Ever consider monitoring the air quality of your home? With the cost of sensors coming way down, it’s becoming easier and easier to build devices to monitor pretty much anything and everything. [AirBoxLab] just released open-source designs of an all-in-one indoor air quality monitor, and it looks pretty fantastic.
Capable of monitoring Volatile Organic Compounds (VOCs), basic particulate matter, carbon dioxide, temperature and humidity, it takes care of the basic metrics to measure the air quality of a room.
All of the files you’ll need are shared freely on their GitHub, including their CAD — but what’s really awesome is reading back through their blog on the design and manufacturing process as they took this from an idea to a full fledged open-source device.
Did we mention you can add your own sensors quite easily? Extra ports for both I2C and analog sensors are available, making it a rather attractive expandable home sensor hub.
To keep the costs down on their kits, [AirBoxLab] relied heavily on laser cutting as a form of rapid manufacturing without the need for expensive tooling. The team also used some 3D printed parts. Looking at the finished device, we have to say, we’re impressed. It would look at home next to a Nest or Amazon Echo. Alternatively if you want to mess around with individual sensors and a Raspberry Pi by yourself, you could always make one of these instead.
15 thoughts on “Watch Those VOCs! Open Source Air Quality Monitor”
Perhaps it’s an oversight or miscommunication but the only sensor listed in the BOM is the Temp/Humidity sensor. Along with an accelerometer, which seems odd for a device designed to be stationary.
If the rest of the sensors have yet to be released it seems a bit disingenuous to advertize them as working features. If they have been released, the BOM needs updated to reflect the supported hardware.
Some claims about the measuring range would be helpful too. If it’s only accurate to the regulated work exposure limits of VOCs and particulates that would be less useful for someone looking to do any sort of long term study, or even just for those who want to ‘collect the all the data’, since work limits are significantly higher than people should be living under. The website is IMO heavy on design and marketing aspects and light on technical details.
But it’s open source, that cures all evils (apparently).
I’m glad I wasn’t the only one wondering what the accelerometer was for. Earthquakes?
I think it replace a switch. All this is over engineered, what the need for a second cpu and a so complex power supply ?
Anyway the sensor used for the main function, air quality, is bad.
If you were to move the device to another room it would make sense to clear any averaging data?
Maybe also to not perform additional sensor sampling while being moved, because data would likely be invalid until it is placed in new location?
Actually, just a case of not reading the relevant articles.
In the demo video they clearly are using tap detection for user input.
Somehow I completely missed the demo video, it makes sense now, thanks! :)
If they used a microphone as an ambient noise sensor, it could do knock detection.
Or use a cheap piezo like everyone else does (which also doubles as a speaker, well beeper anyway).
But accelerometer because reasons. (And the reason is there’s an Arduino library for that.) (We should have a contest for the first person to wire a RGB LED to an Arduino and make it change colour when tapped.) (Underwhelmed by the ‘demo’ video….)
Hello, thank you so much for pointing that out!
You are right I forgot to add sensors to last BOM! Though it’s in schematics, CAD, code and on all assembly photos, so yes it is perfectly supported and not a “soon to come feature”.
Stock supported sensors are:
+ T/RH + Accelerometer
To monitor air quality you should measure the presence of contaminants in the air, such as carbon monoxide (CO), nitrogen dioxide (NO2), ozone (O3), particulate matter (PM2.5 and PM10), sulphur dioxide (SO2), and hydrogen sulphide (H2S).
You can’t do that with a single sensor.
The MQ135 is more a CO leak detector, the GP2Y1010 is a smoke sensor. They are ok for alarms but not enough for monitoring.
But that is a very nice mechanical and electronic design. Thanks for sharing
To get accurate particle sensor , you can consider laser particle sensor , now this sensor is small and cost effective for integrated application . Refer to PM2005
I’m really close to supporting this project because of the elegant design. Very well done.
Two big problems are stopping me:
1) The IAQEngine measures VOCs in “CO2 equivalent” units. After a significant amount of searching, and reading their own documentation, all I can determine is this is engineering black magic. No conversions to actual VOC PPM which all air quality documents use. Just a “oh, look at the spikes when we think the air quality got bad”. I’m willing to get past this considering actual VOC ppm measurements probably cost a lot of money.
2) The posting of data is only documented in two sentences on the front page. There is no mention of supported services/platforms/protocols in even the post.c file in github. I’m not willing to gamble $150 and my time in setting up a platform (I’m guessing… guessing here.. Nimbits or Thingspeak?) on the guess that I can get it working. Especially since I know nothing about IOT protocols and web servers. Documentation is the one thing that makes or breaks an open source project for me.
Looking at the sensors they are using the sharp dust sensor. this sensor is low cost and has very bad accuracy. The sensor lumps all PM sizes into one bin. To get accurate measurements you need to do a lot of calibrations in controlled environments. This can be hard since most people don’t have these testing enclosures.
Now there are some other sensors that can be used that are better but we are still talking about low quality. There is the shinyue sensor and supposedly there is a new sharp sensor. They cost around 12 bucks. They have the analogy to measure both PM 2.5 and PM 10. These are not as user friendly as the cheap sharp sensor. With the cheap sharp sensor it’s a simp ADC to get the value. These more expensive ones need to use a PWM decoder and translate that to a voltage which then corresponds to a concentration.
If interested I can provide all of my research on this topic to anyone. I have compiled a list of documents from other organizations that have made products similar to this one. Some of the papers go into details on ways to dramatic my improve the accuracy of these particle sensors.
Sean – I would be interested in what research you have compiled.
I would be very interested in this, I am doing a senior design project on this and a little help with the research would be very much appreciated
As far as we know ,SHINYEI PPD42 sensor still belong to a taditions , it measure particle >1.0um , when you count PM2.5 , particle number include all particles under 2.5um .How to measure particles less than 1.0um should be solved .
How about consider a laser particle sensor PM2005(http://www.gassensor.com.cn/product_detail_en/Particle_Sensor/PM2005) ,It measure particles from 0.3um . Total accuracy is ±15% , would be a suitable particle sensor for DIY , residental weather station or some OEM project .
Please be kind and respectful to help make the comments section excellent. (Comment Policy)