3 ms·

I recently got into building DIY air quality sensors. Any kind of smoke hangs around in a room for many hours well past when you can smell it. My house periodic

I recently got into building DIY air quality sensors. Any kind of smoke hangs around in a room for many hours well past when you can smell it. My house periodically spikes in air pollutants (still looking for the cause but we are moving soon anyways). The one night when PM2.5 spike from safe levels of below 12 ug/m^3 to over 1000 for one night was the worst sleep everyone in our house got. It seems that it was the equivalent of smoking something like 44 cigarettes in a day. Don’t smoke, especially indoors. It’s a bad time.

Karawebnetwork 6y ago
> Any kind of smoke hangs around in a room for many hours well past when you can smell it. But once the sensors indicate that the PM2.5 is gone, everything is fine, right?
- IgorPartola 6y ago
  If that's sarcasm, you aren't providing enough context for me to pick up what you are putting down. If not, then the honest answer is "I don't know for sure, I'm still researching it." From what I can tell, PM2.5 isn't the end-all be-all metric and sensors can give false readings. That's why I'm building multiple devices with different types of sensors and monitoring multiple metrics. PM10 and PM1.0 are also concerns, as is CO and CO2. I am researching which VOC sensors to get because unfortunately not much info is readily available. But from the research I've read PM2.5 is the most prevalent and damaging in typical households (CO being a big exception, but that's also monitored by regular household CO alarms). I can tell when someone has been frying something in the kitchen from my bedroom sensor for example, and it lingers for a while. I can also tell when outdoor air quality is poor, and then my indoor air filter and closed windows do help. I am still learning about all this, but in general this data has been helpful to figure out when to open windows to avoid headaches. I plan on publishing my findings at some point, but currently I am still waiting on parts and PCBs and working out the software to make it more usable without having to run to grab a USB cable to flash new firmware on the sensors. I was inspired by https://www.airgradient.com/diy/ https://www.airgradient.com/diy/ but those plans are outdated and the dashboard is proprietary so less than ideal. I am working with simply integrating my sensor network with Home Assistant so I have to do very little with frontend stuff. It was very quick to set up notifications to my phone so I don't have to look at sensor screens all the time.
  - adolph 6y ago
    Are you looking at BME680 vs CCS811 vs SGP30 or something else? https://www.jaredwolff.com/finding-the-best-tvoc-sensor-ccs811-vs-bme680-vs-sgp30/ https://www.jaredwolff.com/finding-the-best-tvoc-sensor-ccs8...
    - IgorPartola 6y ago
      I am using the PMSA003 for the PMx sensor, the DS-CO2-20 for the CO2 sensor, and was about to order this sensor breakout for CO and other gasses though currently it looks out of stock: https://www.seeedstudio.com/Grove-Multichannel-Gas-Sensor-v2-p-4569.html https://www.seeedstudio.com/Grove-Multichannel-Gas-Sensor-v2... I might play around with the BME680 but it’s VOC detection is very rudimentary and it does eCO2 which is basically a guess. There is also the MICS-6814 breakout for relatively cheap on AliExpress but reviews on it are mixed, saying it uses the wrong values for its resistors and capacitors and it’s not as accurate as the Grove one. The Grove board is the only one that measures separate gasses separately instead of measuring total gasses and then using a formula to try to divide the number down to expected parts. In practice, who knows. I need to lay my hands on all of them. If I could get all this into a controlled environment to test it all that would really be ideal. Thanks for that link, it was a great read and the comments were helpful as well.
  - Normal_gaussian 6y ago
    are you able to share anything / willing to work with anyone? I'm just getting through the research phase and am looking to coat my house with sensors - its been a struggle finding quality writing in the research and it looks like a graveyard in the world of pcb design.
    - IgorPartola 6y ago
      Absolutely. See my reply to the sibling comment and shoot me an email (in my profile). Happy to share what I found out.
  - ahaucnx 6y ago
    Thanks for mentioning our DIY sensor building instructions. Can you please let me know what exactly is outdated so that I can update it. Also I am more than happy to give people trial access to our dashboard. Just PM me via the support form on our website.

Karawebnetwork 6y ago

> Any kind of smoke hangs around in a room for many hours well past when you can smell it. But once the sensors indicate that the PM2.5 is gone, everything is fine, right?

IgorPartola 6y ago
If that's sarcasm, you aren't providing enough context for me to pick up what you are putting down. If not, then the honest answer is "I don't know for sure, I'm still researching it." From what I can tell, PM2.5 isn't the end-all be-all metric and sensors can give false readings. That's why I'm building multiple devices with different types of sensors and monitoring multiple metrics. PM10 and PM1.0 are also concerns, as is CO and CO2. I am researching which VOC sensors to get because unfortunately not much info is readily available. But from the research I've read PM2.5 is the most prevalent and damaging in typical households (CO being a big exception, but that's also monitored by regular household CO alarms). I can tell when someone has been frying something in the kitchen from my bedroom sensor for example, and it lingers for a while. I can also tell when outdoor air quality is poor, and then my indoor air filter and closed windows do help. I am still learning about all this, but in general this data has been helpful to figure out when to open windows to avoid headaches. I plan on publishing my findings at some point, but currently I am still waiting on parts and PCBs and working out the software to make it more usable without having to run to grab a USB cable to flash new firmware on the sensors. I was inspired by https://www.airgradient.com/diy/ https://www.airgradient.com/diy/ but those plans are outdated and the dashboard is proprietary so less than ideal. I am working with simply integrating my sensor network with Home Assistant so I have to do very little with frontend stuff. It was very quick to set up notifications to my phone so I don't have to look at sensor screens all the time.
- adolph 6y ago
  Are you looking at BME680 vs CCS811 vs SGP30 or something else? https://www.jaredwolff.com/finding-the-best-tvoc-sensor-ccs811-vs-bme680-vs-sgp30/ https://www.jaredwolff.com/finding-the-best-tvoc-sensor-ccs8...
  - IgorPartola 6y ago
    I am using the PMSA003 for the PMx sensor, the DS-CO2-20 for the CO2 sensor, and was about to order this sensor breakout for CO and other gasses though currently it looks out of stock: https://www.seeedstudio.com/Grove-Multichannel-Gas-Sensor-v2-p-4569.html https://www.seeedstudio.com/Grove-Multichannel-Gas-Sensor-v2... I might play around with the BME680 but it’s VOC detection is very rudimentary and it does eCO2 which is basically a guess. There is also the MICS-6814 breakout for relatively cheap on AliExpress but reviews on it are mixed, saying it uses the wrong values for its resistors and capacitors and it’s not as accurate as the Grove one. The Grove board is the only one that measures separate gasses separately instead of measuring total gasses and then using a formula to try to divide the number down to expected parts. In practice, who knows. I need to lay my hands on all of them. If I could get all this into a controlled environment to test it all that would really be ideal. Thanks for that link, it was a great read and the comments were helpful as well.
- Normal_gaussian 6y ago
  are you able to share anything / willing to work with anyone? I'm just getting through the research phase and am looking to coat my house with sensors - its been a struggle finding quality writing in the research and it looks like a graveyard in the world of pcb design.
  - IgorPartola 6y ago
    Absolutely. See my reply to the sibling comment and shoot me an email (in my profile). Happy to share what I found out.
- ahaucnx 6y ago
  Thanks for mentioning our DIY sensor building instructions. Can you please let me know what exactly is outdated so that I can update it. Also I am more than happy to give people trial access to our dashboard. Just PM me via the support form on our website.