Micrometer calibration: procedure, standards and how often to calibrate

What a micrometer calibration checks

A calibration compares the micrometer with traceable reference standards. It records the error at each test point together with the uncertainty of the check. Comparing those errors with a maximum permissible error (MPE) and recording a pass or fail is a verification, and most micrometer certificates carry both.

Four things decide whether an outside micrometer reads true. The first is the zero, the reference point at the start of the range. The second is the flatness of each measuring face and how parallel the two faces are. The third is the error of the screw along the travel. The fourth is a steady measuring force from the ratchet or friction thimble. ISO 3611:2023 defines the length measurement error E and its variation V, with tests that use optical parallels, and it covers the measuring force as well.

This guide covers outside micrometers. ASME B89.1.13 also covers inside and depth micrometers, which need their own setups.

• Calibration vs verification

The standards: ISO 3611, ASME B89.1.13 and ISO 3650

ISO 3611:2023, the third edition, sets the design and metrological characteristics of micrometers for external measurements with analogue or digital indication. It replaced the 2010 edition. The changes include requirements for test methods and a classification system for MPE values. Calibration guidelines sit in an informative annex. According to its introduction, the default decision rules of ISO 14253-1 apply to specifications made to the standard unless another rule is stated.

ISO 3611 also fixes design points that matter in calibration. It recommends a spindle pitch of 0.5 mm or 1 mm and a digital step of 0.01 mm or 0.001 mm. It requires a device on the thimble or speeder that limits the measuring force, such as a ratchet or a friction drive, and a zero adjustment the user can reach.

ASME B89.1.13-2013, reaffirmed in 2022, is the US standard for outside, inside and depth micrometers. It sets out requirements that makers and buyers can share. It asks for length standards with metrological traceability and includes a default decision rule.

Gauge blocks to ISO 3650:1998 are the usual reference. That standard sets limit deviations for a calibration grade K and for grades 0, 1 and 2. NPL's worked uncertainty budget for a 0 to 25 mm micrometer uses grade 1 steel blocks. Whatever the grade, the blocks need their own traceable calibration.

Lengths are specified at 20 °C, the standard reference temperature in ISO 1:2022. ISO 3611 applies its MPE values at exactly 20 °C unless stated otherwise. Test values 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

Besides gauge blocks, a micrometer calibration needs a few special items and a stable room.

• A set of gauge blocks made for micrometer calibration. NPL Measurement Good Practice Guide No. 40, Callipers and Micrometers, describes sets whose sizes are chosen to test both the periodic and the progressive error of the screw. Metric and inch micrometers need separate sets.
• An optical flat for the flatness of each face, and a set of optical parallels of slightly different thicknesses for parallelism.
• A micrometer stand, so your hand does not warm the frame. Use the insulating plates on the frame if it has them.
• Gloves or tongs for handling the blocks.
• A room near 20 °C. NPL advises leaving the micrometer and the work in the room for at least one hour to reach room temperature.
• For a micrometer whose range starts above zero, a setting standard or a gauge block of the starting length.

Zero and measuring force

ISO 3611 treats the reference point as fixed at the start of the range. On a 0 to 25 mm micrometer that means the faces closed together. On larger sizes it means the setting standard or a gauge block of the starting length.

Check the zero the way the micrometer will be used. Clean the faces, for example by closing them lightly on a sheet of clean paper and drawing it out. Then bring the faces together with the ratchet or friction thimble and stop as soon as it slips. NPL's guide warns that turning the ratchet quickly gives false readings, and it suggests repeating the check two or three times. Record the zero as found before you reset it.

The force device is what makes readings repeatable from one user to the next. Check that the ratchet or friction thimble turns smoothly and slips evenly. A sticky ratchet changes the measuring force, and the reading changes with it. NPL lists the measuring force among its daily checks.

Flatness and parallelism of the measuring faces

Lay an optical flat on each measuring face in turn. The interference fringes you see show the shape of the face. A few straight bands mean a flat face, while curved or crowded bands point to wear or a rounded edge. Burrs often sit near the outer edge of a face.

For parallelism, close the faces on an optical parallel with the ratchet and look at the fringes on both faces. A set of parallels of slightly different thicknesses puts the spindle at several positions, so a face that tilts as the screw turns shows up. NPL describes optical parallels as coming in a range of sizes for exactly this reason.

Flat, parallel faces matter most when a round part touches them only at a point or along a line, which NPL points out for spheres and cylinders. A gauge block that covers the whole face can hide a fault that a small shaft would find.

Spindle error across the range

The main test reads gauge blocks at points along the whole travel of the spindle. A standard metric micrometer screw has a 0.5 mm pitch, so the thimble turns once for every 0.5 mm. If every block were a whole number of millimetres, the thimble would stop at the same angle each time. An error that repeats with each turn of the screw would never show.

Calibration sets avoid that. Their lengths are chosen so the spindle stops at different points around the thimble as well as along the travel, which tests the periodic error within one turn and the progressive error along the screw. NPL's example calibration of a 0 to 25 mm micrometer reports errors at 2.5, 10.3, 12.9, 15, 20.2 and 25 mm, and its uncertainty example uses a series of ten blocks.

Put each block between the faces with the micrometer in its stand, close it with the ratchet and record the reading. The error is the reading minus the certified length of the block. Read every point the same way the zero was set.

• As-found and as-left worksheet

Ratchet, thimble and general condition

Before or during the calibration, look over the mechanism. NPL's routine checks cover these points.

• The ratchet or friction thimble turns smoothly and slips evenly.
• The thimble turns freely over the whole travel, without sticking or backlash.
• Clamping the spindle does not move the reading. NPL puts a figure on it: no more than 2 µm.
• The graduations are clear. At zero, the edge of the thimble lines up with a graduation on the sleeve without hiding it.
• The faces are free of scratches and burrs, and the frame shows no sign of a bend.
• A digital display holds a steady reading, and its battery is in good condition.

Daily and pre-use checks

Between calibrations, a short check before use catches damage early. NPL's daily inspection covers the zero, the measuring force and the fit of the spindle, with a fuller check after a drop.

New users can practice reading the sleeve and thimble with the micrometer reading practice tool, which draws a micrometer at a random setting and checks the reading.

• Clean the faces and check the zero with the ratchet, two or three times.
• Turn the ratchet or friction thimble and feel for smooth, even slipping.
• Run the spindle through its travel. It should move smoothly with no play.
• Measure a check standard, such as a gauge block near the middle of the range, and compare the reading with its size.
• After a drop or a knock, check the zero, the measuring force, the fit, the runout of the thimble, the parallelism of the faces and the error of indication before the micrometer is used again.

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

As-found and as-left readings

Record the zero, the face checks and the error at each block before any adjustment. Those are the as-found readings. Resetting the zero is an adjustment too, so note the zero as found first. After any adjustment or repair, repeat the checks and record the as-left readings.

When an as-found error is outside the limit, ISO 9001 clause 7.1.5.2 asks you to determine whether earlier results were adversely affected and to act where needed. The size and sign of the error tell you which parts are at risk. A micrometer that reads high can pass undersize parts and reject good ones near the upper limit. One that reads low can pass oversize parts and reject good ones near the lower limit. Find the parts and lots it measured since its last good calibration, judge the risk and record the outcome.

• As-found and as-left readings

How often to calibrate a micrometer

No standard sets one interval. ISO 9001 asks for calibration or verification at specified intervals, or before use. ILAC-G24 and OIML D 10 (2022 edition) say there is no universal best practice and list the factors to weigh, including wear and drift, how much and how hard the equipment is used, the environment, intermediate checks and its calibration history.

NPL's guide ties the periodic inspection interval for micrometers to usage and calibration history, judged against inspection criteria you define. In practice the drivers are wear of the faces and the screw, drops, and coolant or dirt in the thread, together with the as-found results at each calibration.

Start from a documented interval and adjust it from your own data. ILAC-G24 advises calibrating new equipment more often at first. Long runs of as-found results well inside the limit support a longer interval. A failure, or a drift toward the limit, supports a shorter one.

• 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 B89.1.13 adopted simple acceptance with a 4:1 ratio as its default decision rule, following ASME B89.7.3.1. The introduction of ISO 3611 says the default rules of ISO 14253-1 apply unless another rule is stated. Ask your lab which rule its pass or fail statements follow.

When you measure parts, the micrometer's resolution adds uncertainty because the reading is rounded to its last step. The GUM (JCGM 100:2008, F.2.2.1) gives the step divided by √12 as its standard uncertainty, about 0.29 times the step. For a digital micrometer that reads to 0.001 mm, that is about 0.29 µm.

Labs differ on whether it belongs in the calibration itself. NPL's worked micrometer budget includes the micrometer's resolution. ASME B89.1.13-2013 treats a check against the MPE as a performance test and leaves the micrometer's own resolution out of the test uncertainty, while the resolution of the reference still counts. The other rows in NPL's budget are the calibration and the deviation from nominal of the gauge blocks, repeatability, and the temperature difference between blocks and micrometer.

• TUR and TAR calculator
• Resolution uncertainty calculator
• A worked uncertainty budget for a micrometer
• 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 micrometer's ID, type, range and resolution.
• The reference standards used, including the gauge blocks, the optical flat and the optical parallels, with their certificate numbers and due dates.
• The temperature during the calibration.
• The zero, the flatness and parallelism results and the error at each block, as found and as left.
• 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 micrometer 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 micrometer or decide whether it passes. Those stay with you or your lab.

• Gage management software