Salt Spray Test Hours: How to Read ASTM B117 Results on Coated Aluminum
“1000 hours salt spray” is the most quoted and least useful number in metal finishing. It appears on datasheets without a substrate, without a scribe, without a rating method and without a failure criterion — which makes it unfalsifiable. Two suppliers can both claim 1000 hours on parts that behave completely differently in a coastal enclosure, and neither is lying. This is how to read the number properly: what ASTM B117 actually controls, what the hours mean for each finish family, and what to write on the drawing so the result means something.
What ASTM B117 Actually Controls
B117 is a practice for operating the apparatus, not a pass/fail test. It fixes the fog: 5% sodium chloride solution, pH held between 6.5 and 7.2, cabinet at 35°C, continuous exposure, with a collection rate of 1 to 2 mL per hour over an 80 cm² collecting area. That is the whole scope. Duration, specimen preparation and acceptance criteria come from the specification you are testing against, which is why the standard itself never says “1000 hours” anywhere.
Continuous fog is also an unrealistically aggressive and unrealistically constant environment. Real exposure alternates wet and dry, and dry periods are when corrosion products set up and the next wet cycle does the damage. A coating that survives 500 hours of continuous fog can still fail early in a wet-dry cycle, because the mechanism is different. This is the main reason field correlation is poor — the test is best used to compare two candidate finishes processed the same way, not to predict a service life in years.
Why "1000 Hours" Means Nothing on Its Own
Four variables decide whether a claim is worth anything, and they are almost never quoted alongside the number:
- Substrate and temper — 6061-T6 and 7075-T6 behave differently under the same film; cast alloys with silicon content behave differently again.
- Surface preparation — a chromated or zirconium pre-treatment under a powder coat does more for corrosion performance than another 20 µm of topcoat.
- Scribe or no scribe — an unscribed panel tests barrier properties; a scribed panel tests how far corrosion creeps from damage, which is usually what fails in service.
- Failure criterion — “no red rust” on steel is measurable; “no corrosion” on aluminium is not, because aluminium corrodes by pitting and somebody has to say how many pits and how big.
Typical Hours by Finish Family
The figures below are ranges commonly written into customer specifications for aluminium, not universal constants. Treat them as a starting point for a conversation, and always pair them with the rating method.
| Finish | Commonly specified range | Typical failure mode | Notes |
|---|---|---|---|
| Bare 6061-T6, no finish | <24–96 h to first pitting | Pitting, general etching | Baseline; anything above this is the coating working |
| Chem film, MIL-DTL-5541 Class 1A | 168–336 h | Isolated pits | Under 1 µm, no dimensional change; Class 3 adds conductivity |
| Type II anodize, sealed | 336–1000 h | Pitting through thin spots | Sealing quality drives the result more than thickness |
| Type III hardcoat | 336 h and up | Pitting at edges and sharp radii | 25–63 µm film; edge coverage is the weak point |
| Powder coat over pre-treatment | 500–1000 h+ | Creepage from the scribe | Judged on creepage, not on first rust |
| Electroless nickel | Varies with thickness and heat treatment | Pitting at porosity | Phosphorus content and post-plate bake change the answer |
| PVD / DLC thin films | Substrate-dependent | Defect-driven pitting | Pinholes dominate; not a barrier coating in the usual sense |
Two rows are worth dwelling on. Chem film adds essentially no thickness, so it is the finish of choice when the part has tight fits and a grounding requirement — see MIL-DTL-5541 Class 3 conductive chem film for where that trade-off lands. And powder coating is normally judged on how far corrosion travels from a deliberate scratch, which is a completely different measurement from “hours to first spot”. Background on both sits in the anodize versus chromate conversion coating comparison and the aluminium chem film service notes.
Reading the Result: Creepage, Pitting and Rating
For coated panels, ASTM D1654 is the method that turns a corroded panel into a number. You scribe the panel, expose it, then measure how far undercutting has travelled from the scribe line and rate it. A coating that looks terrible can pass if the creepage is within the specified limit, and a panel with one spectacular pit can fail even though the rest of the surface is clean.
For aluminium without an organic topcoat, rating is usually pitting-based: count, size and location. Pits at an edge, at a machined corner or around a fastener hole matter more than the same pit in the middle of a flat face, because that is where the part is thinnest and most stressed. Always specify which areas are evaluated and which are excluded — cut edges, for instance, are commonly excluded and commonly where failure actually starts.
Where B117 Misleads
- It rewards thick, impermeable films and punishes thin ones, which is not the same as rewarding corrosion performance in a real assembly.
- Galvanic couples behave differently. A coated aluminium panel tested alone says nothing about that panel bolted to stainless steel.
- Geometry is flattened out. Recesses that trap salt solution and flanges that stay wet fail in service regardless of what a flat coupon did.
- Continuous fog suppresses the wet-dry cycle that drives most atmospheric corrosion, so ranking between dissimilar coating types is unreliable.
Where you need a harsher, faster screen for decorative and plated finishes, the copper-accelerated acetic acid variant in ASTM B368 is the usual step up, and cyclic tests that alternate fog, dry and humidity generally track field performance better than either. The honest answer is to run candidates through the same cycle and compare them to each other rather than to a number on a datasheet.
How to Specify It on the Drawing
| Callout | What it prevents |
|---|---|
| Test standard and year, e.g. ASTM B117 | Removes ambiguity about fog conditions |
| Duration in hours | Sets the exposure; meaningless without the items below |
| Substrate alloy and temper | Stops a favourable alloy being substituted |
| Pre-treatment and finish specification | Locks the layer that actually does the work |
| Scribe: yes or no, and method | Decides barrier test versus creepage test |
| Acceptance criterion and rating method | Makes pass/fail measurable |
| Areas excluded from evaluation | Prevents arguments over cut edges and contact marks |
Frequently Asked Questions
Does more hours always mean better corrosion resistance?
No. Hours without a rating method measure nothing, and a finish that reaches 1000 hours on a flat coupon can still fail at a sharp radius, a threaded hole or a weld where coverage is poor. Geometry and pre-treatment usually decide the outcome.
Can salt spray predict years of service life?
Not reliably. The correlation between continuous fog hours and real outdoor or marine exposure is weak and coating-specific. Use it to rank candidates and to catch process drift, not to calculate a warranty period.
Why did a chromated part pass while an anodized one failed?
Usually sealing, or a thin spot at an edge. Chromate films are thin but uniform and self-healing to a degree; an anodic film with poor sealing has open pores that carry chloride straight to the base metal. Compare the hardcoat route when the part sees abrasion as well as salt.
What about steel parts in the same assembly?
Different failure mode — red rust rather than pitting — and different expectations. Zinc plating and electroless nickel are judged on time to white and red rust, and mixing metals in one assembly adds a galvanic variable the coupon test does not capture.
Sending One In
Send the alloy, the finish specification, the environment and the acceptance criterion — not just a number of hours. For outdoor and marine work, coatings for harsh environments are selected around exposure class, and the pre-treatment under them matters more than the topcoat thickness. Where conductivity is required alongside corrosion performance, the trivalent clear chem film and Iridite routes are the ones to quote against MIL-DTL-5541. Powder-coated assemblies are quoted on industrial powder coating with creepage criteria, and hardcoat work at Type III hardcoat anodizing. For thin-film options, the DLC tooling case study shows how a wear-driven coating is evaluated differently from a corrosion one, and the oil and gas corrosion notes cover what happens when chloride and temperature arrive together.
