An undersized compressor starves the plant; an oversized one wastes money twice — at purchase and every month on the power bill. Sizing an industrial air compressor is not guesswork, and it is not nameplate reading either. It comes down to four numbers you can measure or estimate, then a reality check against how the plant actually runs. Here is the method we use, and the checks that keep a sizing exercise honest.
First: measure demand, not nameplate
The number that matters is FAD — free air delivery: the actual volume of air the compressor can supply at working pressure, corrected back to ambient conditions. Motor kW alone does not tell you what comes out of the pipe; two machines with the same motor power can deliver very different airflow depending on design, stage count, and condition. Use real consumption of the equipment connected to the ring: tool and actuator consumption datasheets, flow meter readings if you have them, and duty cycles — a machine that consumes 2 m³/min continuously is not the same as one that draws the same amount for ten seconds per minute.
The four numbers that define the duty point
- 1. Total air demand (FAD). Add up the consumption of everything that runs simultaneously. Summing nameplates of all consumers overestimates the real peak; add measured or estimated duty factors instead. If a flow meter is installed, a week of logged data beats any estimate.
- 2. Working pressure. The highest pressure any significant consumer genuinely requires, plus the pressure drop between the compressor room and that consumer. Setting the whole plant pressure because of one distant machine is expensive — as a rule of thumb, every extra bar of pressure raises compressor energy consumption by roughly 7% (≈1% per 0.1 bar), so ring pressure is a number worth defending.
- 3. Load profile. Steady demand around the clock behaves very differently from a shift pattern with sharp peaks. A plant with a flat load suits a fixed-speed machine; one with wide swings is where a variable-speed drive (VSD) earns its keep, because it follows demand instead of loading and unloading.
- 4. Site conditions. Ambient temperature, dust load, and cooling arrangements derate real output. A compressor sized for a 25°C compressor room will not deliver its table FAD in a 40°C workshop — check the manufacturer’s correction data for the actual site, not the catalogue’s reference conditions.
From duty point to machine: a worked framing
With the four numbers written down, selection becomes a table lookup. Example framing (illustrative, not a substitute for a model table): a plant with a measured demand of 2.5 m³/min at 8 bar, single-shift duty, in a moderate climate lands in small-station territory — think integrated stations in the 3.7–30 kW class. A plant pulling 20 m³/min at 7–8 bar around the clock is in a different world entirely — project-class station territory where the station scope is confirmed per project. The same four numbers that put you in one class or the other also tell the supplier enough to quote the right row.
SIND’s published ranges span that whole ladder: screw compressor families from 7.5 kW (SILVERLUBE) up to 580 kW (PureFlow SP) and beyond, integrated stations in horizontal (SN-H, 2.2–30 kW) and vertical (SN-V2, 3.7–30 kW) formats, project station classes such as SE-020/SE-050 defined by nominal capacity, and specialized low-pressure or oil-free machines when the process calls for them. The selection conversation starts with your duty point, not with a model number.
Five mistakes that turn sizing into regret
- Sizing for growth twice. Adding margin once (20–25%) is sensible planning; stacking margin on top of margin produces a machine that idles. If growth is real, plan a modular path — a second machine that talks to the first — rather than one giant compressor running unloaded.
- Ignoring leaks. In a plant with poor maintenance discipline, leaks can quietly consume around a fifth of everything the compressor produces — some audits put unmaintained systems near 30%. Sizing that ignores the leak budget simply ships compressed air to the atmosphere.
- Confusing pressure with quality. Raising ring pressure to “push more air” wastes energy and does not solve flow starvation — the fix for starving consumers is supply capacity or pipe diameter, not pressure.
- Forgetting the rest of the station. The compressor is one component. Drying (refrigerated or desiccant), filtration, and storage all need to be selected against the same duty point, and each has its own sizing logic.
- Skipping the electrical check. Especially at smaller ratings, supply voltage and available current gate the choice — an integrated 2.2–30 kW station offered at 220/380 V needs a supply that matches before anything else is decided.
Quick sizing sanity-check table
| If your plant looks like… | Typical class to start from | Why |
|---|---|---|
| Workshop, 1–3 m³/min at 8 bar, single shift | Small integrated station, ~3.7–7.5 kW | Covers demand with modest margin; compact footprint |
| Light production, 3–4 m³/min at 8–10 bar | Integrated station, ~11–30 kW, VSD if load swings | VSD follows peaks without unloading losses |
| Continuous process, 15–25 m³/min at 7–8 bar | Project station class (e.g. SE-020/SE-050) | Scope confirmed against site data, not catalogue rows |
| Low-pressure process air at scale | Low-pressure two-stage or oil-free low-pressure machines | Pressure-matched designs avoid throttling 8-bar air down |
Treat the table as orientation: the machine you buy comes from the current model table, checked against your measured duty point and site conditions.
Frequently asked questions
Does higher kW always mean more air?
No. Two-stage and low-pressure designs convert power to airflow differently than single-stage machines. Always compare FAD at your pressure, not motor nameplates.
How much margin should I add?
A modest, single margin of roughly 20–25% over measured demand covers meter error and normal growth. If you expect step-change expansion, plan for an additional machine rather than oversizing one.
Can I just raise pressure instead of buying more capacity?
Raising pressure increases stored and delivered air slightly, but it costs roughly 7% more energy per bar and stresses everything downstream. It is not a substitute for adequate supply capacity.
