Compressor Sizing & Selection Guide (kW, BTUH, Displacement)

Sizing Starts With the Load, Not the Compressor

A compressor is not sized from the room, the pipe size, or the compressor that was there before. It is sized from the load — the heat the system has to remove at design conditions — and then matched to the evaporating and condensing temperatures that the rest of the plant will actually produce. Change either one and the required compressor changes with it.

Build the load first. For a cold room or a process chiller, that means adding every term, not just the obvious one:

• Product load — sensible heat above freezing, latent heat of freezing, sensible heat below freezing, plus any heat of respiration for produce.
• Transmission load — panel or wall U×A×ΔT, driven by the design ambient, not today's weather.
• Infiltration and door load — air changes per hour from door openings, plus defrost and door-heater allowances.
• Internal loads — evaporator fans, lights, forklifts, people, pumps and any motor inside the envelope.
• Safety margin — typically 10–15%, applied to the total, and stated explicitly so the next engineer knows it is there.

The Displacement Formula (Swept Volume × Speed)

Once the load is known, the compressor's job is to move enough refrigerant vapour to carry that heat away. The chain is three steps long, and every selection problem is a failure at one of them.

Reading it in words: the load fixes the mass flow, the suction specific volume converts mass flow into volume flow, and volumetric efficiency — the fraction of the swept volume that is actually delivered as useful vapour after clearance, valve and re-expansion losses — converts volume flow into the hardware you have to buy. Volumetric efficiency is not a constant. It falls as the pressure ratio rises: a compressor that delivers 0.88 at a 2.5:1 ratio may deliver 0.62 at 6:1.

Two numbers in those formulas move more than everything else: v₁ (suction specific volume) and ηv. Both depend on the evaporating temperature and the pressure ratio, which is why you re-check the displacement at the worst condition the plant will see — pull-down, high ambient, or a fouled condenser — and not only at the design point. The compressor selection calculator evaluates exactly these three expressions from CoolProp state data, so you can see how the required displacement grows as the suction pressure falls.

kW, BTU/h and Tons: Getting the Units Right

Most sizing errors we see are not engineering errors, they are unit errors. Cooling capacity in this trade travels in three units, and they are all in daily use: kW (metric design work, Europe and most equipment datasheets), BTU/h (North American equipment, and most residential nameplates), and tons of refrigeration (North American load calculations and legacy drawings).

And do not mix up cooling capacity with input power. A compressor with 7.7 kW of motor input might deliver 30 kW of cooling at one condition and 21 kW at another; the nameplate electrical data tells you what the motor draws, never what the system removes.

Why One Compressor Has Many Capacities

A published compressor capacity is a point, not a property. The number in the catalogue is valid at the standard rating conditions — for EN 12900 / ARI 540 that is 7.2 °C evaporating, 54.4 °C condensing, 8.3 K subcooling and 18.3 °C suction return. Move away from those conditions and the same hardware delivers materially more or less. As a rule of thumb for a scroll or reciprocating compressor in the usual range:

Because capacity moves with conditions, a correct selection usually ends with a small table rather than a single number: capacity and required input at design, at pull-down, and at the maximum condensing temperature the plant can see on the hottest day. Add the superheat and subcooling targets from our charging guide to the same sheet and the whole plant is described in one page.

Worked Example: 30 kW (102,000 BTU/h) R-410A System

Take a small industrial chiller: 30 kW of required cooling at 5 °C evaporating and 45 °C condensing, R-410A, 6 K of superheat at the evaporator outlet and 5 K of subcooling at the condenser outlet. Find the mass flow, the required displacement, and a realistic compressor input.

1. Mass flow. The refrigerating effect Δh₀ is the enthalpy difference across the evaporator: suction vapour with 6 K of superheat is about 428.5 kJ/kg, and the liquid arriving at the expansion valve after 5 K of subcooling is about 274.9 kJ/kg, so Δh₀ ≈ 153.6 kJ/kg and

2. Volume flow. Saturated R-410A vapour at 5 °C occupies about 0.0280 m³/kg, so

3. Geometric displacement. At this pressure ratio a scroll of this size runs around η_v = 0.85, so the swept volume you need is

4. Input power and check. With an isentropic efficiency of about 0.70, compression from 428.5 kJ/kg to roughly 45 °C condensing costs about 31.5 kJ/kg of ideal work, which is 0.195 × 31.5 ≈ 6.2 kW ideal and about 8.8 kW at the shaft. That gives a cycle COP of 30 / 8.8 ≈ 3.4. The Carnot ceiling at 5 °C / 45 °C is 278.15 / 40 = 6.95, so this machine sits at about 49% of Carnot — a normal figure for a small packaged chiller, and a useful sanity check. If your selection claims 6.0 COP at these temperatures, the state points are wrong somewhere; the COP guide shows how to reconcile the two.

Selection Procedure, Step by Step

1. Fix the design conditions in writing. Load in kW (and its BTU/h and ton equivalent), evaporating temperature, condensing temperature, subcooling, superheat, refrigerant, and the design ambient that produces them. These are decisions, not outputs.
2. Convert the load into an enthalpy duty. Pull the four cycle state points for the chosen refrigerant, take Δh₀, and get the mass flow — then the volume flow from the suction specific volume. If the tool and the hand calculation disagree, the state points differ, not the physics.
3. Add the worst case. Repeat at pull-down and at maximum condensing temperature. The compressor must start and stay inside its envelope at the hardest point, not only at the design point.
4. Match against real datasheets. Swept volume per revolution and speed give geometric displacement; compare with the OEM performance table at your actual Te/Tc, not at standard rating conditions. Check motor input, current at maximum load, and the operating envelope limits (discharge temperature, pressure ratio, minimum suction superheat).
5. Record the result as a selection sheet. Capacity, input power, COP, volumetric efficiency assumed, displacement required vs. selected, and the correcting factors you applied. That sheet is what makes the next service call, the next refrigerant change, or the next efficiency upgrade cheap instead of mysterious.

Sizing Mistakes and How They Show Up in the Field

These are the failures that survive commissioning and show up as warranty calls. Each one has a signature you can measure, which is the point: sizing mistakes are visible in the cycle, not just on the invoice.

Every one of these is cheap to catch before purchase and expensive after commissioning. Two habits prevent most of them: keep the design conditions and load assumptions on one page, and re-compute the required displacement at the worst condition the plant will meet. If the cycle ever looks impossible for the hardware — COP above Carnot, or a suction volume the compressor cannot move — the state points are wrong, and the modern alternatives to legacy desktop tools will show you which one.

Start from the load, convert honestly, correct for real conditions, and verify the compressor against its own envelope. That is the entire discipline — the arithmetic is just bookkeeping once the assumptions are on paper.