5 ms·
I connected an SDS011 dust sensor (didn't check US prices, but I'm guessing ~20$) to a raspberry pi and then used remote controlled power sockets to switch my a
by Netcob 6y ago
I connected an SDS011 dust sensor (didn't check US prices, but I'm guessing ~20$) to a raspberry pi and then used remote controlled power sockets to switch my air purifier on and off based on some limits I set.
Of course when building the purifier yourself, it might make sense to attach the sensor directly, use something like an ESP32 and maybe control the fan speed based on pollution levels.
Here's some Grafana graphs: https://i.imgur.com/wYjERxW.png https://i.imgur.com/wYjERxW.png
The dust sensor isn't absolutely silent due to the tiny fan, and the purifier obviously isn't either, so I have a "silent mode" that turns off both. There's also a CO2 sensor that shows you when I opened the windows. For some reason, dust didn't spike when I opened them in the afternoon, only in the morning and last night. The purifier is a BlueAir Pure 221 which is built a lot like the one in the article.
- everdrive 6y agoWhat accounts for the CO2 cycling?
- Netcob 6y agoMe opening my windows for a few minutes, but leaving them closed in general.
- alach11 6y agoI’m surprised how quickly the CO2 level rises! Any idea how high it would get if you never open a window?
- Netcob 6y agoI live in an old building that has a lot of cracks and gaps for air to circulate, so it rarely comes close to 2000ppm. When I leave the house for a longer period of time (hasn't happened in a while), you can see the value slowly approaching the outside baseline. And in the night, while I sleep with the windows closed, I produce much less of it, so I don't get super high values there either. But the rise is quite interesting! It differs a lot based on some factors, such as am I asleep, awake, or working out / are any doors open, i.e. how big is the volume of air I'm in?
- jxy 6y agoI never measured indoor CO2 level, but yours approaching 1400 ppm seems to be alarmingly high. I thought any level above 1000 ppm requires some intervention.
- RL_Quine 6y agoIt's also notable that CO2 sensors utterly suck, none of the consumer ones do absolute measurement, they assume a baseline and calibrate their output to that.
- Netcob 6y agoI live in a relatively small apartment. From everything I've read, levels between 1000ppm and 2000ppm seem to be quite normal indoors, e.g. in an office. Not healthy, but normal. One of the reasons may be that houses have become more air-tight in an effort to improve insulation, while the cognitive effects of high co2 aren't quite as easy to detect as a high energy bill. That afternoon slump may be a biorhythm / digestion thing, or maybe nobody in your office bothered to open the windows since yesterday. The concentration also rises pretty fast. I think I could make a (slightly delayed) intrusion alarm just by reading CO2 values... or a window-open-detector for automatically turning off the heating...
- musingsole 6y agoCross analysis of smart-home sensors to arrive at secondary or validation checks for other sensors (like home invasion) is the feature that will actually make a smarthome useful. Individual sensors trying to detect an event get too many false positives.
- sbierwagen 6y agoI also have a CO2 meter, and it's not hard at all to get above 1500. 5000+ is where you start getting concerned, but the air is noticeably stuffy way before that with all the other organic compounds humans exhale.
- everdrive 6y agoI measure mine, and if I don't leave my windows cracked, it can easily get up to 2,000 in my house. Keeping them cracked usually keeps CO2 between 500-900.
- fpgaminer 6y agoCurious. I have a HPMA115S0-XXX dust sensor and I can't hear any noise from its fan at all.
- grad8888 6y agoIs this the dust sensor the same as a particle counter? All the particle counters I could find were in thousands of dollars.
- bryceacc 6y agoLook up the PMS7003, I think I got it for $40 bucks on aliexpress and it's used in many air filters to monitor PM2.5 and PM10 etc sized particles. Easily run by a raspi and i output to a regular segmented display for my particle counts averaged every 5 minutes to see the california fires and how quickly my air filter could clear the room. For reference the winix 545 ($110 from costco iirc) on full blast got me down to about 0 particles in 15 minutes. edit: oh and regarding fan noise, the PMS7003 has a tiny centrifugal fan in it but I can barely hear it with my ear on top of it. The winix air filter on full blast though... crazy loud compared to the other low/med/high settings it's kind of hilarious how big of a jump it is in fan speed