Circularity

Meaning and scope

Circularity limits the radial separation of two concentric circles enclosing a section. Its form evaluation does not reference a datum. Hexagon: circularity, PC-DMIS 2023.1

Assume an ideal elliptical section with semiaxes a = 10.010 mm and b = 10.000 mm. Concentric circles centered on the ellipse enclose it with radial separation a − b = 0.010 mm, or 10 µm. Its major and minor diameters differ by 2a − 2b = 0.020 mm, or 20 µm. For this symmetric ideal ellipse, those centered circles give the minimum zone.

Circularity is a form control without a datum reference. It does not establish the section's location relative to another feature. A round section can still be eccentric to a functional shaft axis. KEYENCE categorizes roundness among datum-independent form characteristics. KEYENCE: form tolerances

Use the requirement where circular form matters to the part's function, such as a seating or rotating surface, while specifying size and other relationships separately as needed. Do not assume that a circularity result also demonstrates the required fit, orientation or alignment of the complete part.

Governing standards

Read the drawing's governing framework and edition. ASME lists Y14.5-2018 (R2024) for dimensioning and tolerancing. ISO 1101:2017, edition 4, provides the geometrical-specification language and interpretation rules. Neither title alone selects an inspection program's settings. ASME record, ISO 1101 record

For ISO roundness terminology and evaluation, ISO 12181-1:2011 addresses vocabulary and parameters and ISO 12181-2:2011 addresses specification operators. Both records give publication date April 2011, confirmation in 2021, and published status with systematic review underway at this review. Their scopes concern complete roundness profiles of integral features. Do not extend that scope to an incomplete arc without determining the applicable requirement and evaluation. ISO Part 1, ISO Part 2

Circularity versus runout and size

Circular runout evaluates variation during rotation about the specified datum axis. KEYENCE's example limits radial indicator variation at each section perpendicular to that axis; total runout addresses the entire indicated surface. These controls include a datum relationship that circularity does not. A spindle or chuck used by an instrument is therefore not automatically the drawing's datum axis. KEYENCE: runout tolerances

Keep the inspection questions distinct:

Two-point size readings alone do not establish the complete circumferential profile. KEYENCE's micrometer tutorial acknowledges that its two-point method averages opposing measurements and has accuracy limitations. Use such readings within their demonstrated capability rather than treating a maximum-minus-minimum diameter calculation as a universal roundness evaluation. KEYENCE: measuring roundness

Association and software choices

Mitutoyo distinguishes least-squares, minimum-zone, minimum-circumscribed and maximum-inscribed reference circles. Least squares minimizes squared deviations; minimum zone minimizes the radial spacing of enclosing concentric circles. The other methods use enclosing or enclosed reference circles. Record the association used rather than comparing results merely labelled “roundness.” Mitutoyo Quick Guide, printed page 60

PC-DMIS 2023.2 documents a concrete software distinction: legacy form dimensions use the feature solution, defaulting to least squares, whereas Geometric Tolerance Form commands use a min/max algorithm. Its guidance notes that the resulting values can differ. Preserve the command type and calculation settings when comparing or migrating routines. This is a bounded implementation example, not a claim about every CMM package. Hexagon: calculation list

Do not choose a different association simply to make a borderline part pass. Resolve the specified evaluation and any permitted alternatives before reporting conformity. A measurement report should identify both the requirement and the method producing its numerical result.

Sampling and filters

PC-DMIS 2023.1 explains that a cylindrical feature's circularity is evaluated by sections, with the worst measured section reported. It recommends multiple sections to improve the chance of finding the worst actual section, and illustrates how insufficient point coverage can underestimate error. No fixed point count guarantees detection of every possible form defect. Hexagon: section evaluation

Plan circumferential spacing and axial section locations against the likely process errors and functional surface. Include regions where tooling, support or transitions may change the form. Preserve the actual sampling strategy so another operator can reproduce the result. A visually smooth plot can conceal sparse sampling or display smoothing.

Mitutoyo shows that filter cutoff changes measured roundness and defines UPR as undulations per revolution. Its guide also explains centering and leveling adjustments: eccentricity and inclination can introduce measurement effects. Specify the filter type and passband, stylus arrangement and setup conditions; do not prescribe one universal UPR cutoff for all parts. Mitutoyo Quick Guide, printed pages 60–61

Original bounded calculation

This example distinguishes radial zone width from diameter variation. Halving a measured diameter range happens to give the zone width for this assumed ellipse; it is not a general method for lobed, asymmetric or incompletely sampled profiles. The calculation assumes exact geometry without noise, filtering, fixture distortion or uncertainty.

Inspection and acceptance record

Clean and support the part without introducing relevant distortion, establish the intended section orientation, and verify instrument suitability for the tolerance. Account for spindle error, probing effects and environmental conditions in the measurement method. If a thin or compliant part requires a specified restraint state, preserve that condition rather than silently changing the support.

For a useful human or agent handoff, retain feature identifier, drawing revision, governing standard, tolerance, section locations, sampling, association, filters, instrument/program version, measured result and applicable uncertainty/decision rule. Preserve the profile data when needed to resolve differing results. Investigate method differences before attributing disagreement entirely to the part.

Source scope

The Mitutoyo PDF identifies itself as Catalog E11003(7) with printing code 2408 (JP12); a more explicit technical revision date was not verified. Hexagon examples are version-specific as stated. KEYENCE's live tutorials have no verified technical revision date here. The ellipse calculation and proposed record fields are original explanatory guidance.