
Key points at a glance
- The control system determines how long a machine operates under load, at no-load, or at a standstill. Part-load operation is generally less efficient than full-load operation – so the control strategy affects consumption, not just comfort.
- A narrower pressure band does not automatically lower the average system pressure. The position of both pressure limits and the actual pressure profile are decisive.
- Load changes, motor starts, and speed changes are three distinct processes with their own limits. The unload pressure is not necessarily the motor’s shutdown pressure.
- The usable buffer results from volume, usable pressure range, consumption, and simultaneous re-delivery – the receiver volume alone says little.
- Multiple machines without coordination can run unnecessarily at no-load, cycle frequently, or operate against excessively high pressure. However, staggered pressure bands alone are not necessarily a fault.
- Before any setpoint change: demand at the consumer, measured pressure, operating states, load changes and motor starts separately, demand profile, and the limits of the entire system.
In many stations, the pressure setpoint has evolved historically. At some point, a consumer caused problems, so the pressure was increased. Later, a second machine was added, which has mostly stood idle since. Both decisions were understandable at the time – together, they result in a station whose consumption can no longer be explained. Whether a variable speed drive changes anything is a separate question and answered elsewhere.
The control system is the point where changes can be made without replacing machinery. It determines at what pressure the system loads and unloads, how long a machine runs on at no-load, when the motor shuts down, and in what sequence multiple machines engage. The available versions of our industrial compressors are shown in the product overview.
This article explains the terms that are regularly confused, shows the limits of readjustment, and identifies the data that should be available before any change.
What the Control System Influences – and What It Doesn’t
Four things are often mixed up in everyday practice: the compressor type, the motor concept, the drive type, and the control strategy. The control system belongs to the last group. It does not determine how efficient a machine is at its optimum point – but it does determine at which operating point and against what pressure the machine operates, and both affect specific consumption. As a rule of thumb: compressors generally operate less efficiently in part-load than under full-load. Therefore, the question is not “which control system is best”, but “which operating points arise with our demand profile”.
For a fixed-speed machine, the control system essentially knows three states. Under load, it compresses. At no-load, the drive continues to run, but hardly any air is delivered. At a standstill, the drive is off. A fixed-speed SMART operates according to this pattern; a variable speed drive BASE VSD instead adjusts the flow rate within its control range. Which series belongs to which category can be found in the product overview – the specific control functions a machine has are detailed in its documentation.
The Pressure Band: Which Limit Shifts
The pressure band is the range between the pressure at which the machine loads and the pressure at which it unloads. It is a setting of your system, not a product characteristic – the machine’s pressure class is something different.
It is a common assumption that a narrower band automatically lowers the average system pressure. This is not strictly true. What matters is which limit moves. Simplified by calculating the average of both limits: 7 to 8 bar has an average of 7.5 bar. If narrowed to 7.5 to 8 bar, this average rises to 7.75 bar. If, however, it is narrowed to 7 to 7.5 bar, it drops to 7.25 bar. The average of the limits is merely a calculated value; the actual time-weighted system pressure depends on how long the system operates in which range, and thus on the operating profile. A narrower band does not automatically lead to a lower pressure or a specific saving.
How Long the Storage Lasts
The amount of air available between two pressure limits can be estimated. A calculation example as a guideline, explicitly not a design specification:
- Assumptions: Receiver volume 1 m³; pressure drop from 8.0 to 7.0 bar gauge pressure, i.e., a difference of 1.0 bar; reference pressure 1.0 bar absolute; same temperature or reference state before and after consumption; no simultaneous re-delivery; constant consumption of 2 m³/min, referred to the same state.
- Calculation: V ≈ Receiver volume × Pressure difference ÷ Reference pressure = 1 m³ × 1.0 bar ÷ 1.0 bar ≈ 1 m³.
- Result: 1 m³ ÷ 2 m³/min = 0.5 minutes, i.e., approximately 30 seconds, until the receiver is emptied from 8 to 7 bar.
- Regarding units: A pressure difference is the same number in gauge pressure and absolute pressure. The reference pressure, however, is an absolute pressure. If the temperatures of the two states differ, a temperature ratio is added.
- What the value is not: not a complete cycle time and not a measure of the truly usable reserve. Pipework volume, simultaneous re-delivery, thermal effects, and pressure losses are not included. In which direction and how strongly they shift the value only becomes clear from the data of the specific system.
The crucial relationship: If the pressure difference is halved, the amount of air stored between the limits is also halved. A narrower band therefore often means more frequent operations – what exactly these operations are is the next point.
Load Changes, Motor Starts, Speed
“Switching cycles per hour” is a collective term that mixes three things:
- Load changes: The machine switches between load and no-load. The drive continues to run.
- Motor starts: The drive is switched off and restarted. This often happens only after a set run-on time at no-load – the unload pressure is therefore not necessarily the shutdown pressure.
- Speed change: For controlled machines, the flow rate changes without a load change in the sense described above.
Different limits apply to load changes and motor starts, and both are found in the documentation of the specific machine, not in a general rule of thumb. Anyone who narrows the pressure band without knowing which of the three processes will become more frequent shifts the problem from energy costs to wear.
The second lever is the run-on time at no-load. Shortening it reduces the no-load proportion and increases the number of motor starts. Here too, the sensible setting results from the measured demand, not from an adopted value. A high no-load proportion is, by the way, an indication and not yet a diagnosis: It can stem from the machine size, from the control settings, from a lack of coordination of multiple machines, or simply from the demand profile. A measured load profile clarifies the cause.
One Machine, Multiple Machines
As long as demand is reasonably constant and matches the flow rate, a single control system copes well. If it fluctuates significantly, either many load changes or a lot of no-load operation will occur.
With multiple machines, coordination becomes an additional factor. Without it, several patterns are possible, all of which cost energy: machines run unnecessarily at no-load, they cycle frequently, or the station operates against a higher pressure than necessary. Two staggered pressure bands are not automatically a fault – a cleanly stepped cascade is a legitimate, simple form of distribution. It only reaches its limits sooner than a higher-level control system that actively coordinates states and sequences.
A common and well-justified distribution: The base load is covered by as many machines as necessary at full load, with an additional machine compensating for the remainder in part-load operation. This means the majority of the station operates at the more favourable operating point. How many machines this entails in individual cases follows from the demand profile – and this is measured, not estimated; the compressed air demand can be determined step by step for this purpose.
Separate from this is the rotation of the lead machine. This distributes operating hours and thus affects wear and maintenance planning. Whether it changes energy consumption depends on which machines then run at which operating points – a saving is not guaranteed, but neither is it ruled out. And it does not ensure redundancy: whether demand remains covered in the event of a machine failure is a question of the installed flow rate. An application report on parallel operation shows how weekly rotation works in practice.
What Should Be Available Before a Setpoint Change
Without these values, any change is an experiment with an uncertain outcome. With them, it can be assessed afterwards whether the change has had an effect.
| Parameter | How to record | Why it changes the decision |
|---|---|---|
| Required minimum pressure at the application | From the consumer documentation, not from a measurement. The critical consumer dictates the value. | This is the requirement. A measured reading shows what is currently arriving – not what is needed. |
| Actual pressure under demand | Measurement at the critical consumer during consumption that counts. | The difference from the requirement shows how much leeway is truly available. |
| Current pressure settings of the control system | Read and note from the control system: unload and load pressure, if applicable, separate values for motor shutdown. | Without the initial value, it will not be possible to assess later whether a change has had an effect. |
| Operating states over time | Load, no-load, and standstill hours from the control system, over a period that reflects the real patterns. | A high no-load proportion is an indication, not proof. It can stem from machine size, control settings, coordination of multiple machines, or the demand profile. |
| Load changes and motor starts, counted separately | Read both counters separately or log them over several shifts. If only one value is available: note which one. | Different limits apply to load changes and motor starts. Anyone who combines both into a single figure cannot check either. |
| Demand and load profile | Flow rate measurement over shifts, production-free periods, and special runs. | Shows whether variable speed drive control, more storage, or coordination of multiple machines addresses the actual problem. |
| Storage, system volume, and usable pressure range | Receiver volume from the nameplate, pipework volume from length and diameter, plus the pressure limits within which the storage is actually allowed to operate. | How long a peak is covered results from volume, usable pressure range, consumption, and simultaneous re-delivery – not from the volume alone. |
| Limits of the entire system | Operating manuals and markings of compressor, receiver, air treatment, and network. Note permissible operating pressure and maximum permissible pressure separately. | The setpoint range results from the weakest link in the chain, not from a single nameplate. |
The pressure requirement comes from the application, not the machine. How a justified operating pressure can be derived from this is described elsewhere with a calculation example. And the difference between the requirement and the measured value is not generally “reserve”: part of it covers real pressure losses, another part can be avoidable overpressure. Distinguishing these two components is the real benefit of measurement.
Frequently Asked Questions
Does a Narrower Pressure Band Bring Savings?
Savings are possible but cannot be derived solely from the bandwidth. A lower average system pressure can reduce energy consumption; additional load changes, no-load times, and motor starts can diminish the advantage. The comparison in real operation is decisive. The permissible limits of the system must be observed.
Is Coordination Worthwhile Even with Two Machines?
It can be worthwhile as soon as both regularly operate on the network simultaneously. For a true standby machine that only runs in case of failure, the benefit is small. The measured load profile, not the number of machines, is decisive.
Does Rotation Replace Planned Redundancy?
No. Rotation distributes operating hours. Whether demand remains covered in the event of a machine failure is determined by the installed flow rate and its distribution.
How Long Should Measurements Be Taken?
Long enough for the period to include the real operating patterns: all shifts, production-free periods, start-up processes, and, where relevant, seasonal differences. A single week is only sufficient if it accurately reflects these.
Have Your Control Concept Reviewed
Whether a station would run better through pressure settings, the distribution of multiple machines, or a different drive concept is decided by the operational data. For an initial assessment, four pieces of information are sufficient: the current pressure settings, the receiver volume, load and no-load hours from the control system, and the minimum pressure of the critical consumer.