I've now been running the Airpi for about a month - or I should say I've been trying to run it for the past month. The software seems to crash rather too often for my liking.
The output from the code is very basic - it just displays the end values, so it's near impossible to tell where and why the python script is halting. I've tried a different Raspberry Pi just in case it was that, as well as 3 different types of power supply, but all without improvement. Nothing is echoed to the screen to explain the crash either.
The temperature recordings appear way to high as well. The temperature values come from one of the sensors (pressure or humidity) and it seems those devices run hot, causing elevated temperature values. It seems strange that this wasn't picked up in development and replaced with a more reliable method for sensing the temperature. The AirPi was extensively tested before production as the Ultra Violet sensor was removed due to inaccurate settings.
There's a few threads on the Airpi Forum about the code crashing, but the forum is now basically dead. There's very few posts and the developers seem to have moved on to other things.
The idea behind the microphone is to record thunder, which is why the Airpi was originally coded to sense every 10 seconds. I'm not exactly sure how often it thunders in the UK, but it can't be that often.
Not really an issue but the Carbon Monoxide, Nitrogen Monoxide and the light level are meaningless. Just values between 1 and a million.
I've now shut the AirPi down until I can get some time to hack it. My plan is to remove the humidy sensor and wire it directly to another GPIO pin. I'll add a DS18B20 1-wire temperature sensor to it on GPIO 4. I'll then update and run my existing weather sensing code which seems to run without issue.