Caliper calibration: procedure, standards and how often to calibrate

What a caliper calibration checks

A calibration compares the caliper with reference standards whose values are known and traceable. It records the error at each test point together with the uncertainty of the check. Most shops then compare those errors with a maximum permissible error (MPE) and record a pass or fail. That second step is a verification against a specification, and a certificate can carry both.

ISO 13385-1 describes a caliper's accuracy with two characteristics. The first, E, covers measurements with the outside jaws when only part of each measuring face touches the standard. The second, S, is the shift error of the other faces: the inside jaws, crossed knife-edge inside jaws, the depth rod and the step faces. The limits assume the zero was set with the outside jaws closed. The other faces can therefore show larger errors, and each needs its own check.

• Calibration vs verification

The standards: ISO 13385-1, ASME B89.1.14 and ISO 3650

ISO 13385-1:2019 is the international standard for the design and metrological characteristics of calipers with a vernier scale, a dial or a digital display. It is the second edition, and ISO confirmed it at its 2025 review. Default MPE values sit in normative Annex B and calibration guidelines in an informative one. Annex B also gives the decision rule for those default values: simple acceptance when the test uncertainty is small enough. For other specifications made to the standard, the default rules of ISO 14253-1 apply unless another rule is stated.

ASME B89.1.14-2018, reaffirmed in 2023, is the US standard for vernier, dial, electronic digital and specialty calipers. It covers how to specify, verify and calibrate them. ASME describes it as a complement to ISO 13385-1, not a contradiction of it.

The usual reference standards are gauge blocks to ISO 3650:1998. That standard sets limit deviations for a calibration grade K and for grades 0, 1 and 2. Whatever the grade, the blocks need their own calibration certificate, traceable to national standards. If you use the certified length of each block instead of its nominal size, the certificate's uncertainty is what counts and the grade tolerance drops out.

Lengths are specified at 20 °C, the standard reference temperature in ISO 1:2022. ISO 13385-1 applies its MPE values at exactly 20 °C unless stated otherwise. Test results are corrected to 20 °C, or the effect of not correcting them goes into the uncertainty.

• Metrological traceability explained
• Gauge block calibration

Equipment and conditions

A caliper calibration needs little equipment. The conditions matter as much as the standards.

• Gauge blocks that cover the range, with current certificates. ISO 13385-1 also names step gauges and setting ring gauges as suitable standards.
• A setting ring gauge, or a gauge block holder with accessory jaws, for the inside jaws.
• A surface plate for the depth and step checks.
• An optical flat to check the flatness of the jaw faces.
• A stable room near 20 °C. Let the caliper and the blocks settle together, and handle the blocks with gloves or tongs so your hands do not warm them.
• Clean jaws and blocks. Wipe the measuring faces with a lint-free cloth and look for burrs and damage before you start.

Outside jaws: test points across the range

Close the jaws and check or set the zero. Then measure gauge blocks at points spread across the whole range. NPL Measurement Good Practice Guide No. 40, Callipers and Micrometers, calls for at least five roughly equally spaced positions covering the range. The error at each point is the reading minus the certified length of the block.

Choose the points with the mechanism in mind. ISO 13385-1 asks for test points that cover the range of the vernier scale or the dial. On a dial caliper it asks for points that leave the pointer at different angles around the dial. That way an error that repeats with each turn of the dial, or a local error on the scale, gets a chance to show.

Measure each block near the tips of the jaws and again close to the beam. A difference between the two shows jaws that are not parallel or a slider that tilts. Repeat a reading with the slider clamped, because clamping should not change it.

Inside jaws, depth rod and step faces

Check the inside jaws against a known internal size. A setting ring gauge works, and so does a gauge block clamped between accessory jaws in a holder. Read the caliper on its inside faces and subtract the known size. ISO 13385-1 gives crossed knife-edge inside jaws their own shift error, because the thickness of the edges and the gap between them affect small bores. If your caliper has them, include a small ring.

For the depth rod, stand a gauge block on a surface plate. Rest the end of the beam on top of the block and run the rod down to the plate. The step faces are checked the same way, on a block standing on the plate. In both cases the reading should match the length of the block.

Jaw condition, flatness and parallelism

Worn or damaged jaws spoil every reading taken near the tips. Close the jaws and hold them up to a light. Light showing between the faces anywhere along their length means wear or a bend. NPL suggests this as a simple daily check, with optical flats for a closer look once wear is suspected.

In the calibration itself, check the flatness of the outside faces with an optical flat. Check their parallelism with gauge blocks at the tips and near the beam, at one point near zero and one near full range. NPL also lists a check of the fixed face's squareness to the beam with a knife-edge square.

Look over the rest of the caliper too. The slider should move smoothly with no play. Scale lines and the dial should be easy to read. A digital display should hold a steady reading and a steady zero, and its battery should be in good condition. A caliper that fails these checks may need repair before its readings mean anything.

Daily and pre-use checks

A full calibration happens at an interval. A quick check before use catches the caliper that was dropped yesterday. NPL's guide makes it part of setting up: confirm the calibration has not expired, inspect and wipe the faces and the slider, and check the zero.

New users can practice reading the vernier scale with the vernier caliper reading practice tool, which draws a caliper at a random setting and checks the reading.

• Clean the jaws and close them. The reading should return to zero, and a digital caliper should zero at the same point each time.
• Hold the closed jaws to a light and look for gaps.
• Measure a check standard, such as a gauge block or a caliper checker, and compare the reading with its size. Use the same standard each day so the readings can be compared over time.
• After a drop or a knock, repeat the checks before the caliper measures another part.

• Calibration log template
• Vernier caliper reading practice
• Micrometer reading practice

As-found and as-left readings

Record the readings at each test point before you adjust, repair or reset anything. Those are the as-found readings. They show the condition the caliper was in while it was measuring parts. If you then adjust it, repeat the checks and record the as-left readings. A caliper that needed no adjustment has the same values for both.

When an as-found reading is outside the limit, the question is what the caliper measured since its last good calibration. ISO 9001 clause 7.1.5.2 asks you to determine whether earlier results were adversely affected and to act where needed. In practice you find the parts and lots it measured, judge whether the error could have passed a bad part and record the outcome. Then repair, adjust or retire the caliper, and review its interval.

• As-found and as-left readings
• As-found and as-left worksheet

How often to calibrate a caliper

No standard sets one interval for all calipers. ISO 9001 asks for calibration or verification at specified intervals, or before use. The joint guidance ILAC-G24 and OIML D 10 (2022 edition) says there is no single best practice. It lists the factors to weigh, among them the tendency to wear and drift, how much and how hard the instrument is used, the environment, the results of intermediate checks and the instrument's own history.

For calipers, NPL's guide suggests a periodic inspection at least once a year, with the exact interval set by how often the caliper is used. A caliper used all day on castings wears faster than one used twice a week on finished parts. Coolant, swarf and drops shorten the interval. So does a history of as-found readings close to the limit.

Justify the interval with your own data. Start from a documented first interval, record the as-found result at every calibration and adjust. If calipers keep coming back well inside the limit, a longer interval may be justified. An as-found failure is a reason to shorten it. ILAC-G24 also advises calibrating new equipment more often at first, to show its trend.

• How to set calibration intervals
• Calibration interval calculator
• Drift rate interval calculator

TUR, the 10:1 rule and resolution

Two rules of thumb come up. The 10:1 rule, sometimes called the rule of tens, says an instrument's resolution should split the tolerance it checks into at least ten parts. The test uncertainty ratio (TUR) looks at the calibration instead. It compares the limit being checked with the expanded uncertainty of the check. For a symmetric limit of ±T and an expanded uncertainty U at about 95 percent coverage, TUR = T / U.

ANSI/NCSL Z540.3 accepts a TUR of at least 4:1 where the probability of a false accept cannot practicably be estimated. ASME B89.1.14 uses simple acceptance with a TUR of at least 4:1 as its default decision rule. ISO 13385-1 sets a similar rule for its default MPE values in Annex B: simple acceptance, as long as the test uncertainty is no more than a quarter of the MPE. For other specifications it points to the default rules of ISO 14253-1. Ask your lab which rule its pass or fail statements follow.

Resolution adds uncertainty to any reading, because the scale or display rounds to its last step. The GUM (JCGM 100:2008, F.2.2.1) treats that rounding as a rectangular distribution one step wide. Its standard uncertainty is the step divided by √12, about 0.29 times the step. In NPL's worked example a caliper that reads to 0.02 mm gets 0.02 / √12 ≈ 0.006 mm from resolution alone.

When a caliper measures parts, its uncertainty also includes repeated readings, the geometry of the faces and its own permitted error. NPL's example for a caliper used on parts combines repeatability, resolution, the flatness and parallelism of the faces and the caliper's specified maximum error. In a calibration against the MPE, ASME B89.1.14 counts the caliper's own resolution and repeatability in the errors found, not in the uncertainty of the check.

• TUR and TAR calculator
• Resolution uncertainty calculator
• How to build a measurement uncertainty budget
• Gage R&R calculator
• Gage R&R and MSA explained

What the certificate and record should show

Whether the work is done in-house or by a lab, the record of each calibration should let someone else check it later.

• The caliper's ID, type, range and resolution.
• The reference standards used, with their certificate numbers and due dates, so the traceability can be followed.
• The temperature during the calibration.
• Each test point on each face, with the as-found and as-left readings and the errors.
• The uncertainty of the calibration, and the decision rule behind any pass or fail statement.
• The date, the person or lab, and the next due date.

• Calibration certificate requirements
• Calibration certificate template
• In-house vs outsourced calibration

Where Axiospec fits

Axiospec is calibration management software. For each caliper it records the calibrations, the as-found and as-left readings, the reference standards used, the certificates and the due dates. It does not calibrate the caliper or decide whether it passes. Those stay with you or your lab.

• Gage management software