Physical Address
304 North Cardinal St.
Dorchester Center, MA 02124
ASME B46.1-2009 — Surface Texture (Surface Roughness, Waviness, and Lay)
🔬 ASME B46.1-2009 (Surface Texture — Surface Roughness, Waviness, and Lay, revision of the 2002 edition) is the governing American National Standard for the terminology, parameters, and measurement of surface texture, covering surface roughness, waviness, and lay. Maintained by the ASME B46 Committee, it is harmonized with the ISO 4287/4288 and ISO 25178 (areal surface) standards. The main body is organized into seven sections and appendices: Section 1 defines the terms related to surface texture, including profile parameters and area-profiling/area-averaging parameters; Section 2 gives a classification scheme for surface-texture measuring instruments; Sections 3 and 4 set out measurement procedures for contact, skidless and skidded profilometers; Sections 5 and 6 cover area-profiling and area-averaging measurement techniques; and Section 7 addresses nanometer surface-texture and step-height measurements by stylus profiling instruments. The Standard also defines the measurement foundations of sampling length, evaluation length, and filtering (cutoff).
💡 Surface texture is not a single index: roughness is the fine unevenness superimposed on waviness, which itself sits on top of form error. B46.1 separates them in the frequency domain with sampling length, evaluation length, and filtering, parameterizing each layer — that is the heart of the measurement methodology.
1. Scope and Terminology 📌
This Standard applies to the characterization of engineered surfaces, covering:
- Roughness: the closely spaced, shallow irregularities produced by the machining method (cutting, grinding, polishing, etc.), described by parameters evaluated within a sampling length;
- Waviness: periodic or non-periodic undulations with spacing greater than roughness, caused by machine vibration, tool deflection, and similar effects;
- Lay: the direction of the predominant surface pattern (parallel, perpendicular, crosshatched, multidirectional, circular, radial, etc.).
Section 1 systematically defines terms: surface profile, sampling length, evaluation length, filter cutoff wavelength, profile filters (such as Gaussian), and parameters Ra, Rq, Rz, Rt, Rsm, Rsk, Rku, and extends into areal profiling and areal averaging parameters such as Sa and Sq, matching three-dimensional surface measurement concepts (ISO 25178 family).
⚠️ Ra is only an “average roughness.” It cannot tell a spiky grooved surface from a plateau surface — two surfaces with the same Ra can function completely differently (a seal face versus a sliding face). Critical functional surfaces should also specify Rz (maximum height) and even Rsk/Rku (skewness/kurtosis) as supplementary parameters.
2. Instrument Classification and Parameter Definitions 🔧
Section 2 classifies surface-texture measuring instruments into three broad classes:
- Profiling instruments: record the surface profile along a line (contact stylus profilometers, non-contact optical profilers), outputting two-dimensional profiles and parameters such as Ra, Rz;
- Areal profiling / areal averaging instruments: record or statistically sample surface information over an area, yielding areal parameters Sa, Sq or equivalent area-averaged values;
- Specialized instruments: atomic force microscopes, confocal laser microscopes, and similar for nanometer-scale measurement; Section 7 specifies calibration requirements for step-height and surface-texture standards for stylus instruments.
| Parameter | Meaning | Units | Purpose |
|---|---|---|---|
| Ra | Arithmetic average deviation (mean of absolute profile deviation from the mean line) | µm / µin | Most common general roughness index |
| Rq | Root-mean-square deviation (RMS) | µm / µin | More statistically meaningful; precision analysis |
| Rz | Maximum height (largest peak-to-valley within a sampling length) | µm / µin | Controls damaging peaks; seals/fits |
| Rt | Total height over the evaluation length (full peak-valley range) | µm / µin | Controls deep grooves and burrs |
| Rsm | Mean width of profile elements (texture spacing) | mm | Assesses texture density and directionality |
| Rsk / Rku | Skewness / Kurtosis | dimensionless | Distinguishes spike surfaces from valley/plateau surfaces |
💡 A useful rule of thumb: Ra ≈ Rz / 4–5 statistically. But this is empirical and cannot replace direct measurement — acceptance must be based on the parameters and conditions the Standard defines, to avoid disputes over conversions.
3. Measurement Procedures and Filtering 📏
Sections 3 and 4 prescribe contact-profilometer procedures, distinguishing skidless from skidded instruments by how the reference datum is established: skidless instruments use a precision reference guideway and capture the full profile including waviness; skidded instruments ride a skid on the surface and are used mainly for roughness. Measurement essentials:
- Sampling length and cutoff: choose the sampling length from the expected Ra (e.g., 0.08 mm for Ra 0.02–0.2 µm, 0.25 mm for Ra 0.2–2 µm, 0.8 mm for Ra 2–10 µm); the evaluation length is typically five sampling lengths;
- Filtering: separate roughness from waviness with a standard cutoff; the Gaussian filter is the recommended first choice;
- Measuring direction: measure perpendicular to the predominant lay to capture the maximum roughness;
- Repeatability and location: take multiple traces at different positions on the part and report the average or the worst value.
🚨 Data trap: comparing “roughness” without specifying the filter is meaningless — the same surface yields very different Ra values under different cutoff wavelengths. The drawing, the inspection lab, and the supplier must agree on the cutoff and sampling length; otherwise acceptance is guaranteed to become an argument.
4. Engineering Practice Essentials 🛠️
- 🔍 Specification symbols: use the Standard’s symbols on drawings — Ra value plus sampling length/filter, and lay symbols (‖, ⊥, ×, M, C, R, P) where direction matters; never write a bare Ra number alone.
- 🔧 Parameter by function: seal faces control Rz/Rt to stop leak paths; sliding surfaces control Ra/Rq and Rsk (avoid spikes); base roughness before coatings/plating controls adhesion; bearing surfaces also consider waviness.
- 📊 Process verification: combine in-process sampling and SPC; watch surface texture for signs of grinding burn and tool wear; full parameter measurement on first articles, Ra sampling on production lots.
- 📋 Calibration and traceability: calibrate profilometers periodically with standard roughness specimens (step-height standards, Ra reference blocks); measurement reports must state instrument, cutoff, sampling length, and direction.
5. Frequently Asked Questions ❓
❓ Q: How should Ra and Rz be used together?
A: Ra gives the overall average level while Rz captures the worst-case peaks and valleys. For seal faces, press fits, and other functional surfaces, specify both so that a surface with a good average but a few fatal spikes cannot slip through.
A: Ra gives the overall average level while Rz captures the worst-case peaks and valleys. For seal faces, press fits, and other functional surfaces, specify both so that a surface with a good average but a few fatal spikes cannot slip through.
❓ Q: Why must the measurement direction be perpendicular to the lay?
A: Measured perpendicular to the lay, the stylus crosses all peaks and valleys and returns the maximum, most truthful Ra; measuring parallel to the lay gives a lower value that does not represent the surface’s actual quality.
A: Measured perpendicular to the lay, the stylus crosses all peaks and valleys and returns the maximum, most truthful Ra; measuring parallel to the lay gives a lower value that does not represent the surface’s actual quality.
❓ Q: How are roughness and waviness separated?
A: By the filter cutoff: wavelengths shorter than the cutoff (high frequency) are treated as roughness; undulations longer than the cutoff (low frequency) are waviness. By convention the cutoff approximates the sampling length; exact values are read from the Standard’s tables.
A: By the filter cutoff: wavelengths shorter than the cutoff (high frequency) are treated as roughness; undulations longer than the cutoff (low frequency) are waviness. By convention the cutoff approximates the sampling length; exact values are read from the Standard’s tables.
🔍 Q: How does B46.1 relate to ISO 4287?
A: The two are mutually coordinated in terminology, parameter definitions, and measurement procedures; Ra/Rz/Rq and sampling/evaluation length rules essentially agree, and B46.1 additionally incorporates areal parameters (matching ISO 25178). For international projects, cite either system per the contract — values are mutually recognized.
A: The two are mutually coordinated in terminology, parameter definitions, and measurement procedures; Ra/Rz/Rq and sampling/evaluation length rules essentially agree, and B46.1 additionally incorporates areal parameters (matching ISO 25178). For international projects, cite either system per the contract — values are mutually recognized.
