Home Automation near you

Two Australian mandatory standards sit over corded internal window coverings. Supply and labelling are covered by the Trade Practices (Consumer Product Safety Standard — Corded Internal Window Coverings) Regulations 2010, and installation by the Competition and Consumer (Corded Internal Window Coverings) Safety Standard 2014. The installation standard requires the installer to follow the packaging instructions, attach a label showing their name and contact details, install cleats or cord guides correctly, leave warning labels and swing tags in place, and make sure loose cords cannot form a loop exceeding 220 millimetres at any height below 1,600 millimetres. A cleat used to secure a cord may not be installed lower than 1,600 millimetres above floor level.

The warning wording on the cord tag says what the standard is actually about: young children have died by wrapping loose curtain and blind cords or chains around their necks, and the tag directs people to secure cords with cord guides or wind them around a cleat, and to move cots and furniture away from window covering cords. The standard applies in domestic dwellings, though not to installation in caravans, mobile homes or boats.

Motorisation sidesteps the whole problem because there is no cord to secure, which is why it is the one automation worth doing first in a house with small children. If you are motorising only some windows, the rest still have to meet the standard, and the installer is still required to leave the labels and cleats in place on those.

There is a category of automated device where the failure mode is not inconvenience. An automated gate or garage door is the way vehicles and often people get in and out. An automated lock is the way the house is exited in a fire. Automated pool equipment sits behind a barrier that has its own rules. In each case, the question to ask the integrator is not what happens when it works, it is what a person does when it does not, with no power, no internet and no phone.

The answer has to be physical: a key, a release handle, a manual crank, a switch on a wall. Ask to be shown the override on each device at handover, ask where the key or release tool is kept, and write it down for the house. A release handle nobody can find in the dark is not an override.

This is also the reason to be careful about automating a device that was already the failure point. A garage door with a tired spring does not become more reliable when an app can open it; it becomes a heavy door with one more thing that can go wrong in front of it. Fix the mechanism first, automate second.

Most Australian houses arrive at automation with existing equipment: split systems with infrared remotes, roller blinds on chains, a garage door with a rolling-code remote, a pool pump on a mechanical timer, and a hot water system on a controlled tariff. Each can be brought into a system either by adding a controller alongside it or by replacing the device with one that speaks the protocol natively.

Adding a controller is cheaper and reversible, and it is usually right for equipment with years of life left. It is also less reliable, because an infrared blaster or a relay bolted onto someone else's product has no feedback: the system believes the air conditioner is off because it sent an off command, not because it checked. Native devices know their own state, which is why automations built on them behave predictably and automations built on blasters drift.

The sensible rule is to replace at end of life and adapt in the meantime. Motorised blinds are the exception worth bringing forward, both for the cord safety reason and because retrofitting a motor into an existing roller blind is often more expensive and less satisfactory than replacing the blind outright.

A house lasts generations, a blind lasts decades, a motor lasts many years, and the software controlling both is on a much shorter cycle. That mismatch is the structural problem in home automation and it cannot be designed away, but it can be designed around by keeping the long-lived parts dumb and independent.

In practice that means the motor, the actuator, the valve and the wiring should be standard items that another installer can source and replace, while the cleverness lives in a controller that can be swapped without touching them. A system where the blind motor only works with one discontinued hub is a system where a hub failure becomes a window covering replacement.

Ask the integrator directly: if you are unavailable in five years, what does the next person need to know, and which parts of this can they buy off a shelf? An answer that names generic components and open protocols is worth more than a slicker demonstration.