Scope and Purpose
Salt-spray corrosion testing is the core means of evaluating the resistance of materials and their protective coatings to chloride-containing environments. The test itself only provides an accelerated corrosion “stimulus”; the real technical value lies in the scientific evaluation of the test results. Improper choice of evaluation method or excessive operational error will directly lead to incorrect quality decisions, mis-estimation of service life and even market recalls. Taking current national/international standards as the backbone, this document systematically summarises four categories of result-evaluation methods for salt-spray test chambers and presents quantifiable error-control schemes for testing laboratories, in-house quality departments and third-party inspection bodies.

Terms and Definitions
Substrate metal: the metallic body covered by a protective coating.
Protection rating (Rp): corrosion-resistance grade calculated from the percentage of corroded area; grade 0 is the worst, grade 10 the best.
Appearance rating (Ra): qualitative grade describing visible defects such as discoloration, loss of gloss or blistering.
Corrosion products: visible deposits formed by reaction between the substrate and the corrosive medium, e.g. white rust, red rust.
Systematic error: error component that remains constant or varies in a predictable way under the same conditions.
Result Evaluation Methods
3.1 Rating Method (Protection Rating Rp)
Field of application: quantitative assessment of pitting or general corrosion of the substrate after breakdown of protective coatings such as anodic films, electroplated layers or organic coatings.
Procedure:
a) At the end of the test gently rinse the specimens with de-ionised water to remove residual salt crystals.
b) Dry for 24 h at (23 ± 2) °C and (50 ± 5) % RH.
c) Examine with 7× magnification and 3 000 K white LED; record corroded area.
d) Convert to Rp using GB/T 6461, Table 1:
Rp = 10 – log10(A / 0.1), where A is the percentage of corroded area.
e) Take the arithmetic mean of three parallel specimens; if the range is ≥ 1 level, add two more specimens and re-test.
Acceptance criterion: typical automotive exterior parts require Rp ≥ 7; rail-transport structural parts require Rp ≥ 9. Customer specifications take precedence.
3.2 Gravimetric Method (Mass-Loss Method)
Field of application: metallic sheets or fasteners without protective coatings, or when uniform corrosion rate is required.
Procedure:
a) Before testing, degrease, acid-pickle, passivate and dry to constant mass (difference between consecutive weighings ≤ 0.1 mg).
b) Record initial mass m0 (accurate to 0.1 mg).
c) Test duration in accordance with the material standard, e.g. 48 h, 96 h, 240 h.
d) After testing, ultrasonically clean in 25 % HCl + 3 g/L hexamethylenetetramine for 10 min to remove corrosion products.
e) Dry to constant mass and record residual mass m1.
f) Calculate mass-loss rate Δm = (m0 – m1) / S·t, where S is exposed area (m2) and t is time (h); unit g/(m2·h).
Acceptance criterion: for 304 stainless steel Δm ≤ 0.1 g/(m2·h); for aluminium alloy 2A12 Δm ≤ 0.5 g/(m2·h) after 48 h.
3.3 Corrosion-Appearance Method (Go/No-Go Method)
Field of application: rapid screening, process comparison or on-site customer acceptance.
Procedure:
a) Photograph specimens immediately after test under ≥ 1 000 lx illumination.
b) Visual inspection is primary; use 5× magnification if necessary.
c) Record the first appearance time and morphology of white/red rust.
d) If the specification states “no substrate corrosion within 48 h”, any red rust means “fail”.
Precautions: because of subjectivity, at least two inspectors shall evaluate independently; if they disagree, a third arbitrator is introduced.
3.4 Statistical Analysis of Corrosion Data (Reliability Modelling Method)
Field of application: life prediction, estimation of corrosion-rate distribution and optimisation of sampling plans; not used for direct pass/fail judgement.
Procedure:
a) Collect ≥ 30 data points of corrosion depth or mass loss.
b) Use Anderson-Darling test to check for normal, log-normal or Weibull distribution.
c) Estimate distribution parameters by maximum likelihood and build cumulative distribution function F(t).
d) Calculate characteristic life T0.1 (time at 10 % failure probability) and its confidence interval.
e) Compare T0.1 with the required warranty life and evaluate design margin.
Output: a statement such as “At 95 % confidence the salt-spray resistance life is ≥ 720 h”, providing quantitative support for engineering improvement.
Error Sources and System Control
4.1 Chamber-Related Errors
Salt-fall-out rate deviation: standard requires (1.5 ± 0.5) mL/(80 cm2·h); outside this range corrosion kinetics shift exponentially.
Temperature gradient: a difference > 2 °C between any two points in the cabinet causes uneven condensate distribution.
pH drift: if collected solution pH is outside 6.5–7.2, chloride ion activity changes by > 15 %.
4.2 Correction and Reduction Measures
Metrological calibration: send the chamber every 12 months to a CNAS-accredited laboratory for calibration of temperature, fall-out rate and pH; obtain correction factor C and state expanded uncertainty U (k = 2) in the report.
Reference-sample substitution: insert two CR4 steel panels with known Rp = 9 in every test batch; if measured Rp deviates by > 1 level, multiply the whole batch by correction factor k = 9 / Rp(ref).
Symmetrical placement: if nozzle partial blockage is suspected, place specimens symmetrically about the cabinet centreline and average the left/right groups to cancel systematic offset.
Opposite-error compensation: run test A and test B on the same product; if A shows a positive “over-corrosion” error and B a negative “under-corrosion” error of similar magnitude, average the two Rp or Δm values as the best estimate.
Laboratory Management Requirements
Personnel: operators and evaluators must pass internal training and examination under GB/T 27025 and hold valid certificates; rating evaluations require two independent readings; if the difference is > 0.5 level a third arbitrator is introduced.
Equipment: chambers must have automatic water refill, low-level alarm and continuous fall-out recording; calibration labels must be valid.
Environment: evaluation area illumination 1 000–1 500 lx, colour-rendering index Ra ≥ 90; avoid direct sunlight.
Records: raw data shall be retained for ≥ 6 years and include test curves, photographs, correction factors and uncertainty budgets.
Worked Example
An aluminium alloy bracket (6063-T5) for a new-energy vehicle battery case was required to pass 720 h neutral salt spray. The supplier adopted a dual “rating + gravimetric” approach:
Rating result: Rp = 8.3 (customer limit ≥ 7).
Gravimetric result: Δm = 0.08 g/(m2·h) (customer limit ≤ 0.1).
Uncertainty: U = 0.4 level (k = 2), coverage factor k = 1.02.
Conclusion: pass; predicted warranty life T0.1 = 1 080 h, meeting the 5-year warranty requirement.
Closing Remarks
The evaluation results from a salt-spray test chamber are not only a “medical report” on material corrosion resistance but also the technical passport of a company’s quality reputation. Only by deeply integrating standard methods, statistical tools and error control can data be traceable, conclusions reproducible and risks predictable. Every laboratory should compile its own internal work instructions based on the framework above, adapt them to its equipment status and customer requirements, and periodically conduct proficiency testing to ensure continuous improvement of the technical credibility of salt-spray corrosion testing.