Metal Abrasives Precision Surface Control for CNC Machined Parts

• 1
  What metal abrasives actually do to a machined surface
• 2
  Shot, grit and cut wire: how to pick the media
• 3
  Blast parameters that decide the result
• 4
  Where blasting helps a prototype and where it hurts
• 5
  How to verify the blasted surface before it ships
• 6
  Media and process selection by part condition

Every machined surface carries evidence of how it was cut. End mills leave cusps, turning inserts leave feed marks, and both leave a cold-worked layer with tensile stress near the surface. Metal abrasives precision surface work removes that layer and replaces it with a controlled one. The mechanism is simple: a hard particle hits steel or aluminium at speed, deforms the top few micrometres, and either cuts the peaks away or hammers them flat.

The two families behave differently. Cast steel shot is round. It peens rather than cuts, so material is displaced instead of removed. Cast steel grit is angular, so each particle acts like a tiny cutting tool and removes metal. Cut wire shot is round and uniform, made by cutting wire to length, and produces the least dust with a very even impact pattern. Choosing between them is the first real decision on any blasting job.

Impact energy drives everything downstream. At wheel speeds of 60–80 m/s, a 0.6 mm steel particle carries enough energy to form a dent roughly 5–20 μm deep. That dent raises surface roughness and creates compressive stress underneath. Higher speed or larger media means deeper effect, more roughness and higher risk to thin walls. Lower speed means a lighter etch that may not clean mill scale at all.

The engineering consequence is that blasting is a sizing operation too. You cannot specify a tolerance of ±0.005 mm on a feature and then blast it freely. Either mask the feature, control the erosion to under 0.02 mm, or blast before the finishing cuts. We treat blasting as a process step with its own drawing callout, not as a cleanup after machining.

Cast steel shot comes in graded sizes from S70 through S780 and runs around 40–50 HRC. The round shape produces the highest compressive stress for a given mass and leaves a peened, dimpled surface. Use it when fatigue life matters more than surface roughness: shafts, connecting rods, spring seats, anything cyclically loaded. It is a poor choice when you need to remove sharp burrs, because round media folds burrs over instead of cutting them off.

Cast steel grit (G10–G120, typically 45–60 HRC) has angular particles and cuts. Removal rates reach roughly 0.05 mm per pass on mild steel at 4–6 bar. That makes it the right media for deburring machined edges, stripping heat tint from stainless, and preparing a surface for coating. The trade-off is roughness: grit raises Ra noticeably, so it is usually followed by a lighter shot pass if the drawing calls out a finish value.

Cut wire shot is made from high-tensile wire in the 1,800–2,200 MPa range, cut to 0.3–2.0 mm lengths and conditioned so the particles are uniform. Because it is uniform, it produces a predictable, repeatable surface and very little dust. It is the media we reach for on thin sections, medical and food-contact parts, and any job where cross-contamination between materials is a concern.

Quenched and tempered variants sit between these categories. Tempering changes the particle hardness and wear rate, which matters on long production runs where media breakdown shifts the surface over thousands of cycles. Fresh media and worn media are not the same process; both the roughness and the coverage rate drift as particles round off or fracture.

• 1
  Round shotPeening and compressive stress; poor at cutting burrs
• 2
  Angular gritCutting and cleaning; raises Ra
• 3
  Cut wireUniform, low dust, good on thin walls
• 4
  Check media wearRoughness drifts as particles round off

Pressure and wheel speed set the impact energy. Pneumatic blasting usually runs at 2–6 bar depending on the nozzle and the work. Wheel blasting runs at 60–80 m/s. Doubling pressure does not double cleaning speed; it roughly doubles the damage risk on thin sections. On walls of 1–3 mm, we keep pneumatic blasting at 2–4 bar and accept a slower pass rate to hold erosion under 0.02 mm.

Coverage is the second control. A single pass at normal speed covers most of the surface but leaves 2–5% of the area untouched, which is enough to cause patchy coating adhesion. Full coverage usually needs two or three overlapping passes, or mechanical rotation of the part. Coverage rates around 98% in air blasting at 4–6 bar are achievable with a fixed nozzle distance and a rotating fixture. Hand-held blasting rarely reaches that consistently.

Nozzle distance and angle matter more than most people expect. Past roughly 150 mm, impact energy falls off sharply because the particle stream spreads. Angles beyond about 45° from normal produce a smeared surface rather than a cut one. For internal bores and deep pockets, you need an angled lance or you simply will not treat the bottom of the feature at all.

Contamination control is the last parameter and the easiest to lose. Aluminium dust left in a cabinet will embed in the next stainless batch. Media must be segregated by material family, and the cabinet cleaned between jobs. If your drawing says no ferrous contamination, this is the step that determines whether you pass inspection.

Blasting earns its place when the surface has a functional job. It improves coating and adhesive bond strength, removes tool marks and oxide before anodizing or plating, and raises fatigue life on cyclically loaded parts through compressive stress. On aluminium housings heading for anodize, a light grit pass followed by a short shot pass gives a uniform matte that hides machining marks without erasing the part's form.

It also has hard limits. Blasting cannot fix a surface that is out of tolerance, and it cannot remove deep chatter marks without removing metal you need. It rounds sharp edges, which is fine on a bracket and unacceptable on a sealing face or a dowel pin hole. If a drawing calls out a sharp edge or a press fit, that feature needs masking.

Thin walls and fine features are the other boundary. A 1 mm aluminium rib blasted at high pressure will bow. Threaded holes collect media that is difficult to remove, and trapped particles cause galling on assembly. We plug threaded features and mask sealing surfaces before blasting rather than cleaning up afterwards.

There is also a finish question. If the drawing calls for Ra 0.2–0.8 μm, blasting will not deliver it; that finish comes from fine milling or turning, and blasting will only make it rougher. Blasting is the step before a coating or a functional requirement, not a substitute for a fine machined finish.

You cannot verify blasting by looking at it. The surface has to be measured. Roughness is checked with a portable profilometer at agreed locations, not at a spot the operator chooses. If the drawing calls for Ra 0.8–1.6 μm after blasting, that number goes on the inspection record alongside the pre-blast value.

Coverage is checked visually under low-angle light against a blasted reference coupon. Almen strips, where the drawing requires them, give a numeric read on peening intensity rather than a visual guess. For coating adhesion, ASTM D4541 pull-off testing on a witness coupon is the usual check, with values above roughly 12 MPa indicating a surface that will hold a coating through service.

Residual stress from shot peening is harder to confirm in-house. Subsurface compressive stress in the range of a few hundred MPa is what the process is aiming for on fatigue-critical parts, and it is normally validated once during process qualification rather than on every batch. Once the parameters are locked, the record is the pressure, media size, speed and coverage.

Cleanliness matters as much as roughness. Aerospace and medical work often calls for ASTM A380-level cleanliness, which means no embedded media, no residue and no cross-material contamination. We inspect with magnification and, where specified, with solvent wipe tests. If a job fails here, it usually fails because media was not segregated between materials.