Total runout

Total runout evaluates the entire specified surface relative to a datum axis. Keep variation between axial or radial stations in the result. KEYENCE, run-out tolerances

In the ideal shaft example, every section has 0.008 mm circular runout while total radial runout is 0.028 mm, exceeding a hypothetical 0.020 mm total-runout limit. A sectional pass would require its own specified limit. The coaxial envelope’s diameter difference is 0.056 mm. The example assumes the complete modeled surface and an ideal datum, with instrument error excluded.

Establish the requirement and datum axis

Read the drawing or model's governing standard, edition, units, controlled surface and extent, datum reference sequence, and supplementary requirements. ASME identifies Y14.5-2018 (R2024); ISO identifies ISO 1101:2017, edition 4, confirmed in 2022. Reaffirmation and confirmation dates are not new technical editions. These catalog records identify the publications; use the specified standard for detailed interpretation rather than transferring defaults between systems. ASME Y14.5, ISO 1101:2017

The datum axis must be established from the designated datum features. It is not simply the axis produced by whichever chuck, centers, or rollers happen to support the part. Confirm that the fixture or computational construction represents the required datum reference, including combined datum features where specified. A measurement relative to an unrelated manufacturing axis answers a different question.

Hexagon's PC-DMIS 2023.2 datum guide distinguishes constrained simulator fitting from plain least-squares fitting and documents different ASME and ISO implementations. Record the datum construction and math settings; fitting the controlled surface to itself can remove the very datum relationship being inspected. Hexagon, datum construction

Radial and axial applications

For a nominal cylindrical surface, the zone is between two coaxial cylinders whose common axis is the datum axis. The tolerance is their radial separation, not the difference between their diameters. For a nominal face perpendicular to the datum axis, the zone is between two parallel planes perpendicular to that axis. The face application therefore evaluates axial height variation across the specified face. PC-DMIS 2023.2 describes both constructions; its help is dated August 30, 2023. Hexagon, cylindrical and planar features

GD&T Basics' September 9, 2026 article describes radial indicator alignment on cylinders and axial inspection of perpendicular faces. Its fixture examples are possible arrangements, not proof that any V-block or chuck setup faithfully realizes a particular datum. The inspection direction and controlled surface must be stated together: “radial and axial runout” is not one universal combined scalar measurement. GD&T Basics, total runout

Whole-feature variation is not a sectional average

GD&T Basics' September 14, 2020 comparison explains that circular sections are assessed independently, while total runout tracks the maximum and minimum across the complete surface. In a mechanical check, preserve a common reading reference while moving along the feature and rotating it. Re-zeroing each station discards the differences between stations; averaging their runouts also loses the required whole-feature envelope. GD&T Basics, circular versus total runout

A coaxial taper illustrates the distinction: each circular section can be perfectly round and centered, yet the radius changes along the length. A similar distinction occurs on a dished face whose height varies with radius but remains constant around each circular track. Neither surface is established acceptable merely because each individual rotation produces little indicator movement.

Form, orientation, and size remain different questions

On a cylindrical feature, a datum-centered whole-surface envelope limits combined radial effects from surface form and its relationship to the datum axis. On a face, the perpendicular bounding planes limit both face form and orientation. GD&T Basics discusses these combined effects; a single runout result does not identify which manufacturing error caused them. GD&T Basics, relationship to other controls

Diameter acceptance still needs its own size requirement. As an original geometric observation, adding the same constant to every radius changes diameter without changing the maximum-minus-minimum radius range. A perfectly coaxial, constant-radius cylinder can have zero mathematical runout while being too large or too small. Likewise, face runout does not alone establish the face's axial distance from another feature. Check those separate dimensions rather than interpreting runout as a complete fit specification.

Practical inspection and software checks

Plan circumferential and longitudinal coverage around likely process errors and damage. Retain the acquired coverage: several rings leave intervening regions unmeasured. PC-DMIS illustrates how sparse points can understate the actual value. Hexagon, actual versus measured value

For an indicator setup, assess seating, support deformation, spindle error, probe contact direction, and traverse straightness as measurement influences. Establish a repeatable contact force and support condition for slender or flexible parts. An indicator trace includes setup effects; do not attribute its full variation to the workpiece without considering measurement uncertainty.

Confirm which software command is being used. PC-DMIS 2023.2's older Dimensioning Runout help says that selecting only one feature uses the current origin and workplane as the datum reference, whereas selecting a second feature uses that feature as the datum. This is a documented command behavior, not permission to substitute an arbitrary origin for a drawing datum. Review the saved reference and report construction before relying on the number. Hexagon, legacy dimensioning runout

Original calculation: local versus global range

Define an illustrative cylindrical surface by its radial distance from an ideal datum axis:

r(z, θ) = 10.000 + 0.0002 z + 0.004 cos θ millimetres,

where 0 ≤ z ≤ 100 mm and 0 ≤ θ < 2π. This is a prescribed radial surface model, not a claim that every eccentric shaft follows this equation.

At any fixed station, the radial range is (+0.004) − (−0.004) = 0.008 mm. Across the whole feature, however:

• Global minimum: 10.000 − 0.004 = 9.996 mm at z = 0.
• Global maximum: 10.000 + 0.020 + 0.004 = 10.024 mm at z = 100.
• Total radial runout: 10.024 − 9.996 = 0.028 mm.

Thus a hypothetical 0.020 mm total-runout requirement fails, even though every individual section has only 0.008 mm circular runout. The coaxial envelope's diameter difference is 0.056 mm; confusing that with the radial separation doubles the reported value. These calculations assume the complete defined surface and an ideal datum, without instrument error or a conformity decision rule.

For a separate ideal face example, let axial height be h(r) = 0.00002 r² mm over radii 10–30 mm. Every circular track has zero axial variation during rotation, but the whole-face height range is 0.00002 × (30² − 10²) = 0.016 mm. Radial traversal is necessary to reveal that dish.

Information to retain

Retain the controlled extent, datum construction, measurement direction, angular/axial or radial coverage, common reading reference, fitting and filtering settings, software command/version, support condition, and applicable uncertainty/decision rule. Report sectional results separately when diagnosing a total-runout failure; do not replace the required global result with their maximum or average.

Related reference

Dimensions, tolerances and specifications