Guide
Which anodizing defects make a lot rejectable, and how do you trace anodizing defects back to alloy, machining or the anodizer?
In receiving inspection, anodizing defects are visible or measurable departures from a sound anodic film: burned spots, pits, streaks, contact marks in the wrong place, white spots, soft or powdery film, crazing, and uneven or bled-out color. Some start in the anodizing line, but many start earlier, in the alloy, the casting or the way the part was machined, cleaned and stored, so naming the defect correctly is the first step to deciding who fixes it.
By Saola Manufacturing engineering team · Published · Updated
This guide is for quality engineers, design engineers and buyers who have a lot of anodized aluminum parts on the receiving bench and need to answer three questions quickly: is this a defect under the specification we actually wrote, where did it come from, and do we rework it, use it as is or send it back. It uses the workmanship and lot-acceptance clauses of the military anodizing specification most US drawings cite, adds a defect atlas that maps each symptom to its usual owner, and ends with the drawing and purchase order wording that stops the same argument on the next lot. Color drift between lots is covered separately in the anodize color matching guide; this page is about defects in the film itself.
On this page
What counts as an anodizing defect, and which ones make a lot rejectable?
A defect is rejectable when it breaks a requirement you can point to: the workmanship clause of the cited specification, a contact-mark location on the drawing, or a cosmetic standard named on the purchase order. Powdery or loose film, breaks in the coating and damage that reduces serviceability fail the specification on their own; most purely cosmetic marks fail only if your documents say so.
The workmanship paragraph of the military specification requires the anodic coating to be continuous, smooth, adherent and uniform in appearance, and free from powdery areas, loose films, breaks, scratches and other defects that would reduce the serviceability of the parts (MIL-PRF-8625F, paragraph 3.13). The same paragraph says that appearance differences caused by the base metal itself, such as welds, a component with both cast and machined surfaces, or grain-size differences within a forging, are not cause to reject the coating unless the contract or purchase order says otherwise, and it allows slight discoloration where sealing solution drips or runs out of designed crevices. Contact marks are expected: paragraph 3.13.1 only asks that their size and number be the minimum consistent with good practice, and fixes their location only if the drawing, contract or purchase order specifies one.
That leaves a practical split for anyone writing a rejection. Functional defects, meaning soft or powdery film, coating breaks, burns that have eroded the metal, pits on a sealing face, crazing on a part that must resist corrosion, and bare areas where coating is required, are rejectable against the specification itself. Cosmetic defects, meaning streaks, mottling, a rack mark on a visible face, faint water marks or a shade difference, are rejectable only against a cosmetic standard you supplied: limit samples, marked significant surfaces, or written viewing conditions. If none of those exist, a cosmetic rejection is an opinion, and the anodizer will usually win the argument.
Lot acceptance under the specification is also narrower than many buyers assume. Conformance inspection consists of a visual examination, a color examination for dyed coatings and a thickness check for hard coat, sampled from each lot under MIL-STD-1916 at Verification Level II, with the acceptance criteria to be set in the contract or order (paragraph 4.4). A lot is one part number run in the same tank, by the same process, type and class, and no larger than one shift’s output. If your purchase order does not set them, there is no agreed accept number to reject against.
What causes each anodizing defect, and who usually owns it: alloy, machining or the anodizer?
Burning, soft film, seal smut, chloride pitting and crazing already present at receiving are process defects owned by the anodizer, while streaks, blotchy color, porosity bleed-out and many pits trace back to the alloy, the casting or the machining and handling before the part reached the tank. Contact marks are shared: the anodizer controls their size, but only the drawing can control where they go.
The specification itself names the main upstream causes. It notes that severe attack by the electrolyte on castings or welds can come from unsound castings, poor welding practice, a weld filler that differs in composition from the base metal, or electrolyte retained in cracks, crevices and irregular surfaces, and that contaminants in the electrolyte, particularly chlorides, or improper racking can cause the same attack (MIL-PRF-8625F, paragraph 6.4). It explains burning as a current problem: a larger part needs proportionally more current, and forcing that current through contacts sized for a smaller part may burn it, which is why bigger parts need larger or more numerous contact marks (paragraph 6.14). It also requires parts to be anodized after all heat treatment, machining, welding and forming, and bars iron-bearing abrasives such as steel wool and steel wire for mechanical cleaning before anodizing.
Crazing deserves a separate note because it often appears after the parts leave the anodizer. Anodizing consultant Anne Deacon Juhl, writing in Products Finishing, puts the main cause on the thermal expansion mismatch between aluminum and its oxide, notes that fine cracks can appear when anodized parts reach roughly 80 to 100 °C, and says thicker films crack more readily, sealing lowers the temperature at which crazing starts, and cracks through to the metal are common at sharp corners on hard coat (Products Finishing, crazing and cracking in anodizing). The Qualanod defects catalog adds thermal shock between rinse and hot seal, bending or impact after anodizing, and residual machining stress as triggers (Qualanod, process defects 3.7). A lot that was clean at receiving and crazed after a bake, a cure or a hot wash was cracked by that heat, not by the tank.
| Defect | What it looks like | Most likely causes | Usual owner |
|---|---|---|---|
| Burning | Bright or rough spots ringed by darker film, usually at edges, corners or next to a contact, with little or only soft film in the spot | Current density too high; poor electrical contact between rack and part; too little electrolyte movement or cooling; burrs and sharp edges concentrating current | Anodizer; machining when burrs or knife edges are left on |
| Contact (rack) marks | Small uncoated or thin spots where the part touched the rack or wire | Unavoidable electrical contact; size grows with part area (paragraph 6.14) | Drawing owner for location; anodizer for size and number |
| Pitting | Deep, dark, star-shaped pits with no film inside (formed during anodizing), or shallow white pits covered by film (formed before it) | Chloride in the anodizing electrolyte; chloride-bearing rinse water; wrong etch chemistry; acid splashes or corrosion while parts waited; casting porosity | Anodizer for pits under no film; storage, handling or the casting if the pit was there before the film grew |
| Streaks and banding | Lines along the extrusion or rolling direction, around welds, or following drip paths | Die lines, segregation or uneven iron-bearing phases in the extrusion; extrusion seam welds; weld and heat-affected zones; etch solution drying on the part before rinsing | Bar or extrusion supplier, or the weld (weld-zone differences are not rejectable by default); anodizer for etch and rinse stains |
| White spots and blotches | White or pale patches, often round, sometimes fingerprint shaped | Corrosion before anodizing: fingerprints (sweat carries chloride), coolant or moisture trapped under spacers, damp paper or cardboard packing; iron particles embedded by steel wool or wire brushes | Machining, cleaning and storage, or the anodizer’s pretreatment |
| Soft or powdery film | Chalky film that rubs off on a white cloth; poor wear resistance | Electrolyte too warm or too concentrated, too long in the tank, or current density too low for the film thickness | Anodizer |
| Seal smut or bloom | Hazy, powdery or iridescent surface layer after sealing that can be wiped off, leaving sound film underneath | Aluminum hydroxide crystals growing on the surface during hot sealing; seal pH too high; missing or depleted anti-smut additive | Anodizer |
| Crazing and cracking | Fine spider-web network in the film; cracks at sharp corners on hard coat | Heat after anodizing; thermal shock going from rinse into hot seal; thick or hard film; high current density; sharp corners; bending or impact after anodizing; machining stress | Anodizer, design (corner radii) or downstream heat processes |
| Bleed-out and uneven dye | Stains weeping from pores, joints or blind holes; mottled color on castings | Casting porosity (the specification accepts some bleed-out on cast alloys); electrolyte trapped in crevices, press fits or blind holes | Alloy and casting choice; assembly design |
| Bare or thin areas | Uncoated, thin or lighter-colored zones inside holes and pockets or on faces that hung downward in the tank | Air or gas trapped under downward-facing surfaces; oil left in blind holes; shielding by part geometry; masking in the wrong place | Anodizer racking, or design and cleaning before anodizing |
How do you diagnose an anodizing defect before you reject the lot?
Map where the defect sits on the part and across the lot, then run a few simple checks: a rub test, a thickness reading, a look under magnification and, if staining is the complaint, a seal test. The pattern usually tells you whether the cause is the metal, the machining or the tank before anyone strips a single part.
Work through this checklist on a handful of rejected parts and a handful of good ones from the same lot:
- Where on the part? Edges, corners and thin fins point to burning. Pale or thin areas on faces that hung downward, or inside recesses and blind holes, point to trapped gas or solution. A mark in the same place on every part points to the rack position. Fingerprint-shaped or randomly scattered spots on flat faces point to contamination before or during anodizing. Lines that follow the extrusion or rolling direction point to the metal.
- Where in the lot? Ask the anodizer which parts ran on which rack and load. If the defect sits on one rack, one load or one shift, it is a process event. If it follows one bar heat lot or one casting date code, check the material certificates and trace the parts back to their stock.
- Rub test. Rub the area firmly with a clean white cloth. If a hazy or iridescent layer comes off and leaves a clean, glossy film behind, it is seal smut, a sealing-line issue that can often be cleaned off. If the surface keeps chalking however long you rub, the film itself is soft or powdery, which fails the workmanship clause.
- Thickness. Measure film thickness on the defect and on a sound area of the same face with an eddy-current gauge using the method in ASTM B244, keeping the probe off edges and tight radii. A burn usually reads little or no sound film at the spot; a thin reading across a whole face or a whole load points to the anodizing time, the current or the rack contact.
- Magnification. Under a loupe or low-power microscope, check whether there is anodic film inside each pit. Pits with no film over them, often dark and star shaped, formed during anodizing, typically from chloride in the electrolyte; shallow white pits sealed over by film were already in the surface before anodizing, from rinse water, handling or storage. Crazing shows as a fine spider-web network in the top of the film; cracks that run down to bright metal at sharp corners are more serious.
- Seal check. If the complaint is staining, fingerprints that will not wipe off or poor corrosion results, ask for a dye-stain test of the seal per ASTM B136, or the acid-dissolution weight-loss test per ASTM B680.
- History before the tank. Ask how long the parts waited between machining and anodizing, whether they were stored wet or in coolant, whether they were handled with bare hands or packed in damp paper or cardboard, whether steel brushes or steel wool were used for deburring, and whether different alloys or tempers were mixed in the lot.
- History after the tank. For crazing, ask whether the parts were baked, cured, hot-washed, bent or press-fitted after anodizing, and at what temperature.
Strip one sample part only when the checks above disagree. With the film removed, a pit or void that is already in the metal belongs to the material or casting, while a clean surface under the defect points back to the process. The process and handling entries in the Qualanod anodizing defects catalog, with photographs, are a useful reference when the anodizer and your inspector name the same defect differently. Keep the stripped part, the readings and photos together; they become the evidence in the supplier’s 8D report.
Should a defective anodized lot be reworked, used as is or rejected?
Use as is only for cosmetic marks off the significant surfaces, touch up only small inadvertent damage where the specification allows it, and strip and re-anodize only when every dimension and the fatigue requirement can absorb the metal the stripping removes. Reject outright when the defect is functional and rework would push the part out of tolerance.
Touch-up is tightly limited. Per MIL-PRF-8625F, chemical conversion touch-up with MIL-DTL-81706 materials is allowed only for inadvertent mechanical damage such as scratches, the damaged area must not exceed 5 percent of the total anodized area of the item, hard coat may be touched up only where it will not see abrasion, and contact marks are touched up only when the contract asks for it. Touch-up does not cure burns, soft film or pitting.
Stripping and re-anodizing is the usual rework for process defects, but it is not free. Stripping the old film and re-etching before the new one take some base metal with them, and the second film grows partly into the part again, so bores, threads and tight flats move each time; incomplete stripping also leaves remnants of the old film that show through the new one; the anodizing dimensional change guide shows how film growth splits between build-up and penetration. A burned area that has already eroded metal cannot be restored by re-anodizing. Any rework or use-as-is decision on a nonconforming lot should be recorded as a deviation or waiver signed by the design authority, as described in the supplier deviation request guide.
| Finding | Typical disposition | Check before you decide |
|---|---|---|
| Contact marks off significant surfaces | Use as is; marks are permitted unless a location was specified | Whether the drawing named contact-mark zones |
| Contact marks on a specified no-contact surface | Reject, or touch up only if the contract allows contact-mark touch-up | Corrosion exposure of that face |
| Small scratches from handling | Touch up with conversion coating | Damaged area at or under 5 % of the anodized area; no abrasion on hard coat |
| Burns, soft or powdery film | Reject; strip and re-anodize if dimensions allow | Metal erosion under burns; tightest tolerances after rework |
| Seal smut | Return for cleaning and re-inspect; reject if the film underneath is soft | Rub test after cleaning; whether cleaning scratched cosmetic faces |
| Pitting | Reject on sealing, bearing or fatigue-critical faces; engineering review elsewhere | Whether the pits have film inside (formed before anodizing) or none (formed in the tank) |
| Streaks from extrusion or welds | Weld-zone differences: use as is unless the purchase order excludes them; extrusion streaks: judge against your cosmetic standard | Cosmetic standard and significant surfaces |
| Staining, fingerprints that will not wipe off | Hold for seal test; reject if the seal fails | Dye-stain or acid-dissolution result |
| Crazing found after a downstream heat step | Engineering review; usually a process-sequence fix, not a supplier rejection | Temperature and order of bake, cure or wash steps |
What should the drawing, PO and RFQ say so anodizing defects do not recur?
Name the specification with type, class and seal, mark the significant surfaces and the zones where contact marks are allowed, and write the visual acceptance criteria and sampling plan onto the purchase order. Then fix the upstream causes on the drawing: corner radii, alloy and temper, no mixed alloys in a lot, and a note on cleaning and storage before anodizing.
A workable set of notes, to adapt rather than copy:
- Specification. “Anodize per MIL-PRF-8625, Type II, Class 2, black, sealed” or the hard-coat equivalent with its thickness; the MIL-A-8625 types guide explains each designation.
- Significant surfaces. Mark faces “A” (cosmetic, visible on the product) and leave the rest functional. Without this, every face is judged the same.
- Contact-mark zones. “Rack contact permitted only in zone C” or on a named hole or flat. The specification lets you fix the location; it does not do it for you.
- Viewing conditions. State the viewing distance, the light source and how long the inspector may look at each face, chosen to match how the end user sees the part, and name the defects that are never acceptable on significant surfaces (burns, pits, bare spots, powdery film).
- Sampling. Name the sampling plan and accept number on the purchase order, as the AQL sampling guide describes, so lot acceptance is not decided part by part.
- Geometry. Break sharp edges with a radius on hard-coated parts, avoid blind holes and crevices that trap solution where you can, and anodize components before press-fit or welded assembly; the masking guide covers bores and threads that must stay bare.
- Material and handling. One alloy and temper per lot, no iron-bearing abrasives for deburring, parts cleaned and dried after machining, handled with gloves, and not stored wet or packed in damp paper before anodizing.
- After-anodize heat. If the part will be baked, cured, hot-washed or run hot, say so on the RFQ so the film thickness and seal can be chosen for it, and do any bending or forming before anodizing, as the specification already requires.
For castings, keep expectations in line with the metal. The specification accepts dye bleed-out and some lack of color uniformity on cast alloys, and MIL-PRF-8625F paragraph 6.16 notes that a hard coat of 2 mils (about 50 µm) or more is extremely difficult to obtain on high-silicon die castings such as 360, 380 and 383; the die-cast anodizing guide covers which alloys anodize acceptably.
How are anodizing defects handled on a Saola order?
On a Saola order, aluminum parts are machined at our lead CNC plant and anodized at a partner finishing plant we manage, and the quote states where the work is done and the country of origin. Anodizing defects are judged against the specification and cosmetic standard on your drawing and purchase order, so the clearer those are, the faster a defect question is closed.
CNC turning and milling run at our lead CNC plant, T.K Viet Nam Precision Company Limited, in operation since 2017, described on the CNC milling page. Anodizing on aluminum is applied by a partner plant we manage, as listed on the surface finishing page; partner plants are not named, and they are not necessarily in Vietnam, which is why the quote states where each step is done. That split matters for this topic, because it puts the upstream causes in the atlas, such as burrs and sharp edges, deburring method, cleaning and storage before the parts leave for finishing, on the machining side, and the process causes on the finishing side.
In practice, send the anodize callout with type, class and seal, mark significant surfaces and contact-mark zones, and attach any limit samples or written cosmetic standard; our quality page explains that we work to the cosmetic standard you supply rather than asking you to adopt ours. You review the first article and any cosmetic samples before production. If a lot shows a defect, send photos, its location on the part and the lot identity, and we take it up with the plant that did the work, using the checks in this guide to separate a machining cause from a finishing cause. ISO 9001 certification is in progress, not held. Send the drawing and finish notes through the quote form.
Frequently asked
Are rack marks a defect on anodized parts?
Not by themselves. Every anodized part needs an electrical contact, and the specification only asks that contact marks be as small and few as good practice allows. They become a defect when they land on a surface the drawing or purchase order reserved, so mark the permitted zones.
Can a burned or pitted anodized part be fixed?
Sometimes, by stripping and re-anodizing, but only if the metal under the defect is sound and the part can lose a little material. Burns that have eroded the metal and pits that are in the casting or bar stay visible after rework.
Why does anodizing make machining marks and scratches more visible?
The anodic film is grown from the metal and is translucent, so it follows and often emphasizes the surface underneath. Tool marks, scratches and blend lines show through; if a face is cosmetic, specify its surface texture or a blast before anodizing.
Why do high-copper alloys and die castings show more anodizing defects than 6061?
Copper- and silicon-rich phases do not anodize the way the surrounding aluminum does, so high-copper wrought alloys and high-silicon castings give darker, less even films, and castings add porosity that bleeds dye. MIL-PRF-8625 reflects this: it allows dye bleed-out and reduced color uniformity on cast alloys, and unless otherwise specified it does not allow hard coat on alloys above 5 percent copper or 8 percent silicon.
Who decides whether a defective anodized lot can be used as is?
The buyer’s design authority, not the supplier or the anodizer. Record the decision as a deviation or waiver tied to the lot so the next lot is not accepted on the same basis by default.
Sources
- MIL-PRF-8625F w/Amendment 2, Anodic Coatings for Aluminum and Aluminum Alloy (ASSIST QuickSearch document record), US Department of Defense, 23 November 2020 (paragraphs 3.3.1, 3.3.4, 3.4.3, 3.6, 3.13, 4.4, 6.4, 6.14 and 6.16)
- ASTM B136-23, Standard Test Method for Measurement of Stain Resistance of Anodic Coatings on Aluminum, ASTM International, 2023
- ASTM B680-24, Seal Quality of Anodic Coatings on Aluminum by Acid Dissolution, ASTM International, 2024
- ASTM B244-09(2021), Measurement of Thickness of Anodic Coatings on Aluminum with Eddy-Current Instruments, ASTM International, 2021
- Crazing and Cracking in Anodizing (Anne Deacon Juhl, AluConsult), Products Finishing, 28 June 2022
- Anodizing defects catalog: process defects and handling defects, Qualanod, accessed October 2026
- Anodized Aluminum FAQ (sealing smut), Aluminum Anodizers Council, accessed October 2026