DIY Smart Home: Part 1

I got an Adruino kit for my birthday this year because I've been interested in getting into electronics and programming small devices, primarily with the aim of getting some home made smart home projects off the ground.

As I've argued before, the current landscape doesn't look great for larger scale smart home solutions. At this point, out of all of the smart home products I own, only 2 have never done anything to cause me concern. And frankly, I understand why with both; Ikea and Hue.

My Nest products were of course part of the Google fiasco where they wanted to shut down the old Nest service and replace it with... effectively nothing. Sure, I could still control my devices. But, only through Google. And that simply isn't sufficient.

MyQ has made a number of breaking changes to their API over time.

August was bought out by a larger company, and while they haven't screwed me over significantly as a result, they have made some breaking API changes as well.

Which just leaves Philips Hue and Ikea. I get why the Hue ecosystem has never screwed me over. The products cost enough to pay for the service for ages. I know they use some decent quality parts. But, at a manufacturing level, there is simply no reason why a Hue light costs as much more than a Tradfri one. They're playing it smart and charging enough for their product to sustain their offering.

Ikea on the other hand is cheap. But, they have 2 very big things going for them; it isn't their primary revenue stream, and they aren't supporting any real cloud infrastructure. In short, their operations are comparatively affordable. And it doubles as a draw into their stores where they make more money. I'm sure they have some ongoing operational costs. But, nothing compared to a company, say, providing access to camera feeds remotely, or even to companies providing a cloud based API for remote management.

All of that being said, there is no reason really why even those companies couldn't screw me over. I mean, smart lights aren't the fad any more. If sales drop off, and Philips hasn't been managing their cash well, then I could see them shutting down the devices. And Ikea could easily kill off the division, and they could simply choose to be jerks about it. They don't need too... but then neither did many of the companies which chose to shut those operations.

So. Enter the Arduino. Right now, it isn't all that impressive. It is just a board, connected to a rather large breadboard, with a BLE module and a temperature and humidity sensor on it.

This feeds the data into a Pi Zero W, which in turn pipes it on over to MQTT.

I'll be adding more sensors. But, even what I have right now is impressive enough. A nest temperature sensor costs $40CAD. It requires a Nest Thermostat to work. It doesn't record humidity. The API is totally locked down. About the only thing going for it is that it is more attractive than what I have. If I sourced JUST the components I needed, I could have gotten a programmable board with BLE built-in for $10 or less, and then a pack of those sensors for the same price, I could then splurge on a power solution and have a more functional device which no one is ever going to shut down or charge for.

What I ultimately need to do is package it all up. I'm still thinking about what that looks like. I'd like to have a touch based GUI on each device. I'm thinking that something akin to a Nest Hub is the right design. Large enough to fit a bunch of sensors, a mini PC and screen. And then shove one in each room.

What I'm missing at the moment is a light enough OS to run a GUI and the background stuff on a Pi Zero W. Because, the price point is hard to beat. Then, I also need a case and a display. Though, I probably need to source the display before thinking about a case.

I'd also love to power these off of solar and rechargeable batteries. But, we shall see.