4 ms·

> I highly recommend getting one of these or a similar sensor. Also, I'm looking to find something similar but capable of measuring CO and CO2, if anyone knows

> I highly recommend getting one of these or a similar sensor. Also, I'm looking to find something similar but capable of measuring CO and CO2, if anyone knows let me know. I have some suggestions for bare boards/chips that you can program yourself and run your own self-hosted IoT cloud over WiFi. CO: $8 [0] + [1] and very very easy. Great place to start, but these sensors have a reputation for being somewhat garbage (see downthread discussion for more detail). I wouldn't use it as an emergency carbon monoxide alarm but it'll do sensitive measurements (10 to 500ppm), even if it's not accurate and I'd guess not particularly repeatable over time. There are more expensive ones available which may be more reputable. Just need an arduino or any other micro controller. I personally like [2] because it has built-in WiFi but still very cheap. I run a basic web server on them that exposes the data to Prometheus/Grafana. See [3]. CO2[4]: $60 here and you'll want a $10-20 ESP32 or ESP8266 to go with it. Arduino libraries are already written for it and the Arduion IDE has good support for ESP32 so I doubt it would be much work at all to test on / port to ESP32 and fix any porting bugs when they pop up. Alternatively if you feel like writing the interface from scratch you can go down the $40 for equivalent parts from Arrow, DigiKey, or Mouser. Honestly - interfacing with these things is generally not complicated at all as long as there aren't hardware bugs or undocumented behavior. tVOC: $20 here[5] It looks like your PM 2.5 sensor does not also measure total VOC, so this could be handy. For a very scientific breakdown of how a metal-oxide detecting chip can be used to calculate VOC's, see the following[6]. Thankfully, there is free firmware supplied by sparkfun which does all the heavy lifting, but whenever you're using a chip cross-purposes its good to double-check the math and science. I personally combine these kinds of sensors with the WiFi-enabled ESP8266 or ESP32 microcontrollers, which make for ~$20+ USB-flash drive sized sensors I can litter around the apartment/outside. I like controlling my own cloud, because I can't stand how Nest only provided 7 days of historical data, or reliance on cloud servers for things like itead.cc / EWeLink apps. It's stupidly simple to set up Prometheus and Grafana to self-host your own cloud. See mine here[7] If anyone decides to go this route and gets stuck anywhere, feel free to ping me. Email in profile. 0: (CO sensor) https://www.sparkfun.com/products/9403 https://www.sparkfun.com/products/9403 1: (Board for easy soldering/mounting) https://www.sparkfun.com/products/8891 https://www.sparkfun.com/products/8891 2: (Cheap ESP32) https://www.microcenter.com/product/613822/inland-esp32-wroom-32d-module https://www.microcenter.com/product/613822/inland-esp32-wroo... 3: (HTTP server running on $10 microcontroller) http://esp32.rosskoepke.com/ http://esp32.rosskoepke.com/ 4: (CO2 sensor) https://www.sparkfun.com/products/15112 https://www.sparkfun.com/products/15112 5: (MOx sensor which is programmed to output tVOC -- I would not trust the CO2 though) https://www.sparkfun.com/products/14193 https://www.sparkfun.com/products/14193 6: https://amt.copernicus.org/articles/12/1441/2019/amt-12-1441-2019.html https://amt.copernicus.org/articles/12/1441/2019/amt-12-1441... 7: http://grafana.rosskoepke.com:3000/d/MrL-NKnMz/ross-home-office?orgId=1&refresh=5s&from=1595306888709&to=1595911688709 http://grafana.rosskoepke.com:3000/d/MrL-NKnMz/ross-home-off...

namibj 6y ago
I wish any of the affordable NDIR sensors would work below 400 ppm, and also without exposing them to a reference concentration every once in a while. Considering the effects of even small CO2 level increases on mental abilities, I'm looking forward to an indoor CO2 scrubber to get back below 300 and ideally below 200 ppm in the office/living room.
- rckoepke 6y ago
  You've definitely done your research on these so most of this is for other readers while I kind of pretend I'm narrating to you. You may be aware, but the earth is currently at an average of >400ppm right now, so most hobbyists won't need to measure lower than 400ppm. If you want a "cheaper" sensor that does measure all the way to 0ppm, I could suggest this one [0] for $70 at digikey. Realistically I think any solid-state gas sensor is generally going to require calibration with reference gases --- and that totally sucks. For CO2 there are some affordable options but some other gases I've looked at are pretty unreasonable for hobbyists. I've decided that I'm personally okay with inaccuracy as long as it has good repeatability. Your needs may be different, but for me it's not so much that I need to know the exact number, but rather that I need to be able to chart trends/cycles over weekly/monthly timelines. I also accept that my sensors will drift up/down over long timelines and I try to account for that possibility in any analysis I do. It's less than optimal, but as a hobbyist the sacrifices don't completely kill the endeavor. 0: https://rmtplusstoragesenseair.blob.core.windows.net/docs/publicerat/PSP110.pdf https://rmtplusstoragesenseair.blob.core.windows.net/docs/pu...
durkie 6y ago
for what it's worth, i wouldn't trust any sensor in the form of the sparkfun CO sensor (link 0 in your comment) -- there was a fellow in denmark who did a ton of research with these sensors to create an electronic nose that exploited their huge lack of sensitivity. most of them are sensitive to a huge range of gases beyond what they are ostensibly labeled for, so he would drive them at different voltages and could basically get fingerprints of various substances: the nose could smell the difference between different types of olive oils or detect different spices. they are also very finicky to changes in temperature / humidity.

namibj 6y ago

I wish any of the affordable NDIR sensors would work below 400 ppm, and also without exposing them to a reference concentration every once in a while. Considering the effects of even small CO2 level increases on mental abilities, I'm looking forward to an indoor CO2 scrubber to get back below 300 and ideally below 200 ppm in the office/living room.

rckoepke 6y ago
You've definitely done your research on these so most of this is for other readers while I kind of pretend I'm narrating to you. You may be aware, but the earth is currently at an average of >400ppm right now, so most hobbyists won't need to measure lower than 400ppm. If you want a "cheaper" sensor that does measure all the way to 0ppm, I could suggest this one [0] for $70 at digikey. Realistically I think any solid-state gas sensor is generally going to require calibration with reference gases --- and that totally sucks. For CO2 there are some affordable options but some other gases I've looked at are pretty unreasonable for hobbyists. I've decided that I'm personally okay with inaccuracy as long as it has good repeatability. Your needs may be different, but for me it's not so much that I need to know the exact number, but rather that I need to be able to chart trends/cycles over weekly/monthly timelines. I also accept that my sensors will drift up/down over long timelines and I try to account for that possibility in any analysis I do. It's less than optimal, but as a hobbyist the sacrifices don't completely kill the endeavor. 0: https://rmtplusstoragesenseair.blob.core.windows.net/docs/publicerat/PSP110.pdf https://rmtplusstoragesenseair.blob.core.windows.net/docs/pu...

durkie 6y ago

for what it's worth, i wouldn't trust any sensor in the form of the sparkfun CO sensor (link 0 in your comment) -- there was a fellow in denmark who did a ton of research with these sensors to create an electronic nose that exploited their huge lack of sensitivity. most of them are sensitive to a huge range of gases beyond what they are ostensibly labeled for, so he would drive them at different voltages and could basically get fingerprints of various substances: the nose could smell the difference between different types of olive oils or detect different spices. they are also very finicky to changes in temperature / humidity.