Next generation dust sensor for MySensors



  • I was struggling with the current dust sensor for a bit, and this got me thinking: if I were to create a new MySensors dust sensor today, what would that look like?

    The requirements:

    • Cheap
    • Able to sense PM10 and PM2.5 particles
    • No need to calibrate, and reasonable calibration from the factory.
    • Should have a built-in fan.

    After looking around I suspect these would currently be the best candidates:

    Nova Fitness SDS011 ($20)

    • It has good reviews, kinda.
      "With its size, it is probably one of the best sensor in terms of accuracy"
    • It has a UART interface.
    • It's the most popular one on AliExpress.
    • A new version is out, the SDS 021, which should be compatible. No word on how it does yet, although the test results don't look too good.
    • It's not too accurate though: "While SDS011 seemed to achieve good results in tests I've seen specification says its relative error is maximum of 15% and +/- 10ug/m³ (at room temperature and humidity). While 15% seems good enough to me, +/- 10ug is problematic since we are aiming at measuring values in the range of 5-50 ug/m³ (50 being the 100% of the norm)"

    PMS7003 ($25)

    • Claims to be able to measure even more 'bins' (particle sizes), although suspicions a-plenty that these aren't honest.
    • Some experiments here on MySensors.

    Most of these devices, also the laser ones, are very imprecise at small amounts of dust. Below 50 ug/m3 they claim a possible measurement error of +-10ug, which is a bit ridiculous.

    .

    The other devices have more faults:

    • PMS5003 (and earlier). Good design but cheap 8-bit chip with low resolution.
    • Sharp sensor is bad at [low levels],(http://www.howmuchsnow.com/arduino/airquality/) of dust, and doesn't have a fan.
    • Shinyei is only good if you hack it to remove the resistor and just blow air into it constantly with a fan. Otherwise it can fluctuate a lot.
    • Cubic PM2005 ($40). This is the one used by the VAIR monitor, and it sounds like they know what they're talking about. It's not on Aliexpress though.

    Sources
    http://aqicn.org/sensor
    http://vair-monitor.com/2017/01/19/measuring-dust-levels-measure-part-23/

    .

    So, I'm curious to hear what you think: what's should be the new standard?



  • After some more research and some reaching out, it seems the PMS7003 is a good fit.

    I've ordered one online.

    It's available on both eBay and Aliexpress. It's important to get the version with the little wired plug so that you can connect the wires to the Arduino easily.

    If anyone has any other thoughts: please share them!


  • Admin

    So how do these laser sensors work. Are they running all the time or do you wake them up once an hour taking measurements?



  • @hek I know that the SDS011 has a 'sleep' mode, not sure about the PMS7003. There is a PDF with details, but it's in Chinese :-)

    The Specsheet:
    http://aqicn.org/air/view/sensor/spec/pms7003.pdf

    All good air quality sensors use a fan to blow air past the sensor. This makes air quality sensors not the best option for battery powered use. But nothing would stop you from enabling and disabling the whole thing for a while and then go into sleep. It depends on how long it takes for the fan to spin up. I'll check once I get it.

    Personally I would like to have multiple measurements and then average them. All these cheap sensors are quite noisy.In practise I poll them once a minute, and it then does 30 measurements internally (once a second) which it averages. The other 30 seconds it sleeps. But the fan stays on.


Log in to reply
 

Looks like your connection to MySensors Forum was lost, please wait while we try to reconnect.