What is the advantage of using timer capture over external interrupt? I made this and then I discovered your library and I try to understand.

const byte TRIAC_PIN = 9;
const byte ZC_EI_PIN = 2;

unsigned long topMicroseconds = 9700; // 10000 micros is between zero crossings
int prescaler = 8;
byte prescalerBits = _BV(CS11); // /8

void zeroCrossing() {
  TCNT1 = 0; // reset the timer counter
}

void setup() {
  Serial.begin(115200);
  Serial.println("START");

  attachInterrupt(digitalPinToInterrupt(ZC_EI_PIN), zeroCrossing, RISING);

  pinMode(TRIAC_PIN, OUTPUT);
  uint32_t topPeriod = ((F_CPU / 1000000)* topMicroseconds) / prescaler ;
  ICR1 = topPeriod;
  OCR1A = topPeriod + 1;
  TCCR1A = _BV(WGM11) | _BV(COM1A0) | _BV(COM1A1);
  TCCR1B = _BV(WGM13) | _BV(WGM12) | prescalerBits;
}

void loop() {
  if (Serial.available()) {
    unsigned long microseconds = Serial.parseInt();
    Serial.find("\n");
    uint32_t period = ((F_CPU / 1000000)* microseconds) / prescaler ;
    OCR1A = period;
  }
}

I use Robotdyn Dimmer

This library at present only supports 16MHz ATMega 328p boards like the arduino uno, and only on timer/counter 1, but in theory the exact same code could support a much wider range of cpus that have the same peripheral available.

Split from #11.

Need to add logic to enforce a 'minimum trigger pulse end offset' to ensure that the triac latches on even for dimming values very close to 100%.

In cases where either:

• the mains voltage source used for an application has a significant DC component,
• the zero-crossing detector is biased (i.e. emits its pulse earlier on a positive-to-negative transition than a negative-to-positive transition), OR
• the triac and/or triac driver circuitry used to switch the output is biased (i.e. the triac switches on more quickly in one direction than in the other)

It's possible that better performance and higher accuracy could be achieved by treating alternating half-waves independently from a timing perspective.

As an extreme example, take this:

A reasonably simple solution would be to split all current timing variables into two separate variables, corresponding to the separate positive and negative periods. This would involve some overhead from needing to track the alternating phase and index.
This would eliminate any asymmetry caused by a DC offset.

The other two asymmetry modes would require adding an additional half-wave offset configuration parameter that describes the asymmetry caused by the zero crossing detector and triac characteristics.
We would also need some way to synchronize with the correct half of the phase to ensure the offset is applied in the correct direction on every power-up. Assuming some of the asymmetry is caused by the zero-crossing detector, this could potentially be determined purely based on which half of the phase is longer, but if that is not the case this will require additional input circuitry and another input pin to read the phase directly.

Note that at present these issues are purely conjectural possibilities. Before trying to remedy this hypothetical problem, we need to take actual measurements on actual hardware to assess how significant these actually are.
We also need to assess whether this edge case is significant enough to be worth the performance penalty a fix would incur against the general case.

wavedrom source for image:

{signal: [
  {name: 'mains', wave: 'du.d...u.d.'},
  {name: 'zero',  wave: 'lPlPl..PlPl',
                  node: '.A.C...E...'},
  {               node: '.B.D...F...'},
  {               node: '...IH..L.K.'},
  {name: 'triac', wave: 'l...Pl...Pl',
                  node: '....GM..OJ.'},
  {               node: '.....N..P..'},
],
edge: [
  'A-B', 'C-D', 'B<->D t1',
  'C-D', 'E-F', 'D<->F t2',
  'C-I', 'G-H', 'I<->H 50% t1',
  'E-L', 'J-K', 'L<->K 50% t2',
  'N->M', 'N should be here',
  'P->O', 'P should be here'
],
config: {
  hscale: 2
}}

Split from #11.

Credit to @JAndrassy for his suggestion:

Additionally I use pin 10 as slave select for Ethernet shield and your library doesn't have an option to use only one pin of Timer1.

The library ought to be able to be configured to use only one of the two channels and leave the other free for other uses.

With Robotdyn AC Dimmer (schematics) and basic_example.ino the light is full on and sometimes flickers.
I test with Arduino Nano.
I added some debug prints in ISR(TIMER1_CAPT_vect) after last_icr = ICR1:

  Serial.print(micros());
  Serial.print("\t");
  Serial.print(last_icr);
  Serial.print("\t");
  Serial.println(TriacDimmer::detail::period);

output is

1408	2793	2793
11420	22824	20031
12168	22824	0
21440	42855	20031
31460	62905	20050
32208	62905	0
40448	17381	20012
50464	37423	20042
51212	37423	0
60476	57439	20016
70500	11952	20049
80512	31984	20032
81260	31984	0
90540	52027	20043

sorry I rewrote this issue twice until I debugged the problem.
the problem is, there are some phantom zero crossings resulting in period 0 and brightness is calculated with this value

Hi, I am attempting to use this with a potentiometer and am not able to get it to work at all. I realize its looking for a floating point value between 0 and 1. Is there a good way to set this up? Or a simple way to modify the source to use an integer input? Unfortunately I am still learning to code, so while I understand the problem, I am not sure of the solution. Thanks :)

I am trying to do dimmer with TriacDimmerCircuit with internet control... Please Help me... IT help lot for Me... Thank you in advance,,

Split from #11.

Presently the output compare interrupts do not correctly configure the timer peripheral to clear the output pins on a compare match as intended.

TriacDimmer.cpp:100 should read:

	TCCR1A &=~ _BV(COM1A0); //clear OC1x on compare match

and likewise TriacDimmer.cpp:112 should read:

	TCCR1A &=~ _BV(COM1B0); //clear OC1x on compare match

Credit to @JAndrassy for spotting this bug.

Hello.
Please add the schematic of the test circuit used with this library.

I can see that there are a limitation for controlling only 2 channels.

Is it possible to adapt this library to control boards with >2 channels, for example Krida 8CH Dimmer board?

If yes, could someone point me to the direction what needs to be updated in code and what are the challenges?