Refrigeration Cycle p-h Diagram and Heat Pump COP Calculator
Explore real-fluid refrigeration and heat pump cycles with seven refrigerants. Calculate COP, capacity, superheat, subcooling and the optimum CO₂ gas-cooler pressure.
Use this refrigeration cycle calculator to study a single-stage vapour-compression refrigerator or heat pump with real refrigerant properties. It is intended for HVAC engineers checking a cycle concept, students learning to read a pressure–enthalpy chart, and people comparing compressor capacity at different operating conditions. Choose R-32, R-290, R-410A, R-454B, R-134a, R-1234yf or R-744 (CO₂).
Start with evaporating and condensing temperatures, or select absolute pressures. Superheat is measured from the evaporator dew point; subcooling is measured from the condenser bubble point. This distinction matters for blends. The default heat pump COP calculator R290 case evaporates at 0 °C and condenses at 45 °C, with 5 K superheat and subcooling. Change one setting at a time to see how the cycle moves.
The p-h diagram calculator marks compressor suction as state 1, discharge as state 2, heat-rejection outlet as state 3, and expansion-valve outlet as state 4. The bold loop sits over a saturation envelope, isotherms and constant-entropy lines calculated from CoolProp. The dashed compression curve is the isentropic reference. Drag the suction or heat-rejection handle to change its temperature or pressure level. The diagram and state table retain their explicitly labelled SI units; inputs and numeric result cards support the metric/US switch.
Select displacement and speed to obtain refrigerant mass flow from the suction density and volumetric efficiency. Alternatively, enter a measured mass flow. Both cooling and heating capacities are reported, along with compressor shaft power and discharge temperature. These COPs use shaft work; motor losses, fans and pumps require a separate electrical energy balance. Enter the external cold and hot reservoir temperatures for a meaningful Carnot comparison.
For an R32 pressure enthalpy chart or another refrigerant, change the refrigerant while keeping the operating temperatures and efficiencies comparable. Capacity at equal displacement also depends strongly on suction density. A thermodynamic comparison does not establish that refrigerants are interchangeable in equipment.
Selecting CO₂ replaces condensing temperature and subcooling with gas-cooler outlet temperature and pressure. The pressure sweep searches 75–160 bar for maximum cooling COP and refines the best interval. Heating COP reaches its maximum at the same pressure because it equals cooling COP plus one. The Liao correlation is shown separately as a literature comparison.
Play the circuit to follow refrigerant through compression, heat rejection, throttling and evaporation. Pipe colours show calculated refrigerant temperature. Motion illustrates circulation direction, not residence time. The model excludes pressure losses, heat-exchanger sizing, refrigerant charge, defrost, oil return and compressor operating envelopes. R-454B saturation calculations are limited to 55 °C because of the available mixture flash solver. Installation work belongs to a qualified person.
How it is calculated
The steady-flow energy balances follow the single-stage cycle treatment in ASHRAE Handbook—Fundamentals (2025), Chapter 2, Thermodynamics and Refrigeration Cycles. Every state uses CoolProp 8 through fluid(); no constant heat capacities or ideal-gas refrigerant substitutions are used.
- Evaporating temperature denotes dew point; condensing temperature denotes bubble point. Pressure inputs are absolute. State 1 is saturated vapour plus the entered superheat. State 3 is saturated liquid minus subcooling, or the specified CO₂ gas-cooler outlet temperature.
- Evaluate and . Expansion is isenthalpic: . The mixture compressor flashes are solved in temperature with bracketed roots; valve quality is found by an equilibrium pressure–quality flash. R-410A pressure–temperature, pressure–enthalpy and pressure–entropy queries, including the diagram and sweep, use phase-aware brackets. Solve on the stable liquid or vapour density branch; enthalpy/entropy inversions add a bounded temperature root. Two-phase states use CoolProp pressure–quality flashes. This avoids fragile native inversions while retaining the same equation of state.
- Specific refrigeration, shaft work and heat rejection are , and . Multiply by or the specified mass flow. Then and .
- For external isothermal reservoirs, and , with absolute temperatures.
- For CO₂, evaluate the real-fluid cycle over 75–160 bar (161 default points), then maximise COP with a bounded Brent search inside the best interval. The separate literature comparator is Liao, Zhao and Jakobsen (2000): , with pressure in bar and temperatures in °C. It does not replace the numerical optimisation.
Assumptions
Steady single-stage operation; pure refrigerants or CoolProp's fixed-composition blends; adiabatic compressor; constant entered isentropic and volumetric efficiencies; negligible kinetic/potential energy and pressure loss; equilibrium throttling. Superheat is included in useful evaporator heat and subcooling in heat rejection. The full compressor displacement is per revolution. Shaft COP excludes motor, fan, pump, crankcase-heater and control power. R-744 uses a transcritical cycle; subcritical CO₂, ejectors, economisers, internal heat exchangers and multi-stage compression are outside this model.
Where it applies
Evaporation is restricted to −45…25 °C. The high-side saturation state must remain below the refrigerant critical region. For R-454B, the CoolProp mixture dew/bubble solver becomes unreliable near its upper envelope; this tool restricts condensation and its displayed envelope to 55 °C, leaving the critical cap outside the chart. It uses the actual R32/R1234yf mixture, not a surrogate refrigerant. State points and COP are solved directly.
Pure/pseudo-pure charts use the selected diagram resolution (81 points by default). To keep R-454B near a three-second calculation, its isotherms use three single-phase pressure points per branch, entropy lines use four pressure levels and three superheated temperatures plus six equilibrium quality points, the compression illustration uses three pressure levels, and its pressure sweep uses three solved pressures. PCHIP interpolation produces smooth displayed curves. The resulting blend background lines and sweep are qualitative between solved points; increasing diagram sampling refines the dome and display, not those fixed sparse blend meshes. R-454B heat-exchanger colours use separate PCHIP interpolants for liquid, two-phase and vapour branches, with 13 real-fluid knots per phase by default and the exact cycle inlet/outlet states included as knots. The advanced profile-resolution input refines those branches independently of the background diagram; convergence checks compare 13 with 25 knots. This preserves the physical slope changes at saturation. For a single-phase span below 0.001 K, its two endpoint knots suffice. States 1–4 and reported capacities do not use these chart interpolations. R-410A's bracketed flashes retain the entered cycle conditions up to the existing saturation limit, 0.2 K below its CoolProp critical temperature; density brackets select the stable branch and do not extrapolate the equation of state.
Diagram axes are fixed for each refrigerant over the admitted input envelope, including maximum superheat and minimum compressor efficiency, with space for state labels. Moving a pressure/temperature handle does not rescale the axes. Isolines stop at equation-of-state temperature limits. The compression path between its endpoints illustrates a constant overall efficiency; it is not a compressor chamber simulation. Circuit geometry and particle speed are schematic. Liao's Equation (15) assumes constant isentropic efficiency, matching this tool. It is an empirical fit within the paper's simulated range of −10…20 °C evaporation, 30…60 °C gas-cooler outlet and 71…120 bar heat rejection pressure; the study considered 0…20 K superheat, but Equation (15) neglects its explicit dependence. The numerical optimum retains the entered superheat and current CoolProp properties, so it can differ from that fit even under the same constant-efficiency assumption. Outside the comparison range, use the numerical sweep with its endpoint warning. Neither optimum includes a compressor pressure/temperature envelope.
The Copeland guide's R-134a COP/work/discharge row is inconsistent with its stated 0.6 compressor efficiency. Its check therefore verifies only saturation pressures and evaporator enthalpy difference at the published inputs; the conflicting COP and discharge targets are excluded and explained in that case's source. No cycle equation is adjusted. Refrigerant selection also requires equipment compatibility, safety-class, environmental and local regulatory assessment.
A state beyond the equation-of-state temperature range suppresses the results with an input error. Internal endpoint guards only prevent failed property calls while that error is shown; they do not extend the valid domain.
References
- Bell, I.H., Wronski, J., Quoilin, S. and Lemort, V. (2014). Pure and Pseudo-pure Fluid Thermophysical Property Evaluation and the Open-Source Thermophysical Property Library CoolProp. Industrial & Engineering Chemistry Research 53, 2498–2508. https://doi.org/10.1021/ie4033999
- ASHRAE Handbook—Fundamentals (2025), SI edition, Chapter 2: Thermodynamics and Refrigeration Cycles. https://handbook.ashrae.org/Handbooks/F25/SI/f25_Ch02/f25_Ch02_si.aspx
- Çengel, Y.A. and Boles, M.A., Thermodynamics: An Engineering Approach, Example 11–1, p. 613; numerical reference transcribed in the IIT Bombay FOSSEE textbook companion. https://tbc-python.fossee.in/convert-notebook/Thermodynamics_An_Engineering_Approach_by_Michael_A_Boles_Yunus_A_Cengel/Chapter11_1.ipynb
- Liao, S.M., Zhao, T.S. and Jakobsen, A. (2000). A correlation of optimal heat rejection pressures in transcritical carbon dioxide cycles. Applied Thermal Engineering 20(9), 831–841. https://doi.org/10.1016/S1359-4311(99)00070-8
- Copeland/Emerson, CO₂ Product Guide 2021, TGE175-EN-2102, p. 8, reference-cycle comparison. https://media.copeland.com/693d2049-68af-411e-8b72-b16b010424f7/TGE175-CO2-Product-Guide-EN-2102.pdf
- CoolProp 8 mixture definitions and models (R-454B predefined mixture). https://coolprop.org/fluid_properties/Mixtures
- Liao correlation reproduced in the KTH CO₂ cycle study, equation 4-1: 5.3 °C evaporation and 35 °C gas-cooler outlet give approximately 87 bar. https://www.diva-portal.org/smash/get/diva2:10578/FULLTEXT01.pdf
Inputs
| Input | Symbol | Unit | Default |
|---|---|---|---|
| Refrigerant | refrigerant | R-290 · propane | |
| Set saturation levels by | spec | Temperature | |
| Evaporating dew temperature | T_evap | °C | 0 |
| Condensing bubble temperature | T_cond | °C | 45 |
| Evaporating pressure · absolute | p_evap_in | bar | 4.75 |
| Condensing pressure · absolute | p_cond_in | bar | 15.35 |
| Gas-cooler pressure · absolute | p_gc | bar | 90 |
| Gas-cooler outlet temperature | T_gc | °C | 35 |
| Suction superheat Measured above the evaporating dew temperature. | SH | K | 5 |
| Liquid subcooling Measured below the condensing bubble temperature. | SC | K | 5 |
| Compressor isentropic efficiency | eta_is | 0.75 | |
| Refrigerant flow from | flow_spec | Displacement and speed | |
| Volumetric efficiency | eta_vol | 0.8 | |
| Displacement per revolution | V_disp | cm^3 | 10 |
| Compressor speed | speed_rpm | rpm | 3000 |
| Refrigerant mass flow | m_spec | kg/s | 0.05 |
| Cold reservoir temperature | T_source | °C | 10 |
| Hot reservoir temperature | T_sink | °C | 35 |
| Gas-cooler sweep points | N_sweep | 161 | |
| Diagram sampling points R-454B uses fewer solved single-phase points and smooth interpolation. | N_plot | 81 | |
| Refrigerant circulation | t | s | 1.5 |
| Blend profile points per phase Refines the liquid, two-phase and vapour temperature profiles separately. | N_hx | 13 |
Results
With the default inputs above, the calc gives:
| Result | Value |
|---|---|
| Cooling COP · shaft | 3.672 |
| Heating COP · shaft | 4.672 |
| Cooling capacity | 1.115 kW |
| Heating capacity | 1.419 kW |
| Compressor shaft power | 0.304 kW |
| Refrigerant mass flow | 0.004 kg/s |
| Discharge temperature | 62.882 °C |
| Compressor pressure ratio | 3.234 |
| Evaporating pressure · absolute | 4.745 bar |
| High-side pressure · absolute | 15.343 bar |
| Carnot cooling COP | 11.326 |
| Carnot heating COP | 12.326 |
| Cooling COP / Carnot COP | 0.324 |
It also shows: Optimum CO₂ gas-cooler pressure, bar; Optimum CO₂ cooling COP; Liao correlation pressure · CO₂, bar.
Checked against sources
Each case below was calculated by this calc and agrees with the value from its source (within the tolerance where one is shown).
Çengel and Boles · ideal R-134a cycle
Source: Çengel and Boles, Thermodynamics: An Engineering Approach, Example 11–1, p. 613, as independently transcribed in IIT Bombay FOSSEE Chapter11_1. R-134a at 0.14/0.8 MPa, saturated inlet/outlet, 0.05 kg/s. Checks enthalpy differences, avoiding reference-datum offsets. https://tbc-python.fossee.in/convert-notebook/Thermodynamics_An_Engineering_Approach_by_Michael_A_Boles_Yunus_A_Cengel/Chapter11_1.ipynb
Inputs: Refrigerant = R-134a; Set saturation levels by = Absolute pressure; Evaporating pressure · absolute = 1.4 bar; Condensing pressure · absolute = 8 bar; Suction superheat = 0 K; Liquid subcooling = 0 K; Compressor isentropic efficiency = 1; Refrigerant flow from = Specified mass flow; Cold reservoir temperature = 0 °C; Hot reservoir temperature = 25 °C; the others as above.
| Result | Source | This calc |
|---|---|---|
| Cooling COP · shaft | 3.97 ± 0.01 | 3.968 |
| Cooling capacity | 7.18 ± 0.02 kW | 7.184 kW |
| Compressor shaft power | 1.81 ± 0.02 kW | 1.811 kW |
Copeland · transcritical CO₂ reference cycle
Source: Copeland/Emerson, CO2 Product Guide 2021, TGE175-EN-2102, p. 8, reference cycle. CoolProp differs from the published property tabulation by about 1%: absolute tolerances are 0.05 COP, 0.8 K discharge temperature, 1.5 kJ/kg cooling effect and 1 kJ/kg work. The guide's 97 bar is an assumed reference condition, not this model's computed optimum.
Inputs: Refrigerant = R-744 · CO₂; Set saturation levels by = Absolute pressure; Evaporating pressure · absolute = 30.47 bar; Gas-cooler pressure · absolute = 97 bar; Gas-cooler outlet temperature = 38 °C; Suction superheat = 6 K; Hot reservoir temperature = 30 °C; the others as above.
| Result | Source | This calc |
|---|---|---|
| Cooling COP · shaft | 1.98 ± 0.05 | 1.998 |
| Discharge temperature | 103.4 ± 0.8 °C | 103 °C |
| Specific refrigeration effect | 135.6 ± 1.5 kJ/kg | 136.5 kJ/kg |
| Specific compressor shaft work | 68.53 ± 1 kJ/kg | 68.32 kJ/kg |
Copeland · R-134a pressures and cooling effect
Source: Copeland/Emerson CO2 Product Guide 2021, TGE175-EN-2102, p. 8. Checks only pressures and evaporator enthalpy difference at the stated −8/40 °C, 6 K superheat and 0 K subcooling; property tabulations differ by less than about 1%. The published COP 3.22 / work 43.98 kJ/kg / discharge 60.67 °C are inconsistent with the stated isentropic efficiency 0.6. Real-fluid isentropic work is about 33.2 kJ/kg; the published work implies efficiency about 0.755. This model keeps 0.6 and correctly predicts COP about 2.577; the conflicting COP and discharge numbers are not reference targets.
Inputs: Refrigerant = R-134a; Evaporating dew temperature = -8 °C; Condensing bubble temperature = 40 °C; Suction superheat = 6 K; Liquid subcooling = 0 K; Compressor isentropic efficiency = 0.6; the others as above.
| Result | Source | This calc |
|---|---|---|
| Evaporating pressure · absolute | 2.17 ± 0.02 bar | 2.169 bar |
| High-side pressure · absolute | 10.16 ± 0.02 bar | 10.166 bar |
| Specific refrigeration effect | 141.6 ± 1.5 kJ/kg | 142.6 kJ/kg |
Liao correlation · published 87 bar example
Source: KTH CO2 cycle study, Cycle Working Conditions, equation 4-1, printed p. 33: Liao's correlation gives 87 bar at 5.3 °C evaporation and 35 °C gas-cooler outlet. https://www.diva-portal.org/smash/get/diva2:10578/FULLTEXT01.pdf. This verifies the published algebraic correlation example. Liao, Zhao and Jakobsen (2000), Equation (15), https://doi.org/10.1016/S1359-4311(99)00070-8, assumes constant isentropic efficiency, matching this tool; it fits a limited operating range and omits explicit superheat dependence, whereas the numerical optimisation retains the entered superheat.
Inputs: Refrigerant = R-744 · CO₂; Evaporating dew temperature = 5.3 °C; Cold reservoir temperature = 15 °C; Hot reservoir temperature = 30 °C; the others as above.
| Result | Source | This calc |
|---|---|---|
| Liao correlation pressure · CO₂ | 87 bar | 86.997 bar |
Regression · entered 4.75/4.76 bar cycle
Source: Independent review supplied with this calc: R-290 at 4.75/4.76 bar, zero superheat/subcooling and ηis = 0.75 gives cooling COP 2914.38. The exact high-side pressure expectation additionally checks input preservation; a numerical guard must not change an accepted pressure.
Inputs: Set saturation levels by = Absolute pressure; Condensing pressure · absolute = 4.76 bar; Suction superheat = 0 K; Liquid subcooling = 0 K; Cold reservoir temperature = 0.05 °C; Hot reservoir temperature = 0.075 °C; the others as above.
| Result | Source | This calc |
|---|---|---|
| High-side pressure · absolute | 4.76 bar | 4.76 bar |
| Cooling COP · shaft | 2,914 | 2,914 |
Regression · 0.004 K reservoir lift
Source: Analytic Carnot limiting-case regression identified by the independent review: cold/hot reservoirs at 24.993/24.997 °C are 298.143/298.147 K, separated by 0.004 K. The reversible cooling COP is 74535.75 and heating minus cooling COP is one. This checks accepted input preservation and the exact thermodynamic identity.
Inputs: Cold reservoir temperature = 24.993 °C; Hot reservoir temperature = 24.997 °C; the others as above.
| Result | Source | This calc |
|---|---|---|
| Reservoir temperature lift | 0.004 K | 0.004 K |
| Carnot cooling COP | 74,536 | 74,536 |
| Carnot heating minus cooling COP | 1 | 1 |
R-410A · accepted 71 °C condensation
Source: Independent review reproduction: R-410A at 0 °C evaporation, 71 °C bubble-point condensation, 5 K superheat and 5 K subcooling. The outlet must be 66 °C by definition, and the computed constant-pressure temperature profile must reach both actual cycle endpoints. This regression exercises the full accepted input including the former CoolProp near-critical liquid p–h flash failure; it is not a separate measured-performance reference. CoolProp property model: Bell et al. (2014), https://doi.org/10.1021/ie4033999.
Inputs: Refrigerant = R-410A; Condensing bubble temperature = 71 °C; the others as above.
| Result | Source | This calc |
|---|---|---|
| Heat rejection outlet temperature | 66 °C | 66 °C |
| Heat rejection profile endpoint closure | 0 ± 1 × 10⁻⁶ K | 0 K |
R-410A · near-critical liquid branch
Source: Extension of the independent review's R-410A 71 °C accepted-input regression to 0.001 K subcooling. The 70.999 °C liquid outlet follows exactly from the bubble temperature minus subcooling. This specifically exercises the density-based liquid flash just outside saturation, where CoolProp's p–T liquid inversion can also fail. Bell et al. (2014), https://doi.org/10.1021/ie4033999.
Inputs: Refrigerant = R-410A; Condensing bubble temperature = 71 °C; Liquid subcooling = 0.001 K; the others as above.
| Result | Source | This calc |
|---|---|---|
| Heat rejection outlet temperature | 70.999 °C | 70.999 °C |
| Heat rejection profile endpoint closure | 0 ± 1 × 10⁻⁶ K | 1.137 × 10⁻¹³ K |
R-410A · 71.14 °C repeatable liquid and vapour branches
Source: Independent review reproduction: the accepted R-410A 71.14 °C bubble-point input previously failed in its isentropic or liquid flash. With 5 K subcooling the outlet is 66.14 °C by definition. The heat-rejection temperature profile must close at both directly solved cycle states. This is a numerical regression, not a measured COP reference. CoolProp EOS: Bell et al. (2014), https://doi.org/10.1021/ie4033999.
Inputs: Refrigerant = R-410A; Condensing bubble temperature = 71.14 °C; the others as above.
| Result | Source | This calc |
|---|---|---|
| Heat rejection outlet temperature | 66.14 °C | 66.14 °C |
| Heat rejection profile endpoint closure | 0 ± 1 × 10⁻⁶ K | 5.684 × 10⁻¹⁴ K |
R-410A · 48.8 bar pressure input
Source: Independent review reproduction: the accepted 48.8 bar absolute R-410A pressure input previously failed before producing a cycle. Check preservation of the entered pressure and heat-exchanger endpoint closure through the real-fluid calculation; no performance value is inferred from this regression. CoolProp EOS: Bell et al. (2014), https://doi.org/10.1021/ie4033999.
Inputs: Refrigerant = R-410A; Set saturation levels by = Absolute pressure; Condensing pressure · absolute = 48.8 bar; the others as above.
| Result | Source | This calc |
|---|---|---|
| High-side pressure · absolute | 48.8 bar | 48.8 bar |
| Heat rejection profile endpoint closure | 0 ± 1 × 10⁻⁶ K | 5.684 × 10⁻¹⁴ K |

Responses