Key takeaways
- One gram per square metre of oil is a 1.1 µm film, invisible on an aluminium panel and enough to stop a zirconium conversion coating from forming.
- Keep the alkaline degrease at pH 9.5–10.5 and 50–55 °C for 3–5 minutes; a pH 13 cleaner at 60 °C dissolves 1–2 µm of aluminium per minute and leaves a smut that blocks adhesion.
- Desmut in nitric or a proprietary deoxidiser for 1–2 minutes after any etch, because copper, silicon and magnesium from the alloy stay on the surface as a dark film.
- Use a zirconium conversion at 20–60 mg/m², not iron phosphate: phosphate builds no usable layer on aluminium and the powder keys to a weak film.
- Cap cure metal temperature at 190 °C for 6061-T6 and 6082-T6, because artificial ageing starts near 175 °C and yield strength falls above 190 °C.
One gram per square metre of oil is a 1.1 µm film. On an aluminium panel it is invisible, it survives a wipe with a rag, and it is enough to keep a zirconium conversion coating from forming across the whole face. The powder then bonds to the oil rather than to the metal, and the failure appears at the first scratch or the first monsoon.
Aluminium makes the cleaning harder than steel in one specific way: it is amphoteric. A hot alkaline cleaner that works well on CRCA will dissolve aluminium, and the dissolved alloying elements come back out of solution as a smut that sits exactly where the conversion coating needs to form. Degreasing aluminium parts is therefore a different cycle, not a stronger version of the steel one.
This page sets out that cycle for laser-cut and formed aluminium from 0.8 mm to 10 mm, the chemistry limits that keep the etch under control, and the tests that show the surface is ready for powder. The alloy and temper side of the specification is covered in the powder coating guide for aluminium parts.
What is on an aluminium part before it reaches the line?
Five things, and each needs a different mechanism. Mill oil or protective film sits at 0.5–2 g/m². Cutting fluid or forming lubricant is heavier and often contains additives that resist a mild detergent. The laser-cut edge carries a remelt layer and an oxide a few micrometres thick. Handling leaves salts from skin and gloves. And if the part has been etched, it carries smut.
The oxide is the one that differs from steel. Aluminium forms a native oxide within minutes of cleaning, and the laser edge oxide is thicker and uneven. That is not a problem for adhesion in itself — the conversion coating is designed to grow on oxide — but it is a problem where the oxide is loose, or where dross from the cut has not been removed.
Sort the parts before cleaning, as you would for steel. A laser-cut panel with light oil needs degrease, rinse, conversion, rinse and dry. A part that has been etched needs a desmut stage between the etch and the conversion. A part with heavy dross needs mechanical removal first, because no bath will dissolve a lump of resolidified metal.
Degreasing aluminium parts: the cycle and its limits
| Stage | Typical condition | What goes wrong | Ganesh limit or note |
|---|---|---|---|
| Degrease | pH 9.5–10.5, 50–55 °C, 3–5 min | caustic attack, smut, patchy clean | aluminium-safe chemistry in the seven-tank line |
| Rinse | two stages, DI water final | salts dry on the face and block conversion | overflow rinse, checked each batch |
| Etch, where specified | 3–6 µm removal | over-etch thins the section and loads the bath | controlled by time and temperature, not by eye |
| Desmut | nitric 30–50%, 1–2 min | smut left in place, conversion does not form | required after every etch |
| Conversion | zirconium 20–60 mg/m² | iron phosphate builds no layer on aluminium | coating weight and colour checked |
| DI rinse | conductivity below 10 µS/cm | water spotting and patchy film | final rinse before the oven |
| Dry-off | 80–100 °C | damp parts, water marks under the powder | dried before the booth |
| Powder | 60–80 µm face film | thin edges at 40–60% of face film | DFT gauge every batch |
| Cure | 180–190 °C metal temperature | overbake softens 6061-T6 and 6082-T6 | cap at 190 °C for T6 parts |
| Adhesion | ISO 2409, Gt0 to Gt1 at 2 mm | tested on the flat face only | test at an edge and a hole as well |
| Corrosion | ISO 9227 neutral salt spray | hours quoted without acceptance limits | method and limits agreed at enquiry |
| Material range | 0.8–20 mm cut, 10 mm formed | heavy plate needs different handling | 12 kW laser, 40 ft × 16 ft bed |
Why the alkaline stage has a pH ceiling
The chemistry that removes oil also removes metal
Alkaline cleaning works by saponifying oil and lifting it from the surface. On steel, a pH of 11–13 is an advantage: it handles heavy drawing lubricants and it does not attack the substrate. On aluminium the same bath dissolves the metal, and the reaction is fast enough to measure.
At pH 13 and 60 °C, a caustic cleaner removes 1–2 µm of aluminium per minute. A five-minute cycle takes 5–10 µm, which is 0.4–0.8% of a 1.2 mm panel and enough to change a critical dimension on a thin part. It also produces hydrogen: 5 µm over one square metre is 13.5 g of aluminium, 0.5 mol, which generates about 17 L of hydrogen at standard conditions. That is why an etch tank needs extraction, and why an aluminium etch is a different installation from a steel degrease.
Keep the cleaner at pH 9.5–10.5, 50–55 °C, for 3–5 minutes. Silicate-based cleaners are common on aluminium because the silicate protects the surface from attack, but silicate residues are hard to rinse and can interfere with the conversion coating. A non-silicate, aluminium-safe cleaner with a longer contact time is easier to control on a line that also runs steel.
Etch only when the drawing needs it
An etch is not a cleaning step. It removes a controlled amount of metal to level the surface and remove the deformed layer from forming, and it is specified for cosmetic parts and for parts that need a uniform matte finish. If the drawing does not need it, do not add it: every etch stage adds smut, hydrogen and dimensional change.
Where an etch is specified, control it by removal rate rather than time alone. A 3–6 µm removal is typical for architectural work. Measure it on a coupon from the batch — weigh a 100 × 100 mm coupon before and after, and convert the mass loss to thickness using 2.70 g/cm³. A 100 × 100 mm coupon losing 0.081 g has lost 3 µm.
Desmut is not optional
Etching and alkaline cleaning leave alloying elements on the surface. Copper from 2024, silicon from 6061 and magnesium from 5083 do not dissolve at the same rate as the aluminium matrix, so they concentrate as a dark film. That film is what blocks the conversion coating.
Desmut in nitric acid at 30–50% for 1–2 minutes, or in a proprietary deoxidiser where nitric is restricted. The part comes out bright, and the conversion coating then forms evenly across the face and the cut edge. A part that skips desmut looks acceptable and fails the cross-hatch test at the edges within one batch.
Conversion coating: zirconium, not phosphate
Aluminium takes a zirconium conversion coating at 20–60 mg/m², applied in the same tank sequence as a steel line but with different chemistry. The film is thin, uniform and largely amorphous, and it gives the powder both a chemical bond and a mechanical key.
Iron phosphate is the wrong choice. It is designed to react with iron and it builds no usable layer on aluminium, so the powder sits on a surface that has been cleaned but not converted. The part may pass a flat-face adhesion test and fail at the first stone chip.
Chromate conversion coatings still appear on aerospace and defence drawings, and they work. They are also restricted in many markets, and a job shop that runs chromate needs a separate line and a waste stream. For general industrial aluminium, zirconium is the practical route and the one the paint shop runs.
Water-break, adhesion and salt spray
The water-break test is the cheapest check on the line and it works on aluminium as it does on steel. Flood the part with clean water: a properly cleaned surface holds an unbroken film for at least 30 seconds, and a contaminated one pulls the water into beads within a few seconds. Run it after the final rinse and before the conversion tank, because that is the last point where you can send the part back without losing the conversion chemistry.
Adhesion is checked by cross-hatch to ISO 2409 at 2 mm spacing and classified Gt0 or Gt1. Test a production part at a laser-cut edge and around a hole, not only on the flat face. Edge film is 40–60% of face film, so the edge is where the coating fails first and where a flat-face test tells you nothing.
Salt spray to ISO 9227 is the corrosion check, and aluminium fails differently from steel: there is no red rust, so the failure shows as blistering, filiform corrosion from a scribe, or a white corrosion product at a damaged edge. Write the requirement as a method with limits — the standard, the duration, whether the panel is scribed, how far filiform corrosion may run, and whether blistering is permitted. A 60–80 µm polyester film over a zirconium conversion will typically hold 500 hours of neutral salt spray with scribe creep inside 2 mm, but the number depends on alloy, edge quality and film build, and it should be confirmed on the first article rather than assumed.
Film build, edges and the temper limit
Sixty to eighty micrometres is the working range for polyester on aluminium. Measure it with an eddy-current gauge: a magnetic DFT gauge needs a ferrous substrate to close its circuit and will not read the film on an aluminium part. Below 50 µm the film thins over laser-cut edges and the substrate shows through; above 120 µm it becomes brittle at a fold and chips under impact, and a chip that reaches metal starts a corrosion cell that spreads under the film.
Cure is the constraint that differs from steel. Polyester needs 180–200 °C at the substrate for 10–15 minutes, but 6061-T6 and 6082-T6 start to overage near 175 °C and lose yield strength above 190 °C. Cap the metal temperature at 190 °C for T6 parts and extend the time instead. A rack of mixed parts — a 3 mm bracket and a 1 mm panel — reaches temperature at different rates, so rack similar masses together and record the oven profile.
Edges deserve their own note. A sharp 90° laser-cut edge holds 40–60% of the face film because the powder pulls away as it flows. Radius or chamfer any edge that will be handled, and break the dross line before cleaning so the conversion coating can form on clean metal rather than on oxide.
Aluminium on the Ganesh line
Aluminium parts are cut on the 12 kW fibre laser with the 40 ft × 16 ft bed, formed on the 200 T 5-axis brake, and cleaned on the seven-tank pretreatment line before the 18 ft × 14 ft powder booth and the curing oven. The same plant handles copper, brass and aluminium fabrication, so the cleaning chemistry is not a one-off setup for a single alloy.
The plants hold ISO 9001:2015 and ISO 14001:2015, and coating thickness and cross-hatch adhesion are recorded for every batch. Coating work for enclosure, canopy and lift programmes runs alongside the aluminium parts, and the surface treatment hub collects the substrate-specific guides, including degreasing before powder coating and the mild steel cycle.
What to put on the drawing
- Alloy and temper, by designation: "6061-T6" or "5754-H22", not "aluminium".
- Surface condition on arrival: mill finish, oiled, anodised, or previously coated.
- Whether an etch is required, and the removal in micrometres if it is.
- Conversion coating type and weight, where the customer's specification names one.
- Powder type, colour, gloss and film range, with the faces to be coated.
- Cure limit, expressed as a maximum metal temperature, for temper-controlled parts.
- Masking: threads, earth studs, mating faces and any conductive contact.
- Adhesion and corrosion tests, with the acceptance limits.
- Packing, because a film damaged in transit fails before it reaches assembly.
The powder coating cost estimator shows how masking, a second coat and batch size move the coating price, which is usually the part of the quotation that surprises a buyer.
When this route is wrong
A zirconium conversion plus polyester powder is the right finish for general industrial aluminium: enclosures, panels, brackets and covers indoors or outdoors at C3 to C4. It is not universal.
For a part that will be submerged, buried or washed with strong chemicals, a liquid system with a specified primer and an ISO 12944-2 corrosivity category is the correct specification, and powder alone will not meet it. For a part that needs abrasion resistance above what a 70 µm film gives, anodising or a hard coat is the better route, and powder over anodised aluminium is a poor combination unless the anodised layer is keyed. For castings with porosity, the trapped gas in the casting can blister the powder at cure, and the part may need a sealed primer before coating. And for a part where the aluminium is in contact with a steel fastener, the joint needs isolation, because the coating will not stop galvanic corrosion at a scratch.
Send the alloy, the surface condition and the corrosion requirement with the drawing, and the quotation will state the cleaning, conversion and cure route it assumes.
Frequently asked questions
Can I degrease aluminium with the same cleaner as steel?
No. A steel cleaner at pH 11–13 attacks aluminium, dissolving 1–2 µm per minute at 60 °C and leaving a smut of alloying elements. Use an aluminium-safe alkaline cleaner at pH 9.5–10.5 and 50–55 °C, or a neutral cleaner with a longer contact time.
Why does powder coating blister on aluminium?
Three causes, in order: oil or smut left under the conversion coating, moisture trapped in a lap joint and driven out at cure, and a cure temperature above the temper limit of the alloy. All three show as blisters near edges and holes, where the film is thinnest.
Does aluminium need a conversion coating before powder?
Yes, if the part sees weather or cleaning chemicals. A zirconium conversion at 20–60 mg/m² gives the powder a surface to bond to and slows filiform corrosion at scratches. Iron phosphate is the wrong chemistry for aluminium: it builds no usable layer.
What is desmut and why does it matter?
Desmut is the removal of the dark film left after etching or alkaline cleaning, where copper, silicon and magnesium from the alloy sit on the surface. Nitric acid at 30–50% for 1–2 minutes, or a proprietary deoxidiser, removes it. Left in place, the smut blocks the conversion coating and the powder lifts in patches.
How do I test adhesion on an aluminium part?
Cross-hatch to ISO 2409 at 2 mm cut spacing, classified Gt0 or Gt1, on a production part rather than a flat coupon. Test near a laser-cut edge and around a hole, because that is where the film is thinnest and where failures start. A pass on the flat face alone proves very little.
Sources and further reading
- ISO 2409:2020 — Paints and varnishes, cross-cut test · reviewed 3 Nov 2026
- ISO 9227:2022 — Corrosion tests in artificial atmospheres, salt spray tests · reviewed 3 Nov 2026
- ISO 12944-2:2017 — Corrosion protection of steel structures, classification of environments · reviewed 3 Nov 2026
Sending aluminium parts for coating?
Send the alloy and temper, the surface condition on arrival, the finish and the corrosion requirement. We confirm the cleaning, conversion and cure route before quoting.
Request a coating quote