Here’s the latest guide from our resident home builder, Transparent ^^^^

I really enjoyed it, and I’ve actually already shared it with a dog walking friend as we walked past a Mitsubishi ASHP and we couldn’t work out how mild air can be used to heat water. Good timing! 

As someone who has just had an air-source heat pump fitted through the OVO Smart Homes heat trial I have two useful additions to add to your guide.

1. An experienced and competent installer is required to have an efficient system. Do research them before allowing them to proceed. We have had a mix of installation qualities on the scheme from different installers. Some trialists have had great systems installed, whilst some of us have not had a good experience, our electricity bills have gone much higher than expected and we are still waiting for systems to be configured correctly. 

2. Radiators can be used as good emitters to achieve a good Coefficient of Performance (COP), as long as they are (up)sized correctly as part of a thorough heat loss calculation on the property.

In addition to these points I think its worth reinforcing the point that energy storage, as highlighted above, will be essential to allow a heat pumps to make use of TOU tariffs. Without energy storage they will not utilise cheaper tariffs and will not be a sustainable option for helping us to reach our carbon reduction targets. 

That is interesting, because I have been looking at heat pumps, and I just cannot make the numbers work. You spend a lot of money for installing a heat pump, only to be rewarded with a higher energy bill. The problem is that while gas can be stored (although we decommissioned a lot of the seasonal storage for some strange reason), electricity cannot be stored in significant quantities. 

So storage seems to be the key here, ideally thermal storage. That means a large water cylinder, phase change material, or a solid heat storage (as in the old storage heaters). Any of these take space, so where do you find the space? (I once read about a low energy house that had seasonal thermal storage at the core of it. It was 50t tank or something ridiculous like that, but the house was thermally self-sufficient with thermal collectors on the roof.)

And you need a highly sophisticated control system, looking at weather, occupancy, electricity cost and all the forecasts in combination. Ideally it would also like into photovoltaic, using excess photovoltaic energy together with the battery store. 

PS: You said that the UK is moving towards “time of use” tariffs. I do not see much evidence for that statement. There is exactly one Day Ahead tariff available in the UK, and there are a range of two tariff schemes for a number of providers, which are less flexible than the Economy 7 tariffs of old. As much as I want to see change, I just don’t. 

Some really interesting points raised here, @hydrosam and @MrPuds particularly in the importance and practicalities of energy storage as part of the heat pump system. 

TOU tariffs seem to be the hot topic around here at the moment. If you haven’t already seen it there’s already been a great discussion on the future of these type of plans here.

We’re also suggesting this could be the subject of our next event. If you would be interested in joining the event and discussing this with our resident experts,  you can cast your vote in the topic below -

I’d also like to reflect on the point about air source heat pump efficiency:

It takes electrical energy to operate the compressor, but an optimised system might achieve a Coefficient of Performance (COP) of 4. That means it delivers 4kW of heat output for every 1kW consumed.

This appears to be the marketing/lab tested efficiency being quoted by manufacturers. The reality emerging from the heat pump trial at this early stage is COP values between 2 and 3, with only one trialist so far on the forum reporting a figure above 3.  Its early days in the trial though so we are all hoping our systems can be optimised and start operating at an efficiency that makes the technology viable. 

I thought this was pretty well understood. AIr source heat pump do not in real operation exceed a COP of 3. (And to be honest, neither do boilers reach the stated efficiency in real operation.) To get beyond 3, you need a ground source heat pump. 

The main problem with air source heat pumps is of course that they put extra load on the grid during winter, when electricity is already scarce and expensive. Not only do you need more heating in winter, the COP also drops with a temperature below 0, sometimes massively so when the evaporator freezes. 

Never would I have thought that Charles and Boyle’s law will find a reference in my adult life!

As an electronics engineer by training and barely using those hard-acquired skills, it is refreshing to see @Transparent so coherently slips the principles of physics, entropy and thermodynamics so easily. Well done!

On a more relevant note, I concur home energy efficiency rating has a much bigger role to play in achieving a decent COP. Storage solution combined with A/G-SHP are far fewer in proportion and will take time to achieve a similar scale to solar PV’s installation in the UK. Repeatedly proven incentive lead to scale (FIT, EV Grant, GHG etc) deployment.

@MrPudswrote:

storage seems to be the key here…. electricity cannot be stored in significant quantities

This is one of several active discussion topics within the Smart Home Treehouse area. You are quite correct that a very large energy store is required in order to not use any Grid loading during the early evening period of peak demand (nominally 17:00-22:00).

A: A thermal store using hot water would be larger than could be accommodated in the average-sized UK home.

B: The OVO Smart Home Trial announcement does refer to Sunamp who have a nifty Uniq Heat Battery. But we aren’t yet aware of any houses on the Trial who have had one specified, let alone fitted.

C: The cost of storing electricity is falling rapidly, particularly the LiFePO4 Lithium batteries which run at 3.2v per cell with a capacity of 100Ah-300Ah. The two main Chinese suppliers are currently Eve (usually supplied by Xuba) and Lishen.

Let me re-post here a systems diagram which I published in the Smart Home Treehouse just 3 weeks ago

Having an off-grid energy supply such as this avoids the need for a Distribution Network Operator to provide G100 grid-connection certification. Thus you could always have a Heat pump and an EV charger on the same site.

Such a device does not exist at this time. But if we can persuade @sylm_2000 to blow the dust off his Electronics Engineering qualifications then there’s hope yet!

I wish  to solve world peace!

I think engineers have been at work on this already. I was quite lucky to visit some of the auto manufacturers in Japan a decade ago and they were well-ahead in hybrid, hydrogen and V2G technology. It is no surprise that Nissan Leaf partnered with Kaluza as they have been trialling that technology in Japan for at least 3-4 years before the UK trial.

I know many Chinese companies are trying to establish in the recycled EV battery automotive market, where the weak link is BMS (battery management system). Once a platform/standards are established then Tesla’s market cap would be nothing as compared to the energy storage market.

A domestic energy storage solution enabled by a recycled/coupled EV battery is not far away. 

It’s different battery chemistry, @sylm_2000 

Most EVs use cells based on Lithium-NMC (Nickel Manganese Cobalt; LiNiMnCoO2). They operate at 3.7v, have a higher charge/discharge rate, higher capacity per kWh, but a shorter lifetime - approx 20% that of the LiFePO4 cells I’m suggesting.

I’m still thinking this through, but I feel it is best if the energy store were to have a lifetime equivalent to that of the Heat pump system which it’s supplying.

Feel free to suggest otherwise. But you’ll also need to come up with a good argument to counter the social/moral objections to using Cobalt in a battery which doesn’t need to be light and mobile.

Absolutely agree current generation of EV batteries is not sustainable environmentally. The damage in DCR for rare earth exploitation and explosive price escalation of materials is clearly evident that “once again” countries and people with immense natural resources are not benefiting.

I am counting on many (decently qualified engineers) working on plasma, polymer, solid-state and a variety of other interesting technologies just waiting for a breakthrough.

The next (bigger) Elon Musk will be from the battery storage technology!

@Transparent could you give an idea of the dimensions of a thermal storage cylinder please, referring to your option 1 diagram above. 

Your question is a lot shorter than my answer is going to be @juliamc !

The size of the Thermal Store depends on a number of factors.

Those customers, like yourself, who are on the Zero Carbon Heat Trial, funded by BEIS, have had over-size radiators fitted into your homes. That's a good move for a heat pump installation because you can now operate at a much lower temperature than the 70°C+ which we normally use for radiators supplied by gas boilers.

But it's correspondingly bad news for sizing a thermal store. The lower the temperature of the water, the larger the tank needed to hold the energy you wish to store.

Initial assumptions:

Let's do some rough calculations based on a cold winter's day in a 4-bed house of average size.

A typical gas boiler would be rated at 25kW and might need to deliver 100kWh of energy into the house during the course of that day. Ie it's actually running for a total of 4 hours out of the 24.

So if you need to retain enough energy to keep the house warm for 5 hours of peak-demand (5-10pm), then you'd have to store 5/24 x 100kWh = 21kWh in that tank.

Suppose your heat-pump optimisation is such that it delivers water at 55°C max, and that you're prepared to allow your radiator temperature to fall to 40°C by the end of the 5-hour period. Your Thermal store will exhibit a 15°C drop whilst delivering that 21kWh.

The calculation:

1 watt of power for 1 sec = 1 Joule

Thus 1kWh = 1000 watts  x 60 secs  x 60 mins
and therefore
1kWh = 3,600,000J  (3,600 kJ)

21kWh = 75,600,000J  (75,600kJ)

We need to find the mass of water which will hold that energy.

Energy required = Specific Heat Capacity  x  Temp loss  x  mass in grams
where the Specific Heat Capacity of water is 4.184J/gram/°C

Change the formula around:
Mass (grams) = Energy / (Spec Heat x Temp loss)

= 75,600,000 / (4.184 x 15°C)
= 1204589 grams

Convert to kilograms = 1204.589Kg

1Kg of water is 1 litre capacity, so we need a tank holding 1200 litres
1 litre of water occupies a cube with each side being 100mm.

Conclusion:

Assume that this tank is a cylinder with an internal diameter of 800mm (about the largest domestic hot water cylinder you could buy!), it would then need to be 2.4m tall.

Add at least 40mm of high quality insulation around it, plus the stainless steel strong enough to contain 1.2 metric-tonnes of water. That gives you a tank almost a meter across and the height of the ceiling in an average home.

To help you imagine this better, look at these two photos:

On the left is my newest garden rainwater tank, positioned on top of a plinth 800mm high. It holds 1500-litres.

On the right is the thermal store I installed in my own house. It holds just 280-litres.

I’ve included the calculations because readers can play with the figures if you disagree with my assumptions. As I’m neither a physicist nor an engineer, others may also wish to correct any horrendous clanger I’ve just made!

@Transparent thanks ! Love maths !

As ever some brilliant explanations, and I would challenge your claim not to be an physicist there (an expert amateur if ever I saw one!)

I’ve done my best to follow the maths and just have one figure I wasn’t sure of…

Is this 4.184 figure a constant (ie all water has this Specific Heat Capacity) or could that be vary for any reason (I’m thinking perhaps the hardness of the water table?)

 Conclusion: it would need to be MASSIVE! So follow up question - would it need to be inside? or would placing it outside affect the required insulation of the tank?

@Transparent I see.

Would this work ?  Don’t use the water directly from the thermal store in the radiators (which has to be at the lower temp to match the heat pump output). The thermal store would be better (and smaller) containing water of a higher temperature - such as could be collected from a solar thermal array or the immersion run at times when renewables are available. Then the radiator system could draw off heat via a coil set in the thermal store.

Yes @Jess_OVO that figure of 4.184J/gram/°C for the Specific Heat Capacity of water is a constant.

Compare this with the Specific Heat Capacity of (Ethylene) Glycol at 2.433J/g/°C. We commonly add Glycol to water in order to lower the freezing point, but it carries only 58% of the heat energy compared with water.

I use a 4:1 ratio of water to Glycol in the fluid mix which circulates to my external solar-thermal panel. So that 20% dilution has reduced the Heat Capacity to 3.834J/g/°C.

As a result the circulation pump will operate slightly faster in order to transfer the same quantity of energy to my Thermal Store.

The manufacturer’s recommendation was to use a 40% mix of Glycol, reducing the Heat Capacity yet further. But as I’m based in the West Country it’s unlikely that my thermal solar collector will face temperatures low enough to warrant that (-23.5°C)!

The Specific Heat Capacity of water is not much affected by the level of salts dissolved in it. You’d need to be using water more concentrated than the sea before it might become a factor in the calculations.

If you did want a 1200l storage tank, but chose to put it outside, then you would certainly need to increase the level of insulation.

Think of the insulation thickness as a gradient between the water temperature and the outside air.

Each type of insulation has a stated resistance to the passing of heat. This Thermal Resistance is called the R-value, measured in °C.m²/W (degrees Celsius per square meter per watt).

The higher the R-value, the better is the insulation.

If you want a higher R-value:

• increase the insulation thickness
• or change the material to a better insulator

Many self-builders specify a “plant room” for their house, possibly within a basement. This is usually unheated, but contains all the large tanks, pumps, battery storage and ventilation equipment. As its temperature will still be significantly higher than the external air in winter, it offers a compromise for the levels of insulation required.

@juliamcwrote:

The thermal store would be better (and smaller) containing water of a higher temperature - such as could be collected from a solar thermal array or the immersion run at times when renewables are available. Then the radiator system could draw off heat via a coil set in the thermal store.

The moment you consider operating a thermal store at a higher temperature than the Heat pump delivers, you will thereafter only be able to provide that heat from your Solar Thermal or immersion heater.

As Messrs Flanders and Swann so aptly explained the Second Law of Thermodynamics

You cannot pass heat into a hot water cylinder at 70°C by feeding it from your Heat pump that has a maximum output of 55°C

@Transparent yes I do understand that , but why would you want to ? (Though I understood the thermal store to have a temperature gradient, so in fact the lower level would be less than 70 deg.) I’m suggesting you could supplement the heat pump/radiator setup with the solar thermal etc. This is all about getting over the high demand peak of 5pm - 10pm isn’t it ?

OK @juliamc … so let me throw another bit of maths into my answer.

Assumptions:

It’s a cold day in mid-January with some periods of peak sunlight, but the sun will be low in the sky of course. At 51.5°N (SE England) the sun’s elevation peaks at just 45° and sunset is 16:20.

You want to halve the size of the tank to 600l by adding a further 15°C to the stored water. At 17:00 it will be at 70°C, falling to 40°C over 5 hours.

By noon your heat-pump is supplying at the maximum 55° for which it has been optimised.

Your solar-thermal panel registers 68°C on the roof at noon when the sun is out, dropping to 50°C within 30sec of cloud coverage. You have a 6°C hysteresis setting. Ie the solar-pump operates ON when the temperature of the collector exceeds that of the tank by 6°.

It switches OFF again when ΔTemp falls to 3°C. otherwise the losses in the pipework mean that you will sending warm water out of the tank to the roof.

You have a standard 3kW electric immersion heater.

Calculations:

If you were to use the immersion heater, you could definitely raise the tank temperature from 55°C to 70°C by 17:00 hrs. That’s a 15°C temperature rise.

Energy required = Specific Heat Capacity  x  ΔTemp  x  mass in grams
where the Specific Heat Capacity of water is 4.184J/gram/°C

As the tank is now just 600l capacity, its mass is 600Kg or 600,000g

4.184 x 15 x 600,000 = 37,656,000 Joules

As before, 1kWh = 3,600,000J

So the power we require is  37,656,000 / 3,600,000  = 10.46kWh

Your immersion heater is 3kW, so it can deliver that power in 3½hours

You have from noon to 13:30 to extract something usable from the solar-thermal. Without that, you will need to switch on the immersion heater in order to reach the required 70°C by 17:00hrs.

That would cost you about £1.90 including a reasonable proportion of your standing charge.

Are you feeling lucky?

Always @Transparent the sun always shines in Surrey.

In case you missed it, we’ve made a guide on heat pump pipe insulation, here: 

and in case you missed it @Tim_OVO - that link is already embedded in the lead-article at the top of this page

All very interesting. But I’m afraid totally pointless for the majority of existing housing stock in the UK. Especially here in the the North of England.

Thanks for popping up here in this topic, @knight  - You’ll need to qualify that observation however

What is it about the existing housing stock in the North of England which makes it pointless to consider the use of Heat pumps?