Hello,
I have the BME688 and BSEC running on a custom design based on a Cortex M0+ processor. I am communicating with the sensor fine, and can read temperature, humidity, etc and configure the sensor and BSEC using the bsec_integration.c library found in the examples section.
I am doing an initial test on the gas scan function using the 'coffee or not' config array exported from one of the four models. Upon defining the helper functions (sleep, set_config, etc), running the sensor, and viewing outputs, I am confused by the results.
What I see from a full gas scan result in ambient air is below:
gas estimate 1
gas accuracy 1
gas estimate 2
gas accuracy 2
gas estimate 3
gas accuracy 3
gas estimate 4
gas accuracy 4
raw gas
raw gas index
gas percentage
0
0
0
0
0
0
0
0
91135.6328
0
0
0
0
0
0
0
0
0
0
49545.1914
1
0
0
0
0
0
0
0
0
0
86166.2734
2
0
0
0
0
0
0
0
0
0
53522.8945
3
0
0
0
0
0
0
0
0
0
119486.5781
4
0
0
0
0
0
0
0
0
0
153018.5312
5
0
0
0
0
0
0
0
0
0
201099.7656
6
0
0
0
0
0
0
0
0
0
324358.5625
7
0
0
0
0
0
0
0
0
0
913673.875
8
0
0.9994
3
0.0006
3
0
3
0
3
195345.2812
9
0
This repeats for as long as I run the program. Introducing coffee beans to view any updates does not change any parameter. This gives me some questions:
It is scanning the heater profile from 1-10, but I would expect ambient air to show gas estimate 1 and its accuracy at 0.999, 3 for the whole profile. Is this a wrong assumption?
What could it mean if the sensor is not sensing coffee? Do I need to let the sensor run for 24 hours, or the four day calibration period, before any sensing will occur?
Also, unrelated:
Does any BME688 sensor require any kind of calibration period before it is useful, even if we are not using the adjusted IAQ feature?
Thanks!