As I've been playing with the scheduling features of the Evohome Controller, I've been thinking about the various headline features that smart central heating systems are marketed on, and why they're needed.
Traditionally your central heating system had one job to do: ensure that your home is at a comfortable temperature. Now there's a second job for it: minimise overall energy use. Arguably, through the use of a well positioned thermostat and TRVs on each radiator you could achieve job number one. The recent explosion of smart central heating systems are all focussed on job number two: minimising energy use.
There appear to be two basic approaches to minimising energy use. The first is to turn off the heating when you're not at home. And the second is to only heat the spaces that you're using when you are at home. Different companies are tackling one or both of these approaches. For example Nest, Hive and Tado are using motion detection and geo-fencing to tackle the first approach, while Honeywell and Heat Genius are also zoning your house and controlling each radiator independently in order to tackle the second approach.
The pre-smart approach to turning off the central heating when you're not at home was to use a programmable timer. When you delve into the detail it feels like all of the smart central heating systems are really continuing that approach. Some of the systems still require you to program them with the times and temperatures that you want the central heating to work to. Even the ones that promote a 'learning' capability are just using motion detection and geo-fencing to work out your schedule without you manually entering it yourself. Some of the learning systems benefit greatly from you manually entering your schedule first, so they've got a baseline to start learning from.
So if your routine is entirely predictable then all of the smart central heating systems can ensure that your home is nice and toasty when you're in it. It's the unpredictable that some of them struggle to deal with automatically. Those occasions when the train is delayed, the meeting overruns, the motorway is congested, and the heating system kicks in when you're not actually there. In this case no amount of smart learning algorithm is going to be able to compensate and delay heating the house until you're actually going to arrive. As far as I can see, only the systems offering geo-fencing can automatically cope with the unexpected. The others will have to make do with manually controlling the heating system remotely via an app or the web.
When it comes to zoning the house there's very little differentiation between the smart central heating systems. All produce wireless TRVs that allow a central controller to independently adjust the heat output of the radiators. Some of these TRVs allow the temperature to be adjusted locally as well, and some have displays to report on room temperature or temperature setpoint.
Traditionally your central heating system had one job to do: ensure that your home is at a comfortable temperature. Now there's a second job for it: minimise overall energy use. Arguably, through the use of a well positioned thermostat and TRVs on each radiator you could achieve job number one. The recent explosion of smart central heating systems are all focussed on job number two: minimising energy use.
There appear to be two basic approaches to minimising energy use. The first is to turn off the heating when you're not at home. And the second is to only heat the spaces that you're using when you are at home. Different companies are tackling one or both of these approaches. For example Nest, Hive and Tado are using motion detection and geo-fencing to tackle the first approach, while Honeywell and Heat Genius are also zoning your house and controlling each radiator independently in order to tackle the second approach.
The pre-smart approach to turning off the central heating when you're not at home was to use a programmable timer. When you delve into the detail it feels like all of the smart central heating systems are really continuing that approach. Some of the systems still require you to program them with the times and temperatures that you want the central heating to work to. Even the ones that promote a 'learning' capability are just using motion detection and geo-fencing to work out your schedule without you manually entering it yourself. Some of the learning systems benefit greatly from you manually entering your schedule first, so they've got a baseline to start learning from.
So if your routine is entirely predictable then all of the smart central heating systems can ensure that your home is nice and toasty when you're in it. It's the unpredictable that some of them struggle to deal with automatically. Those occasions when the train is delayed, the meeting overruns, the motorway is congested, and the heating system kicks in when you're not actually there. In this case no amount of smart learning algorithm is going to be able to compensate and delay heating the house until you're actually going to arrive. As far as I can see, only the systems offering geo-fencing can automatically cope with the unexpected. The others will have to make do with manually controlling the heating system remotely via an app or the web.
When it comes to zoning the house there's very little differentiation between the smart central heating systems. All produce wireless TRVs that allow a central controller to independently adjust the heat output of the radiators. Some of these TRVs allow the temperature to be adjusted locally as well, and some have displays to report on room temperature or temperature setpoint.