Back in January, I wrote about putting my toe into the waters of Home Automation. As I said at the time, there’s a bewildering array of products and standards out there. Over the past couple of months, I’ve been taking a look at some of the HA solutions out there, and I think I’ve landed on one that seems to fulfil my requirements pretty well.

I had decided back in January that my solution would be one that used Z-Wave, and that’s still the case. I’ve been able to purchase Z-Wave HA devices from several different manufacturers, and they interoperate as promised. I’m satisfied that the choice of Z-Wave is a good one for me.

As I said in January, there is a wide range of possibilities in the choice of the controller for a HA system. I could buy an off-the-shelf unit such as the Fibaro Home Center 2, the Zipato ZipaBox, a VeraEdge controller, or a HomeSeer controller. Or I could buy just the controller software, such as HomeSeer, and install it on a PC or a Rasberry Pi box.

In the end, I decided not to purchase an off-the-shelf unit. Fibaro still seem to be struggling to deliver a stable version of their latest software for the Home Center 2, and not being able to trial the other hardware products meant that purchasing one would be a gamble. It seemed to me that a better approach would be to trial a software solution, using an existing computer.

I ended up taking a look at the following software solutions:

• HomeSeer
• Charmed Quark
• HomeGenie
• Domoticz
• OpenHAB

The first two are commercial products, the rest are open source projects, funded by donations.

For my testing, I purchased a variety of devices:

• Fibaro Wall Plugs – the neatest device around, with both a visual display of power being used and a built-in meter to collect data. I bought several of these to control and monitor a variety of devices from lights to refrigerators and computer servers.
• Z-Wave.Me Outdoor Wall Plugs – used to control the water pumps on our garden ponds.
• Fibaro Relay Insert – used to control outside lighting
• Fibaro RGBW Module – used to control LED lighting in the living room
• Everspring SP103 – Outdoor motion sensor
• Aeon Labs Multi Sensor – Outdoor motion/temperature/lux/humidity sensor.
• Aeon Labs ZStick Series 2 – a ZWave controller in a USB dongle that interfaces with controller software.

With the devices installed, and included in the ZWave network (registered in the ZStick controller), I connected the ZStick to a variety of software controllers in turn…

HomeSeer has been around for a while. That means it’s comprehensive (it can control a wide range of devices using a variety of standards), and it has a wide range of third-party add-ons. However, it still strikes me as being expensive: $249.95 for the basic version of the controller software and $199.95 for the UI designer software. If I were to go with HomeSeer, it would probably make more sense for me to buy the basic (linux-based) HomeTroller Zee controller at $199.95. I installed the trial software of HomeSeer version 3, and used it for a week or two. It worked, and the diagnostic information was copious. However, the interface struck me simultaneously as being rather basic, and yet somewhat complex in places. I decided not to proceed further with HomeSeer.

I admit I only took a cursory look at both Charmed Quark and OpenHAB. I found both difficult to set up, and got the impression that I would spend more time fiddling with them than using them. OpenHAB, in particular, seemed aimed at programmers and developers, rather than end-users at this stage. As a result, I moved on.

HomeGenie is also something that will delight programmers and developers at the moment. Nonetheless, I was able to get it up and running very easily on Windows, and it works well. Here’s the “dashboard” that I see for the devices I currently have in my HA network:

It’s primarily the result of the efforts of one developer, and he’s done a very good job. It’s still at a fairly early stage, so, for example, if you want to develop automated control of your devices, you will find yourself very rapidly at the limit of what the built-in “Wizards” are capable of, and have to turn to grown-up programming to get the job done. That’s all very well, if that’s your thing, but it’s really not what I want to do any more. I made a donation to the project, because I appreciate what has been achieved, and I hope it continues to develop. There’s a small (around 350 members) community forum where ideas are exchanged and bugs highlighted for solving. I could certainly make use of HomeGenie, if there were not other, and for my purposes, better solutions available.

Domoticz is another open-source project, and while it is primarily led by one developer, there are others actively involved, and the community forum is large (around 3,270 members) and active. Domoticz and HomeGenie are similar in many ways, but there are a few differences, which can be both strengths and weaknesses, depending on where you stand. Here’s the Domoticz dashboard:

The strength of Domoticz is its maturity; it already has solutions and support for a lot of hardware. In particular, it supports the reading and display of data supplied by our “Smart Meter” for our gas and electricity consumption. With the simple purchase of a cable, I was able to connect our smart meter to the Domoticz system and get real-time graphs of our energy consumption. Here’s what I see for our electricity usage:

It not only records our consumption (blue), but also the electricity generated by our solar panels (green) that is returned to the electricity grid. Gas consumption can also be tracked:

HomeGenie, at the moment, does not have this connection to Smart Meters built-in. The programmatic interfaces are there, and such an interface could be built, but I’m not the one to do it.

For me, the strength of HomeGenie is that it is a complete solution, in that the developer has written software to control Z-Wave devices directly. Domoticz, on the other hand, hands off the control of Z-Wave devices to another piece of software, developed by a separate open source project: Open-ZWave. That means that there is the chance of issues arising out of effects caused by the fact that there are two different software components developed by two different groups. I’ve already come across an issue with my Z-Wave sensors, which may be caused by a bug in Domoticz, a bug in Open-ZWave, or some combination of the two.

Nonetheless, I can live with that issue, and the quirk that the Fibaro wall plugs don’t always show the correct status in Domoticz. There are two reasons for this:

• The fact that I can track our energy consumption directly in Domoticz (as shown above), and
• the fact that Domoticz supports Blockly for building automation programs.

Think of Blockly as Lego for programming. It’s wonderfully easy to use, and I’ve already programmed the motion sensors on the driveway and by the front door to turn on the outside lights if someone comes along during the hours of darkness (which have been defined, using Blockly and a “virtual device” in Domoticz that I defined: IsDark):

So for the moment, I’m going to stick with Domoticz for our Home Automation system, with HomeGenie held in reserve as my fallback position.

And while I tested both on Windows, both Domoticz and HomeGenie are available for the Raspberry Pi. I’ve also stuck my toe into the waters of using a Raspberry Pi, and Domoticz is currently running very happily on it. But that’s the subject of another post, I think…

A little while back, I blogged about the fact that broadband internet in our neck of the Netherlands is like a piece of wet string. In trying to drum up local support for improving the situation, I’ve been looking into scenarios where real broadband internet (that is, speeds of at least 10 Mbps, and preferably 20+ Mbps) are going to be required.

From a purely selfish perspective, I’m uncomfortably aware that given my age, there’s going to come a time when I may need to depend on healthcare services delivered through broadband internet direct to our home. One of the aspects of such services is support for home automation (or Domotica, as it’s often called here in the Netherlands). While HA is usually thought about in terms of ease, security and energy efficiency, there’s also a healthcare aspect to it as well. For example, remote monitoring can allow patients with dementia to continue to live at home in the environment that they are comfortable with.

So I thought that I should start exploring the possibilities of HA for our home. I’d start simple, for example, have certain lights come on at around sunset, and turn off at midnight, or install a motion sensor in our driveway to get early warning when we have visitors; and at night the sensor could also turn on lights for the driveway. That would in turn be a convenience for visitors and a deterrent for intruders.

Of course, these simple scenarios could be realised with a few timers and lights controlled by motion sensors, but the real advantages start to come when individual items are linked together into a system. An individual neuron doesn’t do much – intelligence is the emergent property that arises out of the interconnection between billions of them. While I’m not looking to build a brain, a flexible method of controlling the environment and security of our house would be nice.

However, when I started researching the technologies available for Home Automation, I soon realised that there’s a dog’s breakfast of competing products and standards out there. Some have been around for years. The X10 standard for example was developed in 1975, and while popular and used by an installed base of millions of devices, is beginning to show its age and limitations. Other newer products, while technically impressive, rely on proprietary technology unique to the vendor. Examples are the Insteon or the Loxone systems. Navigating through the shoals of reefs and whirlpools of Home Automation was not going to be an easy matter. As the Automated Home site says:

There are a multitude of Home Automation systems available, from budget plugin modules that are easily retro-fitted into existing properties, to professionally designed bespoke installations that require a CI (customer installer or integrator) and structured wiring at time of build.

I think I can forget about the professionally designed bespoke installations with their structured wiring – I’m going to be looking at something that can be retro-fitted easily into our farmhouse. That means that I’ll be looking at wireless systems as much as possible. I’d also prefer to go for products that share a modern de facto standard, rather than rely on a single vendor. As a result, I’ve decided that devices that implement the Z-Wave wireless communications protocol are probably my best bet, given that Z-Wave is supported by over 250 manufacturers worldwide.

I’ve started small, and invested in one switchable power plug (the Fibaro Wall Plug) and a motion sensor (the Everspring SP103) – both Z-Wave devices.

The next step is to make a choice about the controller for the HA system. Once again, there is a wide range of possibilities here. I could buy an off-the-shelf unit such as the Fibaro Home Center 2, the Zipato ZipaBox, a VeraEdge controller, or a HomeSeer controller. Or I could buy just the controller software, such as HomeSeer, and install it on a PC or a Rasberry Pi box. There are also open source projects for Home Automation software, such as HomeGenie and Domoticz.

I’m still exploring the possibilities here. I’ve come across a few issues so far. For example, while the Fibaro Home Center 2 looks good on paper, judging from the user support forum, Fibaro are struggling to deliver a stable version of the controller software. The ZipaBox relies on a Cloud service to provide much of the controller functionality, and that’s a design choice that I personally would be less comfortable with. The HomeSeer software has been around for a while, and is now in its third generation. That does mean that it is very comprehensive; it can control a wide range of Home Automation hardware – far more than I would ever need or use. It also has a wide range of third-party plugins. However, its user interface can best be described as old-school utilitarian. There is an additional software product that can be used to design custom user interfaces for smartphones and tablets. And the HomeSeer software strikes me as being pricey: $249.95 for the basic version of the controller software and $199.95 for the UI designer software. If I were to go with HomeSeer, it would probably make more sense for me to buy the basic (linux-based) HomeTroller Zee controller at $199.95.

So I’ll probably spend the next month or two trying out some of the controller software that’s available for Z-Wave networks, and hanging out in the user support forums to read about the experiences of others who are using controllers, both packaged and software-only solutions. Watch this space.