OpenEnergyMonitor Heat Pump Case Studies and Resources

The following case studies and associated resources discuss what we have learnt so far at OpenEnergyMonitor about heat pump design, installation and resulting real-world performance.

Electric heat pumps are a key technology for the decarbonisation of heating, enabling the efficient conversion of electricity to heat. To unlock the highest efficiencies and lowest energy costs, care needs to be taken to ensure good design, installation and operation.

We are interested in an approach that aims for a tight feedback loop between design, installation and monitored results.

Case study 1: End-terrace, Samsung Gen6 5kW

End-terrace solid stone house in North Wales with a Samsung Gen6 5kW air source heat pump. This case study is available as a youtube video here:

YouTube: 1. How I fitted an Air Source Heat Pump ASHP: Samsung Gen6 5kW

  • 0:00 Introduction

  • 3:25 Testing the system

  • 6:54 Heat loss calculation

  • 8:15 Fitting the heat pump

  • 12:35 Frost protection

  • 14:12 Gas boiler removal

  • 20:00 Hot water tank installation

  • 22:34 Hot water tank update

  • 30:38 Hot water cycle data

  • 32:04 Heat pump performance

At 6:54 Glyn discusses how he did his heat loss calculation using HeatLoss.js an open source tool that we have developed at OpenEnergyMonitor.

Glyn covers briefly the importance of designing for low flow temperatures (35-40C) when it is -3C outside (design temperature) and the related radiator upgrades that he made to enable running the system at these lower flow temperatures.

Note that Glyn used a particularly low air-change rate for his calculation of 0.3 ACH based on the result from a blower door test. We have since performed further in-use CO2 monitoring which suggested a range of air change rates from 0.3 to 0.6 ACH. The minimum whole dwelling ventilation rate in the part F UK building regulations for a 2-bed house is 25 L/s which works out to 0.6 ACH for Glyn’s building volume.

Part F regulations are there to ensure healthy air quality. Ventilation rates should ideally be responsive to demand e.g higher ventilation rates when humidity and CO2 rise too high and lower when levels are healthy. See thread: Measuring air change rates with CO2 sensors for heat pump heat loss calculation for a wider discussion of this topic.

Even 0.6 ACH is far lower than the values given in MCS/CIBSE guidance for pre-2000 properties which are around 1.5 ACH for properties of this age (range 1.0-3.0 ACH) and so we will still be avoiding most of the risk of over-estimating heat loss.

YouTube: 2. Air Source Heat Pump 1st Winter Performance in Solid Stone Welsh Cottage

  • 0:00 Introduction

  • 3:29 Winter Test

  • 4:45 Monitoring System

  • 5:45 Heat Pump Performance

  • 7:02 Heat Pump Controller

  • 8:49 Day 3 Performance

  • 9:54 Running Costs

  • 13:04 Water Tank

  • 14:16 System Overview

  • 15:36 Coldest Day

  • 16:44 Average Day

  • 19:06 Data Analysis

  • 20:35 Conclusion

In the introduction, Glyn addresses the validation of his space heating demand estimate through monitored results, highlighting the discrepancy with EPC estimates that, in his case, doubled the projected space heating demand.

Glyn discusses the result of his blower door test in more detail at 3:10. The video presents results from the coldest day monitored up to that point, where the average heat input from the heat pump was 3.0 kW, the difference in temperature between inside and outside was 21K.

This agreed well with Glyn’s calculated heat demand of 3.3 kW at -3C using 0.3 ACH. Further colder days this winter suggested a slightly higher heat loss rate closer to 3.6±0.5 kW See HeatpumpMonitor.org heat demand tool. Using 0.6 ACH for the heat loss calculation results in a heat loss of 4.0 kW which is within the upper edge of uncertainty looking at the measured data.

Case study 2: 1980s Bungalow, Vaillant Arotherm 5kW

YouTube: Solid Fuel to Vaillant Air Source Heat Pump: 1980s Bungalow

  • 0:00 Introduction

  • 0:50 Heat loss calculation using the heatpunk.co.uk tool. This part includes a discussion of:

    • Setting lower air change rates again based on the blower door test results.

    • Heat pump selection, why Glyn choose a 5kW Vaillant Arotherm.

    • Radiator upgrades in order to run at lower flow temperatures.

  • 4:58 Removal of old hot water system

  • 6:40 Installation of new cylinder, primary pipework & heat pump.

  • Note anti-freeze valve detail at 10:25 and 11:44.

  • 12:05 Radiators

  • 16:22 Installation complete

  • 16:53 Vaillant controller settings & heating curve (0.5)

  • 19:43 Monitoring & performance

Case study 3: Mid-terrace, Mitsubishi Ecodan 5kW

YouTube: A mid terrace heat pump example

Note from Trystan: This video is a bit old now and reflects an earlier point in my learning and experience with heat pumps. The heat loss calculated was in hind sight a significant over-estimate as air change rates from the SAP/CIBSE guidance were used. A blower door test done after this video and actual monitored heat loss data suggests that the actual air change rate is closer to 0.4-0.7 ACH. The decision to try a non-thermostatic mixer was definitely going over board and Im now more than happy with the choice of the standard cylinder with large surface area coil rather than alternatives such as a thermal store.

  • 0:00 Introduction

  • 2:50 Performance

  • 4:50 Heat loss calculation.

  • 7:40 Low temperature radiator design

  • 11:55 Hot water cylinder, large surface area coil.

  • 13:14 System diagram and cost of materials (2018)

  • 14:20 Monitored annual consumption vs SAP assessment

  • 15:29 January 2020 data example

  • 18:17 Octopus Agile January 2020 cost example

Blog write-up