Here I am responding to Gideon who wants to understand heat pumps. We are only discussing the science and engineering of them, and whether domestic use is a good and economic idea.
Here I am responding to Gideon who wants to understand heat pumps. We are only discussing the science and engineering of them, and whether domestic use is a good and economic idea.
Definition: Heat pump.
This is a machine based on thermodynamic principles (Physics) that can move energy in the form of heat from place to place. The ratio of the input heat to the output heat is called the coefficient of performance (COP) and is directly related to the temperature difference between input and output. Energy is required to operate the machine, which is often electricity but may be any source of mechanical power, this energy is NOT included in the COP, and varies directly with the temperature difference, and for large temperature differences can exceed the recovered energy from the output. In the case of refrigeration, this is normal as the output heat is effectively thrown away although in large systems it may be recovered for other low-grade heating purposes, typically supermarket space heating (M&S).
Let's look at the COP of heat pumps, first from a theoretical perspective.
All thermal machines operate with the Carnot cycle, a thermodynamic description of an ideal “heat engine”, typically your vehicle engine and your refrigerator, or the vehicle air conditioning. The ideal efficiency of all of these depends on two temperatures, the input temperature and the output temperature. The maximum COP of any of these is given only by these two temperatures, and for a heat pump is
Input temperature / (output temperature - input temperature)
measured in absolute temperatures (Kelvin). Zero centigrade is about 273K.
These theoretical numbers sound quite reasonable, for example:
Input 273K (0C), output 293K (20C) gives 273/20 = 13.65
Input 263K (-10C), output 333K (60C) = 4.38
These numbers assume adiabatic conditions (no heat loss) in the machine, In practise this is not possible, and the loss may be as much as 50%, making the numbers much less impressive. All this ignores the mechanical input required and thus the electricity that will be used. This depends on the working fluid (gas), and the best were “designer” fluorocarbons, which are now outlawed. This leaves hydrocarbons and ammonia, propane being a common choice. Ammonia is used in large plants because it is very good, but it has unfortunate properties, it is both very poisonous and explosive under the conditions found when used as a refrigerant, very unsuitable for domestic use.
The effect of increasing temperature differentials across the pump is that the quantity of heat transferred falls and the driving power increases, and neither is desirable. The best solution to this is to transfer the input heat (low temperature) to air, as in conventional air conditioning. However, this is difficult to install in existing properties as ducts are needed to each room, for which there is no obvious space. The air also needs to be isolated acoustically from the fan used to transfer it from the heat exchanger, another inconveniently big box.
Therefore the rather unthinking government suggests heating the water used to transfer the heat (very efficiently) from a gas boiler, but then finds that the COP becomes low! OK, reduce the water temperature they say, but this is foolish. The output heat from radiators (actually convectors mainly) responds to another thermodynamic number, and as the temperature is lowered the heat transfer reduces by somewhere between a square and quadratic law! Therefore MUCH larger radiators will be needed, even if the temperature is only lowered by 10 degrees to improve the COP.
The overall effect is therefore that installation is very expensive, and the achieved COP is quite low, and unless electricity is very cheap the system is much more expensive than gas. All of these things work in exactly the wrong direction, as does the next problem and that is that the installed overall supply capacity to each house is between 1 and 2 kW, and if many change to a 24/7 heat pump the supply will fail. National Grid has an estimate that upgrading all the supplies to 6kW would cost £4.2 TRILLION pounds, and cannot be achieved even at this cost in the suggested timescale. It would need another 30 nuclear power stations like Hinkley C to be built starting now, or 1000 of the “mini modular” type not yet designed, and the electrical infrastructure in the entire country to be replaced. Clearly, this is impossible. You will also realise that wind and solar are very unsuitable to provide this energy, during long cold periods it is either dark or there is very little wind!
Sounds unpromising. I wonder though if in bungys like mine, a workable strategy is to bit the bullet of lifting the floors to install air ducts. In any case, suspended floors appear the hardest, or most disruptive surface to insulate. I always admired the north American build with an accessible basement…
On a detail - clearly a lot of energy is dissapated in the heat pump's motor. It's the compressor, isn't it, so for a system designed primarily for heating, one must assume it is discharged inside the building? Is it well used?
The heat loss in the electric motor is probably about 5% of the electrical input. Remember though that any heat lost from the compressor itself and is not in the gas is largely wasted because it is difficult to collect. The heat of compression you want is the pump output, provided that it is lost in the output heat exchanger to the air or water, etc. It doesn't matter in a fridge, but in a heating pump, it is very significant.
If you carefully read the specifications of the various systems on sale you will see that the COP figure is not well described. There should be a graph of the temperature differential between hot and cold sides and the COP, and ones of the electrical consumption against temperature differential, and also the useful heat output in kW. As usual, you will see the best case COP given, not the one you really need to know for very cold weather!
In large systems used in hot countries, you will see that the hot side has wet cooling towers, similar to a power station. The reason for the water is because the evaporation energy can be used to vastly increase the cooling capacity, just like the cold side of a power station steam turbine. Unfortunately, you cannot do this for heating, as the inside of your house would have 100% humidity, but doing so would help to use the output heat very effectively. Ducted air is obviously the best solution as the temperature differential is small, and consequently the COP high. Fitting ducts under suspended floors is a good idea, but the noise silencing and disruption of probably taking all the floors up to fit insulated quite large ducts would be a problem. Don't forget too that the circulating fan would also take power, probably at least 1kW, so even more electrical consumption! It is likely that a 3kW pump would provide sufficient heating this way for a normal semi without much additional insulation, but it may not be available yet, and I don't see the installation cost as being low, and upstairs would probably have to be heated by circulating air around the whole house, in other words, more ducts in the roof space and then back to the heat pump heat exchanger.
As an example, I have seen a very large system for cooling in Las Vagas, cooling a vast exhibition hall, about 100,000 m2 and 10 metres high, so 1 million cu metres. It cooled the whole place from 35C ish to 20C in about 10 minutes, which feels fantastic, but it did take 2MW! COP theoretically about 20 so 40 MW of cooling, in reality probably 30MW allowing for efficiencies.
We're about to take you to the IET registration website. Don't worry though, you'll be sent straight back to the community after completing the registration.
Continue to the IET registration site
