Technical Guide

Improving Ra and Rz on 3D Printed Parts: How Abrasive Blasting Reduces Surface Roughness

Surface roughness — quantified as Ra (arithmetic mean roughness) or Rz (mean peak-to-valley height) — is the primary measurable output of abrasive blasting in AM post-processing. Every production metal AM facility and most advanced polymer AM operations specify and measure surface roughness on finished parts. Abrasive blasting is the most effective single-step method for reducing as-built AM surface roughness by 60–80%, creating a uniform, controlled surface texture that meets functional specifications for fatigue resistance, corrosion protection, coating adhesion, and dimensional compliance. This guide explains how blasting reduces Ra and Rz, what values are achievable for each AM process and alloy, how to use cascade blasting for finer results, and when to add secondary finishing processes for specifications below Ra 0.8 µm.

1. Understanding Ra and Rz in Additive Manufacturing

Surface roughness is a statistical description of the micro-geometry of a surface. For AM parts, two parameters dominate engineering specifications:

Ra — Arithmetic Mean Roughness

Ra is calculated as the arithmetic mean of the absolute deviation of the surface profile from its mean line over the evaluation length (typically 5 × λc, where λc is the cutoff wavelength). It is the most widely used single parameter for surface quality specification. Ra captures the average surface texture intensity but does not distinguish between peaked surfaces (many small sharp peaks) and pitted surfaces (few deep valleys) — two surfaces with the same Ra can have very different functional properties.

Rz — Mean Roughness Depth

Rz is the average height of the five deepest profile peaks and valleys within the evaluation length. For the same surface, Rz is always greater than Ra. For blasted AM surfaces, the typical relationship is Rz ≈ 4–7 × Ra. Rz is more sensitive to extreme features (deep valleys, high peaks) and is therefore more relevant to fatigue crack initiation (cracks start at valleys) and coating thickness adequacy (the coating must fill valleys without bridging gaps).

Sa and Sz — Areal Parameters

Sa (areal arithmetic mean height) and Sz (maximum height of the evaluation area) are the three-dimensional equivalents of Ra and Rz, measured per ISO 25178 using optical or confocal profilometry. Because AM surfaces are anisotropic — Ra measured parallel to layer lines differs from Ra measured perpendicular to them — areal parameters provide a more complete and representative characterisation of blasted AM surfaces. For high-accuracy surface specification on AM parts, Sa and Sz are preferred over Ra and Rz alone.

For all AM post-processing reporting in this guide, Ra is used as the primary metric (the most common specification parameter in industry), with Rz noted where it provides additional insight.

2. As-Built Ra/Rz Values by AM Process and Orientation

As-built surface roughness in AM is not uniform — it varies significantly with build orientation, process parameters, and alloy. Understanding the starting Ra is essential for selecting the correct blast parameters and setting realistic expectations for post-blast Ra.

The data above illustrates two critical points: (1) down-skin surfaces are substantially rougher than up-skin surfaces for the same part; (2) EBM titanium starts roughly twice as rough as SLM titanium. Blast protocols must account for these starting Ra differences — a single parameter set will not achieve uniform final Ra across all orientations of a complex part without targeted attention to down-skin surfaces.

3. How Abrasive Blasting Reduces Surface Roughness

The Ra reduction mechanism of abrasive blasting depends on the media morphology:

Spherical Media (Glass Beads, Steel Shot, Zirconia): Peening Mode

Spherical particles impact the surface and plastically deform the material at the contact point. The high local contact stress flattens the surface peaks — the highest points on the Ra profile — by cold-working them. The peak material flows laterally into adjacent valleys, partially filling them. The net effect is a reduction in both peak height and valley depth, lowering Ra. This peening mechanism does not remove material in the traditional sense; it redistributes it. After multiple passes, diminishing returns set in as the remaining peaks are progressively smaller and require more energy per unit Ra reduction.

Angular Media (Al₂O₃, Steel Grit): Cutting Mode

Angular particles with sharp edges remove material by micro-cutting. Each impact of a sharp particle removes a tiny chip of surface material from the peak it strikes, leaving a sharp-edged micro-pit. The Ra reduction comes from peak removal, but the remaining micro-pits left by angular impacts create a characteristic surface texture that is rougher on a micro-scale than a peened surface. This is why angular media (Al₂O₃) at equivalent grit size produces a slightly higher Ra than spherical media (glass beads) at equivalent mesh size, despite both reducing Ra from the as-built condition.

Impact of Media Size on Ra

Media size directly determines the scale of surface modification. Coarse media (large particles, low mesh number) creates larger impact zones and removes larger peaks more rapidly, but leaves a coarser surface texture at the micro-scale. Fine media (small particles, high mesh number) creates smaller impact zones, producing a finer micro-texture but requiring longer cycle times. For the lowest achievable Ra in a single-stage blast, use the finest media that achieves adequate coverage in a reasonable cycle time.

4. Ra Achievable by Media Type and AM Material

Note: Al₂O₃ is not used on titanium or stainless steel due to iron-free media requirements. All values are approximate ranges under optimised parameters; actual results depend on specific machine, powder, parameters, and part geometry.

5. The Cascade Blasting Technique

Cascade blasting is the sequential use of two or more blast stages with progressively finer media — starting with a coarser media to rapidly reduce as-built Ra, then following with a finer media to refine the surface to the target Ra. It is the most efficient approach when the as-built Ra is very high (EBM parts, FDM vertical surfaces, SLM down-skin surfaces) and the target Ra is low.

Example cascade for EBM Ti-6Al-4V (as-built Ra 25–35 µm, target Ra ≤ 3 µm):

1. Stage 1: Glass beads 80–100 mesh at 70–80 psi — rapid peak removal; reduces Ra from 25–35 µm to 5–10 µm in 3–5 minutes
2. Stage 2: Glass beads 150–200 mesh at 60–70 psi — surface refinement; reduces Ra from 5–10 µm to 1.5–4 µm in 3–5 minutes

Total cascade time: 6–10 minutes per part, achieving Ra below 4 µm. Compare to using only 150–200 mesh glass beads from the start on the same part: 10–15 minutes to achieve Ra 4–6 µm, without reliably reaching below 4 µm within reasonable cycle time.

The cascade principle also applies to Al₂O₃ blasting for pre-coating preparation:

1. Stage 1: Al₂O₃ 80 mesh at 65–80 psi — aggressive peak removal, cleaning, and oxide removal
2. Stage 2: Al₂O₃ 150–180 mesh at 55–65 psi — refine anchor profile to target Ra 2–4 µm for paint/powder coat

6. Media Sequence and Diminishing Returns

Understanding the diminishing returns of repeated blasting at the same parameters prevents over-processing and media waste:

• First pass: Largest Ra reduction — 50–70% of the total achievable reduction from blasting occurs in the first pass. The as-built peaks (tallest features) are addressed most rapidly.
• Second pass: An additional 15–25% Ra reduction is typical. Remaining smaller peaks are flattened.
• Third pass: 5–10% additional Ra reduction. Diminishing returns are pronounced.
• Fourth pass and beyond: Less than 5% additional Ra reduction; risk of increasing Ra by creating a more irregular micro-texture; unnecessary media and equipment wear.

Practical rule: define blast cycle time and number of passes by Ra measurement on first-article coupons, not by guesswork. Once the first-article Ra target is confirmed at a specific number of passes and parameters, use that as the production specification.

7. When Blasting Is Not Enough: Ra Below 0.8 µm

For applications requiring Ra below 0.8 µm — precision sealing surfaces, optical components, PVD coating substrates, certain medical devices — abrasive blasting must be followed by a secondary finishing process. The blasted surface at Ra 1–3 µm serves as the starting point for these secondary processes:

The blasted surface at Ra 1–3 µm requires significantly less secondary processing than the as-built AM surface (Ra 8–35 µm) to reach the same final Ra. Using blasting as a mandatory pre-step before secondary finishing is the most efficient workflow for tight Ra specifications.

8. Surface Roughness Measurement Standards for AM Parts

Accurate Ra measurement requires correct selection of measurement parameters, particularly the cutoff wavelength (λc) and the evaluation length. For blasted AM surfaces:

• ISO 4287 (profile) / ISO 4288 (measurement conditions): The primary standards for Ra and Rz measurement using contact profilometry. For blasted AM surfaces with Ra 1–10 µm, λc = 0.8 mm is the standard cutoff wavelength.
• Evaluation length: 5 × λc = 4.0 mm (five sampling lengths). For small AM part features where 4 mm measurement length is not available, document the reduced evaluation length and note this on the inspection record.
• Measurement direction: AM surfaces are anisotropic — Ra measured parallel to build layers differs from Ra measured perpendicular to build layers. Measure in the most critical direction for your application, or measure in both directions and report both values for a complete characterisation.
• ISO 25178 (areal): For R&D and critical applications, areal measurement using focus variation, confocal, or white light interferometry provides Sa, Sz, Sku, and Ssk — a complete three-dimensional surface characterisation that is more appropriate for the complex, isotropic blasted AM surface than a single-profile Ra measurement.
• Instrument calibration: Stylus profilometers must be calibrated with traceable reference specimens per ISO 12179. For aerospace and medical QC, calibration records must be maintained and available for audit.

9. Ra Acceptance Criteria by Application

For more on pre-coating specific Ra requirements, see: Pre-Coating Surface Preparation for 3D Printed Parts: Blasting Before Paint, Powder Coat, and PVD. For shot peening and fatigue life, where Ra of the pre-peen surface is a key input parameter, see: Shot Peening 3D Printed Metal Parts: Improving Fatigue Life with Compressive Residual Stress.

Preguntas frecuentes

---

Published July 2026 by Jiangsu Henglihong Technology Co., Ltd. — Specialists in industrial abrasive blasting media for additive manufacturing post-processing.