Understanding Compressors for Heat Pump Water Heaters

The compressor does the mechanical work of moving heat from the air into the water tank. Its type, operating speed, and refrigerant together determine how efficiently it does that work, and under what conditions.

For product design and scheme registration, the central choice is between a fixed speed compressor and an inverter-driven variable speed compressor. Both are used across the current market. The right choice depends on cost targets, refrigerant, control strategy, noise requirements, target climate, and how the product is tested.

COP and temperature lift

A heat pump uses four core components: evaporator, compressor, condenser, and expansion valve. The compressor raises refrigerant pressure and temperature. The condenser transfers that heat to the water. The expansion valve drops the pressure back down, and the evaporator absorbs heat from the surrounding air.¹

The metric that captures how efficiently this works is coefficient of performance (COP). COP is the heat delivered to the water divided by the electrical energy consumed by the compressor. Because the compressor is moving heat rather than generating it, the delivered heat substantially exceeds the electricity input. A COP of 3.0 means three units of heat delivered per unit of electricity consumed.

COP is not a fixed number. It falls as the temperature gap between the heat source and the delivered hot water increases. This gap is called temperature lift.¹ A unit operating in warm ambient air against a cool tank will have a substantially higher COP than the same unit working in cold air against a near-full tank close to the setpoint temperature.

Measured data from Australian field testing shows that instantaneous COP can range from around 4.5 when the tank-to-ambient temperature difference is approximately −10°C, down to around 2.5 when the difference is 40°C.² Compressor power consumption over the same range runs from around 900 W at low temperature differences up to around 1,400 W at high differences.²

This means compressor efficiency should be assessed across the full heating cycle, not at a single test point.

Fixed speed compressors

Fixed speed compressors run at constant speed. When the heat pump is on, they draw roughly constant power and deliver roughly constant heating capacity.

The simplicity is real. Fixed speed compressors are easier to control and design around, generally cheaper upfront, and carry a long reliability track record from heating, cooling, and water heating applications.

The efficiency trade-off appears at the end of the heating cycle. As water temperature rises toward the setpoint, temperature lift increases and COP falls, but the compressor keeps running at full speed. A fixed speed unit spends the final stage of each cycle at the high end of the power range, when each watt of compressor input delivers the least heat.

Fixed speed compressors also have binary control: fully on or fully off. This limits how a product can be tuned for noise, demand response, or grid interaction.

Variable speed compressors

Variable speed compressors, driven by an inverter, operate across a frequency range. In domestic HPWH designs, 30 Hz to 70 Hz is a typical working range.³ The controller adjusts speed in response to operating conditions.

COP and speed. At lower compressor speeds, COP is substantially higher. One study characterising a variable speed HPWH found COP ranging from 3.81 at 70 Hz to 5.93 at 30 Hz across equivalent ambient and water temperature conditions.³ Lower speed reduces the pressure ratio the compressor must maintain, cutting compressor work per unit of heat delivered.

The complication is that optimal speed is not simply “run as slow as possible”. As water temperature rises during the heating cycle, the optimal compressor speed increases. A controller that maintains the same low speed throughout will see poor heat exchanger effectiveness toward the end of the cycle. Characterising a variable speed HPWH for control purposes requires empirical testing at multiple speeds and conditions. Standard static efficiency tables do not capture this dynamic behaviour.³

Minimum speed constraints. Variable speed compressors cannot run indefinitely at very low speeds. Below a threshold, typically around 30% of rated capacity, refrigerant flow becomes inadequate and lubricating oil fails to return to the compressor reliably.³ Compressors also require periodic operation at higher speeds to circulate oil through the system. A controller that tries to sustain minimal compressor speed for extended periods will cause reliability problems.

Efficiency gains in practice. Without PV or time-of-use control, variable speed operation can reduce annual electricity consumption by around 28% compared to a fixed speed unit running at constant 50 Hz.³

Where variable speed adds less value than expected is in solar self-consumption. When a timer-controlled fixed speed unit is set to operate only during high-PV periods (for example, 10 am to 4 pm), the efficiency difference between fixed and variable speed narrows considerably. The tank’s inherent thermal storage provides most of the demand-shifting benefit, regardless of compressor type. Under modelled South Australian conditions, the lifetime cost advantage of variable speed over a well-controlled fixed speed unit with timer control is around 2.5%.³

Noise and platform flexibility. Variable speed operation can reduce noise during lower-load periods, which matters for products installed near living areas or boundary fences. Frequency flexibility also helps manufacturers adapt a platform across different electrical markets: a fixed-frequency compressor operating point needs to be reconsidered when crossing between 50 Hz and 60 Hz grids.

Compressor type and refrigerant

Compressor architecture depends on the refrigerant, because different refrigerants have different pressure-temperature characteristics and different compressor design requirements.

R290 and R134a. Most domestic HPWHs using R290 or R134a use rotary or scroll compressors, which suit the moderate operating pressures of these refrigerants. R290 can achieve around 10% better COP than R134a at ambient temperatures above 20°C, but experimental testing found no advantage and lower COP at 10°C ambient in at least one study. Compressor design and cold-condition performance are critical to whether the improvement materialises.⁴ R290 compressors are not straightforward R134a substitutes and must be designed for propane operation.

CO2/R744. CO2 operates at much higher pressures than other common HPWH refrigerants. Japanese Eco Cute products, a mature residential market with millions of units installed since the early 2000s, use swing compressors, two-stage compressors, sub-compressors, and in advanced designs, scroll expanders integrated into the refrigerant circuit.⁵ These designs manage CO2’s high operating pressure and extract additional work from the high-pressure refrigerant stream.

CO2’s thermodynamic advantage in water heating comes from its transcritical cycle. Above the critical point, CO2 releases heat across a temperature glide rather than at a fixed condensing temperature. This glide aligns well with heating cold inlet water across a large temperature rise to a high setpoint, which is the condition CO2 is well suited to. A thermodynamic analysis of a residential CO2 HPWH prototype estimated a technical COP limit of around 6.0 under Japanese shoulder-season conditions, heating water from 17°C to 65°C at 16°C dry-bulb air.⁶ The trade-off is design complexity. CO2 compressors and system components are engineered for high pressure, which narrows design options and increases cost relative to R290 or R134a.

R32. R32 is present in Australian and New Zealand hydronic heat pump products and is emerging in some HPWH platforms. Its operating pressure sits between R134a and CO2. Direct experimental COP data for R32 in domestic HPWH applications is limited in the published literature. Performance comparisons with R290 or CO2/R744 should not be inferred from the broader R32 space conditioning market without product-specific data.⁴

What this means for registration

For incentive scheme registration, the compressor type and control description are important inputs into AS/NZS 5125.1 testing and AS/NZS 4234 simulation.

If a variable speed compressor is used, EnergyAE needs to understand the frequency or frequencies at which the unit is tested. A unit described as “variable speed” but tested at a single fixed frequency is, for the purposes of that test, being assessed as a fixed speed product. The tested configuration must match the product supplied to the market.

Under the draft AS/NZS 5125.1 Appendix H physical test method, the unit operates through a standardised 24-hour tapping cycle at a fixed climate condition.⁷ Whether the compressor modulates or runs at fixed speed during this cycle affects the result and must be documented accurately.

Research gaps

The following areas are not well covered in the available published literature.

Compressor performance at Australian and New Zealand ambient conditions. Most published compressor characterisation data comes from Japanese or European test conditions. The Clift et al. (2023) study uses South Australian field data but for one variable speed product. Performance across the ambient range in the draft AS/NZS 5125.1 Appendix H test, which uses 9°C ambient, is not well characterised for the broader Australian product range.

Minimum speed thresholds across compressor families. The approximately 30% of rated capacity minimum speed threshold is documented for specific products but varies across compressor designs, refrigerants, and manufacturers. Publicly available data is sparse. This matters for demand response and PV control strategy design.

R32 in domestic HPWH. R32 is present in the market but direct experimental COP data across a range of ambient temperatures, comparable to what exists for R290 or CO2/R744, has not been identified in the available literature. Performance claims for R32 HPWH products should be assessed against standardised test data, not assumed from the R32 air conditioning and heat pump market.

Compressor type and noise in installed HPWH. Manufacturers cite noise figures in product materials, but independent comparative data on how compressor type (scroll vs rotary vs reciprocating) affects noise at equivalent HPWH conditions is limited.

Compressor control and aggregated demand response. How variable speed compressor control interacts with demand response aggregation, particularly when many units receive the same signal simultaneously, has not been well characterised at distribution network scale.

References

Moran, M.J., Shapiro, H.N., Boettner, D.D. & Bailey, M.B. (2018). Fundamentals of Engineering Thermodynamics (8th ed.). Wiley, pp. 383–412. Heat-pump COP is the heating effect divided by net work input; the maximum theoretical COP falls as the temperature gap between cold source and warm sink increases. ↩ ↩²
Morrison, G.L., Anderson, T. & Behnia, M. (2004). Seasonal performance rating of heat pump water heaters. Solar Energy, 76(1–3), 147–152. Measured instantaneous COP and compressor power across a range of tank-to-ambient temperature differences for air-source HPWHs under Australian conditions. ↩ ↩²
Clift, D.H., Leerson, J., Hasan, K.N. & Rosengarten, G. (2023). Maximising the benefit of variable speed heat-pump water heater with rooftop PV and intelligent battery charging. Solar Energy, 265, 112049. doi.org/10.1016/j.solener.2023.112049 ↩ ↩² ↩³ ↩⁴ ↩⁵ ↩⁶
Lee, A. & Cheng, S. (2023). A comparative study of R134a and propane (R290) as refrigerants in heat pump water heaters. Open Access Journal of Energy and Power Technology. Also: Kleefkens, O. (2019). Refrigerants for Heat Pump Water Heaters, IEA HPT TCP Annex 46. Heat Pump Centre. ↩ ↩²
Zhang, J.-F., Qin, Y. & Wang, C.-C. (2015). Review on CO2 heat pump water heater for residential use in Japan. Renewable and Sustainable Energy Reviews, 50, 1383–1391. ↩
Saikawa, M. & Koyama, S. (2016). Thermodynamic analysis of vapour compression heat pump cycle for tap water heating and development of CO2 heat pump water heater for residential use. Applied Thermal Engineering. ↩
Commonwealth of Australia (2026). Decision Regulation Impact Statement for Heat Pump Water Heaters. Department of Climate Change, Energy, the Environment and Water (DCCEEW), April 2026. ↩