BME680 strange outside values

[micha_pr]

Hi,

I'm using the BME680 for outside air quality control and expecting since 2 weeks about strange values.

As you can see in graph the CO2 values are in the morning about 500ppm but in the evening all the time about 6000ppm.

I'm not sure what this can be, because the air seems (for my nose 🙂 ..) more or less "fresh" , I'm living outside a city in a smaller town.
And yes, right, if the air was "contaminated", i.e. someone made a fire in the vicinity, the sensor recognized this - with values between 7000 and 15000.
(the same for cigar smoke etc.)

But what now happen do not know.

I made a "reset" at April 5th in the morning (in graphics a falling line to 500) but in next time the same again.

Somebody have an idea what kind of air component this could be?
I made an additional test with other indoor sensor BME680 - if I will open the window in the evening the value is getting worse (from 650 to 900 - then closed, and value was going back)

Would be nice to understand it - even if it not easy ... any idea welcome....

Maybe it is simple not working temperature/humidity compensation? But in this wide range I don't want to believe it...
(compensated IAQ have same graph)

Thanks,

Michael

[micha_pr]

I will add the values from sensor for better understanding for Bosch experts.

Please remember that there was a hard reset at Apr.,5th 8:02 - you see it in data file timestamp.

Maybe it is really a temperature issue - but then why can an open window affect the data so much? (second sensor is near the window - see first post)

[handytech]

Thank-you for providing the complete data-log including raw sensor signals. I have shared your data internally already, and hope that we will be able to further process it to provide extra feedback.

Based on quick overview of raw gas output, I could observe significant contrast between the longer phase of higher concentration, and shorted periods of low concentration. Therefore the overall trend observed in your graph seems consistent with the VOC concentrations observed by the sensor.

Unfortunately, it is difficult to comment on the scaling factor you observe outside. I personnaly do not have a lot of experience with outdoor data, and therefore I do not really know what a CO2 reference sensor would/should output. Also, the BME680 itself is a relative tVOC sensor and not a CO2 sensor. BSEC outputs were tuned using a bVOC mixture, and the CO2-equivalent/bVOC-equivalent outputs was later added on popular demand (based on lab tests and field studies). Therefore bVOC-eq and CO2-eq outputs are expected to perform optimally where the main source of VOCs in the environment comes from human activity (typically indoor).

[handytech]

Hello @micha_pr

Just a quick update to let you know even after artficially playing with the temperature/humidity coefficients based on your datalog, the trend in gas signal always remains. Therefore as per your original observations, we do not believe the output has to do with temperature/humidity compensation, but rather with some actual gas concentrations in the environment Maybe you have already found some clues about your observations in the meantime