## Heat Pumps#1

Installation of an air source heat pump (ASHP) in my own house is sadly on hold while the installers await delivery of a part. So I thought I would take this opportunity to update you with one or two things I have learned about how real-world heat pumps operate.

What is a heat pump?

I am preparing an article about how heat pumps work internally, but considering only their operational behaviour, they work like the device illustrated below.

Click for a larger version.

• Cold water enters the the heat pump.
• Warm water flows out.

The engineered ‘miracle’ of a heat pump is that 1 kWh of electrical energy can extract between 2 kWh and 4 kWh of heat energy from the air.

It might seem that nothing could be simpler or more wonderful? But the engineering reality behind the ‘miracle’ requires that the heat pump be operated carefully.

The problem

The key problem is that heat pumps require a high flow of water through them in order to enable efficient operation of the heat exchangers which extract heat from the air. Typical flows are in the range 20 to 40 litres per minute of water.

For my 5 kW heat pump, this can warm such a flow of water by only 2 or 3 °C. So how can such a device heat water to 55 °C?

Domestic Hot Water

When the heat pump is configured to heat domestic hot water – for sinks and bathrooms – then the circuit looks like the figure below.

Click for a larger version. Schematic diagram of how a heat pump heats domestic hot water. See text for further details.

In DHW-mode, the water in the heat pump circuit is passed through a steel tube wound inside an insulated water storage cylinder. This acts as a heat-exchanger between the water in the heat pump circuit, and the water in the cylinder.

But remember, the ‘hot’ water in the heat pump circuit is just a degree or two warmer than the returning ‘cold water. So how can this ever heat the domestic water to 55 °C.

The trick is having a smart heat-pump controller and low losses in the connecting pipework.

The heat pump controller first sets the heat pump operating parameters to warm the water returning from the DHW by a few degrees.

As the DHW tank warms, the returning water also warms, and the controller slowly adjusts the operation of the heat pump to increase the temperature of the water it supplies to the DHW tank. Eventually the controller detects when the water in the DHW tank has reached its set temperature.

So for example, if the outside temperature is 10 °C, and the water returning from the DHW tank is initially at 20 °C, then:

• Initially the controller configures the heat pump to heat the flowing water to (say) 22 °C. Pumping heat from air at 10 °C to water at 22 °C can be done much more efficiently than pumping heat from 10 °C to 55 °C.
• At first the temperature of the water returning from the DHW tank will be only slightly above 20 °C. But as heat is transferred to the DHW tank the temperature of the water returning from the DHW tank increases.
• In response to this increase in the temperature of the returning water, the controller re-configures the heat pump to an incrementally higher temperature.

• The first part of the heating can be done with higher efficiency – perhaps resulting in 4 units of heating for each unit of electrical work.
• The later part of the heating is less efficient and might only results in 3 units of heating for each unit of electrical work.
• So overall – depending on the maximum temperature required – the so-called coefficient of performance (COP) is usually somewhere between 3 and 4.

Space Heating

When the heat pump is configured for room heating – so called ‘space heating’ in the lingo – then the circuit looks like the figure below.

Click for a larger version. Schematic diagram of how a heat pump heats radiators. See text for further details.

I was surprised to find that in this mode of operation the water from the heat pump is not passed directly through the system of radiators.

Instead, most of the water passes through a short section of tubing called a ‘low loss header’ and goes straight back to the heat pump. This allows the heat pump to operate at high flow rates.

The water used in the radiators is drawn from the top of the ‘low loss header’ and returns – cooler – to the top of the ‘low loss header’.

However there is almost no pressure difference between the top and bottom of the ‘low loss header’ – and so very little water would naturally flow through the radiators. So a hydraulic pump is used to push water through the radiators.

The cooled water from the radiators now mixes with the main flow at the bottom of the ‘low loss header’ and returns to the heat pump.

Click for a larger version. Schematic illustration of a ‘low loss header’ See text for further details.

So for example, if the heat pump is supplying 20 litres per minute of water at 55 °C to the ‘low loss header’:

• The hydraulic pump draws perhaps 4 litres per minute of water at 55 °C leaving 16 litres per minute to flow straight through the header.
• The return water from the radiators is cooled to (say) 45 °C.
• From this one can calculate that the radiators have provided heating of 2.8 kW.
• So at the bottom of the ‘low loss header’ there is a mixture of:
• 16 litres per minute of water at 55 °C
• 4 litres per minute of water at 45°C
• When mixed together this makes 20 litres per minute of water at approximately 53 °C which is returned to the heat pump.

At first I was puzzled by this arrangement, but then I realised it was clever trick.

• It allows the heat pump to operate at high flow rates and yet heat water only over small temperature differences.
• And it allows the radiators to operate with lower flows and bigger temperature drops.

For those with experience of electronics, it is analogous to the ‘impedance matching’ effect of a transformer.

It’s complicated…

Things are more complicated than these diagrams would suggest.

Firstly, the heat pump can only operate in one mode at a time.

So the heat pump controller changes modes by operating a valve to direct the water from the heat pump either to the DHW storage tank or the radiators.

Secondly, there are numerous features incorporated for reasons of safety or maintainability.

Some of these guard against the effects of thermal expansion of the water, some guard against the (low risk) of Legionella infection, and some are filters or energy monitoring components.

But I hope the explanations above come close to getting to the gist of heat pump operation.

I have lots more to say about heat pumps: so stay tuned!

Tags: ,

### 4 Responses to “Heat Pumps#1”

1. Peter Says:

On my pressurised system the hot water cylinder has a thermostat which is set to say 55 ⁰C but once a week the thermostat heats the tank to closer to 70 ⁰C to kill legionella. How is this setup mimicked by ASHP? And how does it effect efficiency?

If I wanted to check if my previously installed radiators are sufficient for a ASHP conversion, do I simply pick a low running temperature for my radiators and see if my house gets too cold?

• protonsforbreakfast Says:

Hi Peter.

In my understanding, the ASHP controller exactly mimics the behaviour you described. But an ASHP cannot reach 70 °C and so uses an immersion heater to do this. To heat 200 litres of water through 15 °C requires 12.6 MJ of energy, or 3.5 kWh. This 3.6 kWh comes from wherever you get your electricity – solar/battery/mains. How it affects your efficiency depends on how this 3.6 kWh compares with your use during the rest of the week.

For my wife and I, we use about 4 kWh/day to heat DHW so that’s 28 kWh a week. Using an ASHP with a COP of 3 that corresponds to about 9.3 kWh of electrical energy. So a weekly expenditure of 3.6 kWh to stay ‘legionella-secure” is quite a big hit – about 38% of weekly electricity used for heating. In the summer – that will be mainly solar electricity – but in the winter it will be a direct hit.

Regarding radiators: yes: exactly. That is what my wife and I did and convinced ourselves this would work – and it was very cold for quite a few days. I am still not convinced so in this coming winter I will see if we need to upgrade any of the existing radiators by adding extra panels or fans. I think when its really cold it would be fine to just use a fan heater. If you design an ASHP to cope with the most extreme temperatures then you may be over specifying your system.

Dominic Zapaman (@Zapaman on Twitter) monitored the temperature of his radiators as he adjusted the set point. He then monitored the fraction of the time the radiators were on. When the radiators are just meeting the heating requirement, hey will be on all the time. In this way you can see how close you are to that point. Roughly, if the radiators come on half the time, then you have quite a bit of leeway to lower the flow temperature.

Best wishes

Michael

2. Simon Duane Says:

Wondering if you still have an inclination to offer Heat Pumps#2 – if only once summer’s gone, there is a need for heating and, no doubt, some data to digest …

3. Craig Stewart Says:

Michael, if the floor through the radiators is a separate circuit, what is it about microbore which causes installers to avoid it? Can the hydraulic pump simply be configured to be suitable for the follow achievable, or is a suitable for taste simply not achievable?