Skip to main content

ATtiny85 timer programming using the Watchdog Timer

This Arduino sketch uses the watchdog timer on the ATtiny85 to interrupt every second to pulse the LED:

 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
#include <avr/io.h>
#include <avr/wdt.h>
#include <avr/sleep.h>
#include <avr/interrupt.h>

bool led = true;

void setup() {
  // Misc setup
  pinMode(1, OUTPUT);
  set_sleep_mode(SLEEP_MODE_PWR_DOWN);

  // Setup watchdog timer to interrupt every second
  MCUSR &= ~_BV(WDRF);
  WDTCR |= (_BV(WDCE) | _BV(WDE));
  WDTCR = _BV(WDP1) | _BV(WDP2);
  WDTCR |= _BV(WDIE);
  sei();
}

ISR(WDT_vect) {
  sleep_disable();
  digitalWrite(1, led ? HIGH : LOW);
  led = !led;
  // Execution continues in loop()
}

void loop() {
  sleep_enable();
  sleep_cpu(); // MCU goes to sleep here; wake up by watchdog timer interrupt
}

The advantage of this approach is I can use the SLEEP_MODE_PWR_DOWN mode to put the MCU to sleep to save the maximum amount of power when there is nothing to be done. The watchdog timer clocks on its own 128kHz oscillator, and can be setup to wake the MCU every second via an interrupt, something that cannot be done using Timer0 or Timer1.

Unfortunately, it appears the 128kHz oscillator is not terribly accurate and can range between between 96 and 119kHz. In my test, the LED pulsed slightly slower than 1Hz, and start losing track after about 15 second ticks. There is also no way to calibrate the watchdog timer oscillator as far as I can tell.

ESPCLOCK1 / ESPCLOCK2 / ESPCLOCK3 / ESPCLOCK4

Comments

Popular posts from this blog

Update: Line adapter for Ozito Blade Trimmer

Update (Dec 2021): If you access to a 3D printer, I would now recommend this solution , which makes it super easy to replace the trimmer line. I have been using it for a few months now with zero issue.

Attiny85 timer programming using Timer1

This Arduino sketch uses Timer1 to drive the LED blinker: 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 /* * Program ATTiny85 to blink LED connected to PB1 at 1s interval. * Assumes ATTiny85 is running at 1MHz internal clock speed. */ #include <avr/io.h> #include <avr/wdt.h> #include <avr/sleep.h> #include <avr/interrupt.h> bool timer1 = false , led = true ; // Interrupt service routine for timer1 ISR(TIMER1_COMPA_vect) { timer1 = true ; } void setup() { // Setup output pins pinMode( 1 , OUTPUT); digitalWrite( 1 , led); set_sleep_mode(SLEEP_MODE_IDLE); // Setup timer1 to interrupt every second TCCR1 = 0 ; // Stop timer TCNT1 = 0 ; // Zero timer GTCCR = _BV(PSR1); // Reset prescaler OCR1A = 243 ; // T = prescaler / 1MHz = 0.004096s; OCR1A = (1s/T) - 1 = 243 OCR1C = 243 ; // Set to same value to reset timer1 to

3D Printer Filament Joiner

I have been looking at various ways of joining 3D printing filaments. One method involves running one end of a filament through a short PTFE tubing, melting it with a lighter or candle, retracting it back into the tubing and immediately plunging the filament to be fused into the tubing: One problem with this method is that you can't really control the temperature at which you melt the filament, so you frequently end up with a brittle joint that breaks upon the slightest bend. Aliexpress even sells a contraption that works along the same line. As it uses a lighter or candle as well, it suffers from the same weakness. I am not even sure why you need a special contraption when a short PTFE tubing will work just as well. Another method involves using shrink tubing/aluminium foil, and a heat gun: But a heat gun is rather expensive, so I wanted to explore other alternatives. The candle + PTFE tubing method actually works quite well when you happen to melt it at the rig