4 ms·

This one of those things that sounds silly but is actually very easy to test and see it can be a real issue. Tom Scott did a great video on it a while back ht

by wh-uws 6y ago

This one of those things that sounds silly but is actually very easy to test and see it can be a real issue. Tom Scott did a great video on it a while back https://youtu.be/1Nh_vxpycEA https://youtu.be/1Nh_vxpycEA Do yourself a favor and get an air quality monitor like an awair https://getawair.com/ https://getawair.com/ and monitor the levels. Making sure they are good in whatever indoor areas you spend a good amount of time in is a night and day difference cognitively.

zbrozek 6y ago
Anybody know of something similar which doesn't exfiltrate data from my home?
- detaro 6y ago
  For CO2 specifically, one offline option is the "CO2Meter RAD-0301", also available as "TFA Dostmann AirCO2ntrol Mini" and possibly other labels. They have an unofficial USB interface too: https://news.ycombinator.com/item?id=22764488 https://news.ycombinator.com/item?id=22764488
- ohm 6y ago
  IQAir AirVisual Pro works offline. It's expensive though.
- Lammy 6y ago
  Open your windows at least a few inches. Let some breeze through. Don't worry about the AQI number :)
throwquestion24 6y ago
Testing is fun. But what do you do then? Open the windows? Remove the cars from the freeway down the street?
- jaggederest 6y ago
  Anything that moves the air is fine. And yes, opening windows would be a sensible thing to do in that context. That said, I'm not sure I buy the thesis as a whole. As far as I am aware, last time this came up, the Navy has done pretty detailed tests on CO2 concentration as a result of submarine warfare and found no significant impairments up to 1200ppm.
- Arainach 6y ago
  Opening the windows is a solid choice. I keep a CO2 monitor in my office. So long as I keep the door open, levels usually stay below 800ppm, but if I close the door they can climb rapidly within an hour or two. I have a CO2 meter tied to an alarm to remind me to open the window if they get too high. I probably should have it set to 800, but I found that to trigger too often so I upped it to 1200. I use a RAD-0301, which works fine as a monitor and an alarm by itself, but which I connect to a Raspberry Pi for dashboard purposes using the code shared here: https://news.ycombinator.com/item?id=22764488 https://news.ycombinator.com/item?id=22764488
- kevingadd 6y ago
  If you have central heat/ac your thermostat might have a switch to convert it from 'auto' to 'on'. You can combine that with turning off the heat/cooling such that it's just a house/apartment-wide fan. I do that at night before going to bed to deal with poor airflow in my bedroom and according to my CO2 sensor, it works. It works correctly even if I leave it set to cool or heat, the thermostat still controls whether it's actually heating or circulating air. Likely what you have is the same.
- Lammy 6y ago
  I wish there was a freeway down the street. Then the cars would be here and gone instead of idling in our local streets for fifteen minutes waiting to merge. Here's a real-world study of before and after they tore down two exits of the Central Fwy and built the horrible Octavia Blvd that necessitated a bunch of sidewalk closures and keeps killing cyclists: https://archive.sfcta.org/sites/default/files/content/Planning/CentralOctaviaCirculationStudy/Final%20Report%20Executive%20Summary.pdf https://archive.sfcta.org/sites/default/files/content/Planni...
xoa 6y ago
I don't know about the awair specifically, but it's worth being cautious about cheap air quality monitors. There seems to be little good standardized testing of them, and sometimes even major brands can have abysmal accuracy. See for example the EPA study "Evaluation of Emerging Air Sensor Performance" [1], some of them have very low or even zero correlation with reference monitors. It'd be nice to see some more uniform independent lab testing and marking standards there, or maybe there already are? I've rarely been able to find comprehensive evaluations however. At least it does look like the EPA is looking at them more, last year there was a very interesting webinar overview [2] of research being done on them. There are plenty of fully dumb plain monitors too, but since you picked a smart IOT one I'd personally really want to make sure it'd work LAN-only and supported a wide array of integrations. From a quick glance it's not clear the Awair does that, and it does not have HomeKit support for example. ---- 1: https://www.epa.gov/air-sensor-toolbox/evaluation-emerging-air-sensor-performance https://www.epa.gov/air-sensor-toolbox/evaluation-emerging-a... 2: https://www.epa.gov/research-states/low-cost-air-quality-sensors-webinar-archive https://www.epa.gov/research-states/low-cost-air-quality-sen...
jack12 6y ago
Anyone getting an air quality monitor specifically because of an interest in CO2 needs to be a bit wary, a lot of them claim to have a CO2 sensor but are actually just predicting CO2 levels based on other sensors designed to measure human occupancy. CO2 sensors are/were very popular for HVAC systems to use as a proxy reading of human occupancy levels and general air quality to meet air quality requirements. Now more modern (cheaper) air quality sensors are available which measure a mix of VOCs exhaled by people instead of CO2 but have an output mode that emulates the reading from a CO2 sensor at a similar occupancy level so that HVAC systems wouldn't have to redesign their control systems / formulas, usually marketed as "equivalent CO2" (or eCO2) readings. Because they're called eCO2 sensors while being cheaper and lower power than real CO2 sensors, many air quality monitors have been using them and then surfacing their readings as "CO2 levels" under the mistaken belief that they can use the sensor to measure CO2. But it won't measure environmental CO2 at all, just predict the amount of CO2 the humans in a room would typically generate. Which is all you need for air quality monitoring, but just be aware you're not actually measuring CO2 if that's what you're curious about. There are monitors with real CO2 sensors, but you'll want to look for a teardown to see what sensor is actually being used before you purchase. The Awair 2 seems to have a real NDIR CO2 sensor (Telaire T6703), for example: https://wiki.liutyi.info/display/CO2/Awair+2+inside#Awair2inside-CO2Sensor:AmphenolTelaireT6703-5K https://wiki.liutyi.info/display/CO2/Awair+2+inside#Awair2in...
totetsu 6y ago
Kurtis Baute's full experiment from Tom's video was very interesting. He tried to seal himself in a green house with plants to generate oxygen for a few days. https://youtu.be/PoKvPkwP4mM https://youtu.be/PoKvPkwP4mM
throwaway9d0291 6y ago
If people want a more flexible, less crazy expensive option, it's fairly straightforward to build a WiFi-connected sensor with an ESP32 devkit [0], an off-the-shelf sensor like the SCD30 [1] and ESPHome [2]. This is simple to assemble, will give you fairly good, repeatable readings, auto-calibration that performs pretty well and an open-source platform with full access to your data rather than a proprietary cloud application. Full cost should be ~$70, I'd expect assembly and flashing to take ~1h the first time. The most complicated thing you'll have to do from an electronics standpoint is solder on the pin header to the sensor, other than that it's just jumper wires and flashing over USB. No need for adding resistors or capacitors or anything like that. If you're not comfortable with soldering, I _think_ the evaluation kit for the SCD30 [3] comes with pre-assembled headers. There's also a third-party variant [4] that comes with a cable header you can break out [5]. I strongly recommend that you use the SCD30. I've tested a variety of sensor components from a variety of manufacturers and the SCD30 was by far the best performing. Other sensors have problems with auto-calibration in a residential environment so either give nonsense readings or drift without periodic manual calibration. However if you want something _really_ cheap, you can replace the SCD30 with the ~$20 MH-Z19 and the total price will be ~$30. The auto-calibration on these is utterly terrible however, so you'll need to manually calibrate it with fresh air every couple of months to compensate for drift. And as others have warned, beware of retail products. A lot of them have fake "CO2" or "eCO2" sensors that don't measure CO2 but try to guess it based on other measurements. These are completely useless and are not at all indicative of actual CO2 levels. [0]: Purchase: https://octopart.com/esp32-devkitc-32d-espressif+systems-92405929 https://octopart.com/esp32-devkitc-32d-espressif+systems-924..., info: https://www.espressif.com/en/products/devkits/esp32-devkitc/overview https://www.espressif.com/en/products/devkits/esp32-devkitc/... [1]: Purchase: https://octopart.com/scd30-sensirion-89692236?r=sp https://octopart.com/scd30-sensirion-89692236?r=sp, info: https://www.sensirion.com/en/environmental-sensors/carbon-dioxide-sensors/carbon-dioxide-sensors-co2/ https://www.sensirion.com/en/environmental-sensors/carbon-di... [2]: https://esphome.io/ https://esphome.io/ [3]: https://octopart.com/sek-scd30-sensor-sensirion-101025983?r=sp https://octopart.com/sek-scd30-sensor-sensirion-101025983?r=... [4]: https://www.seeedstudio.com/Grove-CO2-Temperature-Humidity-Sensor-SCD30-p-2911.html https://www.seeedstudio.com/Grove-CO2-Temperature-Humidity-S... [5]: https://www.seeedstudio.com/Grove-4-pin-Female-Jumper-to-Grove-4-pin-Conversion-Cable-5-PCs-per-PAck.html https://www.seeedstudio.com/Grove-4-pin-Female-Jumper-to-Gro...

by wh-uws 6y ago

zbrozek 6y ago

Anybody know of something similar which doesn't exfiltrate data from my home?

detaro 6y ago
For CO2 specifically, one offline option is the "CO2Meter RAD-0301", also available as "TFA Dostmann AirCO2ntrol Mini" and possibly other labels. They have an unofficial USB interface too: https://news.ycombinator.com/item?id=22764488 https://news.ycombinator.com/item?id=22764488
ohm 6y ago
IQAir AirVisual Pro works offline. It's expensive though.
Lammy 6y ago
Open your windows at least a few inches. Let some breeze through. Don't worry about the AQI number :)

throwquestion24 6y ago

Testing is fun. But what do you do then? Open the windows? Remove the cars from the freeway down the street?

jaggederest 6y ago
Anything that moves the air is fine. And yes, opening windows would be a sensible thing to do in that context. That said, I'm not sure I buy the thesis as a whole. As far as I am aware, last time this came up, the Navy has done pretty detailed tests on CO2 concentration as a result of submarine warfare and found no significant impairments up to 1200ppm.
Arainach 6y ago
Opening the windows is a solid choice. I keep a CO2 monitor in my office. So long as I keep the door open, levels usually stay below 800ppm, but if I close the door they can climb rapidly within an hour or two. I have a CO2 meter tied to an alarm to remind me to open the window if they get too high. I probably should have it set to 800, but I found that to trigger too often so I upped it to 1200. I use a RAD-0301, which works fine as a monitor and an alarm by itself, but which I connect to a Raspberry Pi for dashboard purposes using the code shared here: https://news.ycombinator.com/item?id=22764488 https://news.ycombinator.com/item?id=22764488
kevingadd 6y ago
If you have central heat/ac your thermostat might have a switch to convert it from 'auto' to 'on'. You can combine that with turning off the heat/cooling such that it's just a house/apartment-wide fan. I do that at night before going to bed to deal with poor airflow in my bedroom and according to my CO2 sensor, it works. It works correctly even if I leave it set to cool or heat, the thermostat still controls whether it's actually heating or circulating air. Likely what you have is the same.
Lammy 6y ago
I wish there was a freeway down the street. Then the cars would be here and gone instead of idling in our local streets for fifteen minutes waiting to merge. Here's a real-world study of before and after they tore down two exits of the Central Fwy and built the horrible Octavia Blvd that necessitated a bunch of sidewalk closures and keeps killing cyclists: https://archive.sfcta.org/sites/default/files/content/Planning/CentralOctaviaCirculationStudy/Final%20Report%20Executive%20Summary.pdf https://archive.sfcta.org/sites/default/files/content/Planni...

xoa 6y ago

I don't know about the awair specifically, but it's worth being cautious about cheap air quality monitors. There seems to be little good standardized testing of them, and sometimes even major brands can have abysmal accuracy. See for example the EPA study "Evaluation of Emerging Air Sensor Performance" [1], some of them have very low or even zero correlation with reference monitors. It'd be nice to see some more uniform independent lab testing and marking standards there, or maybe there already are? I've rarely been able to find comprehensive evaluations however. At least it does look like the EPA is looking at them more, last year there was a very interesting webinar overview [2] of research being done on them. There are plenty of fully dumb plain monitors too, but since you picked a smart IOT one I'd personally really want to make sure it'd work LAN-only and supported a wide array of integrations. From a quick glance it's not clear the Awair does that, and it does not have HomeKit support for example. ---- 1: https://www.epa.gov/air-sensor-toolbox/evaluation-emerging-air-sensor-performance https://www.epa.gov/air-sensor-toolbox/evaluation-emerging-a... 2: https://www.epa.gov/research-states/low-cost-air-quality-sensors-webinar-archive https://www.epa.gov/research-states/low-cost-air-quality-sen...

jack12 6y ago

Anyone getting an air quality monitor specifically because of an interest in CO2 needs to be a bit wary, a lot of them claim to have a CO2 sensor but are actually just predicting CO2 levels based on other sensors designed to measure human occupancy. CO2 sensors are/were very popular for HVAC systems to use as a proxy reading of human occupancy levels and general air quality to meet air quality requirements. Now more modern (cheaper) air quality sensors are available which measure a mix of VOCs exhaled by people instead of CO2 but have an output mode that emulates the reading from a CO2 sensor at a similar occupancy level so that HVAC systems wouldn't have to redesign their control systems / formulas, usually marketed as "equivalent CO2" (or eCO2) readings. Because they're called eCO2 sensors while being cheaper and lower power than real CO2 sensors, many air quality monitors have been using them and then surfacing their readings as "CO2 levels" under the mistaken belief that they can use the sensor to measure CO2. But it won't measure environmental CO2 at all, just predict the amount of CO2 the humans in a room would typically generate. Which is all you need for air quality monitoring, but just be aware you're not actually measuring CO2 if that's what you're curious about. There are monitors with real CO2 sensors, but you'll want to look for a teardown to see what sensor is actually being used before you purchase. The Awair 2 seems to have a real NDIR CO2 sensor (Telaire T6703), for example: https://wiki.liutyi.info/display/CO2/Awair+2+inside#Awair2inside-CO2Sensor:AmphenolTelaireT6703-5K https://wiki.liutyi.info/display/CO2/Awair+2+inside#Awair2in...

totetsu 6y ago

Kurtis Baute's full experiment from Tom's video was very interesting. He tried to seal himself in a green house with plants to generate oxygen for a few days. https://youtu.be/PoKvPkwP4mM https://youtu.be/PoKvPkwP4mM

throwaway9d0291 6y ago

If people want a more flexible, less crazy expensive option, it's fairly straightforward to build a WiFi-connected sensor with an ESP32 devkit [0], an off-the-shelf sensor like the SCD30 [1] and ESPHome [2]. This is simple to assemble, will give you fairly good, repeatable readings, auto-calibration that performs pretty well and an open-source platform with full access to your data rather than a proprietary cloud application. Full cost should be ~$70, I'd expect assembly and flashing to take ~1h the first time. The most complicated thing you'll have to do from an electronics standpoint is solder on the pin header to the sensor, other than that it's just jumper wires and flashing over USB. No need for adding resistors or capacitors or anything like that. If you're not comfortable with soldering, I _think_ the evaluation kit for the SCD30 [3] comes with pre-assembled headers. There's also a third-party variant [4] that comes with a cable header you can break out [5]. I strongly recommend that you use the SCD30. I've tested a variety of sensor components from a variety of manufacturers and the SCD30 was by far the best performing. Other sensors have problems with auto-calibration in a residential environment so either give nonsense readings or drift without periodic manual calibration. However if you want something _really_ cheap, you can replace the SCD30 with the ~$20 MH-Z19 and the total price will be ~$30. The auto-calibration on these is utterly terrible however, so you'll need to manually calibrate it with fresh air every couple of months to compensate for drift. And as others have warned, beware of retail products. A lot of them have fake "CO2" or "eCO2" sensors that don't measure CO2 but try to guess it based on other measurements. These are completely useless and are not at all indicative of actual CO2 levels. [0]: Purchase: https://octopart.com/esp32-devkitc-32d-espressif+systems-92405929 https://octopart.com/esp32-devkitc-32d-espressif+systems-924..., info: https://www.espressif.com/en/products/devkits/esp32-devkitc/overview https://www.espressif.com/en/products/devkits/esp32-devkitc/... [1]: Purchase: https://octopart.com/scd30-sensirion-89692236?r=sp https://octopart.com/scd30-sensirion-89692236?r=sp, info: https://www.sensirion.com/en/environmental-sensors/carbon-dioxide-sensors/carbon-dioxide-sensors-co2/ https://www.sensirion.com/en/environmental-sensors/carbon-di... [2]: https://esphome.io/ https://esphome.io/ [3]: https://octopart.com/sek-scd30-sensor-sensirion-101025983?r=sp https://octopart.com/sek-scd30-sensor-sensirion-101025983?r=... [4]: https://www.seeedstudio.com/Grove-CO2-Temperature-Humidity-Sensor-SCD30-p-2911.html https://www.seeedstudio.com/Grove-CO2-Temperature-Humidity-S... [5]: https://www.seeedstudio.com/Grove-4-pin-Female-Jumper-to-Grove-4-pin-Conversion-Cable-5-PCs-per-PAck.html https://www.seeedstudio.com/Grove-4-pin-Female-Jumper-to-Gro...