Air demand profiling for compressed-air systems

What is an air demand profile?

A demand profile is a time record of delivered airflow, usually logged alongside system pressure and compressor running state, over a period that represents normal operation. Plotted against the clock, it shows a plant’s daily rhythm: the shift start surge, steady production load, lunch-break dip, weekend or night base load and short spikes from large intermittent users.

Installed compressor capacity tells you what the station can deliver. Adding up tool ratings tells you what the equipment could consume if everything ran at once. Neither tells you what the plant really uses. Most sites that are profiled for the first time find that average demand is well below installed capacity, that a large share of night-time flow is leakage and that the worst pressure dips come from a handful of short events rather than a general shortage of air.

Which decisions does a demand profile support?

Start by writing down the question the profile must answer, because it decides where you measure, how long and how finely. One good dataset can serve several decisions, but each conclusion must stay within the period, boundary and resolution actually measured.

Where should demand be measured?

Mark the measurement point on a current piping diagram before installing anything. A meter at compressor discharge, one after the dryer, one at a header entering a production bay and one at a single machine all describe different things, and all can be right.

• Station discharge (before treatment): captures everything the compressors produce, including dryer purge, filter drains and receiver losses.
• After treatment: shows the air actually sent to the plant. The difference from discharge flow is treatment consumption, which matters for heatless desiccant dryers.
• Zone or branch entry: separates one department’s demand. Check for cross-connections and ring mains fed from two directions; if only one path is metered, the zone figure is incomplete.
• Single large user: needed when one blow-off, blast room or packaging line dominates the peaks.

A precise reading at an incomplete boundary is often less useful than a slightly less precise measurement at a correctly defined boundary. Record temporary hoses, bypass valves and any second compressor that feeds downstream of the meter.

How flow is expressed: FAD, normal and actual

Flow meters on compressed-air lines usually report either actual volume at line conditions, or a normalised volume referenced to a stated pressure and temperature (often called Nm³/min). Compressor capacity is normally quoted as free air delivery (FAD) at the inlet conditions defined by the test standard. These are not the same quantity, even when all three are displayed as m³/min or cfm.

Before comparing measured demand with available capacity, confirm the meter’s configured reference pressure and temperature and the basis of the compressor data. Use absolute pressure and absolute temperature for any gas-state conversion. If the basis cannot be established, report the figures separately and say what needs clarifying rather than applying an unexplained correction factor. Our airflow and FAD reference page covers the conversions in more detail.

Choosing and installing the flow meter

Thermal-mass, vortex, differential-pressure and ultrasonic meters are all used on compressed air. The right choice depends on pipe size, expected flow range, the lowest flow you need to see, pressure, temperature, moisture and oil content, and how much straight pipe is available.

• Turndown: a meter sized for the full peak may lose accuracy at night-time leakage flow, while one sized for normal flow may saturate during peaks. Flag over-range readings in the data rather than letting them clip silently.
• Straight runs: bends, valves, reducers and tees upstream disturb the flow profile. Follow the meter’s installation instructions and record how the chosen location meets them.
• Air condition: wet or oily air upstream of the dryer can foul some sensors. Measure after treatment where the decision allows it.
• Hot-tap and insertion meters: allow installation without a shutdown, but the insertion depth and pipe internal diameter must be entered correctly or every reading will carry the same error.

How long to log, and at what resolution

The logging period must include every operating pattern that matters to the decision. For most single-shift or two-shift plants that means at least one full working week plus a weekend or holiday; for continuous-process plants, a week that covers routine batch cycles and changeovers. Seasonal plants, such as those with summer cooling loads or monsoon-related production changes, may need a second campaign.

Resolution matters as much as duration. Fifteen-minute averages hide the short peaks that cause pressure dips and that receivers are sized to absorb. A practical approach is to log at one to ten seconds and summarise to one-minute values for the main analysis, keeping the raw data for event investigation. Synchronise the clocks of all loggers and note the time zone and any clock drift at the start and end.

1. Agree the question and the measurement boundary with the plant team.
2. Check access, pipe size, pressure rating and the permit needed for installation.
3. Install flow, pressure and compressor-state loggers with synchronised clocks.
4. Keep a production diary: shifts, breaks, batch starts, maintenance and abnormal events.
5. Log for the agreed period, checking data quality after the first day.
6. Remove loggers, back up raw files and record any gaps or instrument faults.

Reading the profile: base load, peaks and idle flow

Once the data is clean, sort it into a few practical numbers that a decision can rest on:

• Base load: the flow that is almost always present during production. This is the duty a fixed-speed base machine can serve efficiently.
• Typical peak and rare peak: separate the peak that occurs every shift from the one that occurs once a week. Sizing supply for the rare peak can often be replaced by storage or by fixing the event that causes it.
• Swing: the range between base and typical peak, and how long the plant spends in each band. A wide, frequent swing points towards a VSD trim machine or better sequencing.
• Non-production flow: flow when lines are stopped but the system is pressurised. It is mostly leakage, drains, purge and equipment left open, and it is often the cheapest saving on the site.
• Rate of change: how fast flow rises at the start of a large event. A fast rise with a falling pressure trace usually means the supply or storage cannot respond quickly enough.

A load-duration curve, which sorts all logged values from highest to lowest, is a clear way to show how many hours the plant spends at each demand level. It makes compressor-mix discussions much easier for management to follow.

Common mistakes in demand studies

• Sizing a compressor from the sum of tool ratings, with no diversity or utilisation factor.
• Logging a holiday week, or a week with a major line down, and treating it as normal.
• Using compressor load hours alone as a proxy for flow, without knowing the machine’s actual delivery at site conditions.
• Ignoring pressure: demand rises when pressure is raised, because unregulated users and leaks consume more air at higher pressure.
• Treating artificial demand from high header pressure as genuine production need.
• Comparing a normalised meter reading directly with a FAD rating.
• Missing a second supply path or a temporary compressor that feeds downstream of the meter.

How Reddy Compressors helps

We carry out demand profiling as part of compressed-air audits for plants of all sizes, independent of compressor brand. Our engineers install flow, pressure and machine-state logging at the agreed boundary, keep the production diary with your team and turn the data into a plain profile: base load, peaks, non-production flow and the compressor running pattern behind them.

Because we sell, rent and service ELGi, Atlas Copco, Ingersoll Rand, KAESER, Chicago Pneumatic and other makes as an independent multi-brand specialist, our recommendations are not tied to one manufacturer. The result may be a different compressor mix, a VSD trim machine, a receiver, a sequencer setting, a leak programme or simply a lower pressure set point. Where you need temporary air while changes are made, our own rental fleet of 50–2,000 cfm diesel and electric machines can cover the gap.

Our office is at Tarnaka, Secunderabad, and our service teams cover the whole of Telangana and Andhra Pradesh. For larger projects elsewhere in India we can arrange site support where feasible. Call +91 99590 60718 or +91 92468 88812.

What to send us

• A simple layout or piping sketch showing compressors, dryers, receivers and main headers.
• Make, model and nameplate photo of each compressor and dryer.
• Shift pattern, production days and any planned shutdowns in the next few weeks.
• The decision you are facing: expansion, replacement, pressure problems or energy reduction.
• Main pipe sizes at the proposed metering points and whether a shutdown is possible for installation.
• Any existing flow, energy or controller data you already have.

Frequently asked questions

How do I calculate the actual air demand of my plant?

The reliable way is to measure it. Install a flow meter at the station outlet or after treatment and log flow, pressure and compressor state for at least a normal working week. Adding up tool ratings overstates demand because tools rarely run together, while compressor load hours alone ignore the machine’s real delivery. Measured data gives base load, peaks and leakage separately, which is what sizing and control decisions need.

How long should an air demand study run?

Long enough to cover every operating pattern that affects the decision. For most plants that means at least one full production week including a weekend or non-production day, so base load and leakage can be seen. Continuous-process plants need a week that includes batch cycles and changeovers. Where demand changes with season or product mix, a second, shorter campaign is worth considering before committing to major equipment.

Can I size a new compressor from my existing compressor’s rating?

Only as a starting point. Your current machine may be oversized and running unloaded much of the time, or undersized and supported by a second unit you have forgotten about. A demand profile shows how much air the plant actually uses at each hour. That lets you choose capacity, pressure and control type for the real duty, often with a smaller base machine and a better trim arrangement.

What logging interval is best for compressed-air flow?

Log at one to ten seconds where the instruments allow, then average to one-minute values for the main analysis. Fifteen-minute averages hide the short peaks that cause pressure dips and that receivers are sized to absorb. Keep the raw high-resolution data, because when a pressure complaint arrives you will want to see exactly what happened in the few seconds before it.

Why does my plant use air when production is stopped?

Air used while lines are stopped is mostly leakage, open blow-offs, timer drains, dryer purge and equipment left valved in. A demand profile shows this as a steady flow during nights, breaks or weekends. It is often a significant share of total consumption and is one of the most cost-effective things to tackle, starting with a leak survey and isolating idle zones.

Does a demand profile help decide between a VSD and a fixed-speed compressor?

Yes. A VSD compressor earns its keep when demand swings widely and spends long periods at part load. A profile shows how many hours the plant sits in each flow band. If demand is steady near one level, a correctly sized fixed-speed machine may be the better base unit, with a smaller VSD or good sequencing handling the swing.

Related pages

• Compressed-air audit and measurement hub
• Power and flow logging
• Pressure profiling
• Airflow, FAD and reference conditions
• VSD integration and control
• Air receiver tank sizing