Salt Spray Hours Do Not Equal Years in Service
Share
A salt-spray report can be useful evidence, but its hour total is not a clock that runs faster than outdoor service. “No red rust at 720 hours,” for example, describes an observation under a defined laboratory exposure. It does not by itself predict how many months or years a coated part will last in coastal mud, road salt, standing water, or an abrasive undercarriage environment.
The practical comparison therefore starts with the reports, not the largest number on their cover pages. A buyer or quality team must identify the test method, the specimens, the exposure history, and the rule used to evaluate change. Only then can it decide whether two results are comparable, conditionally comparable, or not comparable. This is also why a discussion of corrosion evidence should be kept separate from the existing comparison of zinc-flake and phosphate track-bolt coatings: coating selection and report comparability answer different questions.
Identify the method and the product requirement
Begin by copying the complete method designation from each report, including its edition and any cited amendment. ISO 9227:2022 covers neutral salt spray (NSS), acetic acid salt spray (AASS), and copper-accelerated acetic acid salt spray (CASS), while Amendment 1:2024 is separately listed by ISO. Those atmospheres are not interchangeable labels. ASTM B117-26 addresses operation of salt-spray apparatus, but its public scope likewise does not supply the product specimen, exposure period, or interpretation for a particular purchase.
The chamber-practice reference must therefore be paired with the product or coating specification, drawing, purchase order, and any customer requirement that states what was requested. Record the required atmosphere, exposure duration, inspection intervals, failure criterion, and acceptance authority. Capture the report issue and revision, laboratory and site, plus every recorded deviation. A method can describe how an exposure is produced while a product requirement defines what result matters. Treating the method as the product acceptance specification collapses those two roles.
A statement such as “salt spray, 1,000 h” is incomplete. It does not reveal whether the atmosphere was NSS, AASS, CASS, or a different continuous or cyclic regime. It does not say which edition governed, what surfaces counted, whether intermediate observations were scheduled, or who had authority to accept the result. Until those fields are recovered, the duration is only an isolated number.
Match the specimens and coating condition
Next establish what physically entered the chamber. A flat panel and a finished threaded part can carry the same coating name yet present different edges, recesses, contact points, and drainage behavior. Identify whether each specimen was a production part or a panel, its part number and lot, substrate or material, coating chemistry and supplier, pretreatment, and coating thickness together with the thickness-measurement method. A shared color, trade name, or nominal thickness does not prove equivalence.
Preparation history matters as much as the label. Record cure, bake, and aging before exposure; cleaning and handling; specimen quantity and replicates; geometry and orientation; and the pretest condition supported by identified photographs where available. Note whether test racks or suspension created contact points. For threaded parts, state whether threads, recesses, bearing faces, or other geometrically distinct areas were exposed and evaluated.
Edges, masking, and intentional damage often decide whether two reports answer the same question. One laboratory may mask cut edges while another includes them. A scribe may be present on one specimen but absent on another, or made with a different tool or at a different location. The report should identify the important surfaces, excluded zones, cut-edge treatment, scribe rule, and any other intentional damage. If one result counts edge corrosion and the other excludes it, equal hours and similar coating descriptions do not make the outcomes directly comparable.
Compare the exposure conditions and laboratory record
After matching the specimens, compare how the exposure was produced over time. The method and atmosphere are the first fields, followed by the solution and reagents, temperature, pH, and any required collection, fallout, or corrosivity checks. Then compare specimen angle and position, chamber loading, shielding between specimens, and the schedule for inspections and removals. Calibration or verification references, anomalies, deviations, interruptions, and door openings belong in the evidence package because they help explain what actually occurred.
Continuous salt spray and a cyclic corrosion test with wet, dry, or humidity stages are different histories even when both reports show the same elapsed hours. For a cyclic method, record the sequence, stage durations, humidity conditions, and transition or ramp behavior. A technical case published by Q-Lab shows how incompletely specified wet/dry transitions can contribute to interlaboratory differences in cyclic testing. That case supports checking the transition record; it does not prove that transitions explain every disagreement or that one equipment brand is required.
Also determine whether actual exposure time equals nominal test time and how time outside the chamber was handled. Washing or drying before evaluation can affect what the observer sees, so the applicable instruction and actual practice should be recorded. ASTM B117-26 notes that results depend on operating variables and on specimen and evaluation criteria. It follows that a standards citation alone cannot reconstruct an undocumented log or establish that two laboratories delivered an identical corrosive dose.
Read the failure criterion and result evidence
The result needs a named phenomenon, surface, time, and evaluation method. “Corrosion” may refer to white corrosion products, red corrosion or other base-metal attack, blistering, scribe creepage, loss of adhesion, or another defined change. Keep base-metal corrosion separate from coating corrosion and record the rating unit and method edition. ASTM D1654-24e1 covers post-exposure treatment and evaluation of painted or coated specimens for several types of deterioration, but it is not automatically applicable to every report and does not itself define the exposure or the product pass criterion.
Observation wording must be read literally. “No red rust at 720 h” is a censored result: the stated event was not seen by the end of the test under the stated observation rules. “First red rust observed at 720 h” says that the event was seen at that scheduled observation. Neither statement establishes the exact onset time between inspection intervals. Reports should preserve those intervals, results for every specimen, replicate spread, raw observations, deviations, posttest cleaning, observer identity, and report approval.
Photographs help only when they are traceable. Look for specimen identity, exposure time, scale, lighting, surface or zone, and whether the image was taken before or after cleaning. A photograph cannot silently replace a quantitative rating when the requirement names a rating method. Likewise, a close view of a favorable area cannot represent all important surfaces unless the sampling and evaluation rule allows it.
The 2026 ISO technical report ISO/TR 19852 adds a useful caution. Its interlaboratory NSS study concerned two zinc-based coatings on M6x50 bolts and involved 39 participating laboratories testing 75 boxes of bolts. It documented variation among laboratories and cabinets and reported low reproducibility despite nominal ISO 9227 conditions. That evidence strengthens the case for retaining laboratory and replicate context. Its numerical findings must remain limited to that study population; they are not universal scatter values for every coating or component.
Classify whether the reports are comparable
Use a structured record to prevent the hour totals from dominating the decision. Each row below represents a comparison field, not an automatic acceptance rule. “Normal” means the evidence aligns for the intended comparison. “Missing” means a decision-critical field has not been documented. “Conflict” means the reports describe materially different conditions or rules. The final status and any required clarification or retest belong to the named decision owner under the governing product requirement.
| Comparison field | Normal | Missing | Conflict |
|---|---|---|---|
| Purpose and governing requirement | Same stated purpose, product criterion, and acceptance authority | Product or customer requirement is absent | Reports answer different qualification, control, or investigation questions |
| Method and atmosphere | Full designation, edition, amendment, and atmosphere align | Edition or NSS/AASS/CASS/cycle is unstated | Different atmospheres or continuous and cyclic methods are compared by hours |
| Specimen and coating condition | Part or panel, substrate, coating, preparation, geometry, and replicates are defensibly matched | Lot, preparation, thickness method, or pretest condition is absent | Unlike specimens or coating systems are treated as equivalent |
| Edges, scribes, and important surfaces | Masking, damage, included zones, and exclusions use the same rules | Edge, scribe, thread, or important-surface rule is undocumented | One report counts a surface or damage mode that the other excludes |
| Exposure history | Conditions, actual time, interruptions, inspections, and preparation for evaluation align | Chamber log, transition history, or actual time is incomplete | Materially different cycles, loading, or handling are treated as equal dose |
| Endpoint and evidence | Same phenomenon, rating method, observation convention, and traceable records | Rating edition, observation interval, replicate result, or photo identity is absent | Different endpoints or first-observation rules are ranked as one measure |
| Supported disposition | Comparable for the narrow stated purpose, subject to the product authority | Conditionally comparable pending named clarification or evidence | Not comparable; align the requirement and retest if the decision requires it |
Two aligned reports may support a narrow conclusion, such as whether results obtained under the same documented control plan changed between lots. A missing edge rule may leave the reports conditionally comparable for unaffected surfaces but unable to answer the product decision. Conflicting atmospheres, specimens, or endpoints make an hour-based ranking unsound. Do not average unlike results or erase the original records to manufacture agreement.
ISO 9227 states that its methods are not intended for ranking different materials or predicting long-term corrosion resistance. ASTM B117-26 similarly warns about correlation with natural performance. These boundaries do not make controlled testing useless: a fixed laboratory method can reveal changes and support a properly defined product requirement. They do mean that the legitimate conclusion must remain within the documented specimens, conditions, endpoints, and decision authority. The test report alone does not release a lot.
Connect laboratory evidence to service conditions cautiously
A field-performance claim requires a separate evidence chain. Define the geography and microenvironment, salt deposition and time of wetness, temperature and humidity cycling, UV exposure, abrasion and impact, soil, mud, chemicals, crevices, and galvanic contacts. Add maintenance and washing practices, installation damage, and the actual geometry of water and debris retention. These factors can interact in ways a stand-alone salt-spray exposure does not reproduce.
Then look for relevant field exposure or service history, corroborating long-term data, a defined population, and appropriate confidence limits. The scope of inference should match the parts, coating process, environment, and maintenance represented by that evidence. A qualified corrosion engineering assessment may connect several evidence types, but it still needs to state assumptions and uncertainty.
A production quality-control test can be valid for detecting change under fixed conditions while remaining unable to predict years on an excavator exposed to coastal salt, abrasive soil, impact, and intermittent washing. Preserve that distinction in purchasing and quality records. Compare reports only after their decision-critical fields align, and treat service life as a separate question supported by service-relevant evidence rather than by an accelerated-test multiplier.