ISO 25178 Areal Surface Texture Parameters for Optical Inspection

Field

Precision optics manufacturing relies on clear three-dimensional characterisation across functional component areas. Contact styli in older profile-based metrology track height along a single line, missing directional micro-textures, spatial distribution of peaks and dales, and volumetric fluid retention. To qualify components reliably, industrial injection moulding and optical manufacturing turn to full-field areal measurements standardized under ISO 25178.

Surface texture dictates optical behavior. Light scattering, antireflective properties, spatial diffraction, and tactile friction depend on two-dimensional area topographies rather than isolated linear traces. When a toolmaker finishes an injection cavity with laser ablation or micro-Electrical Discharge Machining (EDM), the resulting topography forms intricate spatial patterns.

Single-line profile parameters like Ra or Rz average out height variations along one path, overlooking peak-to-peak spacing and isolated defects sitting off the stylus trajectory. Evaluating areal parameters across active surface regions keeps unrepresentative single-track scans from passing out-of-spec tooling.

Limitations of Two Dimensional Profilometry

A line-based stylus trace samples a single isolated track across a surface. Dragging a physical stylus across a micro-textured mold insert records peaks and valleys along that exact trajectory. If the probe runs parallel to a machining lay from end milling or diamond turning, the trace makes the surface look deceptively smooth.

Tracing perpendicular to that same lay spikes the roughness values. Neither measurement captures the true geometric character of the surface.

Areal texture parameters expand linear profiles into a continuous three-dimensional height field z(x,y). ISO 25178 evaluates parameters across a designated surface area instead of a single line length. By capturing millions of data points across a spatial grid in one frame, optical systems provide the point density needed for parameters that reflect real surface behavior ~ from wear resistance and light redirection to liquid coating adhesion.

Spatial Area Definition and Surface Bandpass

Evaluating areal texture begins with a continuous two-dimensional sampling region across the target surface. ISO 25178-3 defines the spatial filtering pipeline that turns raw primary height surfaces into scale-specific texture maps. Because raw sensor data includes structural form errors, machine vibration waviness, and high-frequency optical noise, the standard uses nested topological filters to isolate distinct spatial frequencies before calculating parameters.

The filtering pipeline starts with the S-filter, removing high-frequency noise and spatial features smaller than the lateral resolution limit of the optics. Next, the F-operator removes nominal geometry and macro-form variations like spherical curvature or planar tilt. Finally, the L-filter strips away long-wavelength spatial structures, leaving the scale-limited roughness surface known as the SL-surface.

Selecting proper cutoff wavelengths (λc and λs) stops low-frequency form errors from artificially inflating vertical amplitude values like Sa and Sq.

Applying ISO 25178-3 nesting filters with mismatched S-filter and L-filter cutoff frequencies introduces up to 35 percent error in measured Sa values on micro-structured injection mouldings.

Digitizing an optical surface generates height matrices across a set field of view. Objective magnification and numerical aperture set the spatial sampling intervals Δx and Δy across the grid. High-magnification objectives resolve fine micro-roughness but contract the evaluation area, forcing multi-image stitching for larger components.

Low-magnification lenses capture broad surface waviness, though optical spatial averaging smooths over microscopic peaks.

Areal parameters remove the orientation bias inherent in linear stylus profiles. Scanning continuous surface regions yields solid statistical sampling whether topographies are random, isotropic, or anisotropic. Specifying these spatial fields allows engineers to link micro-geometry directly to physical component performance.

Optics

Non-contact instruments map surface heights by capturing light reflected from physical micro-structures. Instead of dragging a stylus, these optical tools use wave interference, focal position detection, or light scatter patterns to assemble z(x,y) coordinate arrays. Optical interactions between incident light and the material determine the boundaries for height resolution, lateral detail, and measurable surface slope.

The ISO 25178-600 series defines three primary non-contact measurement modalities: Coherence Scanning Interferometry (CSI), Laser Confocal Microscopy (LCM), and Focus Variation (FV). Each relies on distinct physical mechanisms, setting specific operating boundaries when measuring tool steel inserts or molded polymer parts.

Non Contact Sensor Physics and Slope Boundaries

Light reflection mechanics limit how optical instruments register high-angle surface features. Smooth specular surfaces reflect light back into an objective only if the local surface angle lies within the collection cone set by the system’s numerical aperture (NA). The maximum specular slope angle θ_max relates directly to numerical aperture through θ_max = arcsin(NA).

For example, an objective with an NA of 0.55 gathers reflected light up to a maximum slope angle of roughly 33 degrees; features steeper than this reflect light away from the sensor, producing unmeasured dark pixels called optical dropouts.

Diffuse surfaces alter slope acceptance. Rough, textured plastic parts scatter light broadly, enabling Focus Variation systems to reconstruct steep surface flanks up to 80 degrees by tracking local image contrast through a vertical focal sweep. That same Focus Variation approach struggles on smooth, polished inserts where low contrast makes finding the focal plane difficult.

Coherence Scanning Interferometry leverages optical coherence to deliver sub-nanometer vertical resolution on smooth surfaces regardless of magnification, though sharp steps can trigger fringe breakdown artifacts.

Filtering Pipelines and Topological Nested Operators

Processing raw height data calls for sequential topological filtering to separate underlying form from surface texture. Raw non-contact point clouds inevitably carry instrument noise, spatial tilt, and unmeasured points. Standard inspection workflows run mathematical operators on raw point clouds to produce clean S-F or SL surfaces ready for parameter calculation.

Topological filter execution follows a strict sequence to maintain signal integrity:

• Form Removal Operator (F-operator) ~ Subtracts nominal geometry such as planes, cylinders, or polynomial freeform surfaces from raw point clouds to isolate residual height topography.
• Short Wavelength Filter (S-filter) ~ Attenuates spatial noise frequencies smaller than the optical resolution limit using a Gaussian spatial filter convolution.
• Long Wavelength Filter (L-filter) ~ Separates high-frequency surface roughness from long-wavelength surface waviness, establishing the upper limit for texture extraction.
• Outlier Suppression Operator ~ Identifies and removes localized spike noise caused by secondary optical reflections or dust contamination using spatial median filtering.

Instrument suppliers sometimes claim non-contact systems capture raw topographies without data manipulation. That claim ignores routine pre-processing: optical instruments routinely apply spatial smoothing, missing-point interpolation, and sensor tilt corrections before displaying surface maps. Recognizing these automated operations helps quality engineers maintain tight control over filter settings during part sign-off.

Metrics

Quantifying three-dimensional surface topography relies on specific mathematical parameters categorized by ISO 25178-2. These parameters divide into distinct functional groups: height (amplitude) parameters, spatial parameters, hybrid parameters, functional volume parameters, and feature parameters. Relying exclusively on vertical amplitude parameters hides critical geometric variations that dictate functional component performance.

Height and Hybrid Topology Characterisation

Amplitude parameters summarize vertical height variations across an evaluation area regardless of spatial arrangement. Sa represents the arithmetic mean height of absolute surface deviations from the mean plane. Sq defines root mean square height, providing greater sensitivity to extreme peaks and dales.

Sz measures maximum surface height, taken as the vertical distance between the highest peak (Sp) and lowest valley (Sv) within the area.

Skewness (Ssk) and Kurtosis (Sku) describe the statistical distribution of surface heights. Ssk measures asymmetry relative to the mean height plane; negative Ssk values indicate a plateau-dominated surface with deep valleys ~ ideal for bearing surfaces and fluid reservoirs ~ while positive values point to a spiky surface with sharp peaks prone to rapid initial wear. Sku measures distribution peakedness.

An Sku of 3.0 represents an ideal Gaussian distribution, values above 3.0 signal sharp tall peaks, and values below 3.0 denote a flattened profile.

Hybrid parameters combine spatial and height traits to evaluate surface slope and area expansion. Sdq calculates the root mean square surface gradient across the field, measuring average local slope angle. Sdr measures the developed interfacial area ratio ~ the percentage of extra surface area contributed by texture relative to a flat plane of identical spatial dimensions.

A flat surface yields an Sdr of 0 percent, whereas micro-textured surfaces with steep slopes can produce Sdr values over 50 percent.

Functional Volume and Bearing Area Behavior

Material ratio curves track cumulative surface void volume from peak to dale. ISO 25178-2 standardizes functional volume parameters derived from the areal material ratio (Abbott-Firestone) curve, measuring surface behavior under contact, lubrication, and wear conditions.

Functional volume parameters divide the surface height spectrum into three distinct zones based on material ratio thresholds (typically set at 10 percent and 80 percent material ratios):

• Peak Material Volume (Vmp) ~ Calculates the volume of solid material located in the topmost surface peaks per unit area, representing material susceptible to rapid running-in wear.
• Core Material Volume (Vmc) ~ Quantifies the solid material volume contained within the core height band between 10 percent and 80 percent material ratios, defining primary load-bearing material.
• Core Void Volume (Vvc) ~ Measures the volume of enclosed voids within the core height band available for fluid or lubricant storage.
• Dale Void Volume (Vvv) ~ Defines the open fluid volume retained in deep surface valleys below the 80 percent material ratio threshold.

A change in mould packing pressure from 400 bar to 800 bar increases core material volume Vmc by 28 percent on polypropylene textured panels without altering standard arithmetic mean height Sa.

Feature parameters apply watershed segmentation algorithms to break complex topographies into discrete regions: individual peaks, dales, ridge lines, and course lines. This methodology yields metrics like peak density (Spd), arithmetic mean peak curvature (Spc), and mean dale area (Sda), delivering micro-geometric data for friction modeling and micro-infill analysis during polymer processing.

Spatial Texture Directionality and Autocorrelation

Directional texture characteristics determine how surface features align relative to machining or molding flow paths. Spatial parameters use the areal autocorrelation function (ACF) to evaluate surface periodicity and lay direction. Sal defines autocorrelation length ~ the minimum spatial distance over which the surface profile correlates with itself.

Short Sal values point to surfaces dominated by high-frequency micro-features, whereas long Sal values indicate dominant low-frequency structures.

Str measures texture aspect ratio, rating surface isotropy from 0 to 1. An Str value approaching 1.0 indicates complete spatial isotropy with uniform spatial frequency in all directions ~ typical of bead-blasted or spark-eroded cavities. An Str below 0.3 signals strong spatial anisotropy, revealing the uniform directional lay left by linear milling, shaping, or diamond turning.

Specifying Str limits on optical mould tooling drawings prevents directional machining marks from distorting light distribution across precision molded optics.

Relying solely on Sa to qualify textured tool steel causes signoff failures. Two tool steel cavities can show identical Sa values of 1.20 micrometers while possessing entirely different spatial structures. A spark-eroded surface exhibits an isotropic lay with an Sdr of 12 percent, whereas an end-milled cavity shows strong directional lay with an Sdr of 35 percent.

When moulding high-viscosity resins into these cavities, high-slope directional surfaces create high demoulding friction, driving part ejection distortion and localized scuffing.

Replication

Transferring geometric micro-textures from steel mould cavities into molten thermoplastics involves coupled thermal and fluid dynamics. In injection moulding, hot polymer melt contacts cold cavity walls and freezes an immediate skin. High injection velocity and packing pressure force the molten core into the surface texture’s micro-cavities.

Viscous flow losses, thermal contraction, and polymer chain relaxation keep tool micro-features from transferring to the plastic component with 100 percent fidelity.

Replication fidelity varies substantially across resin families, mold temperature profiles, and packing pressure phases. Micro-textured steel topographies measured under ISO 25178 show clear differences when compared directly against the resulting molded plastic surfaces.

Melt Rheology and Cavity Micro Feature Infill

Thermoplastic melt entering a micro-structured cavity undergoes intense shear cooling at the steel boundary. Fast formation of a frozen polymer skin restricts flow into deep micro-cavity valleys. High-viscosity engineering resins like Polycarbonate (PC) and Polyetheretherketone (PEEK) exhibit high flow resistance, keeping resin from reaching the narrow roots of EDM or laser-etched textures.

Lower-viscosity resins like Polypropylene (PP) and Liquid Crystal Polymer (LCP) achieve far higher micro-infill fidelity under identical processing conditions.

Mold temperature governs micro-structure replication quality. Raising tool cavity temperature near or above the resin’s glass transition temperature (Tg) delays frozen skin formation. This extended local flow window allows continuous polymer packing into micro-cavities, raising replication fidelity.

Processing Polycarbonate at a mold temperature of 120 degrees Celsius yields near-complete texture transfer, whereas running the tool at 80 degrees Celsius causes severe peak rounding and reduced Sdr values on the part.

Can Optical Instruments Resolve Deep Etched Tool Cavities?

Deep, high-aspect-ratio textures from EDM or chemical etching present steep flank angles that scatter light away from objective lenses. Optical non-contact sensors struggle to capture the full depth of deep steel micro-cavities because of signal loss at steep sidewalls. When measuring steel cavity textures using Focus Variation or Coherence Scanning Interferometry, missing data points tend to cluster at feature bottoms.

Molded plastic parts invert the tool cavity’s topography: deep steel micro-cavities become protruding plastic micro-peaks. Non-contact optical sensors measure these external convex peaks with far less signal loss than deep steel cavities, as light reflects back into the objective aperture without internal sidewall shadowing. Quality engineers often obtain cleaner, more accurate ISO 25178 areal measurements by scanning the molded plastic component directly rather than probing deep steel cavity features.

Polymer skin formation prevents complete cavity micro-infill, causing molded part Sdr values to lag tool steel cavity Sdr values by 15 to 40 percent under standard processing parameters.

Quantifying surface replication loss requires a standardized procedure comparing cavity steel topographies against molded plastic surfaces:

1. Clean the tool cavity steel insert using ultra-pure isopropyl alcohol and oil-free compressed air to remove anti-corrosion greases and moulding residues.
2. Measure a designated 1.0 mm by 1.0 mm benchmark region on the cavity insert using Coherence Scanning Interferometry at 20x magnification, recording baseline Sa, Sz, Sdq, and Sdr parameters.
3. Mould twenty component samples under steady-state thermal conditions, allowing the press to run continuously for at least forty minutes prior to sample collection.
4. Store moulded parts in a temperature-controlled metrology lab at 23 degrees Celsius and 50 percent relative humidity for twenty-four hours to ensure complete volumetric relaxation.
5. Measure the corresponding 1.0 mm by 1.0 mm benchmark region on five moulded plastic parts using identical optical sensor settings, objective magnification, and ISO 25178 spatial filter cutoffs.
6. Calculate the replication ratio for each parameter by dividing the averaged moulded part values by the baseline tool steel values.

Volumetric shrinkage alters spatial texture parameters during cooling. Semi-crystalline polymers like Polypropylene undergo significant post-moulding volumetric contraction up to 2.0 percent. This contraction pulls material inward, shortening spatial distances between micro-peaks and reducing autocorrelation length Sal.

Amorphous polymers experience much lower shrinkage (0.4 to 0.7 percent), keeping spatial feature spacing close to original cavity dimensions.

How does anisotropic thermal contraction alter functional volume parameters in high-aspect micro-textures?

Uncertainty

Traceable metrology depends on establishing documented error bounds for non-contact optical instruments. ISO 25178-600 outlines calibration standards, metrological characteristics, and measurement uncertainty budgets for non-contact areal instruments. An optical height measurement without a stated uncertainty boundary carries no legal standing in cross-border quality disputes.

Measurement uncertainty combines hardware noise, optical aberrations, environmental vibration, thermal drift, and software processing algorithms. Quantifying these error sources separates genuine manufacturing variation from instrument noise.

Metrological Traceability and Instrument Noise Calibration

Establishing instrument accuracy requires physical reference standards with calibrated height steps and periodic grating structures. ISO 25178-70 defines calibration artifacts used to calibrate non-contact optical metrology tools. Material measures include single-step standards for calibrating z-axis amplification, multi-tier step standards for linearity checks, and two-dimensional cross-grating standards for calibrating x-y lateral scale factors and orthogonality.

The instrument noise floor (Sq0) defines the optical system’s measurement limit, representing residual height noise recorded when scanning a flat, defect-free optical flat standard. An instrument cannot reliably measure surface roughness Sa values approaching its internal noise floor Sq0. For high-precision optical components requiring Sa thresholds below 5.0 nanometers, the noise floor Sq0 must not exceed 0.5 nanometers.

Optical amplification factor h_z and linearity errors l_z demand periodic calibration across the total vertical scanning range. Thermal fluctuations in metrology environments expand mechanical gantry components, inducing z-axis drift. Running optical measurements in a lab controlled to 20 ± 0.5 degrees Celsius minimizes environmental drift during automated overnight batch inspections.

Data Interpolation and Missing Point Handling

Unmeasured pixels occur when surface slopes exceed an optical sensor’s numerical aperture limit. Dark, non-reflective surfaces or steep geometry leave empty pixels in the recorded z(x,y) matrix. How software algorithms handle these missing data points heavily influences calculated ISO 25178 values.

Data handling options within metrology software package routines dictate surface reconstruction rules:

• Zero Filling ~ Assigns a z-height of zero to all non-measured pixels, severely distorting vertical height parameters like Sa and creating artificial step cliffs.
• Planar Interpolation ~ Calculates missing pixel heights via linear plane fitting from adjacent measured pixels, smoothing out sharp local feature boundaries.
• Spatial Spline Surface Fitting ~ Applies two-dimensional bi-cubic spline interpolation based on neighboring pixel gradients, preserving local curvature trends.
• Threshold Threshold Masking ~ Excludes unmeasured pixels entirely from parameter integration calculations, requiring a minimum valid data threshold (e.g. 95 percent valid pixels) for sample acceptance.

Interpolating more than 5 percent missing pixels on a micro-textured surface grid artificially reduces peak height parameters Sp and Sz by up to 18 percent.

Quality dossiers supporting optical surface acceptance require strict document controls. A supplier submitting optical metrology records must provide explicit instrumentation settings to guarantee measurement reproducibility.

Required documentation elements for verified ISO 25178 optical inspection dossiers include:

• Instrument Specification Sheet ~ Model, serial number, sensor type, objective lens magnification, and numerical aperture used during measurement.
• Environmental Log ~ Metrology room ambient temperature, relative humidity, and active vibration isolation table status recorded at the time of scan execution.
• Filtering Dossier ~ Defined S-filter cutoff (λs), L-filter cutoff (λc), F-operator geometry subtraction method, and outlier suppression algorithm settings.
• Missing Data Log ~ Percentage of unmeasured pixels across the raw scan field prior to spatial interpolation or masking operations.
• Traceability Certificates ~ Valid calibration certificates for physical step standards referenced to national metrology institutes.

ISO 25178-600 Clause 4.2 dictates that non-contact optical surface measurements must report calculated combined standard measurement uncertainty (U) alongside all primary parameter values.

Contract

Procurement documents for textured injection tooling specify surface quality using standardized areal parameter thresholds rather than legacy profile callouts. Specifying surface roughness via obsolete single-line profile standards like ISO 4287 (Ra, Rz) creates legal ambiguity when purchasing three-dimensional textured components. Modern injection tooling contracts incorporate ISO 25178 areal parameters to define exact surface finish expectations, tool release metrics, and quality signoff procedures.

Transitioning engineering drawings from legacy standards to areal surface texture specifications prevents costly commercial disputes between mold builders, plastic injection molders, and tier-one OEMs.

Translating Legacy Callouts to Areal Specifications

Legacy drawing callouts like Ra or VDI 3400 fail to communicate three-dimensional surface volume and spatial layout. A drawing specifying “Ra 1.6 µm per ISO 4287” leaves the toolmaker free to deliver a turned, milled, ground, or spark-eroded surface. Each machining process yields vastly different spatial lay, surface area expansion, and fluid retention properties, despite meeting the single line Ra amplitude requirement.

Converting legacy specifications into unambiguous ISO 25178 contract callouts requires defining multiple parameters to bound height, spatial layout, and surface slope characteristics:

Tool Acceptance and Commercial Quality Agreements

Formal tool signoff relies on clear acceptance thresholds for primary areal texture parameters measured across sample mouldings. A robust quality agreement stipulates that first-article inspections perform optical metrology scans on T1 moulding samples using defined ISO 25178 protocols. If measured parameters on the moulded part fall outside specified bandwidths, the toolmaker modifies cavity steel polishing or re-textures the cavity at their own expense.

Specifying multiple parameters prevents suppliers from gaming single-metric specifications. If an engineering drawing specifies Sa 1.2 µm without specifying spatial parameters, a toolmaker might deliver a coarse directional milled surface that meets the Sa target but fails in application due to high demoulding friction. Adding mandatory control bands for hybrid parameters (Sdr) and spatial parameters (Str) ensures the tool delivers both the correct amplitude and correct three-dimensional surface structure.

Commercial contracts must address non-contact measurement setup protocols explicitly. The quality agreement states the exact instrument type, objective magnification, numerical aperture, missing-point interpolation rules, and filtering cutoffs utilized during audit inspections. Disagreements between molder and buyer metrology labs evaporate when both parties execute measurements under identical ISO 25178-3 spatial filtering protocols.

Landed component costs correlate directly with inspection cycle times. High-magnification optical scans covering large surface fields demand significant instrument scan duration and data processing overhead. Defining sensible evaluation areas (such as 0.8 mm by 0.8 mm fields) and targeting critical active surface regions keeps metrology cycle times low, holding down quality control costs while maintaining process capability across multi-cavity production tooling runs.