Instrument calibration and measurement evidence

Why calibration alone does not make a reading correct

Air audits, energy claims and acceptance tests all rest on measurements of pressure, flow, power, dew point and temperature. A calibration certificate tells you how an instrument compared with a reference, in a laboratory, on a certain date. It does not tell you whether that instrument was suitable for the job, installed correctly, set to the right units or time-synchronised with the other loggers. Each of these can produce errors larger than the improvement being measured.

We therefore ask four separate questions of every measurement that supports a published or reported result.

1. Identity. Which instrument, by serial number, produced the value?
2. Calibration. How did it compare with a traceable reference, when, and across what range?
3. Field suitability. Could its range, installation and operating conditions measure the intended quantity?
4. Claim suitability. Is the total uncertainty small enough for the statement being made?

A yes to one question does not imply a yes to the others.

The chain from claim to instrument

Every reported figure should trace back through an unbroken chain: the statement, the raw data file, the timestamp and channel, the measurement location, the instrument serial number, its configuration and installation, its calibration or verification record, and the reviewer who accepted it. If any link is missing, the figure is weaker than it looks. An instrument model name without a serial number is not enough for any important claim.

What to check for each type of measurement

Pressure measurement in practice

State whether pressure is gauge or absolute, and give it in both bar and psig where helpful (1 bar ≈ 14.5 psi). Record the tapping location, sensor range, resolution and logging interval. For pressure-drop work, a differential pressure sensor across a filter or dryer is usually far more accurate than subtracting two readings from wide-range transmitters. When comparing station and user pressure, log both at the same moment; a few minutes’ offset during a demand swing can create an apparent loss that does not exist.

Flow and the reference-condition trap

Flow figures are only comparable when they share a reference basis. Compressor capacity is normally stated as free air delivery (FAD) at inlet conditions, while many thermal-mass flow meters report normal or standard flow at a fixed temperature and pressure. The difference can be significant in Indian summer conditions. Check the meter’s reference conditions, the pipe internal diameter entered in its settings, the straight-run lengths before and after the sensor and whether wet or oily air may affect the sensing element.

Electrical power: measure power, not just current

Compressor energy claims need true power measured on all three phases, with correct current-transformer ratios and orientation and a proper voltage reference. Current-only logging misses power factor, which changes between loaded and unloaded running. Define what the meter covers: the compressor main motor only, or also the fan motor, the dryer and the cooling system. Specific power in kW per m³/min is only meaningful when power and flow are measured at matching boundaries.

Calibration certificates: what to read

• Serial number matches the instrument actually used.
• Calibration date and the interval relative to the measurement date.
• Range of points tested covers the range used on site.
• As-found and as-left results, so you know whether the instrument was in tolerance before adjustment.
• Stated uncertainty of the calibration and traceability to a national standard.
• Any limitations or conditions noted by the laboratory.

As-found results matter. If an instrument is found out of tolerance at calibration, measurements taken since the previous calibration may need to be reviewed. Keep certificates with the audit file, not only in the instrument store, so that anyone reviewing a result later can check them.

Field checks between calibrations

Calibration intervals depend on the instrument, how it is used and the maker’s guidance, and there is no single correct interval. Between calibrations, simple field checks catch most problems: comparing two pressure sensors at the same point, checking a flow meter reads close to zero with the line isolated, confirming a power logger agrees with a fixed panel meter, and checking a dew-point sensor against a reference sensor where available. Record each check with date, result and who did it.

Uncertainty and the decision threshold

Every result should carry an estimate of its uncertainty, combining instrument accuracy, installation effects, resolution, logging interval and the natural variation of the plant. The key question is whether the effect being claimed is clearly larger than that uncertainty. If a claimed reduction in pressure drop is 0.1 bar and the combined uncertainty is 0.15 bar, the measurement cannot support the claim. Report results with sensible precision; quoting specific power to three decimal places when the flow measurement is uncertain to several percent misleads the reader.

Time synchronisation and logging intervals

An audit usually runs several loggers at once: pressure at the station and at critical users, flow in the main header, power on each compressor. If their clocks differ by even a few minutes, a demand peak seen by one logger appears at a different time on another, and pressure losses or load changes are misread. Set every logger to a common time source at the start, check them again at the end and record any drift.

The logging interval must suit the events being studied. Load and unload cycles on a screw compressor can last well under a minute, so a ten-minute average hides them completely. Short intervals capture control behaviour; longer averages are adequate for weekly energy totals. State the interval and whether values are instantaneous, averaged or peak readings.

Common failure modes

• The certificate belongs to a different serial number from the sensor used on site.
• The sensor range is far wider than the quantity measured, so resolution swamps the result.
• Flow references are mixed between FAD, normal and standard conditions.
• Power was calculated from current with an assumed power factor.
• Dew-point sampling was taken from a dead leg or after the sample line warmed up.
• Loggers were not time-synchronised, so events do not line up.

How Reddy Compressors helps

We carry out air audits and measurement work by the ISO 11011 assessment method for plants across Telangana and Andhra Pradesh, logging pressure, flow, power and dew point with identified instruments. Our reports state which instrument took each measurement, its location, the reference conditions and the limits of the result. We do not issue conformity certificates; we report what was measured and how.

As an independent multi-brand specialist working on ELGi, Atlas Copco, Ingersoll Rand, KAESER, Chicago Pneumatic and other makes, we can also review measurements made by others, such as a supplier’s commissioning figures, and tell you whether they support the claim being made. Our office is at Srinilayam, 12-13-200, First Floor, Street No. 1, Tarnaka, Secunderabad 500017; call +91 99590 60718 or +91 92468 88812.

What to send us

• The claim or decision the measurement must support
• Instruments already on site: type, range and serial numbers
• Any existing calibration certificates and logged data
• A sketch showing compressors, dryers, receivers and measurement points
• Pipe sizes at the proposed flow measurement locations

Frequently asked questions

How often should compressed-air instruments be calibrated?

There is no single interval. It depends on the instrument type, the maker’s guidance, how often and how harshly it is used, and how critical the measurement is. Many sites calibrate portable audit instruments annually, but the as-found results should guide the interval: if an instrument is regularly found out of tolerance, shorten it. Field checks between calibrations help.

Why does my flow meter reading not match the compressor’s rated FAD?

Often because they use different reference conditions. Rated free air delivery is stated at inlet conditions, while many flow meters report normal or standard flow at fixed temperature and pressure. Wrong pipe diameter settings, insufficient straight pipe, wet air or the compressor running below rated pressure can also cause differences. Check the reference basis first.

Is a calibration certificate enough to trust an audit result?

No. The certificate shows how the instrument compared with a reference at one time. You also need to know that the certified instrument is the one used, that its range suited the measurement, that it was installed correctly, that loggers were synchronised and that the overall uncertainty is small compared with the effect claimed. All these links are needed.

What is the difference between pressure dew point and atmospheric dew point?

Pressure dew point is the temperature at which water condenses at the actual line pressure, which is what matters for the air system. Atmospheric dew point refers to the same air expanded to atmospheric pressure and is always lower. Reporting one as the other makes dried air look drier than it is. Always state which basis is used.

Can I measure compressor power with a clamp-on ammeter?

A clamp-on ammeter shows current, not power. Compressor power factor changes between loaded and unloaded running, so current alone gives an unreliable picture of energy. For audit and saving claims, use a three-phase power logger measuring voltage and current on all phases with correct current-transformer orientation, and state which loads the measurement includes.

How accurate does a pressure-drop measurement need to be?

Accurate enough that the combined uncertainty is clearly smaller than the drop you need to detect. Pressure drops across filters, dryers and pipework are often a fraction of a bar, so wide-range transmitters may not resolve them reliably. A differential pressure sensor across the component, with simultaneous logging, is usually the better method.

Related pages

• Compressed air project evidence centre
• Before and after comparison method
• Dew-point measurement
• Power and flow logging
• Measurement uncertainty worksheet
• Motor input and electrical measurement worksheet