EN 12900 Explained: How a Compressor Selector Calculates Performance Data

For pumps, a performance curve shows which head is reached at which flow (see Performance Curves Need a Home and How to Calculate the System Curve). For refrigerant compressors, that role is played by a polynomial with ten coefficients. It is standardised in EN 12900 (Europe) and AHRI 540 (USA) and is the basis of almost every compressor selection — from the datasheet to the selector on the web.

This article shows how the polynomial is built, what it does and does not tell you, and which calculation steps a selector needs beyond it.

The 10-coefficient polynomial

Every performance value of a compressor is given as a function of two temperatures: the evaporating temperature t₀ and the condensing temperature t_c, both in °C. The third-degree polynomial reads:

For each compressor and each refrigerant, the manufacturer publishes one set of coefficients per quantity:

The advantage: ten numbers are enough to calculate any operating point within the envelope, without interpolating tables. And because the format is standardised, a selector can read the data across manufacturers. The coefficient of performance follows directly: COP = Q₀ / P.

What the polynomial assumes: the rating conditions

The coefficients only apply to the rating conditions under which the manufacturer measured. Two values are decisive: the suction gas temperature or superheat at the compressor inlet, and the subcooling of the liquid refrigerant before the expansion valve.

Under EN 12900, typical rating values are a suction gas temperature of 20 °C and a subcooling of 0 K. Many manufacturers additionally publish data for 10 K superheat. AHRI 540 works with US units (°F, Btu/h, lb/h) and its own rating conditions.

Two datasheets with the same operating point are only comparable if the rating conditions are the same as well. A selector must therefore store and display them.

Zeotropic blends: dew point or mean temperature?

For refrigerant blends with temperature glide such as R407C, R449A, R454C or R455A, the temperature is not constant during evaporation and condensation. EN 12900 refers t₀ and t_c to the dew point. Some planners calculate with the mean temperature — with a glide of 5 to 7 K, that can lead to clearly different results. The selector should state which definition applies and convert where needed.

Converting to actual operating conditions

In a real system, superheat and subcooling almost always differ from the rating conditions. The usual correction runs via mass flow and enthalpies:

• 1 · Correct the mass flow via suction gas density. The compressor delivers an almost constant volume. If the suction gas temperature changes, so does the density — and with it the mass flow: ṁ_real = ṁ_rating · ρ_suction,real / ρ_suction,rating
• 2 · Calculate the cooling capacity from the enthalpy difference: Q₀,real = ṁ_real · (h_evaporator outlet − h_evaporator inlet). The inlet enthalpy follows from the liquid temperature before the expansion valve, i.e. from the subcooling. More subcooling means more cooling capacity at the same mass flow.
• 3 · Power input changes only slightly. It is usually taken from the polynomial or adjusted slightly with the mass-flow ratio — the method depends on the manufacturer.

Steps 1 and 2 require refrigerant property data: density and enthalpy as functions of pressure and temperature. In practice, property libraries such as CoolProp (open source) or NIST REFPROP are used.

Checking the operating envelope

The polynomial also produces numbers outside the permitted range — and those are worthless. Every compressor therefore has an operating envelope: a polygon in the t₀–t_c diagram, depending on the refrigerant and often on speed, superheat or additional cooling.

For every operating point, a selector has to check whether it lies inside the polygon. Technically that is a simple point-in-polygon test. What matters is what happens next:

• Point outside (C): the compressor is not offered, or clearly marked.
• Point close to the limit (B): a warning that there is no margin for load changes or higher ambient temperatures.
• Additional conditions: some areas are only permitted with additional cooling, liquid injection or a minimum speed.

Showing the polygon with the operating point drawn in helps planners a lot to follow the selection.

Variable-speed compressors

For compressors with a frequency inverter there is not one polynomial but one set of coefficients per speed or frequency step — for example 30, 50 and 70 Hz. Between the steps, values are usually interpolated linearly, and the envelope can change with speed as well.

For the design, that means two additional questions: which speed is needed for the required cooling capacity — and how does the COP behave in part load, which in practice accounts for most of the operating time?

Special case: transcritical CO₂ (R744)

Above the critical point of CO₂ (about 31 °C) there is no condensation. The condensing temperature t_c is no longer available as an input. Performance data for transcritical operation therefore refer to the high pressure and the gas cooler outlet temperature, and manufacturers publish their own data formats for it.

In addition, every gas cooler outlet temperature has an optimum high pressure at which the COP is at its maximum. A selector for CO₂ has to calculate or specify this pressure — which clearly sets it apart from a selector for conventional refrigerants.

How a compressor selector works

In summary, a compressor selection runs through six steps:

1. InputRequired cooling capacity, refrigerant, t₀, t_c, superheat, subcooling, voltage and frequency
2. CalculationEvaluate the polynomial for every candidate compressor
3. CorrectionConvert from rating to operating conditions
4. CheckOperating envelope — for variable-speed compressors per speed
5. SelectionCompressors with sufficient capacity, sorted by COP, size or price
6. OutputDatasheet with operating point, performance data, envelope diagram and rating conditions

New refrigerants or new series mean one thing above all: new coefficients and new polygons. Why so many series are being converted right now is explained in Refrigerants in Transition: F-Gas Regulation, PFAS and What It Means for Design. If this data is maintained centrally, a conversion is possible without a new software release (see Product Rules Belong in Data and Roll Back One Series, Not the Whole System).

The digirom configurator and selector also works for compressors and refrigeration units.

Frequently asked questions