PRISME-FAIR Program

Fire propagation along long cable trays

The first focus area of the FAIR programme aims to better understand fire propagation along cable trays, whether vertical or horizontal, in confined and mechanically ventilated compartments.

An initial series of large-scale tests focused on fire propagation along vertical cable trays. These tests were carried out in 2025 at the DIVA facility, using a 6-metre-long vertical cable tray, in order to study both the spread of the fire and its consequences in confined, ventilated spaces. 

A second series of tests, also on a large scale, focuses on fire propagation along long horizontal cable trays. This complements the campaign conducted as part of PRISME 3 CFP (Cable Fire Propagation), which was carried out in the DIVA facility’s test tunnel using 6-metre-long horizontal cable trays.

These tests have shown that the use of longer cable trays is more suitable for avoiding any edge effects and thus allows for more thorough validation of analytical models. Taking into account feedback regarding the DIVA safety criteria, new tests using 9-metre-long cable trays will be carried out in 2026 in the DIVA facility corridor.

Effects of cable ageing on fire behaviour

The FAIR project also examines the effects of ageing on electrical cables, and in particular on sheathing materials (insulation), on fire behaviour and fire spread. Two test strands are proposed :

• Part 1: natural ageing

This phase aims to analyse the fire behaviour of naturally aged cables in order to obtain representative results. The main challenge is to be able to collect suitable cables from nuclear facilities: cables that are neither contaminated nor activated, and for which the fire resistance standards were initially known and complied with.

• Part 2: artificial ageing

This part aims to analyse the fire behaviour of artificially aged cables. This allows for a wide selection of cables and the choice of ageing sources. However, the main challenge lies in specifying representative accelerated ageing test protocols.
A selection of candidate cables has been made in accordance with the above specifications. The first ageing tests will take place in 2026.

Combustion in hot and vitiated atmospheres

The second focus area concerns the study of fire behaviour in hot and vitiated environments, characterised by low oxygen concentrations and high temperatures.

The aim is to analyse the combined influence of oxygen depletion and increased gas temperature, or external heat flux, on fire development as well as ignition and extinction conditions. 

To this end, the FAIR programme employs a dual experimental approach :

• large-scale experiments, designed to replicate realistic fire scenarios
• medium-scale experiments, with a more analytical focus, to support the understanding of large-scale test results and contribute to the development of a modelling methodology.

An initial series of tests carried out in 2024 includes large-scale experiments in a mechanically ventilated chamber at the DIVA facility. The aim is to replicate scenarios in which external heat fluxes to the fuels are significant and can offset the negative effect of oxygen contamination. Typical examples include fires in cable trays or electrical cabinets beneath the ceiling, submerged in the smoke layer. 

Ignition of unburned gases

Building on the lessons learned from the PRISME 2 project, the FAIR project aims to gain a better understanding of the smoke explosion phenomenon, and more specifically, the conditions that promote the formation of a premixed gas mixture and those that lead to deflagrations.

In fact, smoke explosion phenomena were observed during large-scale cable tray fire tests conducted as part of the PRISME 2 experiments. 
Several deflagrations, characterized by overpressures exceeding 100 hPa and rapid flame propagation, occurred without any sudden change in ventilation.

Consequently, smoke explosions cannot be excluded from fire safety analyses of nuclear facilities. It is therefore planned to conduct medium-scale experiments, which are more conducive than large-scale experiments, to characterize the conditions under which deflagrations occur and develop.

Propagation between discrete sources

The third research area focuses on propagation between several discrete sources within the same environment, leading to the movement of the fire source within the compartment or the occurrence of several simultaneous fires.

This campaign will include large-scale experiments conducted on the GALAXIE platform, focusing on both real-world configurations and academic scenarios (confined atmosphere using the DIVA facility or open atmosphere using the SATURNE facility). 

The practical configurations will cover two fire propagation scenarios:

• vertical spread, mainly driven by convection due to smoke movement
• horizontal propagation, mainly driven by flame radiation. 

The aim of this approach is to focus on the physical mechanisms involved and to propose academic cases for code validation. 

The fire source will be well controlled, whether it is a spreading fire or a homogeneous solid with known thermal characteristics (critical heat fluxes and ignition temperatures). 

The configuration will involve simple source arrangements, such as sources aligned horizontally or vertically. Testing is scheduled to begin in late 2026.

This area of research enables the assessment of the effects of smoke on the operation of electrical or electronic equipment.

The aim of this test campaign is to study the combustion of fires at height in order to verify the validity of the software used in the field of fire science for carrying out studies to support expert assessments.

Five tests are being carried out in the DIVA experimental facility at the GALAXIE platform. These complement the tests conducted during the PRISME 1 and PRISME 2 programmes, which focused on the spread of smoke from one room to others via vertical openings (open doors) and horizontal openings (chutes).

The tests carried out as part of this research area aim to assess the validity of criteria used in probabilistic safety assessments, which determine whether or not a fire will spread from one electrical cabinet to an adjacent electrical cabinet. These criteria are crucial for evaluating the consequences of a fire in electrical rooms.

Propagation between discrete sources

This campaign will complement the PRISME 2 CFS campaign, the aim of which is to study the effect of oxygen deprivation on fire propagation along electrical cables.

• Source prism: PRS-DI-D1 
  This test involves a 0.4 m² sheet of TPH (hydrogenated tetrapropylene) in a 120 m³ room, ventilated at a rate of 4.7 h-1 (560 m³/h). The supply and extraction vents are located in the upper part of the room.

• Door Prism: PRS-D3:
  This test uses a 0.4 m² sheet of TPH (hydrogenated tetrapropylene) in a 120 m³ room connected to an adjacent room via an open door. Each room is ventilated at a renewal rate of 4.7 h-1 (560 m3/h). The supply and exhaust vents are located in the upper part of the room. Targets, consisting of PVC rods, were placed in the adjacent room. 

The results of two tests are available.

To this end, two types of tests were carried out: 

• Tests designed to study a specific physical phenomenon (known as ‘separate effects’), such as flow patterns around an open door
• so-called comprehensive tests, which consist of fires in a configuration representative of nuclear facilities.

The first three campaigns were dedicated to the study of separate effects: 

• PRISME Source for characterising fire sources
• PRISME Door to study the spread of heat and smoke through open doors 
• PRISME Leak to study the spread of heat and smoke through openings (openings, leaks through a fire door, ducts passing through the fire compartment). 

Finally, the last campaign, PRISME Integral, involved conducting comprehensive tests based on various accident scenarios.
The tests fire source characterization was carried out in an open environment under the SATURNE hood, a calorimeter designed for studying fire sources in a free atmosphere and located at the site ASNR in Cadarache. They involved burning hydrocarbon sheets, electrical cabinets or cables, with a particular focus on determining the fire’s heat release rate.

Five experimental campaigns were conducted between early 2006 and mid-2011, comprising more than 35 full-scale experiments. 
The PRISME Source campaign utilised a single ventilated room. It reproduced the effect of oxygen depletion, caused by atmospheric contamination, on how the intensity of a hydrocarbon pool fire evolves as a function of the ventilation flow rate. 
Subsequent experimental campaigns focused on the flow of smoke and hot gases through open doors between two or three rooms, ventilation openings or leaks in a fire door. Finally, so-called tests were carried out full-scale’, in which all parts of the facility were utilised, involving complex fire sources as well as the activation of fire-extinguishing systems (notably sprinklers) or fire dampers on the ventilation system.

Effect of ventilation on fire intensity

The PRISME program has provided a better understanding of the effect of ventilation on the intensity of a fire that breaks out in a confined, ventilated room, and in particular on the duration of the fire. 
Depending on the ventilation air change rate, the fire may extinguish rapidly due to the decrease in oxygen concentration within the burning room. However, PRISME tests have shown that a balance can be established between the air entering through the ventilation outlet and the intensity of the fire: the fuel then burns slowly and completely. For example, the same fire can last 2.5 times longer than in an open atmosphere in a room ventilated at a rate of 4.7 air changes per hour. The key contribution of developments in cable fire testing is a better understanding of the effect of oxygen deprivation on the evolution of fire intensity.

Mixed convection

The PRISME program has also made it possible to quantify the effect, on smoke propagation, of ‘mixed’ convection, which combines forced convection created by ventilation and natural convection induced by the high temperature of the smoke. 
Mechanical ventilation of the burning room can significantly alter the gas flows that naturally establish themselves at an open door between two rooms. 
Depending on the ventilation system, mechanical ventilation contributes to unbalancing the inflow and outflow from the burning room, and to shifting the position of the neutral plane (the height at which flow velocities are zero due to the reversal of flow rates) at the door.

New models for software

The volume of data collected during this program has enabled the assessment of the software’s ability to simulate different fire scenarios. 
The new models, particularly those for pyrolysis, have been implemented in the software of the various partners (the SYLVIA and ISIS software for the ASNR) and validated using the experimental data obtained during this program.
In addition, software intercomparisons were organised by the ASNR as part of a group linked to the PRISME program.
The various partners compared the results of fire simulations with experimental data. In particular, this group tested and analysed several ‘metrics’ – indicators used during the validation process to objectively assess the discrepancies between experimental data and simulation results. 
Among those recommended by the standards, two metrics proved to be complementary in assessing the software’s ability to simulate a fire: 

• The first metric assesses the relative difference between numerical and experimental results for local values, such as extrema
• The second metric is the normalised Euclidean distance, which allows the discrepancy between numerical and experimental results to be assessed over the duration of the fire. This metric acts as a measure of overall error.