Why this is not working ?

[FvM]

AC type opto ? There is no diode or bridge then how he get ZC pulses.

There is a parallel diode, so surely no AC opto-coupler. Simply assume that the circuit is triggering with halfwave zero-crossing signal on both edges.

 

[milan.rajik]

Is it 1N4148 ?

 

[FvM]

Is it 1N4148 ?

Probably. Anything goes.

 

[milan.rajik]

Ok.

Three buttons dimmer. I will soon post the PCB layout files.


Edit: There is a bug. The negative half cycle is not controlling. Why ?

 

[ALERTLINKS]

The negative half cycle is not controlling

Don't use 220V for Vsin rather use 50V as earlier.

 

[milan.rajik]

Yes, you are right. 220V was the problem in Proteus. Works fine with 50V VSINE. Here is new code.


Edit: Eagle PCB files attached for three buttons dimmer.

---

Code C - [expand]
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
unsigned char milliSec = 0, i = 0;
 
//Timer0
//Prescaler 1:4; TMR0 Preload = 6; Actual Interrupt Time : 1 ms
 
//Place/Copy this part in declaration section
void InitTimer0() {
    OPTION_REG = 0x81;
    TMR0 = 6;
    INTCON |= 0x20;
}
 
void interrupt() {
 
     if(INTF_bit) {          //ZC occurs once every 10 ms
        INTF_bit = 0;
        i = 0;
        if(milliSec == 0) {
             GP5_bit = 1;
             Delay_us(50);    //INTF occurs once every 10 ms. So, it is safe to use 50 us delay
             GP5_bit = 0;
        }
        else InitTimer0();
     }
     
     if(T0IF_bit) {        //Timer interrupt occurs every 1 ms
        T0IF_bit = 0;
        TMR0 = 6;
        //Enter your code here
        if(i++ == milliSec) {
              GP5_bit = 1;
              T0IE_bit = 0;
              Delay_us(80);  // 50 us delay in a 1 ms timer interrupt is not a problem
              GP5_bit = 0;
        }
    }
}
 
void initialize() {
     INTEDG_bit = 0;
     INTF_bit = 0;
     INTE_bit = 1;
     PEIE_bit = 1;
     GIE_bit = ~GIE_bit;
     //milliSec = EEPROM_Read(0);              //There is no option to load eeprom file in Proteus for 12F683, So, uncomment this code while testing in hardware
     //Delay_ms(20);
}
 
void main() {
 
     ANSEL = 0x00;
     CMCON0 = 0x07;
     TRISIO = 0x07;
     GPIO = 0x00;
     IOC = 0x00;
     
     while (1) {
 
            if(GP0_bit) {
                Delay_ms(50);
                while(GP0_bit);
                if(milliSec < 9)++milliSec;
                //EEPROM_Write(0, milliSec);     //There is no option to load eeprom file in Proteus for 12F683, So, uncomment this code while testing in hardware
                //Delay_ms(20);
            }
            else if(GP1_bit) {
                Delay_ms(50);
                while(GP1_bit);
                if(milliSec > 0)--milliSec;
                //EEPROM_Write(0, milliSec);     //There is no option to load eeprom file in Proteus for 12F683, So, uncomment this code while testing in hardware
                //Delay_ms(20);
            }
            
            else if(GP3_bit) {
                Delay_ms(50);
                while(GP3_bit);
                initialize();
            }
     }
}

---

- - - Updated - - -

Edit : Final PCB layout files for POT Dimmer and Three buttons Dimmer attached.

Sunrom 16 bit Dimmer module reverse engineered. Code + Simulation attached.

 

[ALERTLINKS]

Making it with less and cheap parts to reduce cost if mains isolation is not required.

Software need to be modified to work on rising edge and falling edge to detect zero crossing.

 

[paulfjujo]

yes, it is simpler , but you don't control the 2 alternances.

 

[ALERTLINKS]

you don't control the 2 alternances

Now with modified code,

 

[paulfjujo]

is your power supply (for the PIC) made from AC source ?
like trouh a Capacitor reactance ( 0,1µF?) + Rprotect +diode + Capacitor +voltage regulator ?
or by battery + voltage regulator

or simulation only ?

 

[ALERTLINKS]

It is simulation only but 5V can be generated like this

 

[milan.rajik]

@ALERTLINKS

Post the code if possible for POT dimmer. When POT value is min, load should be OFF and when max, full power should be provided to it.

- - - Updated - - -

@ALERTLINKS

Did you write code in assembly language ?

I have attached new versions of codes for

Three button dimmer
Pot dimmer
sunrom 16 bit dimmer (I don't know which 8 pin PIC sunrom dimmer uses. I have used 12F683)

But still in three button dimmer there is a slight varying of firing pulse after every ZC. Why ? Every time the same ISR has to execute and so same time has to be taken to fire TRIAC.

- - - Updated - - -

Edit: See attached microchip application note image. They have used 12C508. Can I connect 20 Meg reisistor to INT pin of other PIC MCus and feed 120V to resistor ? Does all PICs pins have those to ESD protection diodes for the internal MOSFETs ?

I am planning to get 5V for PIC12F683 from transformerless power supply and feed 240V AC to INT pin through a 40 Meg resistor and drive the gate of Triac directly from PIC pin.

- - - Updated - - -

Edit 2:

Nobody is answereing. Here is my low cost dimmer based on PIC12F683. In actual hardware there will be no 1N4148 or I can add it also but there will be a resistor between GP5 and Gate. I have a question related to Proteus simulation. Why gate signal is strange ? It is similar to triac signal. If I disconnect GP5 from gate and check the signal then signal is ok.

 

[milan.rajik]

Four Channel Dimmer using PIC18F45K22. 8 MHz clock is used. Proteus 8.1 SP1 simulation files attached. Use the 2 channel dimmer simulation file. Proteus just can't simulate 4 channel dimmer atleast on my system. I am sure the dimmer will work on hardware.

Press the ON/OFFx button of a channel. Initial value on pressing ON/OFF button is 0 power. Use INCx or DECx button to increment or decrement the power.

- - - Updated - - -

I tested only the firing pulse for 4 channels and they work fine. It was just Proteus simulation problem. It just can't simulate 4 analog signals at the same time. The buttons work fine. I will add eeprom code to hold the milliSec value in memory.

- - - Updated - - -

@paulfjujo

See if you can use this transformerless power supply. It is rated for 30V 1A. Additionally you can add a 7812, 7809 or 7805 to get voltages for PIC and relays.

https://www.electroschematics.com/5499/transformerless-power-supply/

 

[ALERTLINKS]

I use flowcode for programing. I have generated train of pulses for trigger. It helps better firing the triac.I have used 10bit ADC and it is translated into nearly 1000 steps. I think I shall develop it into three stage battery charger using iron core transformer.
By the way, the 16bit dimmer, it is not 16bit but 4bit and 16 steps.

 

[milan.rajik]

Yes, you are right. It is 4 bit dimmer and 16 steps. Please post the flowcode file. I need a flicker free firing of triac.


@ALERTLINKS

Please provide flowcode file or C code for 256 step dimmer.

- - - Updated - - -

Edit 2:

Somebody has done similar to me.

https://arduinodiy.wordpress.com/2012/10/19/dimmer-arduino/
https://forum.arduino.cc/index.php?topic=6970.70;wap2
https://doityourselfchristmas.com/forums/archive/index.php/t-6358.html

 

[ALERTLINKS]

This is good only to simulate and prove the concept. Actually I have used c code in it that's why it does not simulate in flowcode. Without comments, it will be difficult to understand. I personally prefer to write my own rather than trying to understand someone else's code that what is happening.
To make it a real hardware further considerations are needed. For example what happens when mains frequency changes.
It will work worst than a regular passive dimmer when there is a spike on its input. As it is interrupt driven, it will suddenly respond to it and the whole timing will be disturbed.
A remedy to it would be to allow interrupt through a window from 9mS-11mS.
First confirm valid synch to mains frequency for 2=3 cycles and only then apply the PWM. Mains frequency may change a few cycle above or below. Calculations are based on 50Hz. At low level setting this can play a havoc. A more sophisticated system would be to determine mains frequency and then calculate the required timings for PWM.
There should also be remedy against false triggering because it may damage if full voltages are applied to certain systems. Anyway it depends on application.

- - - Updated - - -

C and hex file with better timings.

 

[milan.rajik]

Yes, I thought about mains frequency drifting. I have to think about how to handle that. Here I am attaching code for sunrom 256 step dimmer. Test and tell me if the working is fine.

See image 1. It is the simulation of your pot dimmer. Why the signal is like that. It is as if triac is conducting from 90 degress to 180 degrees and then instead of conducting from 270 degrees to 360 degrees it appears as if it is conducting from 360 degrees to 270 degrees. Why ? Did you do that intentionally ?

Are you firing triac 5 ms after ZC ?


Also see my 256 step dimmer image. It shows triac conducting from some 90 degrees to 180 degrees and then from 270 degrees to 360 degrees.

 

[ALERTLINKS]

I think it is working good except it does not restart when all inputs are once set to zero if it is not intentionally made it it so.
It can withstand hi-voltages from triac's-gate only in simulation.
About my circuit, I have connected bulb on upper side and taking sample from there, so the voltages on scope are reversed.

 

[milan.rajik]

Yes, when extreme values 0x00 and 0xFF are reached then it is not functioning. I have to fix it. If you start the simulation with any value other than 0x00 and 0xFF then it works fine.

 

[milan.rajik]

How should I handle drifting mains frequency in code ? Should I create a lookup table for Timer0 preload values for different frequencies or should I calculate frequency of AC mains and calculate timer0 preload value and use it ?