Elsewhere I’ve covered CO2 sensors from Sonoff and Switchbot to name but two companies, but not really developed it, so I decided I’d have a good look. I managed to get a second CO2 monitor from SwitchBot, and I’m hoping to get a spare from Sonoff. I also have a Sonoff air particle sensor and some general temperature and humidity sensors, so I stuck them all out in the same place today well aware from air movement and people breathing and light variation, etc. I’ve reviewed all of these incidentally.
In the image above, you see the new Sonoff PM2.5 particle sensor air quality monitor on the left, then the new Sonoff CO2 sensor. These are both VERY new products. Then you’ll see a pair of identical SwitchBot CO2 sensors (not so new), and on the right, a general SwitchBot temperature and humidity sensor.
All very nice, but these have been sitting in the same room for quite some time with no interference, nobody breathing on them, no air movement and yet their readings are very different. Because I had a SwitchBot unit and a Sonoff unit, I asked SwitchBot and Sonoff for second units so I could run comparisons. I am still waiting to hear from Sonoff, but today the second unit, the second from the right, appeared from SwitchBot.
So I set the whole lot out in the living room. No air movement. No air conditioning, no light variations. No fires. I walked away, left the units for an hour and came back to take the picture you see above.
As you can see, we go from CO2 of 900 parts per million to 643 parts per million to 744 parts per million and that was read after stabilising for over an hour. So, which one, if any, is correct? The temperature sensor on the right matches up with the others… in fact, the temperature and humidity sensing on all of them is quite reasonable. On the very left, the particle sensor was sitting at nothing. I wondered. I thought that was a bit odd. I never took any more notice and later on looked in from my office, it still said zero, so I thought there might be something wrong, although the air quality here is very good – edge of a small non-industrial town in rural Spain.
I forgot about the whole thing, and my wife came home and decided to cook something in a pan and had a minor mishap, getting some smoke from the pan. I didn’t realise what it was, but then the Sonoff PM2.5 PM2.5 sensor proved its worth – it’s green light had changed to red, and it was beeping away incessantly to let us know that there was something wrong with the air – I could hear it in the next room. I’m quite happy with that, but I’m not happy with the variations of CO2 outputs.
I passed comments back to SwitchBot, and I’m still waiting for a response from Sonoff. The two SwitchBot CO2 values seem to be reasonably close together, further apart than I would have liked, but the Sonoff is miles away. Which one’s right if any? Well, in the next room, similar conditions, door wide open between the two rooms, I have my Apollo Air One sensor, which picks up on just about everything, more or less agreeing with the SwitchBots (somewhere in-between the 643ppm and 744ppm CO2). Certainly more than it agreed with the Sonoff unit.
I hope someone finds this info useful – we could develop this further.