Flatness, Straightness, and Cylindricity: When to Use Each

Form tolerances — flatness, straightness, and cylindricity — control the shape of individual features independent of any datum, and choosing the right one determines whether your part seals, slides, or spins as intended.

Form tolerances are the foundation of GD&T’s hierarchy. They control the intrinsic geometry of a feature — how well it matches its ideal shape — without reference to any external datum plane or axis. This independence makes them both precisely defined and straightforward to apply. However, the three primary form tolerances for linear and cylindrical features — straightness, flatness, and cylindricity — are frequently confused with each other and with size tolerances. This article explains each in depth, compares their control zones, describes measurement methods, and identifies the engineering applications that call for each.

Form Tolerances: No Datum, Intrinsic Control

A fundamental property of form tolerances (straightness, flatness, circularity, cylindricity, and the profile tolerances when used without datums) is that they require no datum reference. The feature is evaluated against its own perfect geometric ideal — a perfectly flat plane, a perfectly straight line, a perfect cylinder — without regard to any external coordinate system.

This also means form tolerances interact directly with ASME Rule #1 (the envelope principle): for features of size (features with a defined size tolerance), the form error cannot exceed the size tolerance. A shaft dimensioned ∅20.0 ±0.1 cannot have a form error larger than 0.2 mm (the total size tolerance range) — otherwise, the shaft could not fit into its maximum-material envelope. Any form tolerance applied must be less than or equal to the size tolerance, or it is redundant (the size tolerance already limits the form implicitly).

Straightness: Line-by-Line Control

Straightness (symbol: —) controls how straight a line element or derived median line must be. It is the simplest form tolerance but has two distinct applications that must not be confused:

Surface straightness (line elements): Applied to a surface, straightness controls individual line elements on the surface in a specified direction. Each line element — imagine slicing the surface into infinitely thin strips — must lie within two parallel planes separated by the straightness tolerance value. Adjacent line elements are not controlled relative to each other; only individual elements are checked.

Feature of size (FOS) straightness — median line control: When the feature control frame is associated with the size dimension (placed below the size tolerance or indicated with a leader to the dimension line), it controls the derived median line of the feature. The median line is the locus of midpoints of all diametrically opposed surface elements. For a shaft, the tolerance zone is a cylinder of the stated diameter within which the entire median line must lie. This allows the shaft to be bent (bowed), but the bow must stay within the cylindrical zone.

Key distinction: surface-applied straightness requires no ∅ symbol; FOS-applied straightness typically uses ∅ and may include the Ⓜ or Ⓛ modifier for bonus tolerance. FOS straightness can actually exceed the size tolerance — it overrides Rule #1 for that specific departure from perfect form.

Measurement of straightness:

• Surface: dial indicator traversed along each line element direction; the total indicator reading (TIR) at each pass must not exceed the tolerance
• Axis (FOS): CMM probing at multiple cross-sections, calculating the median line and verifying it falls within the cylindrical tolerance zone; alternatively, a V-block and indicator setup can approximate median line bow

Practical Applications of Straightness

Straightness is most appropriate when:

• Long shafts or bars: A 500mm drive shaft may have adequate diameter uniformity but still bow enough to cause vibration or bearing wear. | — | ∅0.1 | on the shaft controls the overall bow of the median line without prescribing tighter size tolerances on the diameter.
• Slide ways and guide rails: A machine tool slide way must be straight along its travel direction. Surface straightness in the direction of travel is the appropriate control; flatness would be over-constraining.
• Extruded profiles: Aluminum or steel extrusions need straightness control along their length but not necessarily tight flatness of individual faces (which may have corrugation or texture by design).
• Threaded fasteners: Shaft straightness of bolts or studs ensures they can be inserted without interference from bowing — diameter and pitch are controlled separately by thread tolerance classes.

Flatness: Full Surface Control

Flatness (symbol: ◻) controls how flat an entire surface is — all points on the surface simultaneously. The tolerance zone is the volume between two parallel planes; the entire surface must lie within this zone. Unlike straightness, flatness considers all cross-directions of the surface at once, making it a more comprehensive and more expensive-to-achieve control.

Flatness is inherently a form control — it says nothing about where the surface is relative to any datum (that would be parallelism or perpendicularity). A surface can be perfectly flat but tilted at any angle and still satisfy the flatness requirement. If orientation relative to a datum is also required, parallelism or perpendicularity (with tighter values than flatness if needed) must be specified additionally.

Measurement of flatness:

• Surface plate and height gauge: Part rests on three support points (to eliminate rocking); a height gauge is traversed across the surface. TIR gives the flatness error. For large surfaces, repeatable support point placement is critical.
• CMM: Multiple points are probed across the surface. The CMM software calculates the minimum-zone flatness (MZFL) — the smallest separation between two parallel planes that contain all measured points. This is the true flatness error per ASME Y14.5.
• Optical flat: For very tight flatness tolerances (< 0.002 mm), optical interference bands against an optical flat provide a direct visual measurement of flatness deviation.
• Autocollimator: Scanning the surface in a grid pattern with an autocollimator and integrating tilt measurements gives flatness of precision ground surfaces and machine tool tables.

Practical Applications of Flatness

• Sealing surfaces: Gasket-sealed flanges, hydraulic manifold faces, and cylinder head surfaces need flatness to prevent leakage. A typical hydraulic manifold might specify | ◻ | 0.025 | on mating faces.
• Precision assembly interfaces: Machine tool spindle faces, CMM mounting tables, and precision fixture bases require flatness to ensure accurate part location and repeatable measurement.
• Bearing housings: The face perpendicular to a bore axis often needs flatness to ensure the bearing seats squarely without introducing angular error.
• Welded base plates: Stress-relieved weldments may warp; flatness on the base face ensures the assembly sits stably on its mounting surface.
• Semiconductor and optics: Wafer chucks, optical mounts, and precision stages require sub-micrometer flatness, achieved by lapping and verified with interferometry.

Cylindricity: The Comprehensive Cylinder Control

Cylindricity (symbol: ⌭) is the most comprehensive form control for cylindrical surfaces. The tolerance zone is the annular space between two coaxial cylinders — the entire surface of the cylinder must lie within this zone simultaneously. Cylindricity controls:

• Circularity (roundness) at every cross-section
• Straightness of every longitudinal line element
• Taper (conicity) of the overall cylinder

All three are controlled simultaneously to a single tolerance value. This makes cylindricity both the most restrictive and the most complete specification for a cylinder’s form. A part satisfying cylindricity automatically satisfies circularity and surface straightness individually.

Measurement of cylindricity:

• CMM: Probe the surface at multiple axial positions and multiple angular positions (minimum 4 points per section, minimum 3 sections recommended). The software fits a cylinder and computes the minimum-zone cylindricity error — the radial distance between the two best-fit coaxial cylinders containing all measured points.
• Cylindricity measuring instrument (roundness tester with axial traverse): A precision spindle rotates the part while a probe traverses axially, producing a full surface map. This is the most accurate method for tight cylindricity requirements.
• V-block + indicator: Rotate the part in a V-block while traversing a dial indicator axially. This is an approximation (V-block introduces errors depending on number of lobes), not a true cylindricity measurement — it conflates circularity, straightness, and taper.

Practical Applications of Cylindricity

• Precision bearing bores and journals: A radial ball bearing requires both roundness and straightness of its bore or journal; cylindricity ensures both. | ⌭ | 0.005 | on a journal bearing surface is a typical tight specification.
• Hydraulic cylinder bores: Piston seal life and leakage are directly affected by cylindricity of the bore. Deviation from true cylinder causes uneven seal loading and wear.
• Precision shafts: Grinding a precision shaft to cylindricity ensures even contact with bushings or bearings along the full contact length.
• Gauge pins and plug gauges: Calibration-grade gauge cylinders specify cylindricity in the micrometer range to ensure accurate measurement of the holes they check.

Comparison: Straightness vs Flatness vs Cylindricity

Property	Straightness	Flatness	Cylindricity
Applied to	Lines, axes (FOS)	Planar surfaces	Cylindrical surfaces
Tolerance zone	Two parallel planes (or cylinder for FOS)	Two parallel planes	Two coaxial cylinders
Controls	Individual line elements OR median axis	Entire surface simultaneously	Entire cylindrical surface simultaneously
Datum required	No	No	No
Modifier allowed	Yes (on FOS only)	No	No
Measurement complexity	Low–medium	Medium	High
Typical tolerance range	0.01–0.5 mm	0.005–0.5 mm	0.002–0.05 mm

Interaction with Size Tolerance (Rule #1)

ASME Y14.5 Rule #1 (the envelope principle) states that the actual surface of a feature of size must not violate the MMC (maximum material condition) envelope — the perfect form at MMC size. This means form error is automatically limited to the size tolerance unless a specific form tolerance is called out. When the size tolerance already provides adequate form control, no explicit form tolerance is needed.

An explicit form tolerance is needed when the form requirement is tighter than what the size tolerance allows — which is always the case for precision functional surfaces. For example, a shaft might be ∅30.0 ±0.05 (size tolerance = 0.1 mm), but the cylindricity requirement for the bearing journal might be | ⌭ | 0.005 | — 20 times tighter than what Rule #1 alone would permit.

Conclusion

Flatness, straightness, and cylindricity address different geometric realities and different functional needs. Straightness is efficient for long prismatic or cylindrical features where one-dimensional deviation is the concern. Flatness is the definitive control for sealing and interface surfaces where the entire area must be within a planar zone. Cylindricity is the comprehensive choice when a cylindrical surface’s full geometry — roundness, straightness, and taper together — must be tightly controlled for bearing fits, sealing, or precision location. Matching the right form tolerance to the specific functional requirement avoids both over-tolerancing (which increases cost) and under-tolerancing (which produces functional failures).