What is the real story around heat pumps?

[pcprogrammer]

The house is reasonably well insulated. Average annual energy consumption for heating the house, based on my experience with the wood burner is ~16500 KWh. Worst day average this season is about 8.75 KWh. It is a big house with 240m2 of living surface.

Those numbers don't add up. Assuming that you mean kWh (kilowatt-hours) rather than KWh (kelvin-watt-hours), 8.75 kWh per day every day would add up to only 3194 kWh per year.

I assume that when someone sees k or K before W they will know it means kilo, same as in KB for kilo bytes and Kb for kilo bit, where there is an actual difference for the b. I'm not sure where I picked it up but have always used upper case letters in (digital) technology when shorting Kilo, Mega or Giga.

But that aside. I might be wrong in using the hour when expressing energy usage, but to clarify on a very cold day the system uses on average 31,5MJoule (8.75KWh) every hour, so 756MJoule (210KWh) per day. But this is only when it is -4degree Celsius or less outside. This is an average based on usage over the years.

Per heating season our system uses on average 58781845200Joule (16328,3KWh) to keep the house warm. The season is about 6 months. This number is based on how many times I have to heat up the storage tank in a season and the average amount used form the tank.

This includes the losses from the tank of course, which is maybe 0.5 to 1 degree per hour when the tank is at its top temperature, and drops down when the temperature in the tank goes down.

My issue with it is that we have a 9KVA mains supply and the "Linky", our intelligent electricity meter, might cut us off when the current surpasses the 45A it allows. With a 2KWh water heater, a refrigerator, a freezer and the heat pump possibly running at the same time might be enough to do this.

... or are you possibly using kWh (unit of energy) when you mean kW (unit of power)?

The water heater has is a resistive immersion heater type rated 2KWh, which means it draws ~8.4A when on. Not sure about the ratings of the refrigerator and the freezer, but it all adds up. The water heater is on a clock and only heats during the night. We won't bother with trying to reduce the energy bill on that, because our warm water consumption is not that high.

It might be FUD, but we heard stories about the "Linky" cutting you of directly when the current goes over the set limit. No idea if it is true and it might never happen when the heat pump is running. I just want to be sure it won't.

For one, you could use some sort of lock-out mechanism to stop heating while the water heater is active.

Could be an option but would require sensing that the water heater is on.

.....

The remainder of your post is very educational and I will certainly look into adaptations to my system. At the moment the regulation for the room temperatures is based on a set temperature and a hysteresis between the measurement. The rooms with the radiators start heating when the temperature on the sensor drops half a degree Celsius below the set point and stops when the set point is reached. Depending on the water temperature there is an overshoot, but that is fine for us. For the rooms with the underfloor heating it is 0.1 degree Celsius. The response is much slower but will also give overshoots. On every valve in the system there is an on/off thermo electric actuator that is controlled by the Raspberry PI. Maybe I can change it to run more constantly and regulate the temperatures. Need access to the heat pump system for that though, and that needs aid from the installer and technical data on protocol and interface.

[Bud]

Not sure why is the conversation... According to EU dickheads you should be buying T-shirts and sunglasses, rather than worry about heat pumps.

[nctnico]

As for your house being split between floor heating and radiators: I have no idea whether there are any off-the-shelf solutions for this, but maybe it is possible to run the heating in some sort of interleaved mode? Like, split this into two independent loops, and then heat them alternatingly, one at ~ 30 °C and the other at ~ 55 °C?

There is an off-the-shelve solution for that. It is a simple mixer unit with it's own pump that takes in hot water from the hot water source and mixes that with the cold water coming from the tubes that make the underfloor heating system. These have been in use for decades.

For my underfloor heating installation I have made a boiler simulator which uses a regular Honeywell digital thermostat which tells the simulated boiler what the water temperature should be. All the room temperature control loop intelligence is in the thermostat. The boiler simulator opens the hot water supply until the mixer unit reaches the setpoint temperature.

[zilp]

There is an off-the-shelve solution for that. It is a simple mixer unit with it's own pump that takes in hot water from the hot water source and mixes that with the cold water coming from the tubes that make the underfloor heating system. These have been in use for decades.

That doesn't solve the problem. The problem isn't that the water is too hot for the under floor heating, the problem is that the high temperature is inefficient to generate with a heat pump, and it's kinda bad to have your heat pump generate 55 °C at great expense and then cool it back down to 30 °C.

[Marco]

Google tells me the linky limits current assuming a reference voltage of 200V, so it's going to be closer to 10kW in reality.

PS. is that really still in there in France? The Dutch smart meter regulations removed that pretty much immediately (or rather it's forbidden by government regulations for them to have contactors now). Moving from the sealed main breakers with large safety margin to a hard switch off probably caused way too much problems.

[nctnico]

There is an off-the-shelve solution for that. It is a simple mixer unit with it's own pump that takes in hot water from the hot water source and mixes that with the cold water coming from the tubes that make the underfloor heating system. These have been in use for decades.

That doesn't solve the problem. The problem isn't that the water is too hot for the under floor heating, the problem is that the high temperature is inefficient to generate with a heat pump, and it's kinda bad to have your heat pump generate 55 °C at great expense and then cool it back down to 30 °C.

If the heat pump is optimised for higher temperatures it might not be a problem. Mixing to a lower temperature doesn't matter as no energy gets lost. Still, if it where up to me I'd change to underfloor heating in the entire house.

[Marco]

It's a shame it's getting a bit late in the year to see how far you can get with some DIY modding of the radiators with PC fans and some thermally controlled activation like Heatbooster.

Proper fan coil units aren't that expensive either and will give you a guaranteed heating output so you can just do the math. You could gamble on the DIY solution and replace the radiators with fan coil units if it doesn't work out.

[zilp]

I assume that when someone sees k or K before W they will know it means kilo,

I mean, I guessed as much, but thats not exactly a reason to write it wrong, is it?

same as in KB for kilo bytes and Kb for kilo bit, where there is an actual difference for the b.

That was somewhat of a convention to differentiate the base 2 prefix (K for 1024) from the base 10 SI prefix (k for 1000) when specifying memory or storage sizes because they tend to be in powers of two. But for one I doubt that you are calculating in 1024 Wh units, and also, that convention has been kinda deprecated in favor of the IEC binary prefixes (Ki/Mi/Gi/Ti/...), so the unambiguous way to write that would be KiWh.

But that aside. I might be wrong in using the hour when expressing energy usage,

No, it would be correct for expressing energy usage. Because it is a unit of energy. But it is wrong (and confusing) when expressing power. This is the same as specifying speed in kilometers.

but to clarify on a very cold day the system uses on average 31,5MJoule (8.75KWh) every hour,

I.e., 8.75 kW times 1 hour per 1 hour = 8.75 kW * 1 h / 1h = 8.75 kW. I mean, "8.75 kWh per hour" is also fine, but if you specify just an amount of energy, it is unclear what time span that is for, and especially when you then say something like "the day average is", as that suggests that the time span might be a day.

so 756MJoule (210KWh) per day. But this is only when it is -4degree Celsius or less outside. This is an average based on usage over the years.

Per heating season our system uses on average 58781845200Joule (16328,3KWh) to keep the house warm. The season is about 6 months. This number is based on how many times I have to heat up the storage tank in a season and the average amount used form the tank.

Yeah, that makes sense. And indeed not that bad for 240 m².

The water heater has is a resistive immersion heater type rated 2KWh, which means it draws ~8.4A when on. Not sure about the ratings of the refrigerator and the freezer, but it all adds up. The water heater is on a clock and only heats during the night. We won't bother with trying to reduce the energy bill on that, because our warm water consumption is not that high.

It might be FUD, but we heard stories about the "Linky" cutting you of directly when the current goes over the set limit. No idea if it is true and it might never happen when the heat pump is running. I just want to be sure it won't.

Yeah, I was more thinking of the defrost horror stories and heat pumps supposedly not lasting long and all that. No clue about your smart meter thingy. But you can reactivate the supply, can't you? I mean, if that requires a visit from a technician or something, that would be bad, but otherwise, I guess it might just not be a big problem.

Could be an option but would require sensing that the water heater is on.

Yeah, that's a thing you can buy. But whether that makes economic sense is something that you'll have to figure out, of course.

The remainder of your post is very educational and I will certainly look into adaptations to my system. At the moment the regulation for the room temperatures is based on a set temperature and a hysteresis between the measurement. The rooms with the radiators start heating when the temperature on the sensor drops half a degree Celsius below the set point and stops when the set point is reached. Depending on the water temperature there is an overshoot, but that is fine for us. For the rooms with the underfloor heating it is 0.1 degree Celsius. The response is much slower but will also give overshoots. On every valve in the system there is an on/off thermo electric actuator that is controlled by the Raspberry PI. Maybe I can change it to run more constantly and regulate the temperatures. Need access to the heat pump system for that though, and that needs aid from the installer and technical data on protocol and interface.

I guess that's usually not really officially documented, but you can find the results of people reverse engineering protocols of various models on the internet.

But, yeah, ideally you calibrate the flow of all heating loops in the house such that you don't need any per-room control. Obviously, that is not possible if you need to vary temperatures, but you can try to get as close as possible. You want to have the largest heating surface possible at all times with the largest possible flow to minimize the source temperature that you need to feed the required heating power into the house.

[tszaboo]

I talked with a college, who had geothermal at his new built. It needs to be on constantly, since it would take days to heat up the house from ie. 10 degrees. For high efficiency, you need underfloor heating. So that means you pretty much have to rebuild your entire house, since the floor has to go everywhere, then painting then the work never stops. I estimated 10K for geothermal system, and 30-40K for renovations. The garden will go as well if they dig. Basically you are building a new house. Plus as we know this work lasts months, your house is unlivable in the meantime, people promise deadlines and no show, the usual renovations stuff.
I didn't give up on heatpump heating, but I'm only interested in air sourced, basically airco with reverse operation. As an addition to the district heating.

Not sure why is the conversation... According to EU dickheads you should be buying T-shirts and sunglasses, rather than worry about heat pumps.

Yeah, these politicians simultaneously force you to buy an electric car, and renovate your house with these very expensive systems. I really don't know why they cannot do back of the napkin calculations, to see that they should be working on P2G plants for sustainable energy. That way we don't have to replace the entire infrastructure of basically everything we own. But yes, let's just make simultaneously obsolete our cars and our houses, and this damn inflation.

[Marco]

It needs to be on constantly, since it would take days to heat up the house from ie. 10 degrees.

His house is almost certainly superinsulated to be able to work like that, with his control loop superconservative. No way the system has that little surplus power, it just refuses to use it because of the controller. He probably needs to raise the heating curve ("stooklijn" in Dutch, dunno the exact English term) to heat up fast.

[zilp]

Yeah, these politicians simultaneously force you to buy an electric car, and renovate your house with these very expensive systems. I really don't know why they cannot do back of the napkin calculations, to see that they should be working on P2G plants for sustainable energy. That way we don't have to replace the entire infrastructure of basically everything we own. But yes, let's just make simultaneously obsolete our cars and our houses, and this damn inflation.

Why would that be better than a full calculation that might suggest that P2G at 70% efficiency with burning the result at 95% efficiency might not actually be cheaper than running a heat pump at 400% efficiency, as it increases the demand for electricity by 500%?

[baldurn]

If the heat pump is optimised for higher temperatures it might not be a problem. Mixing to a lower temperature doesn't matter as no energy gets lost. Still, if it where up to me I'd change to underfloor heating in the entire house.

By Carnot's theorem https://en.wikipedia.org/wiki/Carnot%27s_theorem_(thermodynamics) we know that it is always less efficient to make hotter water. Real heat pumps are even worse than the theoretical Carnot limit. You therefore absolutely do not want to make 55 degree hot water and mix it down to 30 degrees. That might halve the COP or worse.

My Panasonic air to water heat pump has a function that can make water for two circuits plus a separate loop for hot tap water (so three circuits/loops in total). It is just a simple remote controlled valve. The heat pump sends a signal when it wants the valve to switch circuit. Simple as that.

[nctnico]

I talked with a college, who had geothermal at his new built. It needs to be on constantly, since it would take days to heat up the house from ie. 10 degrees. For high efficiency, you need underfloor heating. So that means you pretty much have to rebuild your entire house, since the floor has to go everywhere

That is not necessary. You can retrofit underfloor heating in cement floors as the finishing layer is soft by design. In my case I used a mill to make slots in the floor where the underfloor tubing went into.

[Marco]

The insolation mats with ready made grooves for piping only raise the floor by a couple cm too, it's more the sheer amount of labour and mess involved.

[pcprogrammer]

Google tells me the linky limits current assuming a reference voltage of 200V, so it's going to be closer to 10kW in reality.

PS. is that really still in there in France? The Dutch smart meter regulations removed that pretty much immediately (or rather it's forbidden by government regulations for them to have contactors now). Moving from the sealed main breakers with large safety margin to a hard switch off probably caused way too much problems.

Good to know.

I have not looked into regulations around this. The linky for us is fairly recent and when the time came they wanted to install them there came the horror stories about radiation causing cancer, cutting you of when over powering, etc. Did not bothered about that because a: I don't buy into the radiation bullshit and b: our energy consumption has always been low enough to have no "fear" about a cutoff.

But with a heat pump the possibility arises that the consumption goes over.

[pcprogrammer]

As for your house being split between floor heating and radiators: I have no idea whether there are any off-the-shelf solutions for this, but maybe it is possible to run the heating in some sort of interleaved mode? Like, split this into two independent loops, and then heat them alternatingly, one at ~ 30 °C and the other at ~ 55 °C?

There is an off-the-shelve solution for that. It is a simple mixer unit with it's own pump that takes in hot water from the hot water source and mixes that with the cold water coming from the tubes that make the underfloor heating system. These have been in use for decades.

I'm using robot distributors for the underfloor heating.

If there is an option to connect to the heat pump and control the output temperature with my own system I can scrap the mixing valve that is in there now. This most likely improves the performance.

The 55 degree Celsius water temperature is not a must. It depends on how many zones are on at the time. The radiators work fine with 40 to 45 degree Celsius, but when the whole house is demanding heat the input temperature needs to go up to maintain the needed 40-45 degrees on the radiators. But even with lower temperatures on the radiators they manage to keep a minimum temperature in the rooms with extreme cold outside.

For the bedrooms it is not an issue if they go down to 14 degrees or less, as long as they stay above freezing temperature. 
At the moment they are set to 15 and 15.5.

[pcprogrammer]

It's a shame it's getting a bit late in the year to see how far you can get with some DIY modding of the radiators with PC fans and some thermally controlled activation like Heatbooster.

Proper fan coil units aren't that expensive either and will give you a guaranteed heating output so you can just do the math. You could gamble on the DIY solution and replace the radiators with fan coil units if it doesn't work out.

Not going to happen. Fans make noise.

[m k]

That cooling thing of mine was a bit one sided.
What I think is a heat pump is a circulation where material is changing its phase.
So monoblock has a heat pump inside of it and heated water is something else.
So cooling with a water heating radiator is just cooling with a radiator.
Here cooling with air or ground to water system is many times out since the system is also generating the base of warm clean water.
But system for water heating only can be used for cooling, just turn the process around and bypass the reservoir, if needed.
There moisture must be taken seriously, even here, or actually especially here.
If memory serves a human is exhaling 1/5 litres of extra water per day.
So even without cooling humans are actively increasing moisture.

No idea how cold a radiator can be in those cases, but finally wall behind that radiator can probably be close to dripping.
Usually intersection of wall and floor has some kind of a skirting, behind it is a good place for a secret habitat.

We have a machine ventilation with heat conservation.
There is also a possibility to bypass that conservation.
But when that bypassing is actually beneficial, only when outside is cooler than too warm inside.
Here it's nights off and days on, more than a bit tiresome if the operation is manual.

The house here is 150-160m2, two floors and sloped base.
Last year's total electricity bought was 10MWh, Jul-Aug was 1MWh.
Parts of the place had more dropped temps and few m3 of logs were burned.
(dropped temps are not nice, I know, just an issue of rig)
Estimated total energy for 23C room temperature and four persons is around 25MWh, warm water is few MWh.

Costs are also important.
It becomes less stupid to not add a gear when cost is stupid.
My opinion is that some of regulated costs are stupidly high, and artificially.

BTW,
I forgot the shed.
With few bigger power tools the estimated 15A surplus is used in no time.
Mowing can be done with 10A, but leg work increases.
Branch shredding also, up to a point, there time is the issue.
Here renting a machine is practically not available, everybody must have their own, so market is limited.

[zilp]

The 55 degree Celsius water temperature is not a must. It depends on how many zones are on at the time. The radiators work fine with 40 to 45 degree Celsius, but when the whole house is demanding heat the input temperature needs to go up to maintain the needed 40-45 degrees on the radiators.

This means that with a stronger pump, you could lower the temperature. Of course, with a stronger pump, there is a risk of noise if many loops are closed (and thus the flow speed in the remaining loops increases). To prevent that (if it is a problem at all, which you really only can figure out by testing), you need to either plan without individual room control (i.e. (almost) all loops open all the time, calibrated to provide the required heating power, that's how I run my own system), or you need a pump that regulates the output pressure, or you need a bypass valve that limits the pressure by short-circuiting a part of the flow.

[tszaboo]

It needs to be on constantly, since it would take days to heat up the house from ie. 10 degrees.

His house is almost certainly superinsulated to be able to work like that, with his control loop superconservative. No way the system has that little surplus power, it just refuses to use it because of the controller. He probably needs to raise the heating curve ("stooklijn" in Dutch, dunno the exact English term) to heat up fast.

It's designed to be a very efficient system, with the water temperature barely above the setpoint. The higher the temperature difference, the lower the efficiency of a heatpump. You can find some data here: https://en.wikipedia.org/wiki/Heat_pump#Performance, but keep in mind these are going to be the cutting edge numbers that are not really afforable.

I talked with a college, who had geothermal at his new built. It needs to be on constantly, since it would take days to heat up the house from ie. 10 degrees. For high efficiency, you need underfloor heating. So that means you pretty much have to rebuild your entire house, since the floor has to go everywhere

That is not necessary. You can retrofit underfloor heating in cement floors as the finishing layer is soft by design. In my case I used a mill to make slots in the floor where the underfloor tubing went into.

My neighbor did that to get rid of the radiators in his living room. It was months of works, lot of dusts, so much that the rest of the house was unlivable.  It's not just cutting some slots. It's removing all furniture, the floor, isolation, cutting plastering, waiting for it to dry, a new floor, and then the rest that you need to do.

Why would that be better than a full calculation that might suggest that P2G at 70% efficiency with burning the result at 95% efficiency might not actually be cheaper than running a heat pump at 400% efficiency, as it increases the demand for electricity by 500%?

Because heat pump actually doesn't solve the problem. The problem is that we need energy to heat our homes during the winter. And it doesn't matter where it comes from. My house heating comes from a waste energy of a power plant, how much is the efficiency of that? but anyway back to the topic.
You might compare 70% and 500% and say 500% is better. But 70% is during the summer, and 500% is during the winter. If the energy for that 70% conversion is provided by free (as in beers) solar energy, your efficiency doesn't matter one bit.
If we all install solar panels, the upfront cost is 5-10K per house, depending on how big your roof and ambitions are. And it's getting cheaper. And we already see that it can cover 100% of our usage during certain days the summer. In these days electricity is cheap, sometimes even free.
You take this electricity, run it through P2G and make gas out it, that you burn during the winter. Gas storage is a solved issue, unlike seasonal electricity storage. Plus you use CO2 in the process, that you can capture for free at power plants, during the winter. In fact the price of CO2 could be negative, since the power plant doesn't have to pay the carbon tax if it captures the CO2. There just isn't an large industrial process that uses that much CO2 as we make now.
This is just the most sensible way of handling the issue for the next decades.

[nfmax]

The 55 degree Celsius water temperature is not a must. It depends on how many zones are on at the time. The radiators work fine with 40 to 45 degree Celsius, but when the whole house is demanding heat the input temperature needs to go up to maintain the needed 40-45 degrees on the radiators.

This means that with a stronger pump, you could lower the temperature. Of course, with a stronger pump, there is a risk of noise if many loops are closed (and thus the flow speed in the remaining loops increases). To prevent that (if it is a problem at all, which you really only can figure out by testing), you need to either plan without individual room control (i.e. (almost) all loops open all the time, calibrated to provide the required heating power, that's how I run my own system), or you need a pump that regulates the output pressure, or you need a bypass valve that limits the pressure by short-circuiting a part of the flow.

Rather than restricting flow, it might be possible to turn on radiator fans only when extra heat input is needed, and turn off during 'setback' times. The heat pump weather compensation should react appropriately, modulating the compressor to keep the circulating water temperature at its target value. There is then no need to restrict the flow in the hydronic circuit.

Just my thoughts

[tom66]

I'm quite a fan of 'eco technologies' but I am a little skeptical about heat pumps for now for the average British home. For a new build house, built from scratch with excellent insulation, large pipes, and underfloor heating and/or large radiators they might make sense.  We have a 1930s detached house.  It has loft and cavity wall insulation, and double glazing, but that's about all in terms of efficiency improvements.  It has plenty of thermal bridging and uninsulated ground floor (project for this year is to insulate that).  In the depths of winter the boiler is running with flow temperatures of 65C+ to warm up rooms.  Instead of a heat pump I've taken the cheaper approach of zoning the thermal load so that rooms we don't use as much don't get as much heating (and doors are closed to limit rate of heat movement).   The OpenTherm controller means that the temperature of the boiler is kept closely in line with the demand signal from the wireless TRVs, so the boiler hopefully runs as efficiently as it can.

I probably will look at a heat pump in the next 5-10 years and seeing the outcome from people who had them fitted today.  But knowing my luck that will mean all the grant schemes have gone away by then.   I know a utility company here will install one in my house for about £3,000, but that's on top of a £7,500 grant from the government.  The COP would need to be >3.5 to make it cost effective compared to gas. 

Perhaps I'm just a luddite though and will look back on this post in 5 years wondering what I was so concerned about. 

[Miyuki]

The 55 degree Celsius water temperature is not a must. It depends on how many zones are on at the time. The radiators work fine with 40 to 45 degree Celsius, but when the whole house is demanding heat the input temperature needs to go up to maintain the needed 40-45 degrees on the radiators.

This means that with a stronger pump, you could lower the temperature. Of course, with a stronger pump, there is a risk of noise if many loops are closed (and thus the flow speed in the remaining loops increases). To prevent that (if it is a problem at all, which you really only can figure out by testing), you need to either plan without individual room control (i.e. (almost) all loops open all the time, calibrated to provide the required heating power, that's how I run my own system), or you need a pump that regulates the output pressure, or you need a bypass valve that limits the pressure by short-circuiting a part of the flow.

Constant pressure pumps have been here for decades and all efficient ones with permanent magnet motors can do it. There is no reason to use old-style single-phase induction motors.
Another possibility is to change some radiators to bigger or use fan coils. Plain steel radiators cost close to nothing if you do not need some fancy design ones.

The main issue is that many heat pump installation companies are pure scammers and know nothing about it. They choose the wrong overpriced models, and cannot set the heating system right (flow, balance, and so on)
Right-chosen heat pump, even when Air-Water shall not activate resistive heating except on record cold days.

Last year I bought a new fancy wood boiler with lambda control, it is a great thing, can burn anything, big pieces, small pieces, scraps, and dust. From the chimney goes nothing except some condensation when is freezing. With storage tank. It is very convenient.

I also have an Air-Air AC/Heat pump, as it got pretty hot here last years it is handy to have temperature control both ways. And for heating, it has a higher COP than the Air-Water system can reach.
I'm planning to small Air-Water system for days when there is a lot of sun, but for cold days without sun, wood is the right choice for me. And pellets are crazy expensive here nowadays.

[pcprogrammer]

The 55 degree Celsius water temperature is not a must. It depends on how many zones are on at the time. The radiators work fine with 40 to 45 degree Celsius, but when the whole house is demanding heat the input temperature needs to go up to maintain the needed 40-45 degrees on the radiators.

This means that with a stronger pump, you could lower the temperature. Of course, with a stronger pump, there is a risk of noise if many loops are closed (and thus the flow speed in the remaining loops increases). To prevent that (if it is a problem at all, which you really only can figure out by testing), you need to either plan without individual room control (i.e. (almost) all loops open all the time, calibrated to provide the required heating power, that's how I run my own system), or you need a pump that regulates the output pressure, or you need a bypass valve that limits the pressure by short-circuiting a part of the flow.

It will boil down to experimenting with it all.

There is also the option of opening all the zones and use the adjustable return valves to control the flow. Have the system regulate the temperature output of the heat pump to maintain a livable temperature in the rooms. Incorporate the outside temperature in the equation and see what it does.

Another option is the use of a controllable variable speed circulation pump. When only a few zones are open reduce the flow with the pump.

[tom66]

Seems that most heat pump engineers suggest that over-zoning causes problems.  You want the heat pump to be providing constant heat to all rooms, with some exceptions for times when you are out of the home for instance.  So my little trick of not heating rooms (via home automation) when unoccupied, and zoning rooms to different temperatures, would be not recommended.  This is because the heat pump will need to work very hard to get that room up to temperature after returning home or to change temperatures according to the schedule, which means the COP will fall dramatically for the whole house during those periods. 

The general principle of low temperature design is to have the system running more or less continuously keeping the radiators 'ticking over', whereas typical gas/oil boiler CH systems rely on cycling the boiler.  In the boiler case, you oversize the boiler slightly so that even if it is -10C outside and +10C inside the home can get up to +20C within about an hour.   A heat pump would probably struggle with that.