Networking air quality measurements
- 17 January, 2013 23:05
"Air pollution is a problem for all of us. The average adult breathes over 3,000 gallons of air every day. ... Many air pollutants, such as those that form urban smog and toxic compounds, remain in the environment for long periods of time and are carried by the winds hundreds of miles from their origin. Millions of people live in areas where urban smog, very small particles, and toxic pollutants pose serious health concerns. ... Long-term exposure to air pollution can cause cancer and long-term damage to the immune, neurological, reproductive, and respiratory systems."
-- US Environmental Protections Agency "Air and Radiation: Basic Information"
Air quality matters. According to the World Health Organization, urban outdoor air pollution "is responsible for 1.3 million deaths per year worldwide and indoor air pollution is estimated to cause approximately 2 million premature deaths, mostly in developing countries. Almost half of these deaths are due to pneumonia in children under 5 years of age."
While the U.S. government has an extensive network of devices monitoring our air quality, this country is so large that current monitoring only provides information about macro-scale conditions.
For example, I live in Ventura County, Calif., which has a total area of 2,208 square miles and is home to more than 823,000 people. It has hills and mountains, valleys, lakes and ocean front, and to monitor air quality there are just seven, count 'em, seven stations.
While government efforts to monitor and legislate air quality standards are impressive given the scale of the task, detailed data on the majority of local microclimates is mostly unknown (except if you live right next to one of the monitoring stations).
So, have you ever wondered how safe the air is in your local environment?
Last year I backed a Kickstarter Project that aimed to make air quality monitoring accurate and low cost. The project, the Air Quality Egg, produced by Wicked Device, is interesting for a number of reasons: First, it was a hugely popular concept, raising a total of $144,592 from 927 people on a goal of just $39,000!
Second, the project fits in with the "Internet of Things" concept, with all station data being uploaded to the 'Net and displayed on a web-based dashboard that we'll come back to in a minute.
Third, the basic version measures Nitrogen Dioxide (NO2), Carbon Monoxide (CO), temperature and humidity. The choice of NO2 and CO is based on high levels of those gases being indicative of poor air quality, while the temperature and humidity measurements are used to correct the gas sensors readings which are affected by both factors.
Fourth and finally, they delivered! This is in sharp contrast to a few Kickstarter projects, such as the Eyez project, which is now more than a year overdue on delivering and has failed miserably in keeping its backers updated.
My Air Quality Egg (AQE) arrived yesterday and I'm impressed by the engineering but less so by a couple of design decisions.
In the box you get two egg-shaped units each measuring about 5 inches by 5 inches by 3.5 inches, one of which, the base station, has an Ethernet port. Before you power up the base station you have to first go to http://airqualityegg.com/ and enter the MAC address of your AQE to register it with the data capture system called Cosm. The site's home page also shows the locations of all registered AQEs worldwide.
The MAC address is printed on a sticker on the box lid but it would seem obvious that the sticker really should be on the base station itself because, unless you're a pack rat, you're going to throw the box away (I peeled the sticker off the box and luckily it had enough glue to stick to the base station).
Once you've provided the MAC address, a name for your AQE, it's location, elevation, and whether it is indoors or outdoors, and then completed the rest of the registration, you can fire up the base station.
The base station then uses DHCP to get a local IP address as well as DNS and gateway addresses and connects to the data storage and control service called Cosm (formerly called "Pachube").
Cosm provides a secure, scalable platform which accepts, stores and displays data in real time. You can provide data from any source in XML, JSON and CSV data formats using either push or pull via their RESTful API or their socket-server. Cosm can also generate notifications based on simple data value changes via Twitter and HTTP POSTs to any URL you please.
Next you power up the sensor station which registers with the base station (you do this with the base and sensor stations in the same room to ensure that they connect), and then you can relocate the sensor to wherever it should go (I put my sensor outside about 40 feet from the base station, with a number of walls in between, and the two ends seems to have no problem connecting). And that's it, the AQE system is a great example of plug n' play done right.
Sensor data starts uploading immediately, although the gas sensor readings will be unreliable for about 20 minutes while they "warm up."
You can view the data from an AQE via the airqualityegg.com site. This shows a summary of just one AQE at a time, as well as a map showing the location of the device the location of other local AQEs. You can check out my data at http://airqualityegg.com/egg/97759.
This display is derived from data sent to Cosm and, for unknown reasons, my AQE's gas sensors are currently providing spurious data; the acceptable maximums defined by the Environmental Protection Agency for NO2 is 100 parts per billion averaged over one hour while CO should not exceed 35 parts per million averaged over 1 hour, but my current readings are -2147483648 parts per billion and 111375 parts per billion respectively.
Be that as it may, I'm assuming these are teething troubles and there may be a problem with where I've sited the sensor station.
Alternatively, you can examine the raw data for an AQE via Cosm. The Cosm display graphs your AQE data over periods ranging from five minutes to three months (or longer via the system's API). For the gas sensors the raw electrical measurements are also shown, but I can't find an explanation of why these matter. Note that the NO2 level of 101 ppb as shown by Cosm appears to be more realistic than the ridiculous value shown on the airqualityegg.com display.
If you're logged into Cosm, the page for an individual AQE also allows you to set up the previously mentioned triggers for each data stream (these can be for conditions such as less than, greater than, or equal to a value, or when the value changes at all), a map showing the location of the AQE, and the registration details.
Again, you can view my AQE data via Cosm at https://cosm.com/feeds/97759.
So, my complaints ... First, why an egg?! I'm sure it seemed a cool design when the developers first came up with the project, but an egg shape is a pain in the butt to deal with. The Jetson-esque shapes of many consumer devices (I'm thinking of product lines such as most of the Linksys devices and the horrible Motorola NVG510) are bad enough, but a great, lumpy egg shape is just ridiculous. Moreover, neither the base station nor the sensor station have any kind of tag or built-in hook for mounting on walls or anywhere else. They just lie there like egg-shaped lumps.
Next is the issue that the AQE base station indicates it is functioning correctly by showing various colored lights every minute or so. You get a purple glow followed by green then blue. And when I say "glow" I mean it; the entire unit lights up! This is very annoying as I kept getting distracted by the light show, so I kicked it under the equipment rack in my office. It's also pointless given the light sequence apparently doesn't mean anything other than everything is OK.
I'm disappointed that my unit appears to have faulty gas sensing, but given the bleeding edge nature of this project I'm not that disappointed; I'm sure if I have a faulty system it will get replaced.
I'm far more disappointed by the lack of end user documentation (for example, there's no advice about correct placement of the sensor station) and, in reality, just somewhat annoyed by the pointless egg shape and light show.
The Air Quality Egg is priced at $185 and other sensors, including Dust/Particulate Matter ($40), Ozone ($25), Gamma and Beta Radiation ($60), and Volatile Organic Compounds and Hydrocarbons ($25), were apparently provided as options for Kickstarter backers and are promised for general availability.
Did I mention it's all open source as well? And that you can get the electronics of the AQE as a Nanode shield (the Nanode is an open source Arduino-like board with built-in Web connectivity) for just $95?
Awesome! I'm going to give the Air Quality Egg a Gearhead rating of 3 out of 5 (the sensor and documentation issues lower the score), and a huge thumbs up for delivering an ambitious solution to citizen-driven environmental monitoring.
Gibbs is online in Ventura, Calif. Find him at firstname.lastname@example.org and follow him on Twitter and App.net (@quistuipater) and on Facebook (quistuipater).
Read more about software in Network World's Software section.
NSW Police issues warning on 3D printed guns
Box buys iOS app to improve its own
Review: Sony Xperia SP
Coming to a shopping centre near you: 3D body scanners
ASIC debacle: Conroy open to transparency over website blocks