RobertDyball / dust

PMS7003 + Arduino(ESP8266) Air Quality Sensor

Dust: PMS7003 + Arduino(ESP8266) Air Quality Sensor

Implements a PMS7003 and DHT11 based air quality sensor. There are lots of these solutions on the usual sites, but I couldn't tell if they ran the PMS7003 continuously or cycled the sensor on a clock. The laser LED has a 1 year lifetime, so it'd be a bummer to buy a thing that you'd have to chuck every year.

This sensor implements a RESTful server. Put your SSID/password in the right place, build, upload, and the PMS7003 should go into standby (fan off). Hit the IP with a GET on :80, and some artisanal hand rolled JSON should be returned.

$ curl http://dust.local | jq '.'
{
  "t": {
    "humidity_P": 59,
    "temp_F": 62.6
  },
  "a": [
    {
      "SP_1_0": 7,
      "SP_2_5": 9,
      "SP_10_0": 12,
      "AE_1_0": 10,
      "AE_2_5": 12,
      "AE_10_0": 15
    },
    {
      "SP_1_0": 7,
      "SP_2_5": 9,
      "SP_10_0": 10,
      "AE_1_0": 10,
      "AE_2_5": 12,
      "AE_10_0": 14
    },
    {
      "SP_1_0": 9,
      "SP_2_5": 9,
      "SP_10_0": 10,
      "AE_1_0": 12,
      "AE_2_5": 12,
      "AE_10_0": 14
    },
    {
      "SP_1_0": 9,
      "SP_2_5": 9,
      "SP_10_0": 12,
      "AE_1_0": 12,
      "AE_2_5": 12,
      "AE_10_0": 15
    },
    {
      "SP_1_0": 9,
      "SP_2_5": 9,
      "SP_10_0": 13,
      "AE_1_0": 12,
      "AE_2_5": 12,
      "AE_10_0": 17
    }
  ]
}

According the documentation, SP is Standard Particles, CF=1, and AE is Atmospheric Environment. AE_2_5 seems to be the most useful for tracking the level of hazardous particulate matter in the air.

Grafana + influxdb

There's a docker-compose.yml and an accompanying dustd microservith which periodically pulls data from the sensor and posts it to influxdb. To build/install it, just do the following

$ make && make docker && docker-compose up -d && ./db.sh

Obviously you need docker and docker-compose installed. Google is your friend if you don't already have that. Also, you won't get far without go. Again, le Googs will help.

And db.sh will probably not help you unless your docker host is called docker.chaos.local like mine is. Basically create the db on your host using curl like db.sh is doing. Also, take a look at config.json. You probably don't need to change anything. Rather, add whatever changes you want to make as environment variables in env.dustd.

In the end you'll get pretty graphs like the following.

Resources and dependencies

http://aqicn.org/sensor/pms5003-7003/

http://www.handsontec.com/pdf_learn/esp8266-V10.pdf

https://github.com/fu-hsi/PMS

https://github.com/adafruit/DHT-sensor-library