Industries / Mining and materials / Chemicals and petrochemicals / Specialty chemicals / Paint and coatings
Durability testing
Durability testing is the work of finding out how long a paint will keep its color, its gloss and its hold on the surface, and how long it will keep rust off steel, by putting panels outdoors and in machines that speed up sun, rain and salt.
6
Milestones achieved, from 1931 to 2018.
1
To watch for: announced, not yet confirmed outside.
2
Open deliverables. Nobody has shipped them.
The road
AchievedWatch forOpen
Achieved
1931
Florida sun as a yardstick
Atlas sold a lamp cabinet for fading tests from 1915, renamed the Color Fade-Ometer in 1919, and introduced its Weather-Ometer in 1927. Real weather came next: the South Florida Test Service opened near Miami in 1931 to expose painted panels to sun, heat and damp, and Atlas bought it in 1934. Atlas gives the site's mean relative humidity as 78% and its yearly ultraviolet dose as 280 MJ/m², and paint makers still send panels there. Q-Lab, a rival test company, opened its own Miami site in 1995.
Credit
Atlas (now Atlas Material Testing Technology); the South Florida Test Service
1970
Fluorescent lamps speed up weathering
Q-Panel, a Cleveland company founded in 1956 to make standard test panels for paint research, introduced the QUV tester in 1970. It combines fluorescent ultraviolet lamps with periods of condensation, much as sunlight alternates with dew outdoors. By the company's own count it was the most widely used weathering tester in the world by 1983. The University of Pennsylvania's Architectural Conservation Laboratory posts the operating manual of a 1992 QUV model on its website.
Credit
Q-Panel Company (now Q-Lab Corporation)
1973
Beach racks rank zinc primers
At NASA's Kennedy Space Center, salt air from the ocean attacks the steel of launch structures. Zinc-rich primers are paints packed with zinc dust, which corrodes in place of the steel. A NASA engineer exposed 59 commercial zinc-rich primers and 47 topcoated systems on racks about 30 m from the high-tide line for 18 months. The inorganic types, bound with silicates and similar minerals, far outlasted the organic ones bound with resins such as epoxies, and topcoats proved unnecessary or even undesirable over zinc-rich coatings there. In 1994 NASA still qualified coatings at this beach site, with 18 months of exposure for preliminary approval and five years of good performance for final approval.
LessonA test in the real setting can overturn a habit; an extra topcoat over a zinc primer added nothing on this beach.
Credit
William J. Paton, Kennedy Space Center
1998
Corrosion test matched to cars
In 1991 H. E. Townsend and colleagues, working for the North American car and steel industries, compared four cyclic lab tests, which alternate salt, wet and dry periods, with two years of exposure on vehicles in Newfoundland and Quebec. The cyclic tests matched the vehicles well for coated steel sheet, and the authors called the steady salt spray test "totally unsuitable" for ranking these steels. The work led to SAE J2334, a cyclic laboratory test issued in June 1998 that SAE describes as field-correlated, so it can be used to validate products as well as develop them.
Credit
H. E. Townsend and colleagues, with the corrosion task force of the American Iron and Steel Institute; SAE International
2008
Greener paints wear differently
Consumer Reports tests exterior paints on untreated pine boards set facing south at 45 degrees on the roof of its offices in Yonkers, New York, and counts each year there as about three years on a house. In May 2008, after testing more than 40 exterior paints and stains, it reported that finishes newly reformulated, often to meet government limits on volatile organic compounds (VOCs), "might not wear the same as earlier versions": its latest paints did not fend off dirt and weathering as long as some past products. It removed every earlier result for a reformulated product from its ratings, including some of the previous year's top scorers.
LessonA test result belongs to a recipe, not to a brand name, so a product changed to meet a new rule has to be tested again.
Credit
Consumer Reports
2018
Offshore test made harsher
ISO 12944, first published in 1998, is the main international standard for protecting steel structures with paint: a buyer picks a paint system from how corrosive the site is and how many years it must last. Its 2017 and 2018 revision added a "very high" durability class, a category for extreme sites called CX, and a Part 9 for offshore structures that replaced the separate standard ISO 20340. The UK's Institute of Corrosion says the old test for offshore-grade systems, 1,440 hours of hot salt spray and condensation, was "relatively benign" and that product failures followed. Part 9 instead runs 25 weekly cycles of three days in a fluorescent ultraviolet cabinet, three days of hot salt spray and one day at −20 °C.
LessonA test that is gentler than the site lets coatings pass that are not fit for it.
Credit
ISO technical committee TC 35/SC 14
Watch for
Infrared spots hidden coating decay
Researchers at Kyoto University and other Japanese institutions weathered six commercial exterior wood coatings, waterborne acrylic, polyurethane and alkyd types, under xenon-arc lamps. They report that mid-infrared spectra, read by a machine-learning model, flag chemical breakdown before cracking, flaking or fading can be seen, and that they checked the method outdoors in contrasting climates. The trade report gives no accuracy figures or lead time.
Achieved whenanother lab or a coating maker applies the method to its own coatings on outdoor racks and publishes how far ahead, and how accurately, it predicted failure.
Claimed by
Kaho Nishimura, Yoshikuni Teramoto and colleagues, in Resources, Conservation and Recycling, February 2026
Open
01
Paint panels in many climates
A public, multi-site record of how store-bought exterior paints fade and lose gloss outdoors. Paint identical aluminium panels with ten exterior paints, send sets to 25 volunteer hosts such as schools and makerspaces in at least five climate zones, mount them on fixed racks facing the equator, and measure color and gloss once a month. Join each site's readings to its sunlight, temperature and rain from open weather records, and fit a model that predicts color change from the weather each panel received.
The briefCredit · unclaimed02
Open meter for paint films
A low-cost, open-source instrument that measures the impedance of a painted steel panel, meaning how strongly the paint film resists a small alternating current, while it soaks in salt water. Impedance falls as water and salt get through the film, an early sign that the coating will fail. Build it from parts costing under US$300, measure without drilling through the paint by placing two cells on the painted surface, and check it against a commercial instrument.
The briefCredit · unclaimed
Open 01 · the brief
Paint panels in many climates
The first to publish is credited on this line, and the deliverable moves from Open to Achieved.
What to build
A public, multi-site record of how store-bought exterior paints fade and lose gloss outdoors. Paint identical aluminium panels with ten exterior paints, send sets to 25 volunteer hosts such as schools and makerspaces in at least five climate zones, mount them on fixed racks facing the equator, and measure color and gloss once a month. Join each site's readings to its sunlight, temperature and rain from open weather records, and fit a model that predicts color change from the weather each panel received.
Why it is still open
Long outdoor records sit with companies and test firms. Dow's exposure station in Spring House, Pennsylvania, holds about 40,000 panels, and its readings go into an electronic database for Dow's scientists. Consumer Reports exposes exterior paints at a single site, its roof in Yonkers, New York, and publishes ratings rather than measurements. The closest open dataset found, published on Zenodo in October 2026 by Jakub Sandak and Richard Acquah, tracks the color of Norway spruce wood at several European sites, not paint. Searches of Zenodo, Hugging Face and the web found no open multi-site record for paint.
Built with what exists
Aluminium test panels; a drawdown bar to spread even films; a handheld spectrophotometer such as the Nix Spectro 2; a 60° gloss meter; colour-science, an open-source Python library that computes the CIEDE2000 (ΔE00) color difference, a standard number for how different two colors look; the free Open-Meteo historical weather API; the Sandak and Acquah dataset as a template for weather doses.
First step
Paint 30 panels with three exterior paints, measure them, and place sets at three sites in different climates for a two-month pilot.
Done looks like
Ten paints at 25 or more sites in at least five climate zones, measured monthly for 12 months, with the meter checked against a reference tile at every visit. The data are published under an open license, with a model that predicts each panel's 12-month ΔE00 from weather data, scored on sites left out of its training.
Open 02 · the brief
Open meter for paint films
The first to publish is credited on this line, and the deliverable moves from Open to Achieved.
What to build
A low-cost, open-source instrument that measures the impedance of a painted steel panel, meaning how strongly the paint film resists a small alternating current, while it soaks in salt water. Impedance falls as water and salt get through the film, an early sign that the coating will fail. Build it from parts costing under US$300, measure without drilling through the paint by placing two cells on the painted surface, and check it against a commercial instrument.
Why it is still open
Electrochemical impedance spectroscopy (EIS) is the standard lab method; NASA screened zinc-rich primers with commercial EIS equipment in 1994 and recorded values up to about 5.6 million ohms. The open-source ABE-Stat potentiostat (2019) does EIS with about US$105 of parts, but its paper documents it only from under 100 ohms to above 100 kilohms and tests it on solutions and resistor networks, not paint. The two-cell layout that spares the coating was shown in 2009 with laboratory equipment. Y. Kuo, Chung-Ying Lee and Yueh-Lien Lee built a compact coating impedance detector in 2018 and found that it agreed well with a laboratory potentiostat on five coated samples between 10 million and 1 billion ohm-square-centimeters; the paper is not open access and no design files were found. GitHub and Hugging Face searches found no open instrument or dataset for paint films.
ntrs.nasa.goviopscience.iop.orgapi.crossref.orgapi.semanticscholar.org
Built with what exists
The ABE-Stat design files as a starting point; an electrometer-grade amplifier, a type whose input draws almost no current, for the high-impedance range; a Raspberry Pi Pico; impedance.py, an open-source Python package for fitting impedance data; retail primers and paints on standard steel test panels; 3.5% salt water; a university lab's commercial potentiostat for reference readings.
First step
Build or buy an ABE-Stat and measure precision resistors from 1 megohm to 10 gigohms in a shielded metal box, to find where its readings stop matching their marked values.
Done looks like
Over ten weeks of soaking, 20 painted panels give 200 paired weekly readings at 0.1 Hz, and the open meter agrees with the commercial instrument within a factor of two on at least 90% of them across the range from 100,000 ohms to 10 billion ohms. Design files, code and readings are public.