Timeout behavior using Sleep.h and SoftwareSerial.h

Using Arduino Programming Questions
1

Hello everyone,

I ran into a strange problem using two libraries, which seemingly have no interaction, but behave not as intended. The libraries in question are

  • <avr/sleep.h> and <avr/wdt.h>
  • <SoftwareSerial.h>

I condensed my code to visualize the problem. Tested on Arduino Nano (Atmega 328,Old Bootloader

#include <Arduino.h>
// #include <Logger.h>
//   bool DEBUGLOG = true;

#include<SoftwareSerial.h>
#include <avr/sleep.h>
#include <avr/wdt.h>

#define CORE_POWER_INTERRUPT        0x0A
#define CORE_POWER_WAKEUP           0x0B
#define CORE_POWER_NORMAL           0x0C
#define CORE_POWER_CALCULATION      0x0D

SoftwareSerial SSERIAL = SoftwareSerial(2,3);

void setup() {
  Serial.begin(9600);

  SSERIAL = SoftwareSerial(3,4);
  SSERIAL.begin(115200);
 
  Serial.println("--- Lowpower ---");
  delay(1000);
}

uint8_t TIMEOUT_MULTIPLIER = 1;
uint8_t CORE_POWER_STATUS = CORE_POWER_NORMAL;

void wakeupLoop() {
  Serial.println("wakeupX");
  delay(1500);
  CORE_POWER_STATUS = CORE_POWER_NORMAL;
}

void interruptLoop() {
  Serial.println("interrupt");
  delay(5000);
  CORE_POWER_STATUS = CORE_POWER_NORMAL;
}

void calculationLoop() {
  Serial.println("calc");
  CORE_POWER_STATUS = CORE_POWER_NORMAL;
}

// watchdog interrupt
ISR (WDT_vect) 
{
   wdt_disable();  // disable watchdog
}  // end of WDT_vect

void lowPowerMode_Timer() {

    for(uint8_t i=0;i<TIMEOUT_MULTIPLIER;i++) {
        Serial.print("Timer_M = "); Serial.println(TIMEOUT_MULTIPLIER);
        // Serial.print("M= "); Serial.println(i);
        delay(150);

        // disable ADC
        ADCSRA = 0;  

        // clear various "reset" flags
        MCUSR = 0;     
        // allow changes, disable reset
        WDTCSR = bit (WDCE) | bit (WDE);
        // set interrupt mode and an interval 
        WDTCSR = bit (WDIE) | bit (WDP3) | bit (WDP0);    // set WDIE, and 8 seconds delay
        wdt_reset();  // pat the dog

        set_sleep_mode (SLEEP_MODE_PWR_DOWN);  
        noInterrupts ();           // timed sequence follows
        sleep_enable();

        // turn off brown-out enable in software
        MCUCR = bit (BODS) | bit (BODSE);
        MCUCR = bit (BODS); 
       
        interrupts ();             // guarantees next instruction executed
        sleep_cpu ();  
    }    

    sleep_disable();

    CORE_POWER_STATUS = CORE_POWER_WAKEUP;
    Serial.println("\tWakeup");

    // cancel sleep a

}

void loop() {
  
  switch(CORE_POWER_STATUS) {

    case CORE_POWER_WAKEUP: {
      wakeupLoop();
      break; }
    case CORE_POWER_INTERRUPT: {
      interruptLoop();
      break;}
    case CORE_POWER_CALCULATION: {
      calculationLoop();
      break; }
    default: {
      // Serial.println("empty");
      break; }
  }

  lowPowerMode_Timer();
}

Expected Behavior: 8s sleep
Result: Sleep mode is entered, but the device immediately wakes up.

Changing the setup() instruction and removing SSERIAL.begin(115200);

void setup() {
  Serial.begin(9600);

  // SSERIAL = SoftwareSerial(3,4);
  SSERIAL.begin(115200);
 
  Serial.println("--- Lowpower ---");
  delay(1000);
}

OR removing SSERIAL = SoftwareSerial(3,4);

void setup() {
  Serial.begin(9600);

  SSERIAL = SoftwareSerial(3,4);
  // SSERIAL.begin(115200);
 
  Serial.println("--- Lowpower ---");
  delay(1000);
}

results in the correct sleeping time (!). Using both commands results in the strange behavior.

I can not find a logical connection between these functions?

2

Suggest you investigate the use of interrupts within all 3 libraries.

4

I may found the problem.

The Nano was connected to a series of sensors and ICs. The whole setup was (including the Nano) was powered from a LM2596 adjusted to 5.7V. The LM2596 is connected to a 2-Channel laboratory power supply. To measure the power consumption I connected a Multimeter in Series.

Well, the Multimeter was not turned on and the Nano was connected via the serial USB and powered some of the sensors. But the power draw exceeds the amount the Nano can supply. Reestablishing the power connection, the Nano (Sleep, Timeout, Interrupt) worked as intended.

I do not understand the correlation, but since the power connection 'solved' the problem it was probably the inconsistent power draw.

5

This topic was automatically closed 180 days after the last reply. New replies are no longer allowed.