Skip to main content

My first post about this project described where the concept started. My bedroom is poorly ventilated, and I was motivated to create a system that opens and closes the window in response to air quality data. However, while useful, the project itself lacked charisma and wasn't yet telling a clear story beyond itself. In this update, I'll give you an idea of how the concept evolved and where I've got to with the implementation.

Fundamentally this project is about monitoring air quality during sleep. We all need a certain amount of quality sleep every day, which relies on good levels of oxygen in the blood, and thereby air quality (for more see here). The data in this article suggests that this would mean keeping CO2 levels at or below 1000ppm for good quality sleep. Yet for many of us, the concentration is typically well over this figure, reaching about 1800ppm by morning (to say nothing of other indoor pollutants). How well can we really be sleeping?

This is about developing an object to detect whether the air you are breathing is good enough quality to really give you sweet dreams. The bedroom is a dark environment, and if there's any information being displayed by this object, it needs to be intuitive to someone in a half-awake state. I started thinking about lamps - specifically the Davy lamp, a device invented for coal miners in the 18th century by Humphry Davy. The design of this lamp meant that the colour or brightness of the flame would change when dangerous gases such as methane or carbon monoxide were present. This was vital information for miners deep underground, who might be at risk of explosion or suffocation.

Miners' safety lamps

Image: The Davy Lamp (Image source: The National Coal Mining Museum for England, via this article)

I gathered a few other visual references of things that glow in the dark to get a sense of how this might end up looking. I also broke down the project into modular chunks and arranged them in a logical order, so that whatever I was able to build would actually make sense however far I got.

Project flowchart

Early lighting experiments

Rather than simply having the colours always on, and having the AQ data linked to the colour or brightness in a linear way, I wanted to see if I could create a kind of animation with the LEDs that can be modulated by changing a few key parameters. In this way, there would be a sense of constant 'life', and the character of the animation could change according to the AQ data.

I began experimenting with different kinds of time-series, such as sinusoidal and exponential functions. I decided to use the HSV (hue, saturation, value) format to specify the LED values, as an easy way to cycle through the colour space using one number, compared to RGB for which you need to specify three (for a very good tutorial of using HSV see here). The result of this experimentation was a cool simulated flame animation for which I’m able to tweak the values.

I'll have to do a bit more work to turn these basic animations into something you can interpret somewhat easily. This is the 'design' that needs to accompany the engineering. I also realised I'll need to have a light-dependent resistor to detect whether it's day or night and change the brightness accordingly. Overall, though, I'm happy with this as the basis for a night-time air quality display.

LED lights

Testing and coding the LEDs

Materials, shapes, next steps

The LEDs will go into a lantern object. It's tempting to go overboard to create something sculptural and unique, perhaps something that evokes lava lamps and bedroom kitsch. On the other hand, I want to keep this 1) easy to replicate, which means simple, cheap and probably quite minimal, and 2) the kind of thing you might actually want to keep in your bedroom. I also have only fairly basic materials and fabrication facilities available to me at the moment!

Since I’m using LEDs, I won’t want to see the individual lights and I will need to use something translucent like paper or fabric to diffuse their glow. ‘Bedside table’ materials I might use could be wood, paper, fabric, wire, or even ceramics. I’ll be looking for common materials or objects that I can repurpose - and ideally something sustainable!

Creating the flame animation

Overall scheme

Given the scope of all of this, I suspect I’m not going to get to incorporate the actual window opener mechanism, but here is where it would fit for a later stage.

Flow chart

In an upcoming post, you’ll get to see how some of this has turned out…

The future is already here, it's just not evenly distributed yet...