Pressure Calibrator

What is a Pressure Calibrator

A pressure calibrator is an instrument, or a coordinated set of instruments, used to verify and adjust the accuracy of other pressure-measuring devices. Where a process pressure sensor lives permanently on a pipe and reports a value to the control system, a calibrator is a metrology tool that applies a known, traceable pressure to a device under test and compares the device's reading against a reference of much higher accuracy. The output of a calibration is not a control signal but a record: an as-found and as-left table, a deviation curve, and a statement of measurement uncertainty.

Functionally a complete calibrator combines two distinct elements. The first is a pressure-generation source that builds and holds a stable pressure: a manual hand pump, a battery-driven electric pump, or a fully automated pressure controller. The second is a reference measurement standard that quantifies that pressure to a documented accuracy: a precision digital gauge, an interchangeable pressure module, or a deadweight tester (pressure balance). Some products integrate both into one handheld unit, such as the Fluke 719Pro, which carries an internal electric pump rated to 20 bar (300 psi) together with a built-in reference and mA loop functions. Others keep them separate, pairing a precision gauge like the Fluke 700G with a discrete pneumatic or hydraulic test pump.

The defining property of any calibrator is traceability. Every reading must connect, through an unbroken chain of comparisons each with a stated uncertainty, back to a national or international standard, and that chain is documented on an ISO/IEC 17025 accredited calibration certificate. This is why a calibrator's own accuracy is specified far more tightly than that of the instruments it checks: the calibration industry works to a test uncertainty ratio of at least 4 to 1, meaning the reference must be at least four times more accurate than the tolerance of the device under test.

The history of pressure calibration begins with the primary standard. The deadweight tester, or pressure balance, derives pressure directly from physical quantities: calibrated masses act through local gravity on a precisely lapped piston of known effective area, so pressure equals force divided by area with no electronic element to drift. This principle, refined since the nineteenth century, still anchors the top of the traceability pyramid in national metrology institutes. Beneath it, precision digital gauges and automated controllers brought laboratory-grade accuracy to the bench and the field, and the 4 to 20 mA documenting calibrator added the ability to source, simulate, and measure loop signals so that a single technician can certify an entire transmitter loop.

Four engineering attributes determine whether a calibrator is fit for a given task: its reference accuracy expressed correctly as percent of reading or percent of full scale, its long-term stability between recalibrations, its pressure range and the generation medium it uses, and the completeness of its traceability and documentation. A mismatch in any one of these, for instance a reference that is only twice as accurate as the device under test, invalidates the calibration regardless of how precise the hardware appears.

Calibrator Types and Architecture

Pressure calibrators are best classified by how they generate pressure and how that generation is coupled to the reference standard. Four architectures dominate the market, ranging from the low-cost field hand pump to the fully automated laboratory controller. Selecting the wrong architecture is a common and expensive mistake: a deadweight tester on a vibrating plant floor is unusable, and a hand pump cannot deliver the throughput a busy calibration laboratory needs. The table below summarizes the four.

Field documenting calibrators integrate the pump, reference, and electrical functions into one rugged handheld unit. The Fluke 719Pro is representative: a built-in electric pump reaches 20 bar (300 psi), a programmable pump limit prevents over-pressurization, and a full loop calibrator sources, simulates, and measures 4 to 20 mA signals alongside the pressure reading. The Beamex MC6 extends this idea to a multifunction documenting platform that also handles temperature and electrical signals. These instruments trade the very best accuracy for portability, on-board documentation, and one-person operation, which is exactly what plant maintenance and oil-and-gas field service require.

Precision gauge plus test pump separates the reference from the generation source. A precision digital gauge such as the Fluke 700G (0.05 percent full scale, reference class to 0.04 percent of reading) or the Additel 681 (accuracy classes from 0.02 to 0.2 percent full scale) mounts onto a hand pump and reads the generated pressure. The pump and gauge can be swapped independently, so one gauge serves several ranges and one pump serves several gauges. This modular approach is the workhorse of most calibration workshops because it balances cost, flexibility, and accuracy.

Automated pressure controllers close the loop electronically: the user enters a target pressure and the instrument regulates supply and exhaust valves to reach and hold that setpoint within seconds, then steps automatically through a calibration sequence. The WIKA Mensor CPC6050 offers accuracy down to 0.008 percent of span with interchangeable plug-and-play transducers and up to two independent regulating channels, and the Fluke 6270A reaches 0.01 percent of reading over the upper part of its span with modules to 20 MPa. Automation removes operator-induced scatter and is essential where dozens of instruments must be certified each day.

Deadweight testers, also called pressure balances, sit at the top of the pyramid as a primary or transfer standard. They generate pressure by loading certified masses onto a floating piston of certified effective area, so the reference value comes from mass and area, not from electronics that can drift. Hydraulic and pneumatic variants exist; the pneumatic Fluke P3000 series and AMETEK hydraulic Type T testers are widely used. Their strength is exceptional stability and a clean traceability statement; their weakness is slow point-by-point operation and sensitivity to leveling, temperature, and cleanliness.

Reference Standard Grades

Whatever the architecture, the reference measurement standard is the heart of a calibrator, and references are organized into grades by accuracy and role in the traceability chain. Choosing a grade is fundamentally an exercise in the test uncertainty ratio: the reference must be enough times better than the device under test that a passing result is defensible. The table below maps the common grades to their typical accuracy and the class of work they support.

Primary and transfer standards anchor the chain. A deadweight tester reaches about 0.015 percent of reading in routine use and roughly 0.008 percent when software corrects for piston temperature, fluid head height, and air buoyancy. Its piston-cylinder assembly and mass set are dimensionally stable and traceable through an ISO/IEC 17025 accredited certificate to NIST or another national metrology institute. Automated laboratory controllers such as the CPC6050 occupy the same tier when fitted with their highest-accuracy transducers, offering the speed that a deadweight tester lacks while still reaching 0.008 percent of span.

Reference-class digital gauges are the most common bench and field references. The Fluke 700G spans roughly 20 mbar to 690 bar across its model family, stating 0.05 percent of full scale in standard form and 0.04 percent of reading in its reference class, and it can log several thousand readings for later upload. The Additel 681 offers selectable accuracy classes of 0.02, 0.05, 0.1, and 0.2 percent full scale with fully temperature-compensated performance from minus 10 to plus 50 degrees Celsius, and an ATEX intrinsically safe version for hazardous areas. These gauges are accurate enough to serve as the reference for the great majority of process transmitters.

Working and field documenting calibrators trade a little accuracy for integration and ruggedness. They are the instruments a technician carries into the field, and their 0.05 to 0.1 percent class is sufficient to verify the typical 0.1 to 0.5 percent process transmitter at a comfortable test uncertainty ratio. Process check gauges at 0.1 to 0.25 percent are for fast go or no-go verification, not for issuing accredited certificates; they confirm that an instrument has not grossly drifted between formal calibrations.

A practical rule ties the grades together. To certify a 100 psi transducer specified at 0.04 percent full scale, the reference standard must be 0.01 percent full scale or better to hold a 4 to 1 ratio. When a single reference cannot achieve that ratio over the whole span, percent-of-reading instruments help because their error shrinks at low pressures, but the uncertainty budget must still be documented under ISO/IEC 17025 and guard-banded where the ratio is thin.

Generation Media and Standards

The pressure-generation source defines the medium, and the medium constrains both the range and the kind of device that can safely be tested. Two families dominate: pneumatic generation using clean dry air or inert gas, and hydraulic generation using calibration oil or distilled water. The choice is not interchangeable, because gas stores far more energy than an equal volume of liquid and because some devices under test cannot tolerate oil contamination.

Pneumatic generation uses ambient or supplied clean dry air as the calibration medium. A typical pneumatic hand pump kit, such as the Fluke 700PTPK, covers vacuum up to about 40 bar (600 psi). Air is clean, vents instantly, and leaves no residue, which makes pneumatic generation the correct choice for oxygen service, food and pharmaceutical instruments, and any gauge or transmitter that must stay free of oil. The penalty is stored energy: because gas is compressible, a pneumatic source at high pressure holds significant energy, so a programmable over-pressure limit and a controlled bleed valve are important safety features, and very high pneumatic pressures are reserved for purpose-built systems.

Hydraulic generation uses a nearly incompressible liquid, typically a mineral or synthetic calibration oil, and on some pumps distilled water. Hand-operated hydraulic test pumps such as the Fluke 700HTPK reach about 690 bar (10,000 psi), and dedicated high-pressure pumps go higher still. Liquid is stiff, so a hydraulic source reaches and holds a stable high-pressure point with little stored energy and excellent fine adjustment, which is why all high-pressure calibration is hydraulic. The trade-off is cleanliness: the device under test must accept the fill fluid, spills must be managed, and any gas trapped in the lines must be bled out before accurate readings are possible.

The table below maps the common generation choices to their range and the standard reference for each role. The pressure ranges are nominal kit ratings; specific pump and module combinations vary by manufacturer.

The standards that govern the work are independent of the medium. ISO/IEC 17025 sets the competence and traceability requirements for the calibration laboratory and is the basis of every accredited certificate. EURAMET cg-17 and the German DKD-R 6-1 give the detailed procedure for calibrating electromechanical and mechanical manometers, typically prescribing 6 pressure points taken in both ascending and descending directions with a repeatability check near 50 percent of span. EURAMET cg-3 covers the calibration of pressure balances. ASME B40.100 defines the accuracy grades for mechanical pressure gauges used as devices under test in North America. Together these documents fix how points are chosen, how hysteresis and repeatability enter the uncertainty budget, and how the result is reported.

Key Specification Parameters

Comparing calibrator datasheets is harder than comparing process sensors because the accuracy statement itself carries hidden structure. The same instrument can look very different depending on whether its specification is quoted as percent of full scale or percent of reading, and whether temperature and one-year stability are already included. Eight parameters drive a defensible selection: accuracy and its basis, the test uncertainty ratio it supports, temperature effect, long-term stability, range and resolution, generation pressure limit, electrical functions, and traceability. Each is explained below.

Accuracy and its basis is the first thing to decode. Percent of full scale (% FS) fixes the error in absolute terms across the whole span: 0.05 percent FS on a 700 bar gauge is plus-or-minus 0.35 bar everywhere, which is a large fraction of a small reading near zero. Percent of reading (% RDG) scales the error with applied pressure and is far better in the lower range; reference-class instruments often quote it, sometimes only above 30 or 50 percent of span. Always check whether the headline number is a one-year specification that already bundles non-linearity, hysteresis, repeatability, and temperature, or only a best-case linearity figure that must be padded.

Test uncertainty ratio (TUR) is the parameter that links the calibrator to the job. It is the ratio of the device-under-test tolerance to the reference accuracy, and the industry minimum is 4 to 1. A reference that is only twice as good as the device cannot certify it cleanly; either a more accurate reference, guard-banding, or a documented uncertainty budget is required. TUR, not headline accuracy, is the correct buying criterion.

Temperature effect states how accuracy degrades outside the compensated band. The Additel 681, for example, is fully temperature-compensated from minus 10 to plus 50 degrees Celsius; outside that window an additional temperature coefficient applies. For field work in cold or hot environments, the compensated range must cover the actual ambient conditions or the stated accuracy does not hold.

Long-term stability is expressed as the accuracy retained over a stated interval, usually one year. A specification of 0.05 percent FS one year is far more useful than a 90-day or best-case figure, because it tells you the accuracy you can rely on right up to the next recalibration. Deadweight testers excel here: their piston-cylinder and mass references are dimensionally stable and effectively immune to electronic drift.

Range, resolution, and generation limit must be matched to the work. The reference should cover the spans you calibrate with the operating points sitting in its accurate region, the display resolution should be finer than the device-under-test tolerance, and the generation source must reach the highest test pressure with margin. The remaining parameters round out the selection:

• Electrical functions: a documenting calibrator that sources, simulates, and measures 4 to 20 mA, supplies 24 V loop power, and accepts switch contacts lets one technician certify a transmitter loop end to end.
• Generation safety: a programmable over-pressure or pump limit and a fine-adjust vernier protect both the device under test and the operator, especially with compressible pneumatic media.
• Data and documentation: on-board storage, as-found and as-left capture, and PC software for certificate generation turn a measurement into an auditable record.
• Interchangeable modules: plug-and-play pressure transducers (as on the CPC6050 and Fluke 6270A) let one chassis cover many ranges and simplify range expansion later.
• Traceability: an ISO/IEC 17025 accredited certificate with stated uncertainty, not merely a manufacturer test report, is what makes the reference legally and technically defensible.

A final caution: never sum hysteresis, repeatability, and temperature drift into one number unless the datasheet already does so. EURAMET cg-17 and DKD-R 6-1 treat these as separate contributions that combine in the uncertainty budget, and conflating them either overstates or understates the true capability of the calibrator.

Selection Decision Factors

To convert the preceding chapters into a specific purchase, follow the decision sequence below. Most selection errors come not from a single wrong answer but from deciding accuracy before establishing what is being calibrated and to what tolerance. These eight steps work as a fixed RFQ template for a pressure calibrator.

1. Define the devices under test: list the instruments you must calibrate, their full-scale ranges, and their tolerances (for example 0 to 16 bar transmitters at 0.1 percent FS). Everything downstream derives from this list, not from a target instrument.
2. Set the required reference accuracy from TUR: apply the 4 to 1 rule to the tightest device tolerance. A 0.1 percent FS transmitter needs a 0.025 percent FS or better reference; a 0.04 percent device needs 0.01 percent. Decide whether percent of reading helps at the low end of your spans.
3. Choose the architecture: field documenting calibrator for plant loops, precision gauge plus pump for a flexible workshop, automated controller for high laboratory throughput, deadweight tester for a primary reference. Match this to where the work happens and how many instruments per day.
4. Choose the generation medium and range: pneumatic (air) for low and medium pressure and oil-free devices, hydraulic (oil or water) for high pressure. Confirm the pump or controller reaches your highest test pressure with margin and includes a programmable over-pressure limit.
5. Specify electrical and loop functions: if you calibrate 4 to 20 mA or HART transmitters, require source, simulate, and measure of mA, 24 V loop supply, and switch-test inputs so one person can certify the whole loop.
6. Specify environment and safety: ambient temperature range against the calibrator's compensated band, hazardous-area certification (ATEX or IECEx intrinsically safe versions exist for the 700G and 681), ingress protection for field use, and process connection compatibility with your fittings.
7. Plan traceability and recalibration: require an ISO/IEC 17025 accredited certificate with stated uncertainty, set the calibrator's own recalibration interval (typically 12 months, shorter for heavily used field units), and confirm reverse-traceability records so an out-of-tolerance reference can trigger a review of the instruments it certified.
8. Evaluate total cost of ownership: purchase price plus annual accredited recalibration of the reference, consumables and fill fluid, module expansion cost, and the labor saved by automation or documenting features. A cheaper reference that needs frequent recalibration or that fails the TUR test rarely wins over a few years.

One dimension is consistently overlooked: manufacturer serviceability and accredited support. A reference standard is only as good as the laboratory that recalibrates it, so check that the manufacturer or a local partner operates an ISO/IEC 17025 accredited pressure laboratory with reasonable turnaround, stocks spare transducers and piston-cylinder assemblies, and supports firmware and software updates. WIKA Mensor, Fluke Calibration, Beamex, Additel, and AMETEK all maintain accredited calibration and service capability, which determines repair and recertification response over the five to ten year life of the instrument far more than the initial datasheet.