Skip to main content

ATtiny85 - Using capacitor for backup power to persist clock time to EEPROM

As discussed in the initial design post, the idea is to connect a 0.47F capacitor to the VCC and GND pins of the ATtiny85. Then when we lose power, the capacitor will provide the ATtiny85 with enough juice to store the current clock time to its EEPROM. When we gain power again, the ATtiny85 will read the clock time back from the EEPROM and start over. In this way, we avoid killing the EEPROM of the ATtiny85 with too many write operations.

To that end, I purchased something like this over EBay (2 for $3, so works out to about $1.50 each).

The code to check for supply voltage drop looks like this:

void loop() {
  // Execute loop() every second
  if (!timer1) return; else timer1 = false;

  // Measure supply voltage
  // Source: http://digistump.com/wiki/digispark/quickref
  adc_enable();
  ADMUX = _BV(MUX3) | _BV(MUX2); // VCC as reference, band gap voltage (~1.1V) as input
  _delay_us(250); // Wait for Vref to settle
  ADCSRA |= _BV(ADSC); // Start conversion
  while (bit_is_set(ADCSRA, ADSC)); // measuring
  vcc = (int)(1100L*1023L/ADCW);
  adc_disable();

  // Start hibernation when VCC < 3V
  if (vcc < 3000) {
    if (!hibernate) {
      hibernate = true;
      EEPROM.write(0, clockHH);
      EEPROM.write(1, clockMM);
      EEPROM.write(2, clockSS);
    }
  }
  // Normal routine when VCC >= 3V
  else {
    hibernate = false;
    
    // TinyWire requirement
    TinyWireS_stop_check();
  
    // Clock tick-tock
    ticktock();
  }
    
  // Sleep now
  sleep_enable();
  sleep_cpu();
}

Then in the startup() code, we read the clock time from EEPROM:

  // Read clock time from EEPROM if available
  byte hh = EEPROM.read(0);
  byte mm = EEPROM.read(1);
  byte ss = EEPROM.read(2);
  if (hh >= 0 && hh <= 11 && mm >= 0 && mm <= 59 && ss >= 0 && ss <= 59) {
    clockHH = hh;
    clockMM = mm;
    clockSS = ss;    
  }

This configuration has been tested to work quite smoothly.

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