When a coating fails early, the paint gets the blame. Most of the time the cause was settled before the first coat went on, in a blast that met one of three requirements and was assumed to have met all three.
Three Measurements, Not One
A prepared steel surface is judged on three separate things, measured independently and by different means. How much rust, mill scale and old coating remain, which is cleanliness. How rough the surface is, which is profile. And what invisible contamination sits on it, principally soluble salts and dust. A surface can pass any one of these and fail the others, and until all three are checked it is not ready to coat.
The most common specification error follows directly from confusing the first two. Sa 2½ is a cleanliness grade. It says nothing about roughness, which is measured in micrometres under a different standard entirely. A specification calling for Sa 2½ and nothing else has specified how clean the steel must look and left the mechanical key the coating depends on entirely undefined.
Sa 2½ describes what the surface looks like. It does not describe what the coating grips. Those are different measurements and a specification needs both.
Cleanliness, and What the Grades Mean
ISO 8501-1 is a photographic standard. It defines four rust grades, A through D, describing the condition of the steel before work starts, and a set of preparation grades describing the result. For abrasive blast cleaning these run Sa 1, Sa 2, Sa 2½ and Sa 3. For hand and power tool cleaning they run St 2 and St 3. Assessment is by comparison against the photographs in the standard, which is why an inspector carries the book rather than an opinion.
The practical distinctions are narrower than the grade names suggest.
Sa 2½, very thorough blast cleaning: Mill scale, rust and coatings removed such that any remaining traces show only as slight stains in the form of spots or stripes. The workhorse grade for most marine and offshore specifications.
Sa 3, blast cleaning to visually clean steel: The surface free of all visible mill scale, rust, coatings and foreign matter, with a uniform metallic colour. Considerably more expensive in time and abrasive, and specified where the coating system or the service demands it.
St 2, thorough hand and power tool cleaning: Loose mill scale, rust and coatings removed. A much lower standard than any blast grade.
St 3, very thorough hand and power tool cleaning: As St 2 but worked far more thoroughly, so the surface takes on a metallic sheen. Still below blast grades, and used where blasting is not practicable.
Other parts: ISO 8501-2 covers previously coated surfaces after localised removal, 8501-3 covers welds, cut edges and surface imperfections, and 8501-4 covers high-pressure water jetting including flash rust grades.
The 8501-3 point is worth drawing out, because weld spatter, sharp edges and laminations fail coatings disproportionately. Paint thins over a sharp edge and pulls away from spatter, so a facility that blasts the plate beautifully and leaves the edges untreated has built a failure at every corner.
Profile, and Why the Maker States a Band
Surface profile is the peak-to-valley height the abrasive creates, assessed under ISO 8503, with replica tape being the common field method. It provides the mechanical key the coating locks into.
Coating manufacturers state a profile range rather than a minimum, and the reason is that both ends matter. Too shallow and there is insufficient mechanical key, so adhesion depends on chemistry alone. Too deep and the peaks approach or exceed the intended film thickness, leaving high points thinly covered or proud of the coating, which become the first corrosion sites and are invisible on a gauge reading taken elsewhere.
The profile also interacts with the thickness measurement itself, since a gauge reading on a rough substrate includes some of the profile. A specification that sets a profile band and a film thickness without addressing how the gauge is corrected has left an argument waiting to happen.
The Contamination Nobody Can See
Soluble salts are the failure mechanism that makes an otherwise good job fail, and they are invisible. Chlorides, sulphates and nitrates deposit on steel from marine atmosphere, from handling, from contaminated abrasive, and they concentrate in the bottom of corrosion pits where blasting does not reach them.
Blasting removes rust and scale. It does not reliably remove salts, and it can drive them deeper. A surface that grades perfectly to Sa 2½ can carry enough salt to destroy the coating, because salt is hygroscopic: it pulls moisture through the cured film by osmosis, water accumulates at the steel interface, and the coating blisters and disbonds from underneath. The visible symptom appears months or years later and looks like a coating defect.
Salt does not attack the coating. It pulls water through it. By the time a blister appears, the cause has been sitting under the paint since the day it was applied.
Two standards work together to assess it. ISO 8502-6 specifies the extraction, the Bresle method, in which a defined volume of deionised water is held against a known area by a patch or cell, worked in and out, and left for a specified dwell before removal. ISO 8502-9 specifies the evaluation: the conductivity of that solution is measured with a temperature-compensated meter, and the surface density of total soluble salts is calculated from it.
Two features of this method are routinely misunderstood. It is not ion specific, so it reports total soluble salts rather than chloride alone, and results from ion-specific methods cannot be compared with it directly. And ISO 8502-6 sets conditions for the test itself, at 23°C and 50% relative humidity, with deviations recorded and agreed between the parties. A salt report without climate and substrate temperature recorded can be challenged.
The point that matters most for a specification: there is no industry-standard acceptable level of soluble salt contamination. The standards describe how to extract and how to measure. They do not state a pass mark. Any conductivity or salt density limit in a specification comes from the coating manufacturer for that system, or from the project, and a specification that calls for Bresle testing without stating the limit has required a measurement without defining what result is acceptable.
Hold Points, and Why the Signature Has to Come First
Every measurement above becomes unverifiable the moment the next coat goes on. That is the entire argument for inspection hold points: a stage at which work stops, the condition is verified and recorded, and only then does the next operation proceed.
The sequencing is the whole point. A hold point signed after the next coat has been applied is not an inspection record, it is a recollection, and it is worth very little in a dispute two years later when the question is whether the substrate was contaminated on the day.
What to Put in the Specification
A client commissioning coating work can close most of the common arguments before they start, and it costs nothing at the drafting stage.
Name the cleanliness grade and the standard, and name the treatment required for welds, edges and imperfections separately. State the profile as the band the coating system requires, with the measurement method. Require soluble salt testing by extraction and conductivity, and state the limit, since the standards do not supply one. Specify the ambient and substrate conditions under which application may proceed, including the dew point margin, which is a manufacturer requirement rather than a universal figure. Define the hold points, who signs them, and that work does not proceed until they are signed. Require the records to be retained and handed over. And include adhesion testing where the system or the service warrants it.
Every item on that list exists because someone has argued about it after a failure. The coatings themselves carry class and type approvals for their intended service, and those approvals assume application to the standard the maker specified. Where the preparation falls short, the approval is not what failed, and the warranty conversation tends to arrive at that conclusion quite quickly. Our own painting and protective coating work is run this way, with the preparation measured and recorded before any coat goes on.
Frequently Asked Questions
What is the difference between Sa 2½ and Sa 3?
Both are abrasive blast cleaning grades under ISO 8501-1, assessed against photographs in the standard. Sa 2½ requires mill scale, rust and coatings to be removed so that any remaining traces show only as slight stains in the form of spots or stripes. Sa 3 requires a surface free of all visible mill scale, rust, coatings and foreign matter with a uniform metallic colour, which takes considerably more time and abrasive.
Why does a coating manufacturer specify a profile range rather than a minimum?
Because both ends matter. Too shallow a profile gives insufficient mechanical key and leaves adhesion dependent on chemistry alone. Too deep and the peaks approach or exceed the intended film thickness, so high points are thinly covered or proud of the coating and become the first corrosion sites. Profile is measured under ISO 8503 and is a separate matter from cleanliness grade.
What is the Bresle method, and what limit applies?
ISO 8502-6 specifies the extraction: deionised water held against a known area by a patch or cell, worked and left to dwell, then removed. ISO 8502-9 specifies the evaluation by conductivity measurement, from which total soluble salt density is calculated. Importantly, the standards set no acceptable contamination level. Any limit comes from the coating manufacturer or the project specification, so a specification requiring the test must also state the limit.
Why do soluble salts survive blasting?
Because blasting is designed to remove rust, scale and coatings, not dissolved salts, and salts concentrate at the bottom of corrosion pits where abrasive does not reach effectively. Salt is hygroscopic, so it draws moisture through the cured film by osmosis; water accumulates at the steel interface and the coating blisters and disbonds from beneath, often months or years later.
Sources: ISO 8501-1, Preparation of steel substrates before application of paints and related products, Visual assessment of surface cleanliness, rust grades and preparation grades, with ISO 8501-2, 8501-3 and 8501-4 covering previously coated surfaces, welds and cut edges, and high-pressure water jetting respectively · ISO 8502 series, Tests for the assessment of surface cleanliness, including ISO 8502-3 dust assessment, ISO 8502-6 extraction of soluble contaminants by the Bresle method, and ISO 8502-9 field determination of water-soluble salts by conductometric measurement · ISO 8503 series, Surface roughness characteristics of blast-cleaned steel substrates · ISO 4624, Paints and varnishes, Pull-off test for adhesion
MarineCraft Journal
Get the next briefing in your inbox
Safety protocols, offshore training, regulation and market analysis, written for Malaysia’s maritime and offshore industry. No sales pitches.
We use your address only to send MarineCraft Journal. Unsubscribe any time. See our Privacy Policy.