This case study paper describes the advantages of how heat pumps with heat recovery worked to contribute to improving energy performance and reducing the carbon footprint of buildings, with a short payback, thus showing it is a viable technology for energy cost reduction and for improving the environmental impact of buildings within the UK. All actions were taken based on performance for both the current and the future expected needs of the building and are measured against a baseline. The research and results confirm that heat pumps with heat recovery are a viable solution for London office buildings offering an integrated approach to building heating and cooling. The project resulted in a final reduction of 62% on the base building controlled utilities (total building less occupier power). The added benefits will be around the carbon benefits that the resultant 86% reduction in gas will bring due to the gradual de-carbonisation of the electric energy production in the UK.

Practical application: This paper illustrates how to utilise an existing building’s energy use data to determine how relatively simple technology can optimise the provision of simultaneous heating and cooling for comfort. It also details how modelled data can be used to allow the correct sizing of equipment to deliver both good internal comfort and low carbon emissions.

This paper is the result of a study and resultant project around a real office building, 350 Euston Road, which is located at Regent’s Place, London. This building had already achieved significant energy reductions and has been part of awards for best-practice for energy savings by several independent institutions, including CIBSE.

Due to the life cycle of the building, it was necessary to replace the existing boilers and chillers on the property as they were approaching the end of their usable life, this provided a real life situation to study and provide useful information about the efficiencies of Heating, ventilation, and air conditioning (HVAC) solutions .

In order to ensure transparency and quality of results, the University of Padua – Department of Industrial Engineering and Studio Planning have been involved, respectively as an independent research institution and as a qualified building services engineering consultancy with great experience in innovative heat pump applications.

A comprehensive energy model was developed to simulate how the building operates. All actions and prospective results have been designed on the real heating and cooling needs of the building and are measured against a baseline which represents an excellently energy efficient office building in Central London.

With the interest to improve environmental and economic efficiency, this study also took into account one of the priorities highlighted in the London Plan, which is the use of decentralised combined power and heat generation (CHP), as well as the Mayor’s target of 25% of the global heat and power demand of the city being produced by CHP.

Building characteristics

The 350 Euston Road building in North East London is comprised of seven storeys, averaging 1486 m2 each. The ground floor consists of a reception area and two retail areas. The retail areas have an autonomous heating/cooling system, and are not included in this analysis. There are three main technical areas from the first to the seventh floor with two atriums, as shown in Figure 1. The two main surfaces of the envelope are fully glazed: the double skin has one single glass outside and one double glazing surface inside. The double glazing has a U-value equal to 1.78 W/(m2K). The opaque walls have a U-value equal to 0.6 W/(m2K). The real orientation of the building and all projected shadows have been taken into account and the effect of solar radiation has been evaluated.1


Figure 1. Subdivisions into the different zones for ground floor (a) and for floors from 1 to 7 (b).

HVAC system

The existing HVAC system at 350 Euston Road has mechanical ventilation and four-pipe fan coil and was based on: three boilers for heating, total capacity 1380 kW; two air-cooled chillers, total capacity 2180 kW; and two Air Handling Units (AHU) with 40,000 m3/h total air flow. The AHU’s were not originally installed with heat recovery, but do have bypass retrospectively installed which is driven through BMS to decrease the amount of external air supplied.

If needed, outdoor air was cooled and dried by means of the cooling coil, also active during the summer.

For terminal units, the building was equipped with the originally installed fan coils without valves on the supplying pipes. On the refurbished floors new fan coil units with two-way valves had been installed. The primary circuit runs with constant flow pumps while the secondary circuit has inverter driven pumps, which were not automatically controlled and the existing system on the heating side supplied water at 57℃, as is normally the case (even with condensing type boilers).

Operating conditions

Energy consumption had been decreasing in the preceding years thanks to a comprehensive energy reduction plan implemented by the property owners predominantly based on innovative building management policies and limited refurbishments of the existing systems.

The measures applied to the HVAC system before this study were: bypass on the AHU to allow a certain percentage of re-use of returned air; inverter control on secondary pumps; new fan coil units with two-way valves installed on the refurbished floors; boilers off during the summer months and no re-heat supplied to AHU.

Energy consumption in the year 2012 for chillers and boilers has been analysed (that represent the performance systems as previously described) and is shown in Figure 2. The data collected provide information about the 10,400 m2 building and its behaviour. Heating energy consumption was exceptionally low due to the renovation works on a substantial section of the building. However, cooling energy demand was high throughout the year, peaking at 22 kWh/m2 for the month of July that is the same value that would be expected for a whole year’s cooling energy according to CIBSE ECG19.2


Figure 2. Actual 2012 monthly thermal energy consumption (kWh).

This is likely due to the large proportion of glazing in the building envelope as well as to the high internal heat load caused by people and appliances.

There are several months where there are both significant heating and cooling loads, offering an opportunity for energy recovery solutions. These, along with other solutions will be the focus of the improvement options section.

Energy analysis baseline

A thorough and rigorous energy analysis has been carried out following the six-step process described below:

  1. A dynamic energy model based on current energy consumption has been created, in order to have a reliable baseline against which to simulate and measure the energy savings achievable by the proposed upgrades. The present comfort settings have been maintained for the transparency of the comparison.
  2. The model has been created by using software selected for this work was the Transient System Simulation Tool (TRNSYS) (3), because of its advanced HVAC modelling.
  3. Simulations have been carried out based on the Test Reference Year (TRY) of London. They evaluated the hourly profile of net demand from the fan coil units (sensible), from the AHUs, and overall heating and cooling net energy demand. The overall simulated heating and cooling net energy demand is shown in Figure 3.
  4. To validate the model, actual consumption data have been gathered from the energy chiller electricity metres and the gas metres feeding the boilers. Simulated values have been compared with measured values for the year 2011. The measured values used a fixed conversion for the Coefficient of Performance (COP) to provide an estimated thermal energy.


    Figure 3. Overall hourly need of the heating/cooling system (kW).

On average simulated values represent between 84% and 87% of measured data. This confirmed the validity of the model considering that the simulation did not include the losses due to distribution control and emissions. The summary of the peak load values is reported in Figure 4.

  1. The defined model has been applied to a four-pipe system incorporating the present components, namely AHU without heat recovery and chillers plus boilers, for baseline definition.
  2. The defined model has been applied to multiple system improvement options, described in the improvement options and implemented option sections.


Figure 4. Peak power of fan coil units, Air Handling Units and overall peak of the HVAC system (fan coil units and ventilation system).

The main challenge in refurbishing the building with heat pumps is represented by lower supply heating temperatures ideal to exploit heat pump technology at its peak efficiency. As stated in the Model of the Building-HVAC System section, the current system on the heating side supplies water at 57℃, as is normally the case with boilers even of the condensing type. Heat pumps can produce hot water at a temperature of 55℃, but the efficiency performance (COPs) would not be as high as at 45℃. For that reason the possibility to cover the building’s heating load with the fan coil units operating at much lower temperatures has been assessed.

As a precaution, a heating power of the fan coil units reduced by 30% has been considered. Re-assessing the building with 45℃ working temperature, data show that it is possible to guarantee the heating capacity needed by the building. There will be no need to maintain the boilers and the occupied space can be freed. Alternatively it is possible to keep the existing boilers as a precautionary backup, at no extra cost.

These units can produce hot and chilled water at the same time and totally independently, adapting to the variable heating and cooling demands of the building. There are three basic operating configurations, which are totally independent from external temperature conditions: only chilled water production (the unit works as a simple chiller); only hot water production (the unit works as a heat pump); combined production of hot and chilled water (the unit simultaneously and autonomously produces cold and hot water for the two plant sections).

The above working configurations are selected automatically by an on-board microprocessor in order to minimise the absorbed energy and satisfy each thermal request. When simultaneous heating and cooling demand occur, energy can be obtained almost for free for the building’s needs. To measure the performances of such machines a new dedicated tool is needed to assess the global performance of the heat pump, when hot and cold water are produced simultaneously.

For chillers the energy efficiency ratio is EER and for heat pumps it is COP. To validate the efficiency of these combined units a Total Efficiency Ratio (TER) has been used. TER is the combination of the COP and EER in one single index, invented by Climaveneta. In the case of the simultaneous, balanced demand of heating and cooling, these units can achieve efficiency corresponding to TER values between 7 and 8. The superior efficiency is evident considering that 3.2 is the EER for class A chillers.

Improvement options

Multiple HVAC system improvement options considered are presented below. A comparative analysis of the energy efficiency, environmental and economic results of each of the options considered against the baseline has been carried out.

The replacement of existing boilers and chillers with high efficiency boilers and high efficiency chillers has been considered. This solution would provide energy savings vs. standard boilers and chillers, but significantly lower than those achievable adopting heat pumps with heat recovery.

The London Plan guide5 outlines how the CHP systems must be designed to run efficiently and be optimally sized to maximise carbon dioxide savings. For this reason a deep load analysis on each specific case is very important to validate adoption of such systems. If there is a lack of heating demand a CHP system’s efficiency is halved and it is no longer economically and environmentally convenient. The use of CHP was not a viable option in this case because of very variable heating loads. From the data available there was no constant heating demand that could justify the use of CHP. Also micro CHP was not an option because of the lack of large district heating systems to connect due to space availability, low energy efficiency and high management costs for maintenance, local emissions analysis and management.

Suitable and final option

Replacing existing one of the old chillers with a heat recovery heat pump, this ensured significant energy savings as it allowed for ‘energy transfer’, recovering energy that otherwise would be wasted on top of the energy saving ensured by heat pump technology, the other chiller was replace with a high efficiency Turbo Cor, with two boilers retained to provide redundancy heat as back up, and cold morning pre heat.

To properly assess when heating and cooling demand are simultaneous and therefore predict the exact achievable savings, a complete simulation has been run, based on the hourly data gathered from the dynamic model created.

The units selected for the simulations were ERACS2-Q SL/CA 2722. The technical data of a single unit are specified in Figure 5.


Figure 5. Technical data for single ERACS2-Q SL/CA 2722 unit.

Improved controls

To allow such changes in technology to work, there is a requirement to operate the right equipment to suit the internal demands; so a Demand Driven Control strategy is required. Through the installation of more on floor sensors to monitor on floor space temperatures, the control system is able to determine the correct configuration of which equipment to run, the required load of which to run at and also when.

Within the control strategy is also an out of core hours upper and lower limit, by having this the building never gets ‘too hot’ or ‘too cold’.

Calculations and explanations

Savings are measured in kWh of primary energy as a way to compare thermal and cooling energy produced by different sources (electricity and gas), as well as CO2 emissions. Energy is always expressed in primary energy i.e. raw energy before any transformation, in order to enable an efficiency comparison of different generators: for gas boilers, this is the energy contained in the raw fuel and needed by the generator to fulfil the building’s needs (natural gas); for chillers and heat pumps, primary energy is calculated before the production of electric energy and therefore contains the production and transmission losses of the national grid. The conversion factors fact sheet 2013 from the Carbon Trust has been taken as reference for the conversion of kWh of energy to tons of CO2.

Part L4 estimates 15 years as the expected lifetime of the new HVAC system, it makes thus sense to consider the CO2 emissions evolution in light of the scenarios outlined by the National Grid on the 2013 guide ‘UK Future Energy Scenario’.6

The UK government is committed to reducing the CO2 impact of power generation.7 In this paper two scenarios are foreseen: Gone Green, a balanced approach to meeting renewable energy and CO2 emission targets in 2020 and 2030 and; Slow Progression, a slower approach to meeting renewable energy and CO2 emission targets, for example, the UK2020 renewable target is missed and greenhouse gas reductions fall short of the 2050 carbon targets and the fourth carbon budget.

A comparison between the minimum requirements for heat pump units and heat pumps with heat recovery has been calculated. As seen in Figure 6, heat pumps with heat recovery largely exceed minimum required heat pumps performances.


Figure 6. Heat pump with recovery vs. traditional heat pumps.

SPF computation

SPF is the operating performance of an electric heat pump over the season expressed as the ratio between the heat delivered and the total electric energy absorbed over the season. The BS EN 15450:2007 standard considers only standard heat pumps and does not provide a method to calculate the performance of innovative heat pump with heat recovery. Therefore to verify compliance with part L, SPF has been calculated in the most penalising way i.e. considering that all the absorbed electric power is for heating only, neglecting the energy recovered. SPF with this method is 2.7 exceeding the standard by 8%. To correctly represent the performance of a heat pump with heat recovery the calculation should include the share of heat recovered for free by the heat pump. In this case SPF would be 3.34. This value exceeds the standard by 34%, a clear indication of the superior efficiency of this type of unit in London conditions in an office building application.

EER computation

The EER is calculated with an outside temperature of 35℃, which never reached in London. In compliance with The London Plan,5 the overall aim is to achieve an overall reduction in London’s carbon dioxide emissions of 60% (below 1990 levels) by 2025, and the heat pump technologies considered in the simulation would ensure a carbon emission reduction of 90%.

Studied environmental results

This section presents the results of the comparison of the current system and the new system based on heat pumps with heat recovery in terms of CO2 reduction over the life cycle of the systems, based on National Grid 2013 guide ‘UK Future Energy Scenarios’.5

Heat pumps with heat recovery reduce the building’s primary energy consumption by 38% and CO2 emissions by 34.6%. Due to the gradual de-carbonisation of the electric energy production in the UK, the carbon reduction achievable with this technology will improve over the lifetime of the system, which will be discussed in the following section. The new system reduced the building’s CO2 emissions by 34.6% and its primary energy consumption by 38% as shown in Figure 7.


Figure 7. Energy (kWh) and emissions Co2 savings of new heat pumps.

These performance enhancements are due to the energy recovery combined with the higher efficiency of the adopted heat pump technology. The data, shown in Figure 8 show the energy results for each month. In particular during mid-season, when an overlapping cooling and heating demand is more frequent, the actual absorbed energy decreases compared to winter months, even if the total amount of energy (cooling and heating) produced is the same. This shows that a yearly average of 10.4% of the overall energy demand can be recovered for free, instead of being wasted on the heat sink, which is covered more in the Further improvements section.


Figure 8. Heating & cooling energy produced, absorbed energy and percentage of heat recovery.

CO2 emission reduction – life cycle analysis

The study proves that a carbon emissions reduction of 35% is achievable with the adoption of heat pumps with heat recovery based on National Grid 2013 guide ‘UK Future Energy Scenarios’6 and using the data provided, emissions have been estimated in 2020 and 2025.

Increases in the grid efficiency, although definitely very probable, have not been accounted for. Emissions reduction over the life cycle of the new system with heat pumps become even higher if compared to the system including boilers, as emissions linked to natural gas cannot benefit from the government commitment to reducing the CO2 impact of power generation.

In 2020 the building analysed in this paper, if equipped with heat pumps with heat recovery, is predicted to emit between 67.4% and 82.2% less than the emissions of the existing system in 2013. In 2025 the same figure will show a CO2 reduction between 75.4% and 88.4%. In other words, in 2025 the CO2 emissions of the building if equipped with heat pumps with heat recovery, are predicted to be reduced between 50.6% and 55.7%, (for slow progression and ‘Gone Green’ respectively) compared to the chiller plus boiler present solution as seen in Figures 9 and 10.


Figure 9. UK generation by fuel type Gone Green.


Figure 10. UK generation by fuel type Slow Progression.

Actual final utilities results

The data in Figure 11 are that over the previous 5 years, showing that in 2012/13, the total kWh being used by the building for HVAC (shared services) was 1,856,213 kWh, and in 2015/16 the final results were 502,702 kWh, a 73% reduction in input energy.


Figure 11. UK generation by fuel type slow progression.

Economic results

Payback period, based on the Part L guidelines outlined in the Calculations and explanations section, is defined as: the amount of time it will take to recover the initial investment through energy savings. It is calculated by dividing the marginal additional cost of implementing an energy efficiency measure by the value of the annual energy savings achieved by that measure, not taking VAT into account with energy prices derived from the Department of Energy and Climate Change’s statistical data set of September 2013.

Investment costs are based on predictions elaborated with two separate UK installers.

The extra cost of £25,000 for the heat pump installation is due to modification of the existing heating pipe work and pumps shown in Figure 12. The final results on gas savings along; using 3p/kWh gas, resulted in just under £27,000 in gas savings per year.


Figure 12. Equipment costs.

In new developments simplification of the system allowed by this technology not only provides the considerable savings described in this study but results in reduced installation costs compared to a chiller and boiler system.

In refurbishments there are additional, although limited, costs due to adapting existing piping works to the new system, accounting for 5% of the total investment. Given the achievable savings, the payback is less than 2 years with calculations as shown in Figure 13.


Figure 13. Running costs.

Even very prudent calculations of the Net Present Value (NPV) of the total savings at 15 years from the initial investment, assuming the same yearly average energy cost increase of the last 10 years, show a total savings covering 60% of the total cost of the initial investment.

Although significant improvements have been made through the replacement of the existing HVAC system, further improvements were obtained by: advanced plant room monitoring and optimisation software; new terminal units and distribution system. By adopting dedicated optimisation software for the plant room, designed to integrate with the BMS and take full control of the heating and cooling generation as well as pumping, this predicted an additional improvement of about 10% of total HVAC energy efficiency. Refurbishing the current distribution system, completing the replacement of old fan coils with no dynamic water flow control with new fan coils with two-way valves on the heating and cooling coils supply pipes, would allow a reduction in water flow and energy consumption.

To further enhance the system, the secondary circuit were inverter driven as well to reduce the yearly energy consumption and provide fluid demand to the space based on demand.

Through this study and final results highlight what value has been brought in terms of energy costs and use, as well as carbon emissions of using heat pumps with heat recovery in the Central London area. Although initial investment can be higher, especially in regards to retrofit projects, the return on investment is reasonable, if not quick. The implementation of heat pump with heat recovery technology at 350 Euston Road, Regent’s Place, London has shown remarkable results in terms of energy consumed, and therefore energy costs on a buildings HVAC systems can be reduced by 73%. With the changes in the UK fuel type generation, the CO2 benefits will be increased in the future.

As has been shown, the introduction of heat pumps to buildings is beneficial, but the benefits are only possible when the heating and cooling for HVAC purposes is operational simultaneously for more than 25% of the yearly operating time.

We have already identified where the same technology can be used on a water cooled system, with the heat rejection being applied to the heating system, rather than the cooling towers, this is expecting to result in further electrical savings due to reduced pump and motor operation.

This application has also shown that buildings in London could operate on heat provided only from heat pumps, but consideration to the FCU/AHU coils operating on a maximum of 45 Degrees should be considered, but the introduction of suitable controls, can reduce the need increase their size significantly.

Although the performance is already surpassing initial goals, it is likely that by adopting a range of other measures could further improve the energy profile of this property.

A special acknowledgement goes out to the University of Padua – Department of Industrial Engineering and Studio Planning for being the independent research institution and qualified building services engineering consultancy, respectively. Also thanks to British Land and Broadgate Estates, the owner and managers of 350 Euston Road, Regent’s Place, London. An acknowledgement is also sent to Climaveneta the manufacturer of the heat pump technology considered in this study, and Cavendish Engineering for the final delivery of the project.

The author(s) declared no potential conflicts of interest with respect to the research, authorship and/or publication of this article.

The author(s) received no financial support for the research, authorship, and/or publication of this article.

1. Mahdavi, A . Predictive simulation-based lighting and shading systems control in buildings. Build Simul March 2008; 1: 2535.
Google Scholar | Crossref | ISI

2. Energy consumption guide 19. CIBSE, 2000, Appendix B.
Google Scholar

3. TRNSYS, a transient system simulation program, reference manual, vers. 15. USA: Solar energy laboratory University of Wisconsin Madison, 2000.
Google Scholar

4. UK Government. The building regulations 2010 – L2B conservation of fuel and power in existing buildings other than dwellings. 2010.
Google Scholar

5. The London plan. Spatial development strategy for Greater London. Greater London Authority, July 2011.
Google Scholar

6. UK national grid. National grid 2013 UK future energy scenario. 2013.
Google Scholar

7. The carbon plan, delivering our low carbon future. HM Government, December 2011.
Google Scholar