I’m in the process of replacing some existing Raspberry Pi machines… and one of the things I need to do is switch a BME688 sensor from an “oldPi” to a “newPi”… However, in order to avoid any break in data capture, I was fortunate enough to purchase a second ‘688 when I started this particular project back in 2023, so I now have two ‘688 sensors running, one on each Pi.
With my sampling code executing, I now see that there are slight discrepancies between the two sensors in terms of the values they are giving for temperature, pressure and humidity. I’ll put some actual data below for those interested, but the short version is that there are deltas of:-
- 0.4 degrees centigrade on the temperature sensor
- 0.4 millibars on the pressure sensor
- 3.8% on the humidity sensor
Now, to be fair… the sensor fitted to “newPi” has been running about 20 hours, so it still has a way to go to complete the burn-in period.
Both sensors are fitted with 15cm cables to keep them a good distance from their respective Pi hosts and in fact are maybe 2cm apart from each other, so there’s little chance of a difference in the ambient environment of each sensor.
I’d like to ask other users with experience of the BME688 as to their findings with regards to the consistency of the on-board sensors, and any recommendations they might have for calibration? I don’t possess any “super-accurate” measuring devices for temperature, pressure and humidity and I’m not sure that the introduction of a third set of variables [unless I can be very confident of the accuracy of the third set] adds any value.
But I’d like to ask for any recommendations. Obviously once the “newPi” sensor has completed burn in, I can tweak it’s reported values with offsets - and in the grand scheme of things I’m not sure if the variance detailed above is significant or not. I’m very interested to get feedback from other users who may have tried comparisons…
Finally, if I do want to bring these two devices in to alignment, is there any recommended approach? Should I trust the “oldPi” sensor to be more consistent given it’s been running for 3 years?
Thanks in advance.
Old Pi Data:-
Sat Feb 28 22 0 11 22.17 1,016.42 40.781
Sat Feb 28 22 15 11 22.20 1,016.34 40.813
Sat Feb 28 22 30 12 22.20 1,016.22 40.785
Sat Feb 28 22 45 11 22.22 1,016.01 41.374
Sat Feb 28 23 0 12 22.19 1,015.81 41.533
Sat Feb 28 23 15 11 22.19 1,015.49 41.527
Sat Feb 28 23 30 11 22.26 1,015.53 41.352
Sat Feb 28 23 45 11 22.25 1,015.40 42.822
NewPi Data:-
Sat Feb 28 22 0 11 22.56 1016.80 44.642
Sat Feb 28 22 15 12 22.61 1016.72 44.611
Sat Feb 28 22 30 11 22.59 1016.58 44.590
Sat Feb 28 22 45 12 22.61 1016.36 45.249
Sat Feb 28 23 0 11 22.59 1016.15 45.421
Sat Feb 28 23 15 11 22.58 1015.93 45.392
Sat Feb 28 23 30 12 22.67 1015.91 45.197
Sat Feb 28 23 45 11 22.66 1015.79 46.753
