Project OAQ | Open Source Air Quality Sampler for Everyone

by bhuvanmakes in Circuits > Sensors

3333 Views, 63 Favorites, 0 Comments

Project OAQ | Open Source Air Quality Sampler for Everyone

1 (25).png
1 (27).png
1 (12).png
1 (9).png
1 (13).png
OAQM  Open Air Quality Monitor.png


Air pollution kills 7 million people, every single year, and unfortunately the people who are most affected by air pollution are not even aware about its presence. People just quietly suffer this bad air quality without ever finding out about it, maybe that's why Particulate Matter is referred to as a Silent Killer.

We want to change this chronic undetected suffering, however to solve this problem, it first needs to be clearly identified, and the situation with identification is pretty dire.


W.H.O recommends that for accurate measurement of Particulate Matter, there should be atleast 1 sensor every 100 meters, however the current number is about 0.02 sensors per 100 meters, a 100x deficit


Hence, we believe it is necessary to democratize access to air pollution measurement solutions, that includes affordable sensors and AQI meters, as awareness about Local Air Pollution will increase, mitigation solutions will follow.

Supplies

1 (28).png
1 (3).png
1 (30).png
1 (29).png
1 (31).png

So, here is Project Good1, an Open Source AQI Monitoring Sampler, the objectives of Project Good1 were first clearly laid out and accordingly a solution was designed.

  1. Affordable components (Under $25)
  2. NO SMD Soldering (Simple assembly)
  3. Easily source able components
  4. Chargeable ( 18650 ftw )
  5. Customizable Enclosure as per use case
  6. Simple firmware, community driven
  7. Phone app for viewing data and logging functions
  8. Call to actions by sensor fusion.
  9. Integration with Home assistant for easy control of other devices.

Always Start Dirty ! Breadboards It Is

1 (32).png
1 (33).png

After selecting our components, we assembled them on a breadboard to check it the "Logic" works and also if all components are compatible.


So, here is a breadboard version of Project Good1, it worked quite well for how it looked. But if we want Project Good1 to be easily carryable to measure Air quality on demand, the breadboard version will not do.


Hence we designed a perfboard version of Good1 in the smallest possible form factor, this is how it looks like, this soldering took quite some time and skill but it did work with everything literally crammed in !

Now That It Works, Let's Make a PCB

1 (20).png
1 (6).png

So, we started designing the PCB after making sure everything will work together, and as mentioned in the objectives, the PCB was meant to be as small as possible while not using any SMD components.


That's because it's much more difficult to source SMD soldering options for hobbyists and makers, a soldering iron is much more accessible, hence complete Soldering iron assembly was made sure.


The PCB has a footprint of only 52mm*30mm, this helps reduce on fabrication costs and keeps everything super compact, here are some screenshots of the PCB design process.

Which Sensors Should We Use ? All of Them !

1 (6).jpg
1 (5).jpg

Now, let's talk about the most important part of Project Good1, it's sensors, Good1 is meant to measure AQI or Air Quality Index and you cannot do that without the right sensors.


We found that for accurate Air Quality measurement, the following parameters were bare minimum,

  1. Temperature
  2. Relative Humidity
  3. PM 2.5
  4. PM 10
  5. CO2

So, if we got readings of the above 5 parameters, we would be able to accurately measure AQI readings, and hence we started looking up for sensors, and we just kept finding more and more and more sensors.....


See, as for temperature and humidity, we found a clear winner, the DHT22 Temperature and Humidity Sensor, it's accurate, reliable and simply works, also its quite affordable. So Temperature ✅, Humidity ✅


Then came CO2 levels, we found the MQ series sensors were very affordable and reliable enough to detect increasing CO2 levels, so we chose MQ-135.


NOW, we needed a sensor to measure Particulate matter, and this is where we found a number of sensors, we actually ended up buying 10 different Particulate matter sensors from the market ranging from $3 all the way to $60. And the performance of each of them varied.


So we made a decision, we made project Good1 to work with all of them ! Yes, let me explain.

See, Particulate Matter sensors are a little tricky, not all sensors are available everywhere and the cost of these sensors varies quite a lot, so we leave it to you users to select the sensor which best suits your requirements, if you go with the PMS5003 sensor, Project Good1 works, if you decide to use ZH03, that works as well !


Here are all the particulate matter sensors with which Project Good1 is compatible

  1. ZH03
  2. PMS5003
  3. PMS9003
  4. SPS30
  5. SDS011
  6. MPM14
  7. ZH06


After fusing values from all these different sensors, we will determine the AQI readings.

Ok, But What Exactly Is AQI ?

1 (35).png
1 (19).png

I have used the word AQI for a total of 24 times from the start of this article, but what exactly is it ?

Well, AQI ( that's the 8th time ) or Air Quality Index is basically a formula based indicator, and a rather bad one at it, ( I'll speak more about that in some later article)


The Air Quality Index (AQI) formula converts pollutant concentrations into a linear scale based on predefined, pollutant-specific breakpoints.


Here's the formula for AQI :

Let us understand with an example, consider that Good1 sensors measured the following values,

Temperature = 30 °C

Humidity = 60 %

PM1 = 18 µg/m³

PM2.5 = 42 µg/m³

PM10 = 90 µg/m³

VOC = 350 ppb


Now we will apply the given formula on each of these readings ( Note : Temperature and Humidity are not used in AQI, those are meant as a feature and for sensor error correction )


PM 2.5 AQI

42 lies between the range of 35.5-55.4ug/m3, so associated AQI range is 101-150.

So,

I(hi) = 150

I(lo) = 101

BP(hi) = 55.4

BP(lo) = 35.5

Cp = 42

Now, we know that Ip = (I(hi)-I(lo)/BP(hi)-BP(lo))*(Cp-BP(lo))+I(lo)

So, based on the formula, we get AQI for PM 2.5 as 117


Similarly, we get

PM 1.0 AQI as

PM 10 AQI as

VOC AQI as


Now, the Final AQI = Maximum pollutant sub-index

We have,

PM2.5 = 117

PM10 = 68

PM1 ≈ 63

VOC ≈ 65


So, highest value from the above 4 is 117, so AQI is 117, that's it, that's how the AQI value is determined.


Now, everybody has different rules in this calculation, different regulatory bodies try to introduce different rules, but the key principle remains the same.

Time to Sketch

Next I started with sketching the enclosure, I wanted Good1 to be easy to carry and look part of the edc package of almost any person, this meant designing Good1 in the form factor of something which people usually carry in their backpacks and wont raise alarm while carrying and handling it in public spaces.


Basically, Good1 should not look like a tin-foil freak's belonging and look more like something which could be, let's say a....... Power Bank! So, that was it, Power bank was the design inspiration for Good1, that's because the Power bank form factor will help store all the sensors in a clean package while not hindering performance and since Good1 is battery powered with 2200 mAh capacity, it might as well look the part.


Another reason for designing Good1 as a power bank was how strange existing commercial portable air pollution monitors looked like.


These are some of the designs of existing portable Air quality meters and most of them look, very..... Industrial, I personally would not have been comfortable in carrying something which looks like a multimeter on a busy street to monitor my surroundings, it draws in too many eyeballs, and I believe most of us don't want unnecessary attention.


So, I think an inconspicuous design is crucial in ensuring long term adoption of the device.

Here are some sketches of the Good1, now, mind you, I am no Industrial designer, but a man has to somehow put the thought out, right ?

Design With Constraints

1 (2).png
1 (4).jpg

After sketching to get a rough idea of what was in my mind, I started with bringing it to life using Fusion360, you can use any design software of your choice, but I prefer Fusion360 for the ease of 3d printing.


I had to keep in mind the various constraints imposed by different sensors of various sizes and how all the components could snug fit together without using any screws at all, the sensors have been specifically placed to allow easy air intake so as to make sure we get accurate readings and no humidity buildup occurs.


The trickiest part to design were the buttons inside it, it was a little tricky to get the right feedback while making them 3d printing friendly, but after some a little too many more than I would like to admit, we have finally ended up with a functional design.

3D Printing Time

1 (5).png
1 (4).png

Good1 Consists of 3 parts, only 1 of which houses all the electronics and acts as the base carrier plate.


All the sensors, the pcb, the OLED display as well as the 18650 battery attach to this plate, and the two enclosure components snap around this plate.


This means there's infinite scope for design improvement and iterations, and since the files are open sourced, feel free to make design variants of your own.

How Many Prototypes Do You See in This Picture

1 (17).png
1 (16).png

So after designing version 1 of Good1, I 3d printed it, and then I printed more, and more, and more, and I just kept printing, I did not expect to run through so many seemingly minor but capable of breaking down the project type problems throughout the design process, but finally here we are.

Time to Upload Firmware

1 (1).jpg

The firmware for project Good1 is a work in progress with more features on the way and current version does not support OTA updates ( Yes, OTA is incoming ) So you can find the latest version of Good1 on our Github repository.

Here is the link to our github repository : https://github.com/Air-Calibre/OAQ-Open-Air-Quality-Sampler-

I am thinking of exporting a binary file and a .bin uploader to save you guys compiling effort, that will make the project a simple plug and play.

Wait, Why Is It Buzzing ?

1 (11).png
1 (15).png
1 (10).png
1 (24).png

When I uploaded firmware for the first time on my Good1 sensor, it literally started buzzing within like 2 minutes, that's because the AQI in my surroundings was at 140 ! This is considered as very bad air quality and the sampler has been programmed to start buzzing at varying intensities based on the pollution levels.


This is to provide an auditory feedback and warning which urges the user to get out of the bad environment or encourage using mitigation measures to improve the air.


You can turn off or toggle the buzzing feature by a long press of function button for 5 seconds.


We have also made a budget solution to improve the air quality levels of your environment ( Check Last Step of this article )

Application

Components.png
1 (14).png
1 (37).png

The application of project Good1 currently is in its basic stage and basically retrieves data from Good1 using BLE, so no need to connect Good1 to the internet, it works completely offline. Here are some of the preview snapshots of our Application and we are inviting people to join in and contribute to the application effort.


The current version of Good1's application shows AQI as well all pollutant values along with logging features to generate .txt files locally.