Machined surface roughness inspection is a critical quality control process that measures and verifies the micro-geometric texture of machined components, directly influencing friction, wear, fatigue life, sealing performance, and coating adhesion. This guide provides a comprehensive overview of surface roughness parameters such as Ra and Rz, sampling requirements, stylus and optical measurement methods, applicable ISO and GB standards, and typical industrial applications.
Technical Overview
Surface roughness refers to the fine, closely spaced irregularities introduced on a workpiece surface by the machining process itself, including feed marks from the cutting tool, built-up edge fragments, vibration-induced chatter marks, and plowing of the material. It is distinguished from waviness and geometric form error through standardized profile filtering, where the selected cutoff wavelength separates the roughness profile from longer-wavelength components.
In machined components, surface texture governs functional behavior such as friction coefficient, lubricant retention, running-in wear, contact stiffness, fatigue crack initiation, corrosion resistance, sealing effectiveness, and adhesion of coatings. Consequently, machined surface roughness inspection is an essential element of manufacturing quality assurance, ensuring that turned, milled, ground, drilled, bored, reamed, lapped, and electrical discharge machined (EDM) surfaces conform to drawing specifications.
The inspection activity is governed by an established framework of standards, principally the ISO 21920 series, which supersedes and consolidates the classic profile-based standards ISO 4287 (terminology and parameters), ISO 4288 (rules and procedures for assessment), and ISO 1302 (drawing indication). Supporting standards include ISO 3274 for stylus instrument characteristics, ISO 12179 for instrument calibration, ISO 25178 for areal (three-dimensional) surface texture, and the corresponding Chinese national standards such as GB/T 3505, GB/T 1031, and GB/T 10610. JIS B 0601 is referenced for certain legacy parameters used in Japanese supply chains.
Testing Samples
Components Grouped by Machining Process
Surface roughness inspection can be performed on virtually any machined surface, and the expected texture range depends strongly on the generating process:
- Turning and facing: typical Ra from 0.8 µm to 6.3 µm, with directional feed marks
- End milling and peripheral milling: typical Ra from 0.8 µm to 3.2 µm
- Drilling and boring: typical Ra from 1.6 µm to 6.3 µm on bore walls
- Reaming and broaching: typical Ra from 0.4 µm to 1.6 µm
- Cylindrical and surface grinding: typical Ra from 0.1 µm to 0.8 µm
- Honing and lapping: typical Ra from 0.025 µm to 0.4 µm
- Polishing and superfinishing: typical Ra from 0.008 µm to 0.1 µm
- EDM (sinking and wire): typical Ra from 0.4 µm to 10 µm, isotropic non-directional texture
Material Groups
Inspection is applicable to components manufactured from carbon and alloy steels, stainless steels, cast irons, aluminum and magnesium alloys, copper alloys, titanium alloys, nickel-based superalloys, hardened tool steels, engineering polymers, ceramics, and metal matrix composites. The measurement technique must be matched to the material, since soft or delicate surfaces may require non-contact optical methods to avoid stylus scratching.
Sample Preparation and Conditioning
- Clean the surface to remove cutting fluid, oil, grease, swarf, and corrosion inhibitors using solvents that do not alter the texture
- Protect surfaces from handling damage, oxidation, and airborne contamination prior to measurement
- Condition samples at standard laboratory temperature, conventionally 20 °C, to minimize thermal expansion effects on the measurement
- For small series production, witness coupons machined under identical parameters may be submitted when the actual part geometry restricts access
- Identify the measurement location and evaluation direction, which should generally be perpendicular to the lay of the machining marks unless otherwise specified
Testing Items
The inspection scope covers the full set of profile-based roughness parameters defined in ISO 21920-1 and ISO 4287, together with waviness and primary profile parameters where required:
- Ra: arithmetic mean deviation of the assessed profile, the most widely specified parameter
- Rq: root mean square deviation of the profile, more sensitive to extreme peaks and valleys
- Rz: maximum height of the roughness profile within a sampling length; legacy ten-point height definitions are noted for JIS-based drawings
- Rt: total height of the roughness profile over the evaluation length
- Rp: maximum peak height and Rv: maximum valley depth within a sampling length
- Rsk: skewness of the profile, indicating whether the texture is peak- or valley-dominated
- Rku: kurtosis of the profile, describing the sharpness of the height distribution
- Rsm: mean spacing of profile irregularities, relevant for lubrication and sealing behavior
- Rmr(c): material ratio curve parameters, used for bearing-area and running-in evaluation
- Wa, Wz: waviness parameters after roughness filtering, useful for detecting machine tool vibration issues
- Sa, Sq, Sz: areal (3D) parameters per ISO 25178, increasingly specified for critical functional surfaces
Testing Methods
Stylus Contact Profilometry
The reference method for machined surface roughness inspection is skidless stylus profilometry in accordance with ISO 3274 and the assessment rules of ISO 21920 or ISO 4288. A diamond-tipped stylus, typically with a 2 µm or 5 µm tip radius, is traversed across the surface at constant speed, and vertical displacement is converted into a height profile. The roughness profile is obtained by applying a Gaussian filter with the selected cutoff wavelength, and parameters are computed over an evaluation length conventionally equal to five consecutive sampling lengths.
- Cutoff selection follows standardized tables, for example 0.8 mm cutoff for expected Ra values between 0.1 µm and 2 µm
- Traversing length, stylus force, and tip geometry are selected to avoid surface damage and signal distortion
- The measurement direction is normally perpendicular to the surface lay
- Portable and bench-top stylus instruments enable both laboratory and shop-floor measurement
Optical and Non-Contact Methods
Optical techniques are applied where the surface is delicate, soft, or geometrically complex. White-light interferometry provides nanometer vertical resolution over small fields and is suited to polished, lapped, and fine ground surfaces. Confocal laser scanning microscopy and focus variation microscopy extend measurement to steep flanks and larger areas, supporting full areal parameter evaluation per ISO 25178. Light-section microscopy offers a rapid, robust method for coarser machined textures in production environments.
Comparison Specimen Method
For rapid in-process checks, the surface under examination is compared visually and by tactile feel against calibrated roughness comparison specimens manufactured by representative processes such as turning, milling, grinding, and lapping. This method, aligned with the principles of ISO 8503-2 for comparator usage, provides a coarse but economical screening tool; surfaces accepted by comparison should be confirmed by instrument measurement when tighter tolerance verification is required.
Areal (3D) Surface Texture Measurement
Where function depends on texture in all directions, areal measurement per ISO 25178 is performed by stitching optical topography scans or rastering a stylus instrument. Areal parameters such as Sa and Sz eliminate the ambiguity of single-profile direction selection and reveal defects such as chatter marks, tearing, and EDM recast layer texture that a single trace may miss.
Instrument Calibration and Verification
Stylus instruments are calibrated and periodically verified in accordance with ISO 12179 using certified reference standards, including depth-calibration specimens and periodic artifacts. Non-contact instruments are verified with step-height and roughness reference artifacts traceable to national metrology institutes. Accredited third-party laboratories issue inspection reports documenting measured values, applied standards, cutoff settings, measurement locations, and conformity to drawing tolerances, such as a requirement of Ra 1.6 µm maximum with a 16% rule or max rule applied per ISO 21920-2.
Applications
Machined surface roughness inspection is applied across every precision manufacturing sector:
- Automotive: cylinder bores, crankshaft journals, camshaft lobes, valve seats, gear teeth, brake rotor faces, and fuel injection components where texture controls oil film retention and sealing
- Aerospace: fatigue-critical fastener holes, landing gear pins, turbine disk bores, blade root serrations, and honeycomb seal surfaces, where roughness limits govern crack initiation life
- Hydraulics and pneumatics: valve spools, cylinder bores, piston rods, and pump components requiring controlled texture for seal life and leakage control
- Medical devices: orthopedic implant bearing surfaces demanding Ra in the nanometer range, dental abutments, and surgical instruments
- Bearings and power transmission: raceways, rolling element surfaces, and shaft journal seats
- Sealing systems: flange faces, gasket land surfaces, and dynamic seal counterfaces
- Tooling and mold making: mold cavity surfaces for injection molding, die casting, and optical lens replication
- Coating and surface treatment verification: pre-coating substrate texture control for thermal spray, plating, anodizing, and adhesive bonding
- General machining: incoming verification of subcontracted machined parts, process capability studies, and first article inspection
FAQ
What is the difference between Ra and Rz?
Ra is the arithmetic average of absolute profile deviations and describes the overall texture level, while Rz is the maximum height between the highest peak and deepest valley within a sampling length. Rz is far more sensitive to isolated defects and deep scratches, so many specifications limit both parameters simultaneously.
Can Ra and Rz be converted into each other?
No exact mathematical conversion exists because the parameters respond differently to the shape of the profile. For surfaces produced by a stable process, empirical ratios may be observed, but acceptance decisions must always be based on direct measurement of the specified parameter.
How is the cutoff (sampling) length selected?
The cutoff is chosen from standardized values according to the expected roughness magnitude and the process that generated the surface, following the tables in ISO 21920 or ISO 4288. Selecting an incorrect cutoff can mix waviness into the roughness result or truncate genuine texture, producing non-comparable values.
Why must the measurement be taken perpendicular to the lay?
Machined surfaces are anisotropic. Tracing across the lay captures the true peak-to-valley amplitude of the tool marks, whereas tracing along the lay systematically understates roughness. The drawing should specify the evaluation direction where it is functionally important.
How are curved surfaces and small internal bores measured?
Dedicated stylus instruments with bore probes, small-radius stylus arms, or skid-type plug gauges are used for internal diameters, while optical methods with curved-surface correction handle exterior radii. For tightly restricted geometries, comparison specimens or replica materials may be used, followed by laboratory measurement of the replica.
What does a typical inspection report include?
A professional report states the applied standards, instrument and method type, measurement locations and directions, cutoff and filter settings, individual and mean parameter values, the applicable limit rule such as the 16% rule or max rule, and a clear conformity statement against the drawing specification.
How often should roughness instruments be calibrated?
Calibration intervals should be established from usage frequency and required accuracy, with intermediate checks performed daily or before each inspection batch using certified reference specimens. Full calibration per ISO 12179 is typically repeated annually or after any shock, repair, or relocation of the instrument.