Fire certification for leather interiors: IMO FTP Code, EN 1021 and what a tannery can actually certify

The certificate belongs to the seat, not to the hide. Understanding that is the difference between participating in the fire file and being eliminated from it silently.

That asymmetry is why marine-grade vinyl closes deals it should not close. It is fixable, and fixing it costs one properly designed test programme.

What does IMO FTP Code Part 8 actually test?

Part 8 of the 2010 FTP Code (Resolution MSC.307(88)) covers upholstered furniture on board: loose seating, banquettes, built-in benches, lounge and office chairs. Two ignition sources are applied to a mock-up of the real seat — a smouldering cigarette and a match-flame-equivalent butane flame — and the assembly must show no progressive smouldering and no flaming ignition, with the cigarette test held for one hour.

Part 8 cites ISO 8191-1/-2 as its test procedures — the cigarette and match-flame methods that the European contract standards EN 1021-1/-2 mirror almost exactly. The results can then feed type approval, the wheelmark, under Directive 2014/90/EU.

Three consequences follow, and they are the whole point of this guide:

1. The cover cannot pass or fail alone. Change the foam and the certificate is void. Change the leather article and the certificate is void.
2. A land-based EN 1021 result is a strong pre-screen for a marine file, because the ignition sources and pass criteria are equivalent. One test programme informs two markets — it does not replace the marine test.
3. Whoever owns the mock-up owns the certificate — normally the upholsterer or the shipyard, not the material supplier.

Which standard applies to which project?

A superyacht built in the EU for charter and a five-star hospitality refit in London are, from the tannery's side, the same conversation with different paperwork at the end.

What should the tannery supply, then?

It supplies the inputs that make the assembly pass, plus the evidence trail. Ask for these in writing before you approve an article:

• The exact article and finish reference, plus the finish chemistry family — flame-retardant finishing systems exist and must be declared, because they change hand and colour.
• Batch continuity commitment of 12–24 months. A re-test triggered by a discontinued article is the expensive failure mode, not the first test.
• Any existing EN 1021-1/-2 reports on that article over standard CMHR foam, as a pre-screen. It does not certify your seat; it tells you whether your seat is likely to pass before you build ten mock-ups.
• Performance data on the same article: Martindale 50,000+ for high-traffic seating, Blue Wool 5–6 for light fastness, low fogging for enclosed cabins, hydrolysis resistance for humid environments.
• Full traceability to the tannery and the batch — increasingly a shipyard and contract-client requirement in its own right, and the reason LWG Gold audit status shows up in these specifications.

Then build the mock-up with the actual foam, the actual barrier and the actual construction, and test once. Testing a hide over a foam you will not use is money spent on a document nobody can rely on.

Does leather perform better or worse than vinyl here?

Neither claim is safe as a generalisation, and that is the honest answer buyers rarely get. Both leather and vinyl can pass or fail Part 8 depending on the filling behind them and the finish on top. Where they genuinely differ is elsewhere: PU-coated materials hydrolyse in constant humidity, plasticiser migration stiffens PVC over time, and neither can be refinished — which is why the sensible split is exposed zones in technical textile or vinyl and interior zones in leather. That trade-off is worked through in genuine leather vs marine vinyl, and the zone-by-zone specification lives in the marine-grade leather hub.

The commercial point stands regardless: the fire file is decided long before the aesthetic review, and material suppliers who cannot participate in it are eliminated silently.

How long does the certification path actually take?

• Pre-screen on existing EN 1021 data: days, no cost if the tannery has it.
• Mock-up build: 2–3 weeks once article, foam and construction are frozen.
• Part 8 test at an accredited laboratory: typically 3–5 weeks including scheduling; budget as a project line, not a material line.
• Re-test after any substitution: same again. This is why batch continuity is worth negotiating hard at the sampling stage.

Plan for the certificate at RFQ, not at delivery. A yacht interior programme that discovers the fire file eight weeks before handover has already lost the argument to the vinyl supplier who arrived with a folder. The same applies on land: headboards and upholstered bases for hospitality contracts carry the identical logic, as covered in leather for premium bedding brands.

If you need an upholsterer, a test laboratory or a hide supplier who will engage with the fire file rather than dodge it, state the project openly on our open B2B board under Materials.