Smart HOME made

You can smarten up your home yourself, according to your own needs but before you get into the thick of it, it’s worth considering whether it’s worth the effort. Let me save you some time and share my experiences about building my own Smart Home system.

What do we need to consider when establishing a smart home? One of the aspects is the size of the home. A 50 square meter home probably requires less automation than a three-storey building with a cellar and tall ceilings. In most cases, it’s mainly the blinds and lighting that are crying out for automation. We should factor in usability and convenience. It’s far more comfortable to control everything through a mobile app. Nonetheless, this might all depend on the person’s eagerness to make new discoveries, which is a trait that all developers have. This was that motivated me when I started my project in 2017.  

As with all projects, first I sat down and thought about the main points and features that I wanted to incorporate into the system. For example, I wanted it to function without internet access and although it’s important for all devices to run through a single central unit, the heating and lighting should still be independent to the structure. This is important, so my home’s basic functions wouldn’t shut down if the smart control didn’t work for some reason. Furthermore, it should be variable and the firmware should be replaceable in case of a malfunction, and naturally, it should be cheap. However, the home systems developed by manufacturers have a closed nature, which means that new devices from different systems can’t be integrated. Moreover, the prices aren’t that affordable – particularly if it’s for a three-storey house. So, what’s the solution? If inexpensive is what you’re going for, look for the following words: Made in China. I started searching for easily affordable devices from China which could still be adapted to suit my needs despite the low price tag. This is how I came across Sonoff products, as well a community of people that spend their free time re-configuring and hacking them.

Where to start? First of all, I needed a soldering iron. Then I ordered some micro-controllers which I could install into the devices. After the first basic devices showed up, I started building the architecture. I connected to the OpenHAB system (ran on Raspberry Pi) via smartphone or computer; there’s an MQTT server that only costs $9.90, which all the smart devices communicated through (except for the heating system) and the OpenHAB was connected to it as well. I functionally detached the boiler, so it would still run when all the other devices are inactive.

During the installation, I ran into an interesting problem related to the lighting. What if someone manually turns on the lights? Well, in this case, the light switch can’t send a signal to the system, so I have no information on the current state as it can only be turned on and off with a 433 MHz token. So, I had to put together some hardware that would indicate if the lights were turned on. However, this gadget communicated with the system via Wi-Fi, so I had to use two different technologies, the controls of which had to be unified through the MQTT server. This rounded out the system. 

I reached the stage of the actual implementation. I lost countless devices and suffered a lot of “occupational” accidents by the time everything was serviceable. One of the most memorable dead ends was the matter of the video entry phone. I first wanted to replace the software of the device ordered from China, but I quickly realised that the “buggy” software was sending Wi-Fi passwords to a Chinese cloud service. Hmm. Still, I didn’t give up and tried figuring out a way to disable the execution program. After many sleepless nights I finally realised that it’s impossible since I didn’t have the right keys, so the program wouldn’t provide me with access. It was easier to build my own entry phone. This is just one example of the many experiences I had.

An additional fundamental question was how to control the blinds. While in Hungary, a basic blind motor costs around 15,000 HUF per blind, and a smart blind motor is around 50,000 HUF per blind, a Sonoff Dual switch only costs $10. So I chose the latter, which I augmented by ordering installation boxes.

I wanted to install separate temperature sensors in every room, so I picked up quite a few Xiaomi temperature and humidity sensors. The only issue I had was that they could only communicate via Bluetooth. This meant extending the features of the previously installed 433 MHz Gateway, which required new hardware.

At the same time, I had to setup the smart control of the LED lighting, motion sensors and boiler. The boiler was a bit tricky as the adapter for a Vaillant boiler costs approximately 70,000 HUF, which somewhat pushed my psychological monetary boundary. However, I discovered a German forum where engineers build adapters as hobby projects that perfectly integrated with the Raspberry Pi system. They manufactured circuit boards and the related accessories for this in China, which (just like building a model boat) simply requires soldering. All of this cost about €30 and took quite a few hours of work. Another exciting and seemingly whimsical device is the smart plug, which can measure the consumption of electricity and could even tell you when the oil is hot enough to put your frozen fries in the electric fryer. As you can see, I did my best to “smarten up” as many devices as possible.

What have I gained from all of this? I acquired a great deal of useful experience with MQTT, InfluxDB and Watchdog systems and I have mastered the skill of using a soldering iron. Developing the system took me more than three years and during this period, I received over two hundred parcels. I accumulated a vast array of devices coupled with a huge number of hours spent programming, researching and utilising all the IT knowledge I had. How much did all of this cost? The devices I used and applied added up to around $2,000. Would I do it again? I spent over three years configuring my smart home, and there is still room for improvement. Nevertheless, if I could start things all over again, I believe I’d definitely get into the project again. The big question here, is whether this was all worth the effort. While a SMART home can improve your life at home, it also depends of whether developments such as these will increase the value of a house, but for me it’s also about whether projects like this one will inspire others to start a SMART home project of their own.