Wafer Handling — Where Coplanarity Matters More Than Any Single Dimension
Four defects that show up as particles, not as scrap
- Non-coplanar contact pads. Pads machined in separate setups land on slightly different planes. The wafer then touches the high points only, and every touch is an abrasion event.
- Rolled or torn edges on the contact surface. A burr on a wafer-facing edge breaks away over cycles. The part measures correctly and continues to shed material, which is the worst combination.
- Deflection from asymmetric material removal. Lightening pockets remove mass unevenly. Thin arms bow, and the bow is not visible until the part is off the fixture — the same spring-back logic as any thin-wall part, at a smaller scale.
- Damaged datum features. Dowel holes and mounting datums establish where the wafer sits. A reamed hole with a lead-in burr changes the assembled position by an amount that no amount of pad flatness can recover.
If the part holds vacuum against a chamber wall rather than against a wafer, the controlling defect is a leak path rather than a particle — see Vacuum Chamber Component Machining.
What drives tooling on a handling component
| Feature | What usually decides the tool | HENTRY entry |
|---|---|---|
| Contact pads and shallow pockets | Floor flatness and a clean edge, cut in one setup with the datum | PCD Milling Cutter · PCD Square End Mill |
| Vacuum grooves on a chuck face | Groove floor flatness and a clean break at the groove edge | Grooving Tools · PCD T-Slot Cutter |
| Lightening pockets and ribs | Cutting force, because force is what bows a thin arm | High Feed End Mills · Roughing End Mills |
| Dowel and mounting holes | Position and a burr-free lead-in, since the datum sets the assembled plane | Carbide Reamer · Micro Carbide Drills |
| Wafer-facing edge breaks | Cutting the chamfer rather than rubbing it, so no material is pushed over the edge | PCD Chamfer / Spotface · Chamfer Mill |
Which alloy pushes you where
| Material | Route | Why |
|---|---|---|
| 6061-T6 plate and bar | PCD for faces and pads, carbide for holes | Clean shear on large faces and long life where the part runs in volume |
| 5083 aluminum | PCD with a sharp edge, controlled feed | Gummy; a dull edge smears and the smear becomes the particle source |
| Aluminum-silicon-carbide | PCD only, with edge geometry that shears the particle phase | Abrasive enough that carbide life is measured in single parts |
| 316L fixtures and brackets | Carbide throughout | Work hardening; a rubbing pass hardens the next one |
| Engineering polymers | Sharp single-flute or PCD geometry with a defined exit | Melting and fuzzing at the edge rather than a conventional burr |
Boundaries, so you do not read the wrong page
- Chamber bodies, gas lines and seal faces — leak paths and trapped volumes → Vacuum Chamber Component Machining.
- Liquid cooling and micro-channel parts — fin geometry and brazing faces → Semiconductor Cooling Plate Machining.
- Small holes with no flatness or particle requirement → Small-Diameter Holemaking.
- General aluminum housings on line takt → Automotive & EV Powertrain.
What makes a handling-component quote reviewable
- Which surfaces contact the wafer, and which are datum or mounting faces
- Coplanarity or flatness callout, and the inspection method used to verify it
- Particle or cleanliness requirement, and the cleaning step that follows machining
- Wall and arm thickness, plus where mass is being removed
- Whether pads and datums can be machined in one setup
- Downstream finish — anodise, coating — and its surface expectation
Quick answers before you RFQ
Why does machining one setup matter so much on a chuck face?
Because coplanarity is a relationship between features, not a property of any one of them. Pads cut in two setups are each flat and still sit on different planes, and no measurement of a single pad reveals it. Where the print demands coplanarity, the setup plan is part of the tooling review rather than a detail left to the shop.
How do you avoid burrs on a wafer-facing edge?
By not creating one. Edge geometry, cut direction and the exit condition decide whether material shears or folds, and a chamfer cut with a sharp geometry removes the folded material instead of pushing it over the edge. Where a burr is unavoidable, it becomes a defined operation with a defined tool rather than a hand finish.
Is PCD worth it on a one-off end effector?
Rarely. PCD earns its cost through life and through avoiding built-up edge on a long run. On a single part, the geometry and the setup matter far more, and a carbide cutter with the right corner radius will usually give the better result for less.
Can thin arms be machined without bowing?
Usually, but only if the force and the clamping are treated as the problem. Symmetrical removal, supported clamping and a lower-force cutter path all reduce the bow. Where the geometry forces asymmetric removal, the part is roughed, released and finished — the same release-then-finish logic used on any thin-wall component.
Send the print with wafer-contact surfaces and datums marked
Primary tooling entries for handling parts: PCD milling cutters and square end mills for pads, grooving and T-slot geometry for vacuum grooves, high-feed cutters for lightening pockets. Coplanarity is reviewed with the setup plan.