Wrap-up
67%
Surface prep — the hidden critical step
Two identical punches, two coating houses, very different lives. Why surface prep determines coating performance more than any other variable.
Tying it together
The three adhesion inhibitors
Per Northeast Coating and Semicore Equipment's prep guides, pretreatment exists to remove three categories of contaminant:
- Oils — cutting fluid, anti-rust film, fingerprint oils, hydraulic film from press handling
- Particles — grinding swarf, dust, polishing media residue
- Oxides — surface oxide layers, rust spots, scale
All three have to be gone before the coating goes on. Coating houses run cleaning, etching, and degreasing cycles to handle this — but they can only handle what's possible to remove. Embedded contaminants (oil that's soaked into a porous polished surface, abrasive media that's lodged in microcracks) cannot be cleaned out in the coating house.
Surface roughness target
Per KYOCERA's coating-prep guide, the ideal substrate roughness is Ra 0.05 to 0.2 µm. Smooth enough that there are no deep contamination traps, rough enough for mechanical interlocking. Polishing data referenced in the KYOCERA guide shows polished surfaces achieving 60-80 MPa adhesion compared to 30-40 MPa unpolished. That's a 2x improvement in bond strength purely from surface finish.
What kills surface prep
- Silicone-containing polishing compounds. Per Dayton Tool Grind & Coat, silicone residue is one of the worst PVD adhesion killers. Polishes are sold with and without silicone; if your shop polishes parts before sending them out, switch to silicone-free explicitly.
- Excessive grinding heat. Per the same source, heat generated during finish grinding can locally anneal the surface (creating soft spots the coating cracks over) or induce stress cracks. Slow feeds, fresh wheels, good coolant flow.
- Prior incompatible surface treatments. Wet bath platings, black oxide, and nitriding (with the compound layer intact) are explicitly called out as treatments that interfere with PVD adhesion. If a part has been treated, the coating house needs to know — they may decline to coat it, or they may need to strip the prior treatment first.
What this means on the shop floor
- Don't grease, oil, or coat a polished tool before sending it out for coating. The coating house has to remove anything you put on. A clean tool wrapped in clean paper or a vapor-corrosion-inhibitor bag is fine. A tool dripping with way oil is a problem.
- Establish a polish-and-handle protocol once and stick to it. Decide whether your shop polishes or whether the coating house polishes. If your shop polishes, lock the compound choice (silicone-free) and the procedure (no heat, no contamination). If the coating house polishes, ship the part rough-finished and let them do it.
- Communicate prior treatments. If the part has been nitrided, EDM'd, ground, or anything-else'd before coating, the coating house needs the history. They can't see "this was nitrided three years ago" by looking at it.
Next: putting all five process choices together. When do you pick coating? Nitriding? Heat treat alone? Some combination?