- (972) 475-2726
- Follow Us
Menu
Eighteen things worth knowing before your part reaches a coating line — written by the people who run one. Most of what goes wrong in coating is decided long before the part gets here, on the drawing and on the purchase order.
Request a Quote(972) 475-2726Coating adds material. Every rule in this group exists because a part came back that would not assemble.
Any operation that cuts, heats or deforms the part after coating destroys the coating locally, and any weld burns it off entirely. Machining, welding, forming, stress relief and heat treat all come before finishing. If a feature genuinely has to be machined after coating, tell us on the print so we can mask it or leave stock.
Typical powder film build is 2–3 mils, and it lands on both walls of a hole or slot. A 1.000″ bore coated at 3 mils finishes near 0.994″. Liquid systems run thinner — a primer-plus-topcoat aerospace stack is roughly 2.3–3.2 mils total, and a MIL-PRF-23377 primer alone is 0.6–0.9 mil. Cerakote is thinner still, typically 0.75–1 mil, and Micro Slick runs 0.1–0.5 mil.
What we cannot do is guess which one you wanted.
Masking is skilled hand work and it is quoted by the feature. “Mask as required” is not a specification. It also matters why a surface is masked: a thread that must gauge, a bore that must hold a press fit, a flange that must seat flat and an electrical ground point are four different problems with four different solutions.
A masked boundary is a physical edge, not a machined feature. There is a build-up ridge at the mask line roughly equal to the film thickness, and the boundary is a transition zone rather than a line. If you need tighter than that, the answer is a machined-after-coating feature or a tooled mask — tell us early and we will quote the tooling instead of surprising you with it.
Two things happen at a sharp edge. During application, the electrostatic field concentrates and distorts. During cure, surface tension drags the molten film away from the edge toward the adjacent flat, leaving the thinnest film at exactly the point that needs the most. That is where corrosion starts.
ISO 8501-3 preparation grade P3 calls for a minimum 2 mm edge radius, referencing ISO 12944-3. Breaking sharp edges before coating buys more corrosion life than any upgrade to the coating itself.
Every coated part has at least one contact point, because it has to hang from something and it has to be grounded. Rack marks are normal and their location is at our discretion unless you tell us otherwise.
If a surface is cosmetically critical or must be fully coated, call out where we may and may not make contact, or design a hanging hole or tab into a non-critical area. A 1/8″ hole in a flange costs nothing and solves the problem permanently.
Powder is applied electrostatically, and the charge follows the outside of the part. Inside corners, deep recesses and channels see a weaker field — the Faraday cage effect — and coat thinner than the flats. Liquid spray reaches into geometry that powder struggles with, which is one reason we run both.
If an internal surface carries the corrosion requirement, say so. We will change the process, the racking or the gun setup rather than let it run thin and hope.
A cure oven is a heat treatment. Half the coating failures we see are metallurgy, not paint.
Standard powder cures around 10 minutes at 400°F part metal temperature, not oven air temperature. Low-cure grades exist down to roughly 340°F, and specialty low-temperature powders run lower still.
That matters on heat-treatable aluminum. 6061-T6 is artificially aged at roughly 320–350°F for 8–18 hours, and the alloy is generally considered stable to about 300°F. A 400°F cure sits above both. The exposure is short — ten or twenty minutes against an eighteen-hour aging cycle — so the effect is far smaller than a re-age, but it is not zero and the direction is toward softening. Published work on 2024-T3, 6061-T6 and 7075-T6 found a high-temperature cure negatively affected mechanical properties.
If the part is structural or flight-critical in a heat-treatable alloy, the defensible answers are a low-cure powder, a liquid system, or coupon testing signed off by your engineering group. Tell us the alloy and temper and we will tell you which we recommend.
Cast material is porous, and the gas trapped in that porosity expands when the part heats through the coating’s gel point. It blows through a film that is already too viscous to flow back, and you get pinholes, craters and blisters.
The fix is a degas bake before coating — typically heating the part above the subsequent cure temperature and coating it before it cools. Outgas-forgiving powders that flow longer before gelling are also available. Both cost time, and we would much rather build them into the quote than discover the problem on the first article.
Hot-dip galvanizing traps volatile material that erupts during cure, and zinc corrosion products on the surface trap moisture and blister. ASTM D7803 covers the preparation. The bake runs above cure temperature, and the sweep blast has to be gentle — a shallow angle with a soft, fine abrasive — or you strip the zinc you paid for.
Call out galvanized material on the print. We will not guess it from the photo.
A coated thread grows by the film thickness on every flank, and a class-fit thread will not gauge afterward. There are three honest answers and they cost different amounts:
Hydrogen embrittlement is introduced by electroplating and by acid pickling and cleaning — not by blasting or powder coating. It matters here because plate-then-paint sequences and acid-cleaned high-strength parts fall inside the requirement.
ASTM B850’s bake table starts at 1,001 MPa / 145 ksi / 31 HRC and scales the bake duration with strength, all at roughly 375–430°F. Baking has to begin promptly after plating. You must identify the requirement on the purchase order, along with the material condition and the governing specification — we cannot infer tensile strength from a drawing.
Thicker is not automatically better. Excess film build causes sagging, poor edge definition, and gas entrapment that shows up as pinholes. TGIC-free polyester in particular has a real ceiling around 3.5–4 mils because its cure reaction releases water that has to leave the film.
Specify the build the service actually needs, and let the coating manufacturer’s data sheet set the range. If you are not sure, send us the environment and we will recommend one.
Coatings very rarely fail because the coating was wrong. They fail because of what was under it.
SSPC-SP 10 / NACE No. 2 tells us how clean. It says nothing about how rough. Anchor profile is a separate specification, measured per ASTM D4417, and both belong on the print.
For aluminum, stainless and other non-ferrous work, SSPC-SP 16 and SP 17 are the correct standards to cite rather than SP 7. Note that although SSPC and NACE merged into AMPP in 2021, the existing standards keep their designations — “SSPC-SP 10 / NACE No. 2” is still how you write it.
If the profile peaks are taller than the film, the tips of the steel sit essentially unprotected under the coating and you get pinpoint rusting through a film that measures fine everywhere else. There is a common shop rule that film should be about three times the profile depth; treat it as a field heuristic, not a standard — the governing number is the coating manufacturer’s data sheet.
Media matters here. Coarse aluminum oxide produces a much deeper profile than fine; glass bead produces essentially no anchor at all.
Embedded ferrous particles rust, and on stainless they set up galvanic corrosion in a material that was specified precisely to avoid it. Aluminum oxide or garnet for non-ferrous work, always. Cerakote goes further and prohibits round media entirely — glass bead and steel shot peen and dimple the surface rather than cutting the profile the coating needs.
Freshly blasted steel has no oxide protection and enormous exposed surface area. Rust-back can begin in minutes or take weeks, depending entirely on humidity and residual salts. There is no universal number — specifications in common use range from same-shift to 24 hours, and MIL-DTL-53072G requires blasted surfaces to be coated within four hours.
The condition that actually governs it is dew point: final blast should not happen when the steel is less than 5°F above dew point. If visible rust develops before coating, the surface gets re-cleaned. That is not a negotiation, it is the only way the coating holds.
Silicone measures around 15–20 dynes/cm surface tension against 40–50 for a coating. The film cannot wet it, so it retracts and leaves a crater. Concentrations as low as 0.001% are enough to cause widespread fisheyes, and it travels by aerosol — which is why silicone contamination presents as a plant-wide outbreak rather than one bad part.
The usual sources are mould release, sealants, and consumer aerosols including WD-40 and silicone spray. If your upstream process uses any of them, tell us, because remediation after the fact is close to impossible and prevention is the only real control.
Mild steel cut with oxygen assist gas leaves a dull, dark oxide on the cut edge that coatings struggle to adhere to. It passes every flat-panel test and fails on the edges in the field. Brush deburring, blasting, or switching the laser to nitrogen assist all solve it.
Mill scale is worse in a quieter way: it is cathodic to the steel beneath it, so wherever it cracks it drives accelerated corrosion of the exposed metal. Chemical pretreatment does not remove it. That is the real reason a cheap brush-off spec on hot-rolled steel underperforms.
None of this is bureaucracy. Every item here is something that, when it is missing, costs a day or a lot.
The finish spec with class, type, grade and thickness where the spec requires them — and the revision level. We coat to the latest revision in effect unless you fix one on the PO.
Base material, alloy, temper or heat-treat condition, and any prior coating. This drives pretreatment, cure temperature and whether the part needs a bake.
Which features stay bare, whether the requirement is “no coating” or “no coating build,” and the boundary dimensioned from a datum.
A color callout is not a color standard. If appearance is controlled, send a current approved sample on the same alloy, temper and mechanical finish.
Assembly, welding, adhesive bonding, EMI grounding, gasket seating — all of it changes what we should apply and where.
Every part needing stress relief or embrittlement relief, with the bake time, temperature and governing specification.
Anything beyond our standard Certificate of Conformance — salt spray, cross-section thickness, adhesion, FAI to AS9102 — must be on the order at the time of order.
Any drawing, specification or part subject to ITAR, EAR or government flow-down. We are ITAR registered and handle controlled data under access control, but you have to tell us it is controlled.
If your material is not on this list, send the drawing — most of what we run got added because a customer asked.
| Substrate | Powder | Liquid / MIL-spec | Cerakote | What to watch |
|---|---|---|---|---|
| Carbon & alloy steel | ✓ | ✓ | ✓ | Mill scale and flash rust. Blast-to-coat window is the real constraint. |
| Stainless steel | ✓ | ✓ | ✓ | Never blast with carbon steel media. Needs a mechanical profile or conversion coating. |
| Aluminum, wrought | ✓ | ✓ | ✓ | Conversion coating per MIL-DTL-5541 or a chrome-free equivalent. Oxide reforms fast — coat promptly. |
| Aluminum, heat-treated (T6, T3) | ● | ✓ | ✓ | Cure temperature vs. temper. Low-cure powder, liquid or Cerakote are the safer routes. |
| Cast aluminum & iron | ● | ✓ | ✓ | Outgassing. Needs a degas bake and often an outgas-forgiving powder. |
| Zinc die cast | ● | ✓ | ✓ | Porous and heat-sensitive. Low cure temperatures only. |
| Galvanized | ● | ✓ | – | Zinc corrosion products and outgassing. ASTM D7803 prep, gentle sweep blast. |
| Titanium | ✓ | ✓ | ✓ | Straightforward. Cerakote is common here for thin film on tolerance-critical parts. |
| Brass, copper & bronze | ✓ | ✓ | ✓ | Non-ferrous media only. |
| Plastics, polymers & composites | – | ✓ | ✓ | Cannot see a standard powder cure. Cerakote publishes a 150–180°F schedule; liquid cures at ambient. |
| Assemblies with seals, bearings or wiring | – | ✓ | ● | Anything that cannot see 350–400°F is a liquid or air-cure part. |
This table is a starting point for a conversation, not a specification. The governing document is always the coating manufacturer’s data sheet and the specification on your print.
Typical film build is 2–3 mils per surface, so a hole or slot loses roughly twice that on the dimension. Liquid systems generally run thinner — an aerospace primer-plus-topcoat stack is about 2.3–3.2 mils total — and Cerakote is thinner still at roughly 0.75–1 mil, with dry film lubricant down around 0.1–0.5 mil. If a feature is dimensionally critical, dimension for the coating, mask it, or authorize a post-coat chase.
It sits above where 6061-T6 is aged and above the temperature at which the alloy is normally considered stable, and published testing found high-temperature cure negatively affected mechanical properties in 2024-T3, 6061-T6 and 7075-T6. The exposure is short, so the effect is much smaller than a full re-age — but it is not nothing. For structural or flight-critical parts we recommend a low-cure powder, a liquid system, or coupon testing your engineering group signs off on. Send us the alloy and temper and we will make a recommendation.
Yes, and there are three ways to handle them: coat them, mask them, or coat and chase them. Coating and chasing is often cheaper and more repeatable than masking, but it means running a tap through your part, so it has to be authorized on the print. Call out the thread size, class, and which approach you want.
Every racked part has at least one contact point — it has to hang and it has to be grounded. Rack marks are a normal characteristic and their location is at our discretion unless the drawing specifies contact locations. If a surface is cosmetically critical, tell us, or design a hanging feature into a non-critical area.
Cleanliness (SSPC-SP 10, for example) says how much of the original surface has to be gone. Profile says how rough what remains has to be, measured per ASTM D4417. They are independent and both belong on the drawing — a surface can be perfectly clean and far too smooth for the coating to key into, or rough enough that the profile peaks are taller than the film.
If the part is steel at or above roughly 145 ksi tensile or 31 HRC, and it has been electroplated or acid cleaned, yes — per ASTM B850, with the duration scaling by strength. Blasting and powder coating do not introduce hydrogen. Identify the requirement on the purchase order along with the material condition and the governing specification; we cannot infer tensile strength from a drawing.
Usually not. In our experience the overwhelming majority of coating failures trace to preparation — residual contamination, an inadequate or excessive profile, flash rust under the film, mill scale left in place, silicone contamination, or a casting that was never degassed. That is why this guide spends more space on prep than on coatings.
Yes. Liquid systems cure at ambient or force-dry at low temperature, and Cerakote’s C-Series air cures with no thermal load on the part at all. Tell us the temperature limit and we will pick a system that lives under it.
Wall-format chart mapping each process to its governing specifications, typical film build, cure schedule and best-fit applications.
One page: every process we run, the substrates we handle, part-size capacity and the specifications each process meets.
Wet paint, powder and Cerakote side by side — properties, film build, cure, and honest notes on when each one is the wrong answer.
Send the drawing and the service environment. We will tell you what we would run and why — including when the answer is something we do not do.
