Having recently had to struggle with the wiring on my home heating (y plan) to get it working with a nest, I would like to know why central heating systems are not more modern:

Why isn't there a central controller that receives information from all sensors (tank temp, boiler temp etc) and outputs commands to valves/pump/boiler rather than the series style wiring (programmer commands stat commands valve commands boiler).

Is it just a case of if it ain't broke don't fix it? , getting a single standard would be a pain , reliability and ease of maintenance or is there something more to why heating systems are relatively low tech to 'smart' tech?

I ask purely out of interest (not to criticise the existing wiring systems), I am interested to know if with having more localised generation (pv cells etc) if heating systems could become smarter and more efficient; use the pv cells if there is excess generation or if a home battery has charge, switch on an immersion heater instead of gas boiler for one example.

I am happy to admit the simplicity of it does allow for modification (shove an arduino in there with some relays etc).

  • It's the same reason cars didn't have anti-lock brakes and smog control equipment until the gov't forced it--people are too short-sighted to see the benefit and therefore don't want to pay for it.
    – isherwood
    Nov 14, 2017 at 14:02
  • 4
    High tech cost more to install and equipment failures are more frequent with high-tech vs older simpler technology. Yes it can save energy $.
    – Ed Beal
    Nov 14, 2017 at 14:02
  • This should clear that up for you: explainxkcd.com/wiki/index.php/927:_Standards
    – Ecnerwal
    Nov 14, 2017 at 15:30
  • Upvoting your answer because finding a way to link XKCD in a home improvement post, is epic! @Ecnerwal Nov 14, 2017 at 17:04
  • 1
    I feel that the parable of the toaster is relevant here.
    – Pont
    Nov 14, 2017 at 20:28

3 Answers 3


why central heating systems are not more modern

Because the established systems like Honeywell Y-plan or S-plan have many benefits ...

  • Low cost parts, simpler and sold in volume.
  • lower cost maintenance - heating engineers are used to these arrangements and know how to diagnose problems.
  • Fail-safe. A fault in a sensor can't turn on a boiler without turning on a pump.
  • Compatibility - the traditional programmer has a common backplate used by most manufacturers. You can snap in a new programmer without rewiring.
  • Reliability - most failures are due to wear in pumps and zone-valves, or due to lack of maintenance of the boiler.
  • Internet appliance makers have an appalling record for security, privacy etc.
  • I can buy parts and accessories for a 30-year old system. There's no track record for companies like Nest - we don't yet know whether they will be around in 30 years time or whether they will make drop-in replacements for current products.
  • Vladmir Putin's henchmen can't hack your heating to improve GazProm revenue.

So far as I can tell, systems like Nest only have significant benefit if your lifestyle is erratic and you need to remotely instruct your house to keep cold to save money.

Why isn't there a central controller

There seems to be a trend where more of the functionality is incorporated into the boiler. For example it is becoming more common for the boiler to control the pump so that it can run-on after the boiler stops, thus removing residual heat from the boiler.

Of course, systems like Nest don't change much of the above, they are designed to slot into those types of system. Often they function as a combined programmer and room-thermostat replacement with additional features.

Random example showing industry standard backplate interoperability:

enter image description here

  • Thanks for this, why wouldn't it move forwards though? for instance y plan moved towards s plan (I think) which required new parts, admittedly it wires in fine. what's to stop cylinder stats from becoming digital (like the nest room stat)?
    – Alex
    Nov 15, 2017 at 11:07
  • @Alex, it does move forward slowly. But it is just as likely to become more complex (in wiring up new widgets) rather than less. Innovation can sometimes (often?) lead to fragmentation and destruction of standards and therefore to confusion and cost. Nov 15, 2017 at 11:19
  • true, I was thinking along the lines of ODB2 in vehicles (in fairness volkswagen did their own thing) which is across board.
    – Alex
    Nov 15, 2017 at 11:28

Will your high tech solution work in 50 years? That's a conservative guess for the current age of the terrifying hazardous-substance-containing bimetallic thermostat that still works in my house. The terrifying hazardous substance is nicely contained in glass, and does not terrify me. I don't plan to take a hammer to it, and I don't expect it to suddenly break after 50 or more years.

As it happens, it's on a shelf, having been replaced with a battery operated automatic setback unit that still works fine with the two wires that were sufficient for it when it was installed, 50 or more years ago. But if that one fails, the old round honeywell will keep the temperature that's set, quite accurately, using its (quite sophisticated, actually) low technology.

Neither can be hacked over a network. One will survive an electro-magnetic pulse (EMP) attack, though there would be other problems with the overall system at that point.


Easy. The system is simple, modular and failsafe.

Closing switches is as simple as electrical gets, and easy to troubleshoot. You simply shunt (short) wires: shunt R and W to call for heat, R and G for fan, R and Y for cool. You're accustomed to thinking in protocols and APIs. Believe me when you think in both worlds, this is the simplest way. Yes it does constrain features somewhat, such as it's hard to tell an air conditioner with VFD how fast to run.

Anyway, you won't save wire; even CANbus would need 4 wires and lots of houses get it done with 2-3.

Modular is the crux of the matter. Most people don't replace their furnace, condenser and thermostat all in one go. And we wouldn't be having the "upgrade to a Nest" conversation if the standard were not open and simple.

What's more, by being modular, the system is hackable. Want to put 240 V electric heaters on a Nest? Trivial hack with a $12 A/C contactor. Want a Nest to control a millivolt heater? Also easy with a $3 relay.

Failsafe comes out of simple. It means that if the heater has fired, it can only happen because something shunted R and W, not because some glitch happened in some piece of software that only 3 people in India understand. If R and W are not shunted and the heater is on, the heater is defective.

  • Simple also means you don't need a 4 year degree to install/maintain the systems. Which keeps installation/maintenance costs (somewhat) under control.
    – Tester101
    Nov 14, 2017 at 19:55
  • Also for Silicon Valley salaries. Smart, employable people are already in short supply, and shortages are impeding innovation. The last thing the tech industry needs is another industry poaching all their geniuses. Nov 14, 2017 at 20:43
  • @Harper you actually just answered a question I was going to ask: how do I force the heat to run without a thermostat? It is getting cold where I live and I need to hard-wire the heat to run while I paint the wall that the thermostat is on. I was hoping it was this easy. R+W=heat.
    – user4302
    Nov 14, 2017 at 20:59
  • @Snowman yup, twist them together until you start to melt, then unhook them. You might need G also, or not... some furnaces have logic to run the fan once the heater is warm enough. Nov 14, 2017 at 21:31
  • Spot on about the ability to modify the system easily :)
    – Alex
    Nov 15, 2017 at 11:18

Not the answer you're looking for? Browse other questions tagged or ask your own question.