How often should gauge blocks be calibrated?

No standard sets one interval. CEM, the Spanish national metrology institute, suggests one to two years for gauge blocks in normal use and as little as six months for heavy use. Adjust the interval from how much each block changes between calibrations and how often the set is used.

What are the gauge block grades?

ISO 3650 has a calibration grade K and grades 0, 1 and 2, with grade 2 the widest limits. The US standard ASME B89.1.9 uses grades K, 00, 0, AS-1 and AS-2. Each grade limits how far the length may be from nominal and how much it may vary across the face.

What is the difference between grade K and grade 0 gauge blocks?

Grade K blocks have the tightest limit on variation in length, but their length may be as far from nominal as a grade 1 block. They are reference blocks meant to be used at their calibrated lengths. Grade 0 blocks are held closer to their nominal size, so they suit precise work at nominal.

How are gauge blocks calibrated?

Most are compared on a gauge block comparator with a reference block of the same nominal size, at the centre and the four corners of the face. Grade K reference blocks are usually measured by interferometry against the wavelength of light. In both cases the results are stated at 20 °C.

Why are gauge blocks calibrated at 20 °C?

ISO 1 and ISO 3650 define lengths at 20 °C. Gauge block steel expands by about 11.5 µm per metre for each degree, so a 100 mm block is about 1.15 µm longer at 21 °C than at 20 °C. That is nearly four times the grade 0 limit for that size.

Is it gage block or gauge block?

Both name the same thing. Gage block is the common US spelling and the one ASME uses in B89.1.9, while ISO 3650 writes gauge block. NPL's guide notes that they are also called slip gauges.

Compliance and audits

Gauge block calibration: grades, methods and how often to calibrate

13 min read Published November 8, 2026

Gauge block calibration measures the central length of each block and how much its length varies across the measuring face, usually by comparing it on a gauge block comparator with a reference block of the same nominal size, or by interferometry for the best reference blocks. The results are judged against the limits for the block's grade in ISO 3650 or ASME B89.1.9 at the reference temperature of 20 °C.

On this page

What a gauge block calibration measures

A gauge block is a length standard with two flat, parallel measuring faces. ISO 3650 defines its length in a particular way. It is the distance from a point on the free face to a flat plate of the same material and surface texture that is wrung to the other face. The length therefore includes one wringing film. That is why blocks can be stacked without a correction for the films between them.

A calibration reports two things for each block. The first is the deviation of the central length from the nominal size, measured at the centre of the free face. The second is the variation in length, the difference between the longest and the shortest point across the face. It shows how flat the faces are and how they sit relative to each other.

A certificate can give the measured length of each block, a statement that the block meets its grade, or both. NIST's Gauge Block Handbook notes that NIST reports the measured length and does not state a grade. A block you use at its measured length needs those certificate values. A block you use at its nominal size needs to be inside its grade.

Calibration vs verification

Gauge block grades in ISO 3650 and ASME B89.1.9

ISO 3650:1998 is the international standard for gauge blocks of rectangular section from 0.5 mm to 1,000 mm. ISO last reviewed and confirmed it in 2025, so the 1998 edition is current. It sets limits for a calibration grade K and for grades 0, 1 and 2.

In the US the spelling gage block is common, and it is the one ASME uses. ASME B89.1.9-2023, Gage Blocks, sets limits for a calibration grade K and for grades 00, 0, AS-1 and AS-2.

Each ISO grade has two length limits. The limit deviation te is how far the length at any point may be from the nominal size. The tolerance tv caps the variation in length across the face. Grade K stands out. It has the tightest limit on variation in length, but its limit on deviation from nominal is as wide as grade 1. K blocks are reference blocks used at their calibrated lengths, so flat, parallel faces matter more for them than closeness to nominal. NPL's good practice guide on gauge blocks says grade K is mainly used to calibrate other blocks on a comparator, while grade 2 blocks are found in the workshop.

ISO 3650 length limits for gauge blocks up to 100 mm, in µm. te is the limit deviation of the length at any point from the nominal length. tv is the tolerance for the variation in length.
Nominal length	Grade K: te / tv	Grade 0: te / tv	Grade 1: te / tv	Grade 2: te / tv
0.5 mm to 10 mm	±0.20 / 0.05	±0.12 / 0.10	±0.20 / 0.16	±0.45 / 0.30
Over 10 mm to 25 mm	±0.30 / 0.05	±0.14 / 0.10	±0.30 / 0.16	±0.60 / 0.30
Over 25 mm to 50 mm	±0.40 / 0.06	±0.20 / 0.10	±0.40 / 0.18	±0.80 / 0.30
Over 50 mm to 75 mm	±0.50 / 0.06	±0.25 / 0.12	±0.50 / 0.18	±1.00 / 0.35
Over 75 mm to 100 mm	±0.60 / 0.07	±0.30 / 0.12	±0.60 / 0.20	±1.20 / 0.35

Tip: The values are from ISO 3650 as given in the gauge block procedure DI-014 of CEM, the Spanish national metrology institute. ISO 3650 also sets flatness tolerances for blocks up to 100 mm: 0.05 µm for grade K, 0.10 µm for grade 0, 0.15 µm for grade 1 and 0.25 µm for grade 2. If you use each block at its calibrated length rather than its nominal size, the grade limit drops out. The certificate's uncertainty and any drift since the calibration are what count.

Comparison or interferometry?

Interferometry measures a block against the wavelength of light. The block is wrung to a flat platen, and light reflected from its top face and from the platen is compared with light from a reference mirror. It gives the smallest uncertainty and is the usual method for grade K reference blocks. NPL notes that it starts from an approximate length found on a comparator.

Mechanical comparison is the usual method for grades 0, 1 and 2. A gauge block comparator measures the difference between the block and a reference block of the same nominal size. The length of the reference plus that difference gives the length of the block. The reference has to be of the same grade or better. CEM's procedure, for example, calibrates grade 0 blocks against grade K and grade 1 blocks against grade 0 or K.

ISO 3650 ties the two methods together. A block's length is traceable when an unbroken chain of comparisons, each with a stated uncertainty, leads back to a block calibrated by interferometry.

Compare blocks of the same material where you can. If the reference is steel and the block is ceramic or carbide, the probes press into the two materials by different amounts and the two expand differently with temperature. The result then needs a correction for each. NIST's comparator program keeps steel masters for steel blocks and chrome carbide masters for chrome carbide blocks, so in its practice only tungsten carbide blocks need a deformation correction.

Metrological traceability explained

How a comparator calibration is done

Set the comparator to zero on the centre of the reference block, then measure the test block at its centre. The difference is the deviation of the test block from the reference.

Repeat the readings in a pattern that cancels drift, such as reference, test, reference. CEM's procedure takes five readings in that pattern, then turns the block over so that the other face is up and takes five more. NIST uses drift-eliminating designs that compare several blocks in a set order.

For the variation in length, measure the centre and the four corners, about 1.5 mm in from the side faces. Both the DKD guideline for gauge blocks (DKD-R 4-3 Blatt 3.1, published by PTB) and CEM's procedure use these five points. The highest and lowest readings give the variation in length.

Position matters too. ISO 3650 defines the length of blocks up to 100 mm with the block standing vertical on a measuring face. Longer blocks are defined lying on a narrow side face, supported at 0.211 times the nominal length from each end. Record how each block was supported.

Tip: The DKD guideline scales the work to how the blocks are used. A reference standard of the highest accuracy gets a wringing test of both faces and at least three probings of the central length. A working standard gets a flatness check of both faces and at least two probings.

Flatness, wringing and the condition of the faces

Look at each face before you measure. Burrs, scratches, chips and rust stop a block from wringing and can damage the blocks it touches. NPL's guide removes a small burr with an Arkansas stone held flat on the face. A block that stoning cannot fix goes back to the maker for re-lapping or is replaced, and a replacement needs its own calibration.

Check flatness with an optical flat. Under light of one colour the interference fringes act as contour lines, each half a wavelength apart in height. Straight, evenly spaced fringes mean a flat face, and curved fringes mean a convex or concave one. Thin blocks bend, so CEM checks blocks of 2.5 mm and under while they are wrung to a thicker base.

The wringing test says the most about a face. Wring the face to an optical flat and look at it through the glass. A good wring looks an even grey. Bright or coloured patches show where the face does not touch. The DKD guideline treats this test as the surest guide to the condition of the faces and requires it for the smallest uncertainties.

Tip: To wring two blocks, NPL describes holding them crosswise and sliding or twisting them together until they grip, then turning them into line. Keep the faces clean and work low over a cloth. NPL also advises against leaving blocks wrung together for more than 24 hours, because they can become hard to separate.

Temperature and the 20 °C reference

Gauge block lengths apply at 20 °C, the standard reference temperature in ISO 1:2022 and in ISO 3650. Gauge block steel expands by about 11.5 × 10⁻⁶ per °C. NIST's handbook works the example: a 100 mm steel block grows by about 1.15 µm for each degree. That is nearly four times the grade 0 limit of ±0.30 µm at that length.

So the test block and the reference must be at the same temperature when they are compared, and that takes time. As a guide, CEM's procedure gives about 1 hour of soaking for blocks up to 5 mm and about 6 hours for blocks over 50 mm up to 100 mm. Blocks of different materials need a correction as well. NIST puts the expansion of tungsten carbide at about one third of steel and chrome carbide at about two thirds.

The room matters too. CEM compares grade 0 blocks with grade K, and grade 1 with grade 0, in a room held at 20 °C ± 0.5 °C, and uses 20 °C ± 1 °C for the other grades. Handle blocks with gloves or tongs so the heat of your hands does not reach them.

Uncertainty and what limits it

The GUM works through this case in its first example (JCGM 100:2008, H.1). A 50 mm block is compared with a reference of the same nominal length, and the difference is the mean of five readings. The combined standard uncertainty comes to 32 nm, before any coverage factor. The largest part, 25 nm, is the uncertainty of the reference block's own calibration.

The DKD guideline lists the usual contributions. The setup brings the calibration and drift of the reference, the comparator reading, the scatter of repeated readings and the resolution. The block brings the condition of its faces, its variation in length and its expansion coefficient. The room brings the temperature measurement, the departure from 20 °C and any temperature difference between the two blocks.

Tip: Gauge blocks are the reference for other instruments, so their uncertainty carries down the chain. Before a block calibrates a caliper or a micrometer, compare its uncertainty with that instrument's tolerance. A test uncertainty ratio (TUR) puts a number on it.

When a block is out of grade or damaged

When some blocks in a set measure outside their grade, CEM's procedure gives two options. Replace those blocks, or downgrade the whole set if many of them fail. NPL marks a block that is no longer fit as a length standard by cutting off a corner, and it can still serve as a packing block in a jig.

A gauge block is usually a reference standard, so an out-of-tolerance result reaches past the block itself. The instruments it calibrated since its last good calibration may carry its error. ISO 9001 clause 7.1.5.2 asks you to determine whether earlier results were adversely affected and to act as needed. Start by comparing the size of the error with the tolerance of each instrument the block was used to check.

How often to calibrate gauge blocks

No standard sets one interval for gauge blocks. ISO 9001 asks for calibration or verification at specified intervals, or before use. ILAC-G24 and OIML D 10 (2022 edition) list the factors to weigh, among them wear and drift, how much and how hard the equipment is used, the environment and its calibration history.

CEM's gauge block procedure suggests one to two years as a reasonable period. The choice depends mainly on the stability seen in past calibrations and on how often the blocks are used, and heavy use can bring it down to six months. CEM also advises a shorter interval after a large change between calibrations, or after a serious incident such as a repaired edge or misuse.

NIST's handbook observed that most of its customers sent master blocks back every two years. It judged that schedule probably adequate, since most masters see light use in clean, dry rooms. NPL adds that new blocks, and blocks with an unknown history, should be calibrated more often than blocks known to be stable.

Each calibration also shows how much each block changed since the last one. Track that drift block by block. A set whose blocks barely move supports a longer interval.

What the certificate and record should show

Whether the blocks go to a lab or are compared in-house, the record should let someone else follow what was done.

The set or block ID, the material, the grade claimed and the nominal size of each block.
The reference blocks or the interferometer used, with their certificates and due dates.
For each block, the deviation of the central length from nominal and the variation in length.
How the faces were checked, for example with a wringing test or an optical flat, and what was found.
The position of the blocks during the calibration and the temperature.
The uncertainty with its coverage factor, and the decision rule behind any statement that a block meets its grade.
The date, the lab or person, and the next due date.

How to calibrate gauge blocks, step by step

Check that the reference blocks and the comparator have current calibrations and that the room is close to 20 °C.
Clean the blocks with a suitable solvent and a lint-free cloth, and inspect each face for burrs, scratches and rust.
Stone off small burrs, and set aside blocks that need re-lapping or replacement.
Check the flatness of each face with an optical flat, and wring each face to it to judge its condition.
Let the test blocks and the reference blocks soak on the comparator until they reach the same temperature.
Zero on the centre of the reference block and measure the test block at its centre in a pattern such as reference, test, reference. Turn the block over and repeat.
Measure the four corners to find the variation in length.
Correct for any difference in material, then work out each block's deviation from its nominal size and its uncertainty.
Compare the results with the limits for the grade using your documented decision rule, and replace or downgrade blocks that fail.
Record the results, the conditions and the next due date, and update the certificate values wherever the blocks are used at their calibrated lengths.

How Axiospec helps with gauge blocks

Axiospec is calibration management software. It does not calibrate gauge blocks. On an in-house calibration, a final reading outside its tolerance sets the result to Fail, and recording a pass anyway takes a written reason. The tolerance, the decision rule and the interval stay with you or your lab. Axiospec keeps the record. A gauge block set can be tracked as one instrument and marked as a reference standard, with the certificate attached to each calibration and the due date on the register.

When someone logs a calibration of a caliper or a micrometer, they pick the gauge block set from the register as the reference standard. If the set has failed or is out for calibration, the calibration sheet warns them to check it is fit for use. The set's Reverse Traceability card lists the instruments it was used to calibrate. If the set itself fails, a calibration logged as Fail puts it in quarantine and the failed record asks for the out-of-tolerance impact. The live demo has a grade 0 gauge block set with its Reverse Traceability card.

The free plan covers up to 50 active instruments. A new workspace starts with a 45-day trial of Professional, with room for 1,500 active instruments and no credit card. We load your existing records for you at no charge on every plan.

Common questions

How often should gauge blocks be calibrated?: No standard sets one interval. CEM, the Spanish national metrology institute, suggests one to two years for gauge blocks in normal use and as little as six months for heavy use. Adjust the interval from how much each block changes between calibrations and how often the set is used.
What are the gauge block grades?: ISO 3650 has a calibration grade K and grades 0, 1 and 2, with grade 2 the widest limits. The US standard ASME B89.1.9 uses grades K, 00, 0, AS-1 and AS-2. Each grade limits how far the length may be from nominal and how much it may vary across the face.
What is the difference between grade K and grade 0 gauge blocks?: Grade K blocks have the tightest limit on variation in length, but their length may be as far from nominal as a grade 1 block. They are reference blocks meant to be used at their calibrated lengths. Grade 0 blocks are held closer to their nominal size, so they suit precise work at nominal.
How are gauge blocks calibrated?: Most are compared on a gauge block comparator with a reference block of the same nominal size, at the centre and the four corners of the face. Grade K reference blocks are usually measured by interferometry against the wavelength of light. In both cases the results are stated at 20 °C.
Why are gauge blocks calibrated at 20 °C?: ISO 1 and ISO 3650 define lengths at 20 °C. Gauge block steel expands by about 11.5 µm per metre for each degree, so a 100 mm block is about 1.15 µm longer at 21 °C than at 20 °C. That is nearly four times the grade 0 limit for that size.
Is it gage block or gauge block?: Both name the same thing. Gage block is the common US spelling and the one ASME uses in B89.1.9, while ISO 3650 writes gauge block. NPL's guide notes that they are also called slip gauges.

Put it into practice

Got a gage list? Send it over and we load it for you, usually in a couple of business days. Free on every plan. Then log every calibration to a tamper-evident audit trail and produce records on demand.

Create your account

Axiospec is a documentation and workflow tool. It helps you keep clean, traceable, audit-ready records. Certification depends on your own processes, scope and assessor.