Application

Is it really Sa2.5? Four field checks before priming

The conversation at the blast hall door is always the same. The applicator says it is finished, the inspector says it is below grade, both hold up a phone photo and neither convinces the other. It usually ends through relationships or programme pressure rather than data.

This should never be an argument. Sa 2½ is a measurable condition, not an impression. Break it into four items and each one has a tool and a way to record it.

1. The comparator: it judges cleanliness, not profile

The Sa 2½ photographic comparator in ISO 8501-1 judges surface cleanliness: no visible oil, grease, dirt, mill scale, rust, previous coating or foreign matter, with only slight spotting or striping permitted.

A frequent misunderstanding: many assume the comparator also specifies roughness. It does not. The comparator settles clean or not clean; roughness belongs to a different standard, ISO 8503. So "looks like the comparator" answers only half the question.

Lighting matters. Comparison should be made at not less than 300 lux with oblique illumination, otherwise fine rust spots disappear. The common site error is comparing in backlight or at dusk, and grading Sa 2 as Sa 2½.

One more detail: comparators fade. Paper prints that are old, damp or sun-bleached shift in tone, so a shop should replace the ones hanging in the blast hall periodically.

2. Profile depth: without roughness, adhesion is hollow

Blasting does two jobs: it removes contamination and it creates a profile for the coating to key into. ISO 8503 divides profile into fine, medium and coarse, judged in the field against a comparator or measured with a stylus or replica tape.

Medium G is the usual specification, corresponding to an Ry5 of roughly 40 to 85 microns. That range has to match the coating: a 60 micron thin film over coarse profile will have peaks protruding through the film and early rust spotting. Conversely, a high-build product over fine profile will not develop enough adhesion.

Two typical causes of low profile: abrasive that is too fine or has been recycled too many times and is now blunt, so you get matting rather than blasting; and too great a standoff distance, too square an angle or insufficient pressure. Both are fixed by changing abrasive and parameters rather than equipment, and both are frequently overlooked.

Recording: take several points per inspection lot with replica tape or a stylus, photograph them and note the location. Profile is the item most often skipped because instruments are expensive and slow, yet it is the hardest to correct later — once coated, low profile means stripping and starting again.

3. Soluble salts: invisible, decisive

Residual salts on blasted steel, chlorides and sulphates above all, are a leading cause of early coating failure. They cannot be seen or felt, but they create osmotic pressure under the film, drawing water in and causing blistering and loss of adhesion.

Testing follows the ISO 8502 series. The Bresle method is the common one: a patch cell is filled with a measured volume of pure water, worked against the surface repeatedly, then extracted and its conductivity measured and converted to salt density. Conductivity pens and dedicated test kits are also used.

There is no single global acceptance value; a common requirement is chloride below 20 to 50 milligrams per square metre depending on coating type and service. Immersion areas and ballast tanks are the strictest, often under 20, while atmospheric areas can be relaxed. The figure must be written into the specification or acceptance has no basis.

Where field work goes wrong: dust around the test area not cleaned, contaminated syringes, inaccurate estimation of the test area. These handling details affect the result more than instrument precision.

4. Surface dust: the tape method, one minute

The ISO 8502-3 tape method: press a specified length of pressure-sensitive tape onto the surface, smooth it down, lift it and transfer it to white paper or a comparator card, then grade by the number and size of retained particles.

The standard grades dust from 1 to 5, lower being cleaner. Coating work usually requires class 2 or better, meaning few and small particles. That sounds lenient, yet many blasted surfaces fail in practice, especially where dust from the blast hall floor settles on freshly cleaned plate.

This is the item most often skipped because the surface "looks clean". But dust trapped between steel and coating is an artificial release layer, and adhesion figures drop immediately. The failure is also very hard to see: nothing shows at application, and the whole area peels months later.

The fix is simple: blow down with dry, oil-free compressed air after blasting, and vacuum critical areas. It costs almost nothing, but it has to be written into the work card or nobody does it.

5. Edges and hand-prepared areas: how to grade them

This is the most disputed point. Blasting reaches flat plate but not weld toes, the backs of bolts or tight corners. Those areas are usually prepared with power tools to St3. So does the plate count as Sa 2½ or not?

The usual arrangement: accept the main surface as Sa 2½ and edges and local areas as St3, but state this in the method statement in advance and require additional coating reinforcement there, such as a stripe coat. Record by area rather than grading everything together.

The worst case is discovering this after the fact. The applicator says edges are hand-prepared by definition, the inspector says he required Sa 2½ throughout, and neither is wrong because the clause never said.

My advice is blunt: add a table to the specification listing main surface, edges and welds, and non-blastable areas, each with its required grade and acceptance method, signed by all three parties. That table removes eighty percent of the arguments.

6. Turn the judgement into a record

Four items become one inspection sheet: cleanliness (comparator grade plus illuminance), profile (Ry5 value plus points), salts (mg/m² plus method) and dust (class). One number each, with location and date.

With that sheet, the dispute moves from "I do not think it is to grade" to "item three is out of limits, how do we handle it". The first is a matter of position, the second of technique — and technical problems can be solved.

One more point worth making: not every area needs Sa 2½. Immersion, buried service and severe corrosivity justify the cost; secondary members in a dry interior are fine at Sa 2 or even St3 with an appropriate scheme. A standard is a tool, not a dogma — but once written into a contract, it is binding.

Need selection advice?

Tell us your substrate and service conditions

One-to-one coating consultancy: system design, quantity take-off and quotation.

Contact us
WhatsApp