Bug in SenseBenderDefault sketch

[TheoL]

I was wondering why my SenseBender kept reporting every minute, even though the temperature and humidity didn't change.

It's a minor one. But in the sendTempHumidityMeasurements(bool force) method, the difference is determined by dividing with 100 and not 100.0 this causes a wrong rounding and always reports a difference higher than the treshhold. Resulting in a temperature report every minute. Even though the temperature hasn't changed much.

corrected code.

/*********************************************
 *
 * Sends temperature and humidity from Si7021 sensor
 *
 * Parameters
 * - force : Forces transmission of a value (even if it's the same as previous measurement)
 *
 *********************************************/
void sendTempHumidityMeasurements(bool force)
{
  bool tx = force;
  
  si7021_env data = humiditySensor.getHumidityAndTemperature();
  float oldAvgHum = raHum.getAverage();
  
  raHum.addValue(data.humidityPercent);
  
  float diffTemp = abs(lastTemperature - (isMetric ? data.celsiusHundredths : data.fahrenheitHundredths)/100.0); // <-- fixed
  float diffHum = abs(oldAvgHum - raHum.getAverage());

  Serial.print(F("TempDiff :"));Serial.println(diffTemp);
  Serial.print(F("HumDiff  :"));Serial.println(diffHum); 

  if (isnan(diffHum)) tx = true; 
  if (diffTemp > TEMP_TRANSMIT_THRESHOLD) tx = true;
  if (diffHum >= HUMI_TRANSMIT_THRESHOLD) tx = true;

  if (tx) {
    measureCount = 0;
    float temperature = (isMetric ? data.celsiusHundredths : data.fahrenheitHundredths) / 100.0;
     
    int humidity = data.humidityPercent;
    Serial.print("T: ");Serial.println(temperature);
    Serial.print("H: ");Serial.println(humidity);
    
    gw.send(msgTemp.set(temperature,1));
    gw.send(msgHum.set(humidity));
    lastTemperature = temperature;
    lastHumidity = humidity;
  }
}

ps. I'm sorry I don't have a github account.

[mfalkvidd]

nice catch!

[TheoL]

@mfalkvidd Thanx! It took my a while, but got it.

[TheoL]

Also, do I compile the Sketch for 8Mhz or 1Mhz?

[hek]

From what I can tell, this has already been fixed:

https://github.com/mysensors/Arduino/blob/development/libraries/MySensors/examples/SensebenderMicro/SensebenderMicro.ino#L248

Back in October

https://github.com/mysensors/Arduino/blame/development/libraries/MySensors/examples/SensebenderMicro/SensebenderMicro.ino#L248

[mfalkvidd]

@hek so the real problem is too long release cycles, and/or not enough backporting?

[hek]

Yes!

[tbowmo]

@TheoL

the sensebender is runing on 8Mhz as default. Originally I made so that it could switch to 1Mhz in order to lower the current consumption. But since we are sleeping for 99% of the time, where it's only the watchdog timer that is running, I don't see that much of a power savings running at 1Mhz, compared to 8Mhz.

In theory if running on 1Mhz the awaken time is also 8 times longer, than when it's running on 8Mhz.. So that can also contribute to a more "equal" power consumption.

So in short.. Compile for 8Mhz