EN 45545 is the harmonized European standard that sets fire protection requirements for railway vehicles, and it has been mandatory across Europe since March 2016.1Problemy Kolejnictwa. Test Methods and Instrumentation for Assessing Reaction to Fire Properties of Railway Rolling Stock It works by classifying each vehicle into one of three hazard levels based on where it operates and how it is built, then applying tested performance limits to materials, fire barriers, electrical systems, vehicle design, and onboard fire management. The goal running through all seven parts of the standard is the same: prevent ignition, slow the spread of flame and smoke, and buy enough time to evacuate.
The Seven Parts at a Glance
EN 45545 is a family of seven parts, not a single document. Part 1 defines the framework and the hazard classification system every other part depends on. Part 2 sets reaction-to-fire requirements for materials and components.2iTeh Standards. EN 45545-2:2020+A1:2023 Part 3 covers fire barriers such as partitions and doors.3Estonian Centre for Standardisation and Accreditation. EVS-EN 45545-3:2024 Part 4 addresses vehicle-level design measures.4iTeh Standards. EN 45545-4:2024 Part 5 handles electrical equipment, including trolley buses, track-guided buses, and magnetic levitation vehicles. Part 6 covers fire detection, alarm, and management systems. Part 7 governs installations that use flammable liquids or liquefied petroleum gas for traction, auxiliary power, heating, or cooking.
Parts 2 and 3 receive the most attention from material suppliers because they decide whether a given product can go on a train. Parts 4, 6, and 7 matter equally to the people designing the vehicle as a whole.
How the Hazard Level Is Determined
Before any fire testing begins, the vehicle must be classified. Two classifications defined in Part 1 combine into a single hazard level (HL1, HL2, or HL3), and every material threshold downstream flows from that number.
Operation Categories (OC1–OC4)
Operation categories reflect how hard evacuation would be if a fire started. OC1 covers vehicles that can stop immediately with quick access to safety. OC2 and OC3 apply to vehicles passing through progressively longer tunnels or underground sections. OC4 covers vehicles in very long tunnels with severely limited rescue access, where the train may have to keep moving for an extended period before it can safely stop.
Design Categories (N, A, D, S)
Design categories capture risks built into the vehicle itself. Category N is standard passenger stock. Category A is automatic vehicles with no emergency-trained staff on board, meaning passengers must self-evacuate.2iTeh Standards. EN 45545-2:2020+A1:2023 Category D is double-deck stock, where extra height and passenger density lengthen evacuation. Category S is sleeping and couchette vehicles, where passengers may be asleep when a fire starts.
The HL Matrix
Operation category and design category combine in a matrix to give the hazard level:2iTeh Standards. EN 45545-2:2020+A1:2023
- OC1: HL1 for N, A, and D; HL2 for S.
- OC2: HL2 for all design categories.
- OC3: HL2 for N, A, and D; HL3 for S.
- OC4: HL3 for all design categories.
The pattern is intuitive. Sleeping cars bump up a level because unconscious passengers cannot self-rescue. Longer tunnels bump up a level because the fire has more time to develop before anyone reaches safety. A commuter train on open track (OC1/N) sits at HL1. A sleeper through Alpine tunnels (OC3/S or OC4/S) hits HL3, where the allowable limits for heat release, smoke density, and toxicity are far tighter. A material that comfortably passes at HL1 can fail at HL3.
Getting the hazard level right is the mandatory first step. An error here means every test result is measured against the wrong criteria, and the mistake usually surfaces during certification review, when it is expensive to fix.
What Materials Have to Prove: R-Sets and Fire Tests
EN 45545-2 organizes materials into requirement sets labeled R1 through R26, each tied to a specific location or function on the vehicle. The R-set determines which fire tests apply and what thresholds must be met at the vehicle’s hazard level.5CEN. EN 45545-2 Railway Applications Fire Protection on Railway Vehicles Part 2 Assignment depends on where the material sits, how much of it there is, and what it does. A thin decorative strip on a wall faces different tests than the wall panel behind it, even though they are centimeters apart. The granularity is intentional: a material’s fire risk depends heavily on its mass, surface area, and orientation.
Very small components can be exempt from full testing if their total combustible mass falls below defined thresholds, but the exemption analysis itself has to be documented. You cannot simply skip a component without showing the math.
Flame Spread
ISO 5658-2 measures how far and how fast fire travels across a material’s surface. A vertically mounted sample is exposed to radiant heat and a pilot flame at one end, and the test records the critical heat flux at extinguishment. It appears in R-sets for large interior surfaces.6RISE. ISO 5658-2 Fire Testing of Products in Trains
Heat Release
The ISO 5660-1 cone calorimeter test measures how much heat energy a burning material releases. The key output is MARHE, the Maximum Average Rate of Heat Emission, in kW/m². A low MARHE means the material releases energy slowly, giving passengers more time before conditions become untenable.7RISE. ISO 5660-1 Fire Testing of Products in Trains8International Organization for Standardization. ISO 5660-1:2015
Smoke Density and Toxicity
Smoke and toxic gases cause most injuries and deaths in train fires, so the standard treats them with particular seriousness. ISO 5659-2 measures the optical density of smoke released when a sample is heated in a sealed chamber; denser smoke means worse evacuation visibility.9International Organization for Standardization. ISO 5659-2:2017
Toxicity is quantified through a Conventional Index of Toxicity (CIT), which aggregates measured concentrations of carbon monoxide, hydrogen cyanide, hydrogen fluoride, hydrogen chloride, hydrogen bromide, sulfur dioxide, and nitrogen oxides into a single number. The gases collected during smoke density testing are analyzed by FTIR spectroscopy, and the CIT is calculated at both 4-minute and 8-minute intervals. Materials must stay below CIT limits for their hazard level to confirm that the air remains survivable during the evacuation window. A separate CIT calculation covers non-listed products, so novel materials face the same scrutiny as common ones.
Oxygen Index
ISO 4589-2 determines the minimum oxygen concentration that will sustain combustion in a material. The result, the Limiting Oxygen Index, is reported as a volume percentage. Since normal air contains about 21% oxygen, materials with an LOI well above 21% are inherently difficult to ignite under ambient conditions. This test is used primarily for printed circuit boards and certain other requirement sets.
Cables
Electrical cables have their own regime under the EN 60332 series. EN 60332-1-2 checks whether fire propagates along an individual cable. EN 60332-3-24 tests bundles of larger-diameter cables (12 mm and above) for vertical flame spread. At all three hazard levels, the burned portion must not exceed 540 mm, and fire spread must stay below 2.5 meters in the bundle test.
Passenger Seats and the Vandalism Test
Complete passenger seats are tested as whole units rather than material by material, because foam, fabric, plastic shells, and metal frames interact during a fire.5CEN. EN 45545-2 Railway Applications Fire Protection on Railway Vehicles Part 2 Seats also undergo a vandalism test defined in Annex A: a 150 N blade slashes the covering at controlled speed and depth before the fire test runs. A pristine covering can protect the foam beneath it; a slashed one will not. The seat must perform after abuse, because that is the realistic transit scenario. A fabric that passes on an intact seat can still fail once the foam is exposed, and finding this out late forces a material substitution at the worst possible time.
Fire Barriers Under Part 3
EN 45545-3 governs the walls, partitions, doors, and floors that keep fire confined to one section of a vehicle. Barriers are rated on three criteria:10Problemy Kolejnictwa. Fire Resistance Testing of Railway Rolling Stock
- E (Integrity): how long the barrier keeps flames and hot gases from passing. It fails if a gap wider than 6 mm and at least 150 mm long opens, or if sustained flaming appears on the unexposed side for more than 10 seconds.
- I (Insulation): how long the barrier limits temperature rise on the unexposed face. No single point may rise more than 180 °C above ambient, and the average across five measurement points must stay below 140 °C above ambient.
- W (Radiation): usually verified indirectly, because meeting the insulation criterion automatically satisfies it.
Each criterion is established for 10, 15, 30, or 60 minutes. A barrier rated E30/I15 holds flame integrity for 30 minutes and thermal insulation for 15. If integrity fails at 13 minutes, the classification drops to E10. As a reference point, a 2 mm steel sheet typically achieves E60, while a 5 mm aluminum sheet reaches E15.11Efectis. Fire Resistance of Rolling Stock Elements
Vehicle Design and Fire Management
Parts 4, 6, and 7 treat the vehicle as a system rather than a collection of materials.
Part 4 requires that technical areas (engine compartments, underfloor equipment cases, cabs, spaces behind ceiling panels) be locked or otherwise secured against passenger access. Equipment enclosures in passenger areas must keep surface temperatures at or below 60 °C under normal operation, and cooking and catering equipment must be installed so that adjacent surfaces also stay below 60 °C. In higher operation categories, luggage storage is restricted; OC4 vehicles cannot use luggage stacks at all, and double-deck vehicles must enclose lower-deck luggage storage with fire-rated partitions or protect it with fixed firefighting equipment.4iTeh Standards. EN 45545-4:2024
Part 6 covers fire detection and response: automatic detectors, alarm systems, selective shutdown of electrical energy when an alarm activates, automatic closure of held-open fire barrier doors, and fixed firefighting equipment. It also addresses passenger alarm systems, voice contact with staff, and the placement of portable extinguishers.
Part 7 governs flammable liquid and LPG installations used for traction, auxiliary power, heating, or cooking. Its focus is preventing fire from starting through leakage, spillage, or spray generation rather than managing a fire already underway. It does not extend to other flammable gases beyond LPG.
How EN 45545 Fits Into EU Vehicle Authorization
EN 45545 is not a stand-alone rulebook. The Technical Specification for Interoperability for Locomotives and Passenger Rolling Stock (TSI Loc & Pas), established under EU Regulation 1302/2014, explicitly requires that materials used in rolling stock comply with EN 45545 for the appropriate operation category. Compliance is therefore not optional for any vehicle seeking authorization on the European rail network. Notified Bodies assessing vehicles for interoperability certification must verify EN 45545 conformity as part of their assessment, and Recommendations for Use issued to Notified Bodies provide standardized interpretation guidance.
Getting Certified and Staying Certified
Certification starts well before any material enters a lab. The manufacturer identifies the vehicle’s hazard level, maps every component to its correct R-set, and calculates total combustible mass by material group to see whether any small parts qualify for exemption. Safety Data Sheets and technical drawings showing orientation, surface area, and installation context accompany samples to the laboratory.
Testing is performed at laboratories holding ISO/IEC 17025 accreditation, which verifies that the lab operates competently and produces valid results using calibrated equipment and standardized procedures.12International Organization for Standardization. ISO/IEC 17025 The lab runs each test specified by the material’s R-set at the appropriate hazard level and records flame spread, heat release, smoke density, and toxicity data in real time. Timelines depend on how many R-sets are involved and how many samples are needed; a complex interior fitout takes considerably longer than a single-component submission.
When testing is complete, the lab issues a report listing numerical results against required pass/fail thresholds. If the material meets every applicable limit, a Certificate of Compliance follows. That certificate is the manufacturer’s proof that the component is approved for use at the tested hazard level. Records must be retained for the vehicle’s operational life; transport authorities can request them during inspections or quality audits.
Factory Production Control
Initial certification only proves that the tested samples met the standard. Production has to stay consistent with those samples, which is the point of Factory Production Control audits. Under the ISO 17065 Type 5 scheme, auditors periodically inspect the manufacturer’s facilities, review quality-control procedures, and compare current output against the originally certified versions. The audit covers machinery, production-line controls, documentation, and staff competency. In many jurisdictions, ongoing audits are a condition for continued approval; an initial test certificate does not guarantee perpetual compliance if production drifts.
Where Compliance Efforts Most Often Break Down
Misidentifying the hazard level is the most common failure. Because the level governs every downstream threshold, an error here invalidates every test result.
Treating R-set assignment as a formality is another. The specific R-set decides which tests apply and what limits count. A material used as an interior vertical surface (R1) and the same material used as a floor composite (R10) face different protocols. A pass under one R-set does not transfer to another.
Underestimating the seat vandalism test is a recurring problem. A fabric that performs on a pristine assembly can fail on a slashed one, because the exposed foam burns differently. Testing only the intact configuration and finding the vandalism failure late forces last-minute material substitution.
Documentation gaps cause avoidable delays. Labs cannot test a material without its intended R-set, hazard level, and installation orientation. Incomplete submissions get sent back, and every round trip adds weeks.