Ballast tank coatings keep failing? The key PSPC points are these
A ballast tank is the most demanding compartment on a ship for any coating. It combines alternating seawater and empty conditions, complex structure with many spray shadows, few opportunities for repair until dry docking, and failure that means structural corrosion rather than a cosmetic defect.
The PSPC was written for exactly this. But between the text and the site there is a gap.
1. What the PSPC actually controls
The IMO performance standard for protective coatings in ballast tanks rests on a few hard numbers: surface preparation to Sa 2½ with profile between 30 and 75 microns; soluble salts not exceeding 50 milligrams per square metre as chloride; a nominal dry film thickness of 320 microns for epoxy systems; a light colour to make cracks and rust visible; and a coating technical file submitted and approved before work begins.
There are also organisational requirements: qualified supervision of application, inspection by personnel holding recognised certification such as NACE, SSPC or FROSIO level 2, and records for every step forming a complete technical file.
One clause is often overlooked: a target service life of 15 years. That is not an aspiration, it is the design premise of the whole standard — every number serves that 15 years.
So the minimum bar for compliance is not low. But between the bar and the outcome there is a distance, and that distance sits in the following places.
2. 320 microns is nominal, not a floor to be measured once
The commonest misunderstanding is treating 320 microns as a single-point minimum. The text refers to nominal dry film thickness, and acceptance normally follows a 90/10 or 80/20 rule: the great majority of readings at or above nominal, the lowest not below 80% of nominal.
The problem is at execution level. To make the data look good, yards tend to measure the middle of flat plate and skip edges and complex structure. Yet ballast tank failures start precisely at edges, stiffener roots and around drain holes, where thickness is often half the main field.
The PSPC requires a stripe coat for exactly this reason. But stripe coats easily become ceremonial: a thin pass in the same colour as the main coat, invisible at inspection, and hollow when failure comes.
My experience is to watch two things: the stripe coat should be a different colour from the main coat, which many projects do with a contrast shade for visual confirmation, and its thickness should be sampled separately. Get those two right and 320 microns means something.
3. The 50 milligram limit is often calculated, not measured
The chloride limit of 50 milligrams per square metre is tested by the Bresle method under ISO 8502-9. The common site problem is non-standard sampling: test areas not cleaned, syringes reused, area estimated by eye.
Worse is selective sampling — choosing the cleanest-looking spot. Salt distribution in a ballast tank is very uneven, and welds, pit bottoms and areas near the tank floor run much higher. Measure only flat plate and the data will always pass.
The sound approach is stratified sampling by area: flat plate, weld zones and complex structure, several points each, all recorded. If a weld zone exceeds the limit, the remedy is high-pressure fresh water washing and retesting, not deleting the point from the report.
4. Light colour is not for appearance
The light colour requirement exists for inspection. Rust spots, cracks and blistering are visible at a glance on a light coating; on a dark one early rust is nearly undetectable, and by the time it is visible the failure is extensive.
This is often compromised on real projects. Owners sometimes ask for dark ballast tanks because they hide dirt and look tidier, and yards are happy to oblige. Technically it plants a problem: over a 15-year inspection cycle the eye is the primary detection method.
If dark is genuinely required, the alternative is stronger inspection — periodic borescope or hammer survey — which adds operating cost. That arithmetic belongs before the order, not at the first Special Survey.
5. Compliance is not the same as a good coating
This is the point I most want to make. The PSPC is a set of minimum requirements plus process control. It guarantees that if you do it as written, the coating has a good probability of reaching 15 years. It is not a performance warranty.
Two compliant schemes can differ widely in real life. The difference sits in formulation: resin type, whether pure epoxy or phenolic epoxy; the pigment system and choice of anticorrosive pigment; and the water resistance of the curing agent. The standard does not address these, yet they decide behaviour under long-term wet and dry cycling.
So on top of compliance, ask three questions: does this product have a long service record in ballast tanks; how has it performed on similar ship types and trades; and can the manufacturer show cases beyond 15 years.
There is also repair. The PSPC mainly targets newbuildings, but a ballast tank is patched several times over its life. Repair preparation rarely reaches Sa 2½ because of site constraints, so select a more tolerant product rather than copying the newbuild specification.
6. The documents to watch on site
The coating technical file is not a formality. It should contain the coating scheme and products per coat, surface preparation requirements, application condition limits, inspection items and frequency, and the repair procedure. Reviewing it before work starts surfaces most later disputes.
Inspection records must trace to a specific tank and date: cleanliness and profile after blasting, salts, thickness per coat, and temperature and humidity during cure. In a claim or assessment 15 years later, those records are the only evidence.
Finally, people. Check that inspector certificates are in date, and that supervision can explain how stripe coat and overcoating intervals are controlled. This item is the easiest to treat as paperwork, and it determines whether all the numbers before it are true.