Heat Pumps in HEM — Technical Guide

The Home Energy Model (HEM) calculates heat pump performance using a dynamic, variable-COP methodology documented in HEM-TP-12. Rather than applying a single seasonal efficiency figure as SAP does, HEM interpolates the coefficient of performance (COP) at every half-hourly timestep based on real source and sink temperatures, part-load conditions, and defrost losses. This approach — grounded in European test standards EN 14825, EN 15316-4-2, and EN 16147 — captures the genuine performance characteristics of heat pumps across the full range of UK weather conditions, rewarding well-designed systems and penalising poor ones.

HEM-TP-12: Heat Pump Calculation Methodology

The heat pump module is documented in HEM Technical Paper 12 (HEM-TP-12), published as part of the HEM technical documentation on GOV.UK. It describes how HEM determines the electrical energy consumed by a heat pump to meet both space heating and domestic hot water demand at each half-hourly timestep. The methodology draws on three European standards:

• EN 14825 — defines the test conditions and data format for declaring heat pump seasonal performance, providing COP values at specific source and sink temperature combinations
• EN 15316-4-2 — specifies the calculation method for determining heat pump system energy consumption from test data, including part-load corrections and auxiliary energy
• EN 16147 — covers testing of heat pumps for domestic hot water heating, including tapping cycle tests and standby losses

At its core, HEM-TP-12 takes the heat demand calculated by other modules (space heating from HEM-TP-04, hot water from HEM-TP-09) and determines how much electrical energy the heat pump requires to deliver that demand, given the current operating conditions. The result feeds into HEM's electricity balance, which determines total energy consumption and, ultimately, whether the dwelling meets the FHS compliance targets.

Variable COP Modelling

The coefficient of performance (COP) of a heat pump is not a fixed value. It varies continuously with operating conditions, primarily driven by the temperature lift — the difference between the heat source temperature and the required output (sink) temperature. HEM models this variability explicitly at every half-hourly timestep.

Source Temperature

The source temperature is the temperature of the medium from which the heat pump extracts thermal energy:

• Air source heat pumps (ASHPs): The source temperature is the outdoor dry-bulb air temperature, taken directly from the half-hourly weather data. In the UK, this typically ranges from −5°C to +30°C over a year, producing dramatic variation in COP. During a cold January night at −3°C, an ASHP might achieve a COP of only 2.0–2.5. On a mild October afternoon at 12°C, the same unit could deliver a COP of 3.5–4.0 or higher.
• Ground source heat pumps (GSHPs): The source temperature is derived from a ground temperature model. In the UK, ground temperatures at borehole depth (typically 50–150 metres) remain relatively stable at 8–12°C year-round, while horizontal collector loops at 1–2 metres depth show more seasonal variation (approximately 4–14°C). This stability means GSHPs maintain a more consistent COP throughout the year, particularly during peak winter demand when ASHPs are least efficient.

Sink Temperature (Flow Temperature)

The sink temperature is the flow temperature to the heating distribution system. HEM determines this based on the emitter type, design conditions, and the weather compensation curve (if applicable):

• Underfloor heating: Typically designed for 30–40°C flow temperature, producing the smallest temperature lift and therefore the best COP
• Oversized radiators at low temperature: The FHS notional building assumes radiators sized for 40°C flow temperature, which requires radiators approximately 2–2.5 times larger than those sized for the traditional 75/65°C boiler-era design temperatures
• Conventional radiators: If existing radiators require 55°C or higher flow temperature, COP drops significantly. HEM captures this penalty explicitly

Weather compensation reduces flow temperature in mild weather when the full design output is not needed. HEM models this by adjusting the flow temperature at each timestep based on the outdoor temperature, following a user-defined or default compensation curve. On a day when the outdoor temperature is 10°C, the flow temperature might drop to 30°C rather than the 40°C design maximum, improving COP substantially.

COP Interpolation from Test Data

HEM does not use a theoretical Carnot formula to calculate COP. Instead, it interpolates from EN 14825 test data specific to the heat pump make and model. EN 14825 requires manufacturers to declare COP at defined test points:

• Source temperatures: For ASHPs, test points are typically at −7°C, +2°C, +7°C, and +12°C outdoor air temperature. For GSHPs, test points are at 0°C, +5°C, and +10°C brine temperature.
• Sink temperatures: Test data is provided at multiple flow temperatures, commonly 35°C and 55°C for space heating applications.

At each half-hourly timestep, HEM determines the current source temperature from weather data (or the ground model), identifies the required sink temperature from the emitter circuit and weather compensation curve, and then performs a two-dimensional interpolation across the EN 14825 test matrix to derive the operating COP. When conditions fall between test points, linear interpolation is applied. When conditions fall outside the tested range (for example, an ASHP operating at −10°C when the lowest test point is −7°C), extrapolation is used with appropriate safeguards to prevent unrealistic values.

European Test Standards in Detail

EN 14825 — Seasonal Performance Declaration

EN 14825 is the primary standard governing how heat pump performance is tested and declared for regulatory purposes. It defines a set of standardised operating conditions (termed “part-load conditions”) representing typical European climates. For the UK (“average climate” profile), the standard defines four test conditions for ASHPs:

At each test condition, the standard requires measurement of heating capacity (kW) and electrical input (kW), from which COP is derived. Tests are conducted at both low-temperature (35°C flow) and medium-temperature (55°C flow) sink conditions. The resulting dataset gives HEM a matrix of COP values across the operating envelope, from which it can interpolate for any real-world combination of source and sink temperature.

EN 15316-4-2 — Heat Pump System Calculation

While EN 14825 defines the test data format, EN 15316-4-2 provides the calculation method for determining energy consumption from that test data in a real building context. HEM draws on this standard for:

• Part-load correction: Adjusting COP for the actual operating load relative to the heat pump's capacity at current conditions
• Cycling losses: When the heat pump capacity exceeds demand, the unit cycles on and off, reducing effective efficiency
• Auxiliary energy: Accounting for parasitic electrical consumption from circulation pumps, controls, crankcase heaters, and fans
• Standby losses: Energy consumed when the heat pump is not actively heating but remains energised

EN 16147 — Hot Water Heating Test

EN 16147 addresses heat pump performance specifically for domestic hot water production. It defines standardised tapping cycles (sequences of hot water draws representing typical household usage patterns) and measures the heat pump's ability to maintain cylinder temperature while meeting those draws. Key outputs include:

• COP for water heating: Measured over a complete tapping cycle, including recovery time and standby losses
• Recovery time: How long the heat pump takes to reheat the cylinder after a large draw
• Standby losses: Electrical energy consumed to maintain cylinder temperature between draws

HEM uses EN 16147 data alongside the hot water module (HEM-TP-09) to determine the electrical energy for water heating at each timestep. The interaction between space heating and hot water is critical: most domestic heat pumps prioritise hot water over space heating, meaning that during a hot water recovery cycle, space heating may be temporarily interrupted. HEM models this priority logic explicitly.

Air Source vs Ground Source Heat Pumps in HEM

HEM applies the same core methodology to both ASHP and GSHP systems but the source temperature behaviour creates fundamentally different performance profiles. Understanding these differences is important for architects and designers selecting systems for FHS compliance.

The FHS notional building specifies an air source heat pump, meaning ASHPs are the baseline for compliance. A GSHP with its higher COP profile provides additional compliance headroom, which can offset other aspects of the design that fall short of the notional specification. However, the higher capital cost of ground source systems means they are typically only selected where site conditions are favourable or where the compliance margin from an ASHP is insufficient.

Defrost Cycle Modelling

When an air source heat pump operates at low outdoor temperatures (typically below 5°C) with high relative humidity, moisture in the air freezes on the evaporator coil, reducing heat transfer efficiency. The heat pump must periodically run a defrost cycle to remove this ice, which consumes energy and temporarily reduces heating output.

HEM models defrost losses as a function of outdoor temperature and humidity conditions. The EN 14825 test data inherently includes defrost effects at the lower test points (particularly at −7°C and +2°C), so the interpolated COP values already account for typical defrost behaviour. However, HEM applies additional corrections when conditions are particularly adverse — for example, during extended periods near 0°C to +3°C with high humidity, where defrost frequency is typically highest.

For the UK climate, defrost losses are most significant during November to March. HEM's half-hourly resolution captures periods of intense frosting that would be hidden within SAP's monthly averages. A week of freezing fog in January, for instance, can substantially increase defrost energy consumption, and HEM models this period explicitly rather than spreading it across the entire month.

Part-Load Performance

Heat pumps rarely operate at full capacity. For most of the heating season, the outdoor temperature is well above the design condition, meaning heating demand is a fraction of the heat pump's rated output. How the heat pump performs at part load has a major impact on annual energy consumption.

Inverter-Driven (Variable Speed) Heat Pumps

Modern inverter-driven heat pumps can modulate their compressor speed to match output to demand. This avoids on/off cycling and typically improves part-load COP because the compressor operates at a lower, more efficient speed. HEM models this behaviour using the part-load COP data from EN 14825, which includes performance at the four defined part-load ratios (88%, 54%, 35%, and 15% of design capacity).

At very low load fractions (below the minimum modulation range of the compressor, typically 15–30% of rated capacity), even inverter-driven units must cycle. HEM applies a cycling degradation factor derived from the EN 14825 test data to account for the efficiency loss during these periods.

Fixed-Speed (On/Off) Heat Pumps

Fixed-speed heat pumps cannot modulate — they operate at full capacity or not at all. When demand is below rated capacity, the unit cycles on and off, with each cycle incurring start-up losses and reducing average COP. HEM applies a cycling correction factor (C_c) derived from EN 14825 test procedures:

• At full load: no correction (COP as tested)
• At 50% load: effective COP reduced by approximately 5–10% due to cycling losses
• At 25% load: effective COP may be reduced by 15–25% depending on the unit's cycling characteristics

This distinction matters because a well-insulated FHS home has relatively low peak heating demand, meaning the heat pump spends a large proportion of the heating season at low part-load ratios. Inverter-driven units perform significantly better in this operating regime, and HEM's part-load modelling captures this advantage accurately.

Low-Temperature Emitter Design

The sink temperature (flow temperature to the emitter circuit) is one of the two primary drivers of heat pump COP. Designing emitter circuits for low flow temperatures is therefore critical to achieving good heat pump performance under HEM.

FHS Design Conditions

The FHS notional building assumes a 40°C design flow temperature for the emitter circuit. This is substantially lower than the 75/65°C flow/return temperatures traditionally used with gas boiler systems and requires emitters to be sized accordingly:

• Radiators at 40°C flow: Must be approximately 2–2.5 times larger than the equivalent radiator sized for 75/65°C. In practice, this often means double-panel, double-convector (Type 22) radiators on every wall, or very large single-panel units. The increased radiator size has significant implications for room layout and architectural design.
• Underfloor heating: Naturally operates at low flow temperatures (typically 30–40°C) and provides a large emitting surface area. It is inherently well-suited to heat pump systems and typically produces the best COP results in HEM.
• Fan-assisted radiators: Use integrated fans to increase convective output, allowing smaller physical radiators to deliver the same output at low flow temperatures. HEM credits the improved output but accounts for the fan's electrical consumption as a parasitic load.

Weather Compensation

Weather compensation controls reduce the flow temperature when outdoor conditions are mild, taking advantage of the reduced heating demand. HEM models weather compensation as a linear (or user-defined) curve mapping outdoor temperature to flow temperature. For example:

• At −3°C outdoor: flow temperature at 40°C (design maximum)
• At +5°C outdoor: flow temperature reduced to approximately 32°C
• At +12°C outdoor: flow temperature reduced to approximately 25°C

This dynamic adjustment means the heat pump operates at a lower temperature lift for most of the year, improving average COP. HEM calculates the flow temperature at each timestep from the compensation curve and uses it as the sink temperature for COP interpolation. Without weather compensation, the system would run at the full design flow temperature regardless of outdoor conditions, wasting significant energy.

SAP vs HEM — Heat Pump Modelling Comparison

The shift from SAP to HEM represents a fundamental transformation in how heat pump performance is assessed for regulatory compliance. The table below summarises the key methodological differences:

The practical consequence of this shift is that heat pump system design matters far more under HEM. A well-designed system — low flow temperatures, correct sizing, weather compensation, and inverter-driven compressor — will demonstrate significantly better performance than a poorly designed one. Under SAP, these distinctions were largely invisible because the simplified SPF approach could not differentiate between them. For a broader comparison, see our SAP vs HEM overview.

Interaction with Hot Water Production

Most domestic heat pump installations serve both space heating and domestic hot water, typically using a hot water cylinder. The interaction between these two services is modelled in detail by HEM, drawing on both HEM-TP-12 (heat pumps) and HEM-TP-09 (hot water).

Priority Logic

Domestic heat pumps typically operate in a priority mode where hot water takes precedence over space heating. When the cylinder thermostat calls for heat, the heat pump switches to water heating mode, temporarily interrupting space heating. HEM models this priority logic at each timestep:

1. Check whether the cylinder requires heating (temperature below setpoint)
2. If yes, allocate the heat pump to hot water mode and calculate the electrical energy consumed using the water-heating COP (from EN 16147 test data)
3. Determine the duration of the hot water recovery cycle and any resulting interruption to space heating
4. Calculate the space heating shortfall during the interruption and the resulting temperature drop in the heated zones

This interaction is significant because hot water heating typically requires higher flow temperatures (50–55°C to the cylinder) than space heating (35–40°C to emitters), resulting in a lower COP for the water-heating portion. HEM captures the blended effect of these two operating modes, producing a more accurate picture of total energy consumption than SAP's approach of applying separate seasonal factors for space heating and water heating.

Legionella Protection Cycles

Building Regulations require periodic heating of hot water cylinders to at least 60°C to prevent Legionella bacteria growth. For a heat pump, reaching 60°C involves a very high temperature lift, dramatically reducing COP (often to 1.5–2.0). Some systems use a direct electric immersion heater for this cycle instead. HEM models the energy consumed by Legionella protection cycles, whether delivered by the heat pump at reduced COP or by an immersion heater at a COP of 1.0.

Product Data Requirements

HEM's variable COP methodology relies on detailed, product-specific test data. The quality and completeness of this data directly affects the calculated performance of the heat pump in the compliance assessment.

Product Characteristics Database (PCDB)

Heat pump performance data is held in the Product Characteristics Database (PCDB), maintained by BRE. For each heat pump make and model, the PCDB entry should contain:

• EN 14825 test data at multiple source and sink temperature combinations
• EN 16147 test data for hot water performance (if the unit serves DHW)
• Rated capacity at design conditions
• Minimum and maximum modulation range (for inverter-driven units)
• Cycling degradation coefficient (C_c)
• Auxiliary electrical consumption (fans, pumps, controls)
• Standby power consumption

Complete, accurate PCDB data allows HEM to model the specific performance characteristics of the installed heat pump. This is a significant departure from SAP, where the PCDB entry for a heat pump was relatively simple — essentially a seasonal efficiency value with limited product differentiation.

Punitive Defaults for Missing Data

Where make-and-model-specific data is not available in the PCDB, HEM applies punitive default values. These defaults assume significantly worse performance than a typical tested product would achieve, serving two purposes:

• Incentivising data submission: Manufacturers are strongly motivated to submit test data because using defaults substantially disadvantages their products in compliance calculations
• Protecting accuracy: Without real test data, HEM cannot know the actual performance of the unit, so conservative assumptions prevent untested products from gaining an unearned compliance advantage

Sizing Implications Under HEM

Heat pump sizing is more consequential under HEM than it was under SAP, because the half-hourly calculation captures the effects of incorrect sizing that monthly averaging concealed.

Undersizing Penalty

An undersized heat pump cannot meet demand during the coldest half-hours. HEM models this explicitly: when the heat pump's maximum output at the current source temperature is less than the heating demand, the shortfall must be met by backup heating (typically a direct electric immersion heater or resistance element with a COP of 1.0). This dramatically increases electrical consumption during peak demand periods.

Under SAP, this effect was largely hidden because monthly averaging smoothed peak demand into the surrounding period. A heat pump that could not quite meet demand on the coldest nights but was adequate on average would appear satisfactory in SAP. Under HEM, the energy penalty of those specific half-hours is captured explicitly.

Oversizing Considerations

Oversizing also carries penalties under HEM. An oversized heat pump spends more time at low part-load ratios, increasing cycling losses (for fixed-speed units) or operating near the bottom of the modulation range (for inverter-driven units). HEM's part-load modelling captures this effect, meaning there is an optimal sizing range that balances peak capacity against part-load efficiency.

The recommended approach is to perform a detailed heat loss calculation at design conditions (typically −3°C to −4°C outdoor temperature for most of England) following BS EN 12831, and size the heat pump to meet 100% of the design heat loss without backup heating. Modest oversizing (10–15% above calculated demand) provides a safety margin without significant part-load penalties.

Future Homes Standard Context

The Future Homes Standard effectively mandates heat pumps for new-build homes. The notional building specification — against which all designs are assessed — includes an air source heat pump as the primary heating system. While alternative low-carbon heating technologies are not explicitly banned, it is extremely difficult to meet the FHS carbon targets without a heat pump.

Notional Building Heat Pump Specification

The FHS notional building assumes:

• Air source heat pump (ASHP) as primary heating system
• 40°C design flow temperature to the emitter circuit
• Weather compensation control
• Hot water cylinder with heat pump as primary heat source
• Performance based on a “typical good practice” ASHP with EN 14825 test data

To pass FHS compliance, the actual building must achieve performance at least as good as the notional building. This means that heat pump selection, emitter design, and system integration must all be at least as good as the notional specification. Selecting a heat pump with better EN 14825 test data, or designing emitters for a lower flow temperature, generates compliance headroom that can offset other areas where the design falls short.

Implications for the Industry

The transition from gas boilers to heat pumps under FHS requires significant changes in design practice, installer skills, and supply chain capability. Key implications include:

• Architects must integrate heat pump requirements from the earliest design stages, including space for external units, acoustic considerations, and emitter sizing
• Builders need to understand that heat pump performance is highly sensitive to installation quality — refrigerant charge, pipe runs, flow rates, and controls commissioning all affect real-world COP
• Developers must factor in the cost of larger emitters, hot water cylinders, and potentially higher-specification heat pumps when assessing scheme viability
• SAP assessors transitioning to HEM must understand the variable COP methodology and the importance of selecting heat pumps with complete PCDB data

For a comprehensive overview of FHS compliance routes and how the heat pump specification fits into the broader picture, see our compliance pathways guide.

Frequently Asked Questions

Related Pages