Surface Roughness Chart: Ra, Rz, and N-Grade Conversion Guide

Surface Roughness (Ra, Rz, ISO 1302) – A Practical Engineering Guide

Surface roughness is one of the most influential and frequently misunderstood aspects of functional design and manufacturing. It affects how parts wear, seal, carry lubricant, bond with coatings, and even how they look and feel to an end user. This guide explains the most commonly used roughness parameters Ra and Rz, how to select and specify them on drawings in accordance with ISO 1302, typical values by machining process, and how roughness influences coating adhesion. It also provides a compact N-grade to Ra conversion and practical advice for inspection and production.

What Is Surface Roughness and Why It Matters

Surface texture is a combination of roughness, waviness, and form. Roughness comprises the fine, closely spaced irregularities produced by a material removal or additive process; waviness consists of larger, more widely spaced deviations; and form covers overall shape error. In many functional applications—such as sealing, bearing contact, sliding friction, fatigue life, and coating adhesion—surface roughness plays a central role. Specifying surface finish poorly can lead to leaky seals, noisy gears, premature wear, poor paint or powder adhesion, and excessive manufacturing cost.

International standards define parameters and measurement rules to quantify roughness. Historically, ISO 4287 and ISO 4288 governed the definition and evaluation of profile parameters like Ra and Rz. More recently, ISO 21920 has updated those profiles and practices. For drawings, ISO 1302 remains the primary standard that defines how surface texture requirements are indicated. This guide uses the established Ra and Rz parameters and explains how to apply them correctly using ISO 1302 conventions.

Defining Ra (Arithmetic Mean Roughness)

Ra is the arithmetic mean deviation of the roughness profile from the mean line over the evaluation length. Conceptually, imagine slicing the roughness profile into extremely fine strips and taking the absolute height of each point relative to the mean line; Ra is the average of those absolute values. It is a scalar measure of overall roughness amplitude and is widely used because it is stable, relatively insensitive to occasional single peaks or valleys, and easy to measure with a stylus profilometer or optical instrument.

Because Ra effectively averages the surface deviations, it does not describe the shape of the profile or distinguish between infrequent deep valleys and frequent shallow peaks. Two surfaces can have the same Ra but vastly different functional behavior, for example in sealing or wear, because their peak-to-valley structure differs.

Defining Rz (Mean Peak-to-Valley Height)

Rz quantifies vertical distance between prominent peaks and valleys. In the ISO definition, Rz is typically computed as the mean of the five highest peak-to-valley spans within the evaluation length (using the roughness profile filtered per the specified cutoff). As a result, Rz is much more sensitive to occasional high peaks or deep pits than Ra.

Practically, Rz highlights extremes that an averaging parameter can hide. Where Ra might look acceptable, Rz may reveal unacceptable isolated burrs, tears, pits, or plateau/valley structures that matter to sealing and contact mechanics. If you care about the height of asperities that contact a mating part, Rz often correlates better with functional risk than Ra alone.

Ra vs Rz: When to Use Each

Although Ra is ubiquitous, relying on it alone can be misleading. Consider the following guidance:

Use Ra when: You need a simple, stable index of overall finish that relates to cost and general part appearance. Processes like grinding or superfinishing that produce uniform profiles are well represented by Ra. Many legacy specifications and cost targets are expressed in Ra, so it is a valuable baseline.

Use Rz in addition to Ra when: The function depends on peak height or valley depth. Sealing surfaces, bearing rolling contacts, and surfaces that must avoid stress risers or debris entrapment benefit from an Rz limit to control outlier features. Rz can expose deep tool marks from milling, torn material from turning, or pits from casting and blasting that might pass an Ra-only requirement.

Use Rz (or a combination) when: Adhesion, fatigue, wear-in, or plateau honing are involved. For example, cylinder bores often need valleys to retain lubricant yet control peak heights to protect rings; combining parameters (Ra for overall roughness and Rz or Rpk/Rvk for peak/valley) yields better functional control. Where measuring capability permits, consider more descriptive parameters such as Rsk (skewness), Rku (kurtosis), or bearing area curve parameters (Rpk, Rk, Rvk) defined in ISO standards, but do not abandon Ra and Rz if your supply base relies on them.

In short, specify Ra for general control and cost benchmarks, and add Rz if your function is sensitive to outliers, sealing, contact stress, debris trapping, or coating anchoring. Always accompany these with the correct cutoff (sampling) length and evaluation rules to avoid ambiguity.

Cutoff, Filters, and Evaluation Length

Measurement standards define how raw measured profiles are filtered to separate roughness from waviness and form. The cutoff length (λc) is a critical filter parameter that defines the spatial scale of roughness; a too-short cutoff includes high-frequency noise, and a too-long cutoff includes waviness that inflates roughness parameters. Evaluation length is typically five times the cutoff length. If you do not state the cutoff (or allow “default per instrument”), different shops may filter differently, producing incomparable results even when the part is identical.

Common cutoff lengths for machined metals include 0.8 mm and 2.5 mm, but smaller or larger values apply to fine finishes and coarse cast surfaces. In your drawing callout, state the parameter (Ra or Rz), the numerical value, and the cutoff (for example, “Ra 1.6, λc = 0.8 mm”). If inspection is critical, also state the evaluation length or cite the applicable ISO evaluation standard for consistency.

Typical Ra Values by Machining Process

Process capability depends on machine condition, tooling, material, and parameters like feed, speed, and coolant. Nonetheless, the following ranges are widely used planning targets for metallic components. Always validate with trials for your specific combination of material and process.

Process	Typical Ra Range (µm)	Notes
Grinding	0.1 – 0.8 µm	Fine control with wheel selection and dressing; can achieve superfinishes below 0.1 µm with specialized setups.
Turning	0.8 – 3.2 µm	Feed rate and nose radius dominate; wiper inserts can bring Ra near 0.8 µm or better on stable lathes.
Milling	1.6 – 6.3 µm	Step-over and toolpath matter; climb milling and sharp tools improve peaks; 3-axis scallops can raise Rz despite modest Ra.
Drilling	1.6 – 6.3 µm	Chip evacuation and tool wear are decisive; reaming can reduce Ra to 0.4 – 1.6 µm.
Sand casting	6.3 – 25 µm	Mold media, coating, and gating influence; secondary machining needed for tight seals or bearing surfaces.

These bands help set realistic finish requirements early in design. If your part needs, for example, Ra 0.4 µm over a large face, a single-pass face milling operation is unlikely to meet the requirement with acceptable yield and cost; a grinding or lapping step may be necessary. Conversely, do not over-specify an ultra-fine Ra where a moderate Ra suffices, as cycle time and tool cost will rise sharply.

N-Grade to Ra Conversion

Many shops and legacy drawings refer to roughness “N-grades” (also called roughness classes). Each N-grade corresponds to a target Ra value. The table below lists a compact set of widely encountered conversions:

N-Grade	Ra (µm)	Typical Process Context
N1	0.025	Superfinishing, lapping, precision optics
N4	0.2	Fine grinding, honing, fine turning
N6	0.8	Standard grinding or good turning
N8	3.2	General turning, face milling
N10	12.5	Rough milling, as-cast surfaces

N-grades are convenient shorthand, but they can hide important details like the cutoff length and whether alternative parameters, such as Rz, also apply. When precision matters, state the parameter, value, cutoff, and any lay direction. For example: “Ra 0.8 µm, λc = 0.8 mm; Rz 5 µm max.”

ISO 1302 Drawing Callout Symbols

ISO 1302 standardizes how surface texture requirements appear on technical drawings. Understanding the symbol variants and the text that accompanies them ensures designers and manufacturers interpret requirements consistently.

Basic symbol: The fundamental symbol resembles a checkmark of two legs meeting at an acute angle. On its own, it indicates a surface texture requirement applies but does not constrain the manufacturing method.

Material removal required: Adding a short bar on the upper right of the basic symbol indicates that material removal by machining is required to obtain the specified texture. This prevents, for example, leaving an as-cast surface where a machined finish is necessary.

Material removal prohibited: Replacing the bar with a small circle indicates that material removal is not permitted. This is used when, for instance, a functional surface must remain as-molded or as-cast.

Textual fields attached to the symbol: ISO 1302 allows you to place parameter, value, cutoff length, and other modifiers adjacent to the symbol. The most common content includes: - The parameter and numeric limit, such as “Ra 1.6” or “Rz 10”. - The cutoff length (λc) if not default, e.g., “λc 0.8 mm”. - The evaluation method or reference standard if necessary. - Lay direction symbol (e.g., parallel, perpendicular, crossed) or descriptive text, such as “lay ⟂ to datum A”. - Machining allowance (a value indicating how much stock must be removed). - The surface sampling length count or repeat requirements as needed.

Examples of clear callouts in ISO 1302 style:

1) “Ra 1.6 µm, λc = 0.8 mm, lay ⟂ to datum A” attached to a machined seal face. This states the average roughness, the cutoff used for filtering, and that the tool marks must be perpendicular to a datum direction for sealing performance.

2) Symbol with bar (material removal required) plus text: “Rz 8 µm max; Ra 1.2 µm max; λc = 0.8 mm”. This combination caps the peak-to-valley extremes while controlling the average amplitude.

3) Symbol with circle (no material removal) plus “Ra 6.3 µm max, as-cast”. This preserves a casting texture for coating adhesion or aesthetics, with a ceiling on allowable roughness.

Placement and general notes: Place the symbol leader to the surface of interest or specify globally with a general note like “Unless otherwise specified, all machined surfaces: Ra 3.2 µm, λc 0.8 mm, ISO 1302.” If certain critical surfaces require tighter control, override the general note with local symbols and values.

Measurement Methods and Practical Considerations

Several metrology methods can quantify Ra and Rz. Selection depends on surface accessibility, required precision, and budget.

Stylus profilometers: The most common instruments. A diamond-tipped stylus (typical radius 2–10 µm) traverses the surface, generating a profile. They are robust and good for a wide range of roughness values, but can struggle with very soft surfaces, steep slopes, or fragile coatings. The stylus radius limits the smallest features it can track, which can reduce measured Rz on sharp peaks.

Optical methods: White-light interferometry, confocal microscopy, focus variation, and structured light allow non-contact measurement. They are excellent for delicate or wide-area measurements and can produce areal parameters (Sa, Sz) in addition to profile parameters (Ra, Rz). However, reflectivity, transparency, and surface color can complicate results. For coarse cast surfaces with deep cavities, optical shadowing may bias Rz unless multiple angles are acquired.

Sampling and filtering: Regardless of method, apply the specified cutoff and evaluation length. Ensure the traverse direction aligns sensibly relative to lay: measuring parallel to turning lay can underestimate peak heights that matter in cross-contact. For anisotropic textures (typical with turning, milling, and grinding), measure orthogonal to the lay direction for functional assessment unless otherwise specified.

Gauge capability: Match instrument range and resolution to the specified values. For Ra below 0.2 µm, environmental vibration, stylus noise, and filtering choices can dominate results. For Ra above 10 µm, consider larger cutoffs and robust probes. Always verify with calibration artifacts near the target range.

Coating Adhesion and Roughness

Surface roughness strongly influences mechanical interlock and wetting of coatings, especially powders and paints. Too smooth a surface may not provide enough anchor points for a durable bond; too rough a surface can entrap air, cause voids, and lead to thin spots or orange peel.

Guideline for powder coating adhesion: An Ra of 1.5 – 3.2 µm is often ideal. This range offers sufficient microtexture for anchoring without introducing deep valleys that trap air or interfere with flow and leveling.

Too smooth: When Ra < 0.8 µm, especially on polished or ground surfaces, powder particles may not key into the substrate effectively, and wet adhesion can be poor. Light blasting or a controlled mechanical or chemical pre-treatment can raise the effective anchor profile to the target zone.

Too rough: When Ra > 6.3 µm, the risk of air entrapment increases, leading to pinholes, craters, or incomplete cure in valley bottoms. Coating thickness can vary, with peaks overcoated and valleys undercoated, reducing corrosion resistance. In such cases, pre-finish with a light machining, fine blasting, or a sanding step to bring the roughness into a controlled window before coating.

Surface preparation matters: Cleanliness, oxide removal, and chemical compatibility can override roughness effects. Degreasing, conversion coatings, and appropriate blasting media (e.g., aluminum oxide versus glass bead) influence both the micro-geometry and chemistry of the surface. Always validate adhesion with cross-hatch, pull-off, or bend tests on representative coupons that have the same pre-treatment, roughness, and geometry as production parts.

Directional lay and coating flow: Strong directional textures from turning or milling may telegraph through coatings and influence flow. A brief non-directional abrasion (e.g., orbital sanding) or multidirectional blast can reduce anisotropy, improving aesthetic and functional outcomes.

Selecting Roughness for Function

Right-sizing roughness prevents over-processing while achieving performance. Consider the following application-driven guidance:

Sealing surfaces (static seals): Prefer Ra in the 0.4 – 1.6 µm range with an Rz cap to curb occasional scratches that cause leaks. For elastomeric seals, a slight roughness aids wetting and prevents stiction; overly smooth surfaces can trap lubricant films that reduce sealing force. State cutoff and lay relative to seal line.

Dynamic seals (shafts): Specify both Ra and Rz (e.g., Ra 0.2 – 0.6 µm; Rz ≤ 3 µm) with lay oriented circumferentially. Plateau finishes are beneficial; avoid torn metal that accelerates lip wear.

Sliding bearings and wear pairs: Control peak heights to manage run-in and lubricant film formation. Ra alone may be insufficient; add Rz or bearing area curve parameters. For low-friction dry sliding, too rough increases abrasion; too smooth may lead to adhesive wear—validate tribology in representative tests.

Fatigue-sensitive regions: Limit Rz to minimize stress concentrations. Even with acceptable Ra, a high Rz from milling scallops or EDM recast can initiate cracks. Consider edge conditioning and micro-deburring in addition to roughness control.

Adhesive bonding and coatings: Target the 1 – 3 µm Ra region for many structural adhesives and powders, adjusting based on resin viscosity and cure behavior. Validate with joint-level testing.

Cost and Manufacturability Implications

Tighter roughness requirements often increase cycle time, tool cost, inspection effort, and scrap risk. Practical steps to reduce cost without sacrificing function include:

Specify ranges, not just maxima: For example, “Ra 0.8 – 1.6 µm” provides a process window and reduces over-polishing that burns capacity.

Use process-appropriate values: Do not call for Ra 0.4 µm from a roughing milling step if a later grinding step will set the final finish. Put the symbol on the final surface condition.

Control only what matters: Add Rz where it impacts function; avoid stacking numerous unrelated parameters that confuse sourcing and inspection.

General notes with local exceptions: Use a general finish for most surfaces and tighter local callouts for critical areas. This keeps quotes realistic and focuses quality control effort.

Common Pitfalls and How to Avoid Them

Equating Ra and Rz: They measure different aspects. Do not convert between them by rule of thumb; two profiles with identical Ra can have very different Rz values and vice versa.

Ignoring cutoff: Without λc, the same surface can “measure” different roughness on different instruments. Always specify or reference your standard practice.

Measuring along the wrong direction: For strongly directional lay (turned or ground), measuring parallel to lay can underestimate the functional peaks a mating component will see. Unless stated otherwise, measure perpendicular to lay for worst-case contact.

Specifying finishes unachievable by the chosen process: Anchor your requirements to process capability. If you must push the boundary, allow for additional finishing steps and budget time for trials.

Confusing waviness with roughness: Low-frequency form errors (waviness) can cause leaks or vibration even with a good Ra. If waviness matters, add appropriate parameters or controls on flatness and roundness in parallel.

How to Write Clear ISO 1302 Callouts

Good callouts are compact, unambiguous, and practical to inspect. Consider including:

Parameter and value: “Ra 1.6 µm” and, if function demands, “Rz 10 µm max”.

Cutoff (λc): “λc = 0.8 mm” for fine to medium finishes, larger for coarse surfaces.

Lay direction: “lay ⟂ to datum A” for seal faces; “lay ∥ to shaft axis” for journal surfaces.

Process requirement (optional): Use the symbol variant with bar for mandatory machining or circle to prohibit machining, supported by text like “as-cast” or “ground”.

General notes: “Unless otherwise specified, all machined surfaces: Ra 3.2 µm, λc 0.8 mm, ISO 1302.” Then override locally where needed.

Worked Examples

Example 1: Static O-ring groove sealing face. Function needs moderate texture for sealing without directional channels. Callout: Material removal required symbol, “Ra 0.8 – 1.6 µm; Rz ≤ 6 µm; λc 0.8 mm; lay crossed.” Rationale: Crossed lay reduces leak paths; Rz cap prevents scratches from compromising sealing.

Example 2: Powder-coated bracket mounting face. To promote adhesion while avoiding pinholes. Callout: Basic symbol, “Ra 1.6 – 3.2 µm before coating; λc 2.5 mm; as-blasted with Al2O3.” Rationale: Establishes target anchor profile and a reasonable cutoff for coarser blasted textures.

Example 3: Ground shaft journal for lip seal. Callout: Material removal required symbol, “Ra 0.2 – 0.4 µm; Rz ≤ 3 µm; λc 0.8 mm; lay circumferential.” Rationale: Fine finish limits wear; Rz cap controls peak excursions that damage seal lips.

Inspection Planning and Acceptance

Link drawing callouts to inspection plans. Define the number of measurement locations and orientation on each feature, environmental conditions, and acceptable variation. For large surfaces, require measurements at several radial or grid locations and specify acceptance criteria for outliers. If cosmetic appearance matters, include a visual standard in addition to numerical roughness to avoid disputes about sheen or texture patterns that do not affect function.

For supplier alignment, share target process parameters, sampling length, stylus tip radius, and instrument model or standard. Where economically feasible, exchange witness coupons finished alongside parts, then test them for coating adhesion, corrosion, or wear. This decreases risk when tolerances are ambitious or when switching suppliers or materials.

Relating Roughness to Other Specifications

Surface roughness does not exist in isolation. Geometric tolerances (flatness, cylindricity), dimensional tolerances, and material properties (hardness, microstructure) all interplay with texture. Excessively aggressive finishing can alter dimensions (e.g., lapping that reduces thickness) or induce residual stresses. When setting finish requirements, verify there is adequate stock allowance and process control so that the final part meets both surface and geometric requirements.

When corrosion resistance is critical, coordinate roughness with coating thickness and porosity. Very rough surfaces can require higher coating thickness to achieve coverage in valleys; this may conflict with fit or thread tolerances. Consider pre-smoothing to a target Ra before coating and relaxing the as-coated texture to what functionally matters (e.g., “pre-coat Ra 1.6 – 3.2 µm; post-coat cosmetic only”).

Summary and Key Takeaways

Ra and Rz are complementary tools for controlling surface roughness. Ra provides a stable measure of average amplitude, while Rz captures peak-to-valley extremes that can dominate functional risk in sealing, wear, and adhesion. ISO 1302 defines how to place clear and consistent callouts on drawings; use it to state parameter, value, cutoff, lay, and any process constraints. Align your requirements with process capabilities: grinding can reach 0.1 – 0.8 µm Ra, turning 0.8 – 3.2 µm, milling and drilling commonly 1.6 – 6.3 µm, and sand casting 6.3 – 25 µm. For powder coating adhesion, aim for an Ra of 1.5 – 3.2 µm; avoid finishes below 0.8 µm and above 6.3 µm to reduce delamination and air entrapment risks. Where legacy N-grades are used, remember that N1 ≈ 0.025 µm, N4 ≈ 0.2 µm, N6 ≈ 0.8 µm, N8 ≈ 3.2 µm, and N10 ≈ 12.5 µm Ra.

Ultimately, the best surface specification is one that reliably delivers function at the lowest total cost. Combine clear ISO 1302 callouts, realistic process windows, and robust inspection practices to achieve that outcome. When in doubt, prototype and test: small trials to confirm that your chosen Ra and Rz produce the desired sealing, wear, or adhesion performance will save time and money in production.