmer is ok. What is your opinion about driving inductive load like transformer?

Hi,
Here is some application information stated in the datasheet of the MOC3020:



Additionally a snubber circuit across the triac may be needed, e.g. 47 ohms in series with 0.01uF

 

[andre_luis]

the zero crossing detector contains an optocoupler (4N25 or similar),

The opto 4N25 has not a good gain, so that for low angles the circuit will not be able to detect precisely the phase where zero cross happens.

 

[DanyR]

The opto 4N25 has not a good gain, so that for low angles the circuit will not be able to detect precisely the phase where zero cross happens.

Hi André, Thanks.
I must admit I did my actual project with a TIL111 optocoupler, which has a minimum CTR of 50%, while the 4N25 has a minimum CTR of 20%. The reason for the difference is that I doubt if the TIL111 is easily obtainable, and I also have no LTSpice model for it...

Nevertheless, an LTSpice simulation shows it should work with the 4N25 (I assume they have the CTR set tp 20% in the 4N25 spice model...):

 

[DanyR]

The opto 4N25 has not a good gain, so that for low angles the circuit will not be able to detect precisely the phase where zero cross happens.

Hi André,
I see that the 4n25 model in LTSpice has a CTR of 73% (measured in a simulation), so that is probably the reason why the zero crossing detector works well in an LTSpice simulation...

I did the same simulation with a CTR of 20%, and the zero crossing pulses are somewhat wider then, which means the start of the zero crossing is detected too soon, and the end of it detected is too late.
This can result in not be able to trigger the triac at a very early time after the zero crossing, resulting is a slightly less max power for the load.