Guide
What causes filiform corrosion under powder coat on aluminum parts, and what pretreatment should you specify?
Filiform corrosion is a thread-like corrosion that tunnels under an intact paint or powder film, starting from a break in the coating and leaving a trail of white corrosion product that lifts the film into fine raised worms. On powder-coated aluminum it is almost always a pretreatment and edge problem made visible by humid, salty air, so the fix lives in the pretreatment callout and the part design rather than in a thicker powder.
By Saola Manufacturing engineering team · Published · Updated
The usual story: an aluminum housing, bracket or enclosure looks perfect at receiving inspection, then comes back from a coastal site, a wash-down area or a humid warehouse with worm tracks radiating from the edges, the screw holes and every scratch. The coating did not fail on its own. Something under it (a deformed or contaminated metal surface, a missing conversion layer, a salt residue) gave the corrosion a path, and a break in the film let it in. This guide covers how to recognize the problem, the causes in rough order of how often they are to blame, how to test for it, and what to write on the drawing and purchase order so the next lot does not repeat it.
On this page
What does filiform corrosion look like, and how do you tell it from other coating failures?
Filiform corrosion shows as narrow, winding filaments under the film, each with an active head and a tail of corrosion product, growing out from an edge, hole or scratch. Blisters are round and scattered, and adhesion loss lifts whole areas, so the worm-like pattern is the giveaway.
The FAA describes it as a special form of oxygen concentration cell on metal surfaces with an organic coating, recognized by a worm-like trace of corrosion products beneath the paint film, and notes that on aluminum the filaments can cross under one another, which makes the damage deeper than on steel (FAA AC 43.13-1B, paragraph 6-16). If it is not removed and the area treated, the same paragraph warns, it can lead to intergranular corrosion around fasteners and seams.
Three checks separate it from look-alikes on a returned part:
- Pattern. Filaments start at a defect and travel across the face with a fairly constant width. Osmotic blisters from trapped soluble salt or water are domes with no tail. Outgassing pinholes over a die casting are craters in the film itself, not tracks under it.
- Origin. Trace the filaments back. They nearly always lead to a sharp edge, a cut or drilled hole made after coating, a fastener seat, a rack mark or a handling scratch.
- Interface. Lift the film with a knife at a filament. White, powdery product on dull metal under a film that peels easily near the track points to an interface problem; if the film splits within itself, look at the powder or the cure instead.
Photograph the worst areas with a scale and a note of where each filament started before anyone strips or touches up the parts. That record is what settles the root cause later.
What causes filiform corrosion under powder coat on aluminum?
It needs a break in the film, humid air with a trace of salt or acid, and a weak metal-coating interface. Of those, the interface is the one a supplier controls, which is why poor pretreatment tops the list of causes.
ASTM’s filiform guide says some organic coatings on metal develop filiform corrosion when there is a break in the coating film and the relative humidity is in the 70 to 95% range (ASTM D2803). The FAA puts the band at 78 to 90 percent on a slightly acidic surface, and for prevention lists storage below 70 percent relative humidity, coating systems with a low rate of diffusion for oxygen and water vapor, and washing off acidic contaminants (FAA AC 43.13-1B). Coastal, wash-down and condensing service sits inside that band for long periods, so the questions are what lets the attack start and what lets it run. Ranked by how often each is the root cause on machined, cast and sheet aluminum parts:
- Etch skipped or too light. Machining and grinding leave a thin, very fine-grained deformed layer at the surface of aluminum. A University of Manchester thesis on A356 cast alloy found that this layer was electrochemically active and dissolved preferentially, that coated ground panels grew both fast superficial filaments and slower blistering ones, and that once the full cleaning stages of a pretreatment had removed the layer, the fast filaments no longer appeared and filiform resistance rose significantly (Goodall, University of Manchester PhD thesis, 2019). A degrease-only line leaves that layer in place.
- Smut left after the etch. An alkaline etch dissolves aluminum faster than the alloying elements, which stay behind as a dark smut unless an acid desmut removes them. Paint over smut sits on a loose, reactive film.
- Missing, contaminated or aged conversion coat. Parts that skip the conversion stage, see poor rinsing, or wait too long between pretreatment and coating lose the chemical bond that slows a filament down. The architectural QUALICOAT scheme limits that wait to 16 hours and asks for clean gloves when handling pretreated parts (QUALICOAT Specifications 2024, section 3.3).
- Thin film at edges and holes. Powder pulls back from a sharp edge as it flows, so a crisp machined edge carries far less film than the face next to it. Holes drilled, threads chased or edges trimmed after coating are bare starting points by design.
- Salt or acid trapped at the interface. Hand sweat, residual cutting fluid, poorly rinsed cleaner and storage near salt air all leave ionic residue that the humidity later activates.
- Alloy and substrate. Chemistry outside the standard range and porous or silicon-rich cast skin can both change how the surface pretreats. After a small number of coated samples failed early in acetic salt spray and filiform testing, a QUALICOAT working group found the alloy itself was a contributing factor and the scheme added an alloy specification for 6060 and 6063 that can be met with primary or recycled aluminum, provided the chemistry meets the standard range (Specification Online, 2023). Recycled content is not the problem; uncontrolled chemistry is.
How do you diagnose a filiform failure and test for it?
Diagnose the returned parts first: map where filaments start, check film build at the edges, and look at the interface. Then prove the fix with a filiform test, not more hours of neutral salt spray.
On the failed parts, measure dry film thickness on the face and as close to the edge as the gauge allows, cross-section one edge if the coater can, and run a cross-hatch on a flat face; bare metal in the lifted squares points at cleaning or conversion coating, as the ASTM D3359 guide explains. Ask the coater for the pretreatment records of the lot: line type, etch stage, rinse checks and the time between pretreatment and coating. Filaments starting only at drilled holes and trimmed edges point at the sequence of operations, not the chemistry.
For prevention, name a test that actually produces filiform. Continuous salt fog keeps the surface wet and does not drive filaments well, which is why the salt spray guide recommends a dedicated test when filiform is the concern. ISO 4623-2 assesses coatings on aluminum against filiform corrosion that grows from a scribe mark, and its scope says the result applies only to the coating and substrate combination tested (ISO 4623-2:2016). So test panels of your alloy and surface condition, or real parts, not a generic panel.
| Method | What it is for | Key settings and limits |
|---|---|---|
| ASTM D2803-09(2020) | Guide with three procedures for whether filiform forms under an organic coating on metal | Procedure B is equivalent to ISO 4623; filiform occurs at 70–95% relative humidity with a break in the film |
| ISO 4623-2:2016 | Paint or powder on aluminum, attack growing from a scribe | Result valid only for the coating and substrate combination tested |
| EN 3665:1997 | Aerospace filiform test for paints on aluminum alloys | Use when the customer works to aerospace paint specifications |
| QUALICOAT 2024, section 2.18 (ISO 4623-2, modified) | Architectural powder on aluminum, approval of pretreatment systems | Two scribes; 37% hydrochloric acid for 1 minute; 40 °C and 82% relative humidity for 1,000 hours; longest filament 4 mm or less, average 2 mm or less |
| ASTM B117 neutral salt spray | Process control of barrier and scribe creep | Not a filiform test; add it only alongside one of the above |
Which fixes stop filiform corrosion, and what pretreatment should you specify?
Specify an etch that removes the deformed surface layer, a desmut, a conversion coating or thin pre-anodize, and coating soon after pretreatment. Then round the edges and finish all cutting before the powder goes on.
The QUALICOAT architectural specification is the clearest public benchmark for the pretreatment itself. Every chemical pretreatment for powder must include an aluminum etching stage, either acid or dual alkaline-and-acid; the standard pretreatment calls for a total etch of at least 1.0 g/m², and the enhanced pretreatment for its seaside endorsement at least 2.0 g/m², with pre-anodizing qualifying for that endorsement automatically (QUALICOAT Specifications 2024, sections 3.2 to 3.4). The same document sets no etch degree for cast accessories, which is a reminder that castings need their own agreement. You do not need a QUALICOAT coater for an industrial part, but asking for “enhanced pretreatment or equivalent” gives the coater a defined target instead of “pretreat per best practice.”
For the conversion coat, chrome-free systems are the normal choice where RoHS applies, because traditional yellow chromate conversion contains hexavalent chromium. Name the requirement (“chrome-free conversion coating, RoHS compliant”) rather than a brand, and let the coater state the chemistry in the quote.
| What you see or know | Likely cause | Fix to specify |
|---|---|---|
| Filaments on flat faces from small scratches | Degrease-only line; deformed layer or smut left on the surface | Etch plus desmut, total etch at least 1.0 g/m²; at least 2.0 g/m² for coastal service |
| Filaments from every sharp edge | Thin film where powder pulled back | Radius the edges before coating; check film build at the edge on the first article |
| Filaments only from holes and trimmed edges | Machining or trimming after coating | Finish all machining before pretreatment; mask features instead of cutting through the film |
| Film peels easily near the tracks, bare metal under it | Missing or aged conversion coat, poor rinse | Conversion coat required; coat no later than 16 hours after pretreatment |
| Only cast parts fail, machined faces pass (or the reverse) | Pretreatment tuned for one surface | Agree a pretreatment and a test for each surface type |
| Service at 70–95% humidity with salt | Environment drives the attack | Seaside-grade pretreatment or pre-anodize, plus a filiform test on your alloy |
Design changes do as much as chemistry. A rounded edge holds film where a sharp break does not, so replace “break sharp edges” with a defined radius on coated aluminum and size it with the coater; the edge break guide covers how to write it. Put drain paths where water would sit against a coated seam, and keep fasteners from cutting through the film where you can use a masked bare seat or a washer instead.
Then close the gaps after the coating line, because the FAA prevention list (dry storage, washing off acidic contaminants) applies to parts in a warehouse as much as to aircraft:
- Packing and storage. Ship and store coated parts dry and sealed from salt air, with desiccant in the pack if they cross the sea or sit in an unconditioned warehouse; the FAA threshold for storage is below 70 percent relative humidity.
- Installation. Any hole drilled or edge cut in the field is a new bare start point. Tell the installer to seal it with a touch-up primer or paint compatible with the powder before the part goes into service.
- Maintenance. In coastal or wash-down service, rinse salt deposits off coated housings on a schedule and repair chips while they are small, before a filament has started.
What should the drawing, RFQ and PO say so filiform does not come back?
Write the pretreatment, the edge condition, the order of operations and the filiform test into the finish note, with acceptance limits and who tests what. “Powder coat, black” leaves every one of those decisions to the coater.
- Finish note. For example: “PRETREATMENT: ETCH AND DESMUT, CHROME-FREE CONVERSION COATING (RoHS). ENHANCED (SEASIDE) ETCH DEGREE PER QUALICOAT SPECIFICATIONS 2024 SECTION 3.2.1 OR EQUIVALENT. POWDER COAT, COLOR AND GLOSS PER LIMIT SAMPLE.”
- Sequence. “ALL MACHINING, DRILLING AND TAPPING COMPLETE BEFORE PRETREATMENT. MASK THREADS AND BARE CONTACT FACES; NO CUTTING THROUGH CURED FILM.”
- Edges. A defined edge radius on coated edges instead of a generic edge break.
- Environment. One line, such as “OUTDOOR, COASTAL, CONDENSING,” so the supplier knows the part is in the filiform humidity band.
- Verification. “FILIFORM RESISTANCE PER ISO 4623-2 ON PANELS OF PART ALLOY AND SURFACE CONDITION, OR ON PRODUCTION PARTS, AT FIRST ARTICLE; ACCEPTANCE AS AGREED IN THE QUOTE.” Add film thickness at an edge location to the first article record.
For a nonconforming lot already in hand, quarantine it, keep the photographed parts unaltered, and ask the supplier for the pretreatment records before agreeing on strip-and-recoat. Recoating over the same interface only hides the cause; stripping, a full etch and conversion, and recoating is the repair that addresses it.
Who does what on a Saola order for powder-coated aluminum parts?
Machining runs at our lead CNC plant, and pretreatment and powder coating run at a partner finishing plant we manage. The quote names the production site and the country of origin for each step.
Aluminum parts are machined by CNC milling and turning at our lead CNC plant, T.K Viet Nam Precision Company Limited, and when the drawing calls for it, the edge radii, holes and threads are machined there before the parts go to coating. Cast housings come from die casting partners we manage, as described on the die casting page. Pretreatment and powder coating are applied by partner plants under our surface finishing menu, and those partners are not necessarily in Vietnam; the quote states where each step is done.
Salt-spray results are supplied on request, with the test standard and duration confirmed in the quote. A filiform test such as ISO 4623-2 or ASTM D2803 is not on the standard document list, and specialty conversion coatings are not offered as a separate finish, so if your drawing calls for a named pretreatment, an etch degree or a filiform test, put it in the RFQ and get the answer in writing in the quote before tooling. To start, send the drawing with the finish note and the service environment through the quote form.
Frequently asked
Does a thicker powder coat stop filiform corrosion?
Not on its own. Filaments start at breaks in the film and run along the metal interface, so a thicker face does nothing for a thin edge, a drilled hole or a weak pretreatment. Fix the etch, the conversion coat and the edges first.
Can filiform corrosion happen indoors?
Yes, wherever humidity stays high and a film defect exists: unconditioned warehouses, enclosures that condense, and equipment near wash-down areas. Dry, conditioned indoor service rarely sees it, which is why it often appears only after parts reach coastal customers.
Is pre-anodizing better than a chemical conversion coat under powder?
For severe coastal service it is the stronger choice, and QUALICOAT grants its seaside endorsement automatically to the anodic pretreatment route. It adds a process step and cost, so reserve it for parts that really live in salt air and use an enhanced etch with a conversion coat elsewhere.
Why did only the die-cast parts fail when the machined parts passed?
Cast skin and machined faces pretreat differently, and an architectural etch specification does not even set an etch degree for castings. Agree the pretreatment for each surface, and test panels or parts that carry the cast skin.
Can filiform-damaged parts be repaired?
The FAA guidance for aircraft removes the corrosion by bead blasting or sanding, then treats the area and refinishes it. For production parts, stripping, a full pretreatment and recoating is the reliable repair; spot touch-up over live filaments lets them keep growing.
Sources
- ASTM D2803-09(2020), Standard Guide for Testing Filiform Corrosion Resistance of Organic Coatings on Metal (scope, significance and use), ASTM International, 2020
- ISO 4623-2:2016, Paints and varnishes: Determination of resistance to filiform corrosion, Part 2: Aluminum substrates (abstract), International Organization for Standardization, 2016, confirmed 2021
- AC 43.13-1B, Acceptable Methods, Techniques, and Practices: Aircraft Inspection and Repair, paragraph 6-16 Filiform Corrosion, Federal Aviation Administration, 1998, change 1 2001
- QUALICOAT Specifications 2024, master version V01b (sections 2.18, 3.2.1, 3.3, 3.4), QUALICOAT, valid from January 2024
- Control of Cosmetic Corrosion of an A356 Aluminum Alloy, PhD thesis abstract (University of Manchester Research Explorer), M. Goodall, The University of Manchester, 2019
- QUALICOAT exhibits at Zak London (alloy specification, Appendix A13), Specification Online, September 2023