How to tell if a curtain is fire retardant? 4 reliable checks

A curtain delivered with the word "flame-retardant" on its label is not necessarily compliant with the M1 or M2 requirements expected in a public-access building. The difference between the commercial claim and the regulatory proof determines the operator's legal liability during a safety commission visit or after a fire. This article details the four possible checks, their limits and their legal weight.

For the obligations that apply per space, see our guide to public-access building M1/M2 regulations. To understand the fabric itself, read the article on the three families of flame-retardant fabrics. To make a non-compliant curtain compliant, see the three post-manufacturing flame-retardant treatment methods.

Why verify that a curtain is genuinely flame-retardant?

Three concerns justify a rigorous check, especially across a B2B estate.

1. Public-access building compliance. The safety commission inspects the register at every visit (3 years for most public-access buildings, 5 years for category 5 establishments). With no traceable classification report, an unfavourable opinion is possible, regardless of the actual quality of the fabric.
2. Multi-risk insurance. Fire policies almost always include a forfeiture clause in the event of material non-compliance. A fire involving a curtain presented as M1 without documentary proof can lead to a refusal of cover.
3. Civil and criminal liability. In the event of a fire-related accident, the operator must prove that they exercised due diligence, which includes actively verifying the installed materials, not merely trusting the supplier's label.

Four checks cover these three concerns, with very different levels of proof.

Method 1 — The curtain's technical label

The label is the first instinct and the most accessible. On a flame-retardant curtain, it must state at a minimum: the fibre composition (for example "100% inherently flame-retardant polyester"), the class obtained (M1, M2, or Euroclass B-s1, d0), and ideally the reference of the weaver or the yarn used.

What the label proves: that the manufacturer declares a class. It is a commercial declaration, not a regulatory proof.

What the label does not prove: the authenticity of the class, the date it was obtained, whether it has held over time (a fabric treated chemically loses its class after several washes, and the label does not flag this degradation), nor the traceability back to the original classification report.

Before a safety commission, the label alone is never sufficient. At most it can point to a classification report to request from the supplier.

Method 2 — The class M classification report

The classification report is the only legally enforceable proof of a fabric's class in France. It is issued by a recognised laboratory after a standardised test under the NF P 92-507 standard. Three laboratories account for the vast majority of these reports in France.

• CSTB (Centre Scientifique et Technique du Bâtiment) — the historical reference for the building industry.
• LNE (Laboratoire National de métrologie et d'Essais) — a versatile state laboratory.
• IFTH (Institut Français du Textile et de l'Habillement) — specialised in textiles.

A usable report contains five mandatory pieces of information.

1. The name and signature of the issuing laboratory.
2. The test date and the report issue date.
3. The precise identification of the fabric tested (commercial reference, composition, weight).
4. The regulatory result (class M0, M1, M2, M3, M4 and where applicable Euroclass B/C/D).
5. The test conditions, in particular whether the fabric was tested after washing (M1 confirmed after X cycles).

A report that does not state the test date, the issuing laboratory, or the exact fabric reference is not enforceable. Request the full PDF version, not an extract or an attestation written by the supplier.

Method 3 — The at-home wick test

The wick test is a simple physical check: you take a wick of yarn from a non-visible spot on the curtain (inner hem, seam surplus), place it on a non-flammable surface, bring it close to a small flame and observe the behaviour. An M1 fabric does not ignite durably and self-extinguishes as soon as the flame is removed.

What the wick test proves: physical behaviour consistent with a class M. If the wick burns vigorously and self-sustains, the fabric is not M1, full stop.

What the wick test does not prove: the exact class (M1 and M2 cannot be told apart with the naked eye), regulatory compliance (the test has no value before a safety commission), nor whether the class holds after washing.

The test remains useful for spotting blatant fraud (a curtain presented as M1 but that clearly ignites) before starting a more formal procedure to request a classification report.

Method 4 — The direct request to the supplier

For a B2B estate, the most complete method is to question the supplier in writing. Five questions frame the check.

1. What is the exact fibre composition of the fabric? Expected answer: percentage per fibre, polymer type (inherently flame-retardant polyester, flame-retardant viscose, chemically treated cotton or linen).
2. Is the M1 class inherent or by chemical treatment ? Inherent answer = lifetime class; treatment = limited duration to be specified.
3. Which laboratory issued the classification report, and on what date ? Expected answer: the name of the laboratory (CSTB, LNE or IFTH) and a test date less than 5 years old.
4. Can you send me the report in PDF format? Expected answer: sent within 48 hours. Any hesitation or evasive reply is a red flag.
5. What is the guaranteed durability under industrial washing? Expected answer: number of guaranteed cycles and washing conditions (temperature, compatible products).

Keep the answers in writing (email, technical sheet). They form part of the commercial traceability and can be produced in the event of a dispute.

Comparison of the four methods

Method	Type of proof	Value before a safety commission	Cost or effort
Technical label	Commercial declaration	Indicative, never sufficient	Free, immediate
Class M classification report	Regulatory proof	High if dated less than 5 years old	Free if supplied, paid if retested
Wick test	Physical check	None (internal use only)	Free, 5 minutes
Written request to the supplier	Commercial traceability	Indicative, complementary to the report	Free, 24 to 48 hours

The right approach combines all four. The label gives a first reading, the wick test rules out blatant fraud, the request to the supplier reveals the quality of the commercial follow-up, and the report is the key document to keep in the safety register.

Special cases to watch

A curtain inherited from a previous operator

When taking over a business or changing managers, curtains may stay in place with no documentation. Always request the previous operator's safety register. Failing that, have it retested by a laboratory (300 to 800 € depending on complexity) or replace it.

A curtain imported from outside the European Union

Foreign classifications (US NFPA 701, British BS 5867, German DIN 4102) are not equivalent to the French class M. A foreign declaration of conformity does not exempt you from the NF P 92-507 test if the curtain is installed in a public-access building in France. The only case where the European classification (Euroclasses A to F) is accepted is the one provided for by the order of 21 November 2002, with a precise correspondence table.

A report older than 5 years

The NF P 92-507 standard sets no legal validity period for the report, but the inspection bodies (Bureau Veritas, Apave, Socotec) recommend a recent report. A report more than 5 years old may be subject to a retest request, especially if the fabric composition has changed at the weaver since the initial test.

What to do if the curtain fails the check?

Three options open up depending on the severity of the defect.

• The curtain is physically M1 but has no traceable report: order a retest from the CSTB, the LNE or the IFTH. Indicative cost 300 to 800 €. Turnaround 4 to 8 weeks. The report obtained applies to the fabric tested, not to another batch.
• The curtain is not flame-retardant but recent: consider a post-manufacturing flame-retardant treatment (workshop saturation or on-site application) with a report issued by a certified workshop. Our dedicated article details the three methods and their real lifespan.
• The curtain is not flame-retardant and worn: replace it with a new inherently flame-retardant curtain classed M1 from the outset. This is the most durable solution and the least costly over 10 years for a significant estate.

Kurtens traceability in B2B

With every flame-retardant B2B order, Kurtens supplies the named, dated class M classification report from the weaver, as a PDF sent with the delivery. The report is included in the commissioning slip, ready to be attached to the operator's safety register. M1 certification is available on request across all our manufacturing, with any additional cost built into the quote without a separate line.

For a project to replace or upgrade an estate, see the made-to-measure flame-retardant curtains page or request a B2B quote. Reply within 24 hours.

Frequently asked questions about verifying a flame-retardant curtain

How do you read a class M classification report correctly?

Look for five elements: issuing laboratory (CSTB, LNE or IFTH), test date, fabric identification (reference, composition, weight), class obtained (M1, M2 or Euroclass), and test conditions (in particular whether tested after washing). Any report that does not state these five points must be considered incomplete and should be completed with the supplier.

Can the wick test void the fabric warranty?

A weaver's or reseller's commercial warranty does not cover modifications made by the user. Taking a wick from a hem or a seam surplus stays harmless if the curtain is intact elsewhere. Prefer a hidden spot and limit the sample to a few centimetres. Keep the original hem and the installation offcuts for future tests.

My insurer requires an attestation, not a report: which one prevails?

The report is the official regulatory document. An attestation is usually a letter issued by the weaver or reseller summarising the characteristics. It is useful as a complement, never as a substitute. If the insurer requests an attestation, attach both the report and the attestation, and ask for written confirmation that it is accounted for in the policy.

How do you check whether a CSTB report is authentic?

The CSTB does not offer a public register of issued reports, for reasons of commercial confidentiality. Verification is done through a direct request to the CSTB citing the report reference: the service confirms the existence and validity of the document. A report in an unusual format, without an official header or signature, should prompt caution and trigger this check.

Is there a national register of M1 curtains?

No. France does not keep a centralised national register of products classed M. Traceability rests on the pairing of fabric and report, kept in the safety register of each operator. It is precisely this absence of a central register that makes documentary verification essential for every order.

Can a report more than 5 years old still be valid?

Legally yes, the NF P 92-507 standard sets no expiry date. In practice, a report more than 5 years old may be challenged by an inspection body, especially if the fabric composition has changed at the weaver. The prudent rule is to ask the supplier for a recent report with each new order.

Regulatory note: this article presents the practical methods for verifying a curtain's class against the French NF P 92-507 framework and the order of 25 June 1980. It does not replace consulting an approved inspection body. The validity of a class in a public-access building depends on complete documentary traceability and the specifics of the building.

For a project where M1 traceability is required from delivery : request a B2B quote. Reply within 24 hours, CSTB or IFTH report attached to the delivery, M1 certification on request.