Pneumatic machine air consumption acceptance testing

Why machine air consumption needs a formal acceptance test

New packaging lines, filling machines, moulding cells and assembly stations are often bought with a stated air consumption in the supplier’s quotation. That figure then drives compressor sizing, dryer selection, pipe sizing and the running-cost case. If nobody verifies it at commissioning, an under-stated figure only shows itself later as falling header pressure, an extra compressor running loaded and higher electricity bills.

A structured acceptance test answers three practical questions:

• How much free air does this machine draw per productive cycle and per accepted item, under agreed conditions?
• What is its peak flow, and at what minimum inlet pressure does it still perform correctly?
• How much air does it use when it is not producing: standby, idle blow-off, purge and leakage?

The result is a number both the machine supplier and the plant can stand behind, and a baseline against which drift, leaks and later modifications can be judged.

Define the boundary before you measure

Most disputes about machine air consumption come from an unclear boundary rather than a faulty meter. Write down exactly what is inside the measurement:

• The single supply point (or points) feeding the machine, downstream of any shared header.
• Whether the machine’s own filter-regulator-lubricator (FRL), local receiver, dryer or vacuum generators are inside or outside the boundary.
• Any auxiliary users on the same drop: blow guns, cleaning nozzles, ionisers, cabinet coolers, a neighbouring machine.
• Legitimate consumers that are part of the process, such as ejector blow-off, part cooling or bag inflation.

If another user shares the meter, either isolate it during the test or log it separately. A machine cannot be held responsible for air drawn by a hand-held blow gun on the same branch.

Reference conditions: free air, normal air and pressure at the meter

Air quantities must be stated on a declared basis, or two honest measurements can disagree by a large margin. The usual bases are:

State which basis the result uses, the meter’s own reference setting, and the line pressure and temperature during the test. Convert everything to the same basis as the compressor’s FAD before using the figure for sizing. As a reminder, 1 m³/min is about 35.3 cfm, and 1 bar is 14.5 psi.

Instrumentation that gives a defensible result

For most machine tests a thermal-mass insertion or in-line flow meter is practical, because it reads mass flow directly and needs little correction. Vortex or differential-pressure meters can also work if pressure and temperature are logged for correction. Whatever the meter:

• Respect the straight-pipe lengths upstream and downstream that the meter maker specifies, or accept and declare a larger uncertainty.
• Check the meter’s range covers both the standby trickle and the peak flow; a meter sized only for the average may clip short peaks.
• Log at a sample rate fast enough to see individual machine cycles; one-minute averages hide what is happening inside a two-second cycle.
• Log pressure at the machine inlet on the same time base, so you can see whether the machine was starved during peaks.
• Have calibration evidence for the meter and the pressure transmitter, and record serial numbers in the test report.

The machine’s cycle counter, good-part counter and reject counter must be readable at the start and end of each window, or exported on the same clock. Unsynchronised clocks are one of the commonest reasons a test has to be repeated.

Operating states and how to treat each one

A machine does not use air at one steady rate. Classify its states and measure each separately:

The ratio of standby to productive flow is often revealing. A machine that uses a large share of its productive flow while waiting is usually leaking, blowing off continuously, or using ejectors without valves that close when no part is present.

Step-by-step acceptance procedure

1. Agree the test conditions in writing with the machine supplier: product, format, recipe, line speed, supply pressure at the machine inlet and acceptable reject level.
2. Confirm the boundary and isolate or log any shared users.
3. Install the flow meter and pressure transmitter, check zero at no flow, and synchronise all clocks.
4. Run the machine until it is thermally and mechanically stable at the agreed speed.
5. Record counter readings and start the measurement window. Run long enough to cover many complete cycles and at least one full repeat of any longer sequence, such as a pallet layer or reel change.
6. Record counters at the end of the window. Note any stops, jams or faults with times.
7. Repeat the window at least twice more so that repeatability can be assessed.
8. Measure standby and starved states separately, with the machine in its normal idle condition.
9. Carry out a short leak-down check with the machine stopped and pressurised, isolated at its own valve, to estimate internal leakage.
10. Calculate the results and the uncertainty, and issue a signed report.

Calculating air per cycle and per accepted item

The arithmetic is simple once the data is clean:

• Air per cycle = total free air in the window ÷ number of completed cycles in the window.
• Air per accepted item = total free air in the window ÷ number of good items produced in the window.
• Average flow = total free air ÷ window duration.
• Peak flow = highest short-interval flow recorded, with the averaging interval stated.

Worked example (illustrative only): over a 30-minute productive window a meter totals 54 m³ of free air while the counter records 1,800 cycles and 1,764 good items. Air per cycle is 54 ÷ 1,800 = 0.030 m³ (30 litres). Air per accepted item is 54 ÷ 1,764 ≈ 0.0306 m³. Average flow is 54 ÷ 30 = 1.8 m³/min, about 64 cfm. If the logged peak over one-second intervals was 4.5 m³/min while inlet pressure dipped from 6.2 to 5.6 bar (90 to 81 psig), the supply branch and local storage must be judged against that peak, not the average.

Report uncertainty in plain terms: the meter’s stated accuracy at the measured flow, the effect of any installation compromise, and the spread between repeat windows. If the repeat windows disagree more than the meter uncertainty, the machine’s behaviour is not yet stable and the test should not be signed off.

Common reasons a machine fails acceptance

• Continuous blow-off nozzles or air knives running when no product is present.
• Vacuum ejectors with no air-saving valve, drawing air throughout the cycle.
• Leaking cylinder seals, valve exhausts or push-in fittings.
• Regulators set higher than the process needs; air use rises roughly in proportion to absolute pressure for many devices.
• Oversized cylinders or long, large-bore tubing between valve and cylinder, so dead volume is filled and vented every stroke.
• Panel coolers or purge devices left on as standard.

Each of these is fixable, and the acceptance report is the best place to list them so the supplier corrects them before the machine is handed over.

Using the result for compressor and pipework decisions

Once accepted, the figures feed directly into plant decisions. Average flow sets the long-term energy and the base load on the compressors. Peak flow and minimum inlet pressure set the drop size, local receiver and regulator capacity. Standby flow tells you what the plant pays for during breaks and changeovers. Adding a new machine to an existing system should be checked against measured spare capacity, not against nameplate totals. Remember that raising system pressure to cover a starved machine costs about 7% more compressor power per extra bar; fixing the branch or adding local storage is usually the better route.

How Reddy Compressors helps

Since 2002 Reddy Compressors, based at Tarnaka in Secunderabad, has worked with plant engineers on compressed-air supply for machines of every type. We are an independent multi-brand specialist, not an authorised dealer of any brand, and we sell, rent and service ELGi, Atlas Copco, Ingersoll Rand, KAESER, Chicago Pneumatic and other compressors. We have tie-ups with all major and reputed brands. For machine acceptance and the decisions that follow we can:

• Log flow and pressure at a machine or branch and help interpret the data alongside your counters.
• Carry out air audits by the ISO 11011 method, including demand profiling and leak surveys (we do not issue conformity certificates).
• Advise on compressor capacity, dryers, receivers and pipework once the real demand is known.
• Supply new or used/refurbished compressors, and rental machines from our own fleet (50–2,000 cfm, 2–30 bar, diesel and electric) to cover commissioning or trial periods.
• Provide service, AMC and spare parts through more than 20 service engineers and technicians covering all of Telangana and Andhra Pradesh, with equipment and spares supplied across India.

What to send us

• Machine type, maker and the air consumption stated in its quotation or manual, with the basis used.
• Required supply pressure at the machine inlet and its line speed or cycle rate.
• A sketch or photo of the supply drop, FRL and any local receiver.
• Details of the compressors, dryer and header pressure feeding the area.
• Any symptoms seen so far: pressure dips, faults at peak speed, high standby air.

Frequently asked questions

How do I measure the air consumption of a single machine?

Fit a flow meter on the machine’s own supply line, downstream of any shared header, and log pressure at the inlet on the same clock. Read the cycle and good-part counters at the start and end of each window. Divide the total free air by cycles and by good items. Repeat the window at least three times so you can show the result is repeatable before you accept it.

Why does the compressor display not show my machine’s air use?

The compressor display shows what the whole system draws, including every other machine, leaks, dryer purge and changes in receiver storage. It cannot separate one machine from the rest, and its timing does not match the machine’s cycles. Only a point-of-use meter at the machine’s supply, synchronised with its counters, can tell you air per cycle or per accepted product.

What is the difference between scfm, Nm³ and FAD?

All three describe a volume of air referred to a declared pressure and temperature. FAD refers to compressor inlet conditions, normal cubic metres to 0 °C and 1.013 bar dry, and scfm to a standard condition that varies between sources. Always state the basis, and convert the machine’s figure to the same basis as your compressor’s FAD before using it for sizing.

How long should a machine air consumption test run?

There is no universal duration. The window must cover many complete machine cycles and at least one repeat of any longer sequence, such as a reel change or pallet layer, with the machine already stable at the agreed speed. Run several repeat windows. If results still vary more than the meter uncertainty, extend the test until behaviour settles.

Why is my new machine using more air than the supplier quoted?

Common causes are a different supply pressure from the one quoted, blow-off or vacuum ejectors running continuously, leaking seals and fittings, long large-bore tubing adding dead volume, and a different reference basis between the quotation and your meter. An acceptance test with the supplier present usually identifies which of these applies and what correction is needed.

Should standby air be included in machine acceptance?

Yes. Measure it separately rather than mixing it into air per cycle. Standby and starved-state air shows leakage, purge devices and blow-off left running, and it is paid for every break, shift change and stoppage. Many plants find that standby consumption is one of the easiest savings once it is visible in the acceptance report.

Related pages

• Air demand profiling
• Power and flow logging
• Instrument calibration and measurement evidence
• Compressed air for packaging machines
• Pressure optimisation and artificial demand
• Pneumatic application engineering centre