In almost all the plants we visit, there is a compressor room. And in almost all of them, that room belongs to no one. It is not for production, because it does not manufacture anything. It is not for maintenance, because it works. It is not for energy, because it does not appear as a line item on the bill: compressed air is not bought, it is made on-site with electricity, and that is why it does not appear on any line of the supply contract.
This absence is what makes it interesting. An energy vector that is not billed separately is a vector that is not measured, and what is not measured is not managed. Before talking about electrifying processes, heat pumps, or new contracted power, it is worth closing the balance of an installation that is already there, that consumes electricity every hour of production and often does not even have its own meter.
Compressed air is not air: it is transformed electrical energy
When an engineer looks at a compressor, they should not see a machine that makes air, but a converter that transforms electrical energy into pneumatic energy with a modest overall efficiency, and that dissipates the rest in the form of heat. Everything that enters through the electrical panel ends up, sooner or later, as heat in the room or in the cooling water. This is the reading that changes decisions: every improvement in the system translates directly into electrical kWh, and every kW that is not utilised is available heat that is being wasted.
The second correction is boundary. What matters is not the efficiency of the compressor, but the efficiency of the system: generation, treatment, storage, distribution and points of use. An excellent compressor feeding a network with load losses is a bad system. The ISO 11011:2013 makes it explicit: the assessment must be made on the three subsystems —supply, distribution and demand— and not on the isolated equipment.
Four numbers before touching anything
An honest diagnosis always starts from the same place, and it is not the catalogue. They are four magnitudes that must be measured, not estimated:
The actual flow and its profile. Not the specification plate or the installed nominal flow, but the flow that the plant consumes over a representative production cycle, with its variability. The difference between installed power and actual demand is, almost always, the first discovery.
The set pressure and the pressure needed by the last consumer. They are two distinct things, and the distance between them is usually explained by load losses and treatment, not by process need. When the general setpoint is raised to solve a local problem, the entire plant pays the price.
The hours. A compressed air system often operates many more hours than the production that justifies it. Weekend hours, shift change hours, and technical downtime are what turn a small loss into a constant bill: a leak knows no working calendar.
The required quality. ISO 8573-1:2010 classifies the content of particles, water and oil into purity classes. Specifying for the entire plant the quality that only a specific line needs is an expensive and frequent mistake: the treatment has an energy cost and is paid for at all operating hours.
The recurring errors
After many diagnoses, the pattern repeats with surprising regularity.
Leak is confused with failure. Leaks are managed as a living indicator, with periodic detection, recording and closing of work orders. A leak programme without an indicator is an activity, not a measure of efficiency.
Compressed air is used for jobs that do not need it. Blowing, drying, cooling, moving material or cleaning with compressed air is using the most expensive vector of the plant for a task that is often solved by a fan or a low-pressure blower.
It is sized and controlled for the peak. A battery of compressors without a joint control strategy ends up having machines running empty for hours and hours. Cascade control and sequencing are engineering decisions, not default settings.
Heat is forgotten. The heat of compression is the only residual energy from a plant that appears at useful temperature, continuously and at a specific point. If there is demand for hot water or preheating in the same plant, not recovering it is a decision that should be justified in writing.
What the standard is already moving
There is a regulatory framework that is quietly advancing beneath this installation. The Regulation (EU) 2019/1781 sets ecodesign requirements for electric motors and variable speed drives, and the motor that drives a compressor is fully included. The European Commission is currently reviewing these requirements to extend them to material efficiency and circular economy.
The message for industrial management is simple: the replacement of a motor will cease to be an exclusively maintenance decision and will start to have product-related conditions. Anticipating it means arriving with the balance done.
In parallel, those who maintain an energy management system according to ISO 50001 have here one of the cleanest energy performance indicators that exist: electricity consumed per unit of free air volume, with a well-defined baseline and specified normalisation variables. An EnPI that can be measured, verified and defended before a third party.
Requirement, advantage
Compressed air is not a spectacular topic. It does not make headlines in the energy transition nor does it open a board meeting. But it is one of the few interventions that simultaneously improves electricity consumption, process reliability, available capacity at the connection, and the quality of the energy data of the plant.
33 years of observing plants have left us with an uncomfortable conviction: before contracting more power or budgeting for new equipment, it is worth knowing what the one you already have is doing. The first source of energy remains the one that does not need to be produced.
If you are preparing your electrification plan, start with the room that no one considers theirs.