BRAVENT – Large-scale fire tests (part 1): Fire performance for non-fire rated ventilation components In the overall BRAVENT project, the goal is to generate answers and documentation on current issues related to ventilation and fire by examining these with experimental fire tests. The present study aimed to evaluate the fire performance of key non-fire-rated components, mainly DCV dampers and exhaust filters, in a comfort ventilation system by testing the hypothesis that the ventilation components will not be damaged by fire within 30 or 60 minutes. To test the hypothesis, a total of 14 large-scale fire tests with different fuels were carried out. The tests were carried out in a test building with several rooms that are representative of classrooms, offices, and corridors. The fire tests were designed to investigate relevant fire scenarios for school buildings, but the findings from the tests can also be used for other purpose-built buildings. The building was equipped with a full-fledged damper-optimized ventilation system, sized to serve a total of 450 m2 and thus provide a realistic basis for the fire tests. Currently, there is no documentation on how non-fire-rated dampers are affected by high temperatures that occur during a fire. The temperature specifications given for non-fire rated DCV dampers are intended for normal operation. Two different types of DCV dampers were tested. In one type, the airflow was measured with a measuring cross, and for the other, the airflow was measured with sensors integrated into the damper blade itself. In several of the conducted tests the non-fire-rated dampers were not able to sustain their function for the required duration of 30 minutes or longer, and failed completely when the temperature inside or outside the dampers reached about 200 ºC. Misreporting of some temperature measurements in the building management system already occurred at lower temperatures, around 120˚C, without significantly affecting the delivered air flow rate. For the damper type with a measuring cross, the plastic hoses connecting the measuring cross and the measuring transducer for the damper melted when hot smoke was transported through the damper. This failure resulted in the DCV damper measuring too low or no airflow. In several tests, this measurement error meant that the DCV damper opened fully, trying to achieve a large enough airflow. In one of the tests where the supply air damper was placed inside the fire room, such a fault on the supply air damper caused it to close completely. In addition to the damage to the dampers, the power supply to the damper was destroyed, and a fuse for the power supply in the control cabinet was short-circuited. This resulted in the building management system losing contact with all the dampers. This shows that a local error can cause the entire system to fail. For the other damper type, where the sensors were located in the damper blade, the high temperatures caused the entire damper blade to melt. It was not observed that soot in the fire smoke led to problems with the dampers' measuring sensors for any of the damper types examined. This indicates that for the performed test series, high temperature and not soot was the greatest challenge for the dampers in the event of a fire. In addition to examining how dampers are affected by fire, there was also an investigation into how the filter performs during a fire. It was found that the filter could largely capture the soot particles in the smoke. This means that equipment located downstream of the filter is relatively well protected against soot, and the possibility of soot contamination to the supply air side via a rotating heat recovery unit can, therefore, be considered minimal as long as the filter is not damaged. However, when the filter collects so much soot, it shows that the potential for the filter to clog. How quickly this happens depends, among other things, on the materials that burn and the size of the fire in relation to the air handling unit's capacity. This also aligns with results from an earlier BRAVENT project [1]. The air temperature in the unit was in all tests carried out below 60˚C and thus lower than the filters' maximum operating temperature of 70˚C. The conclusion from the tests is that the extraction principle with non-fire-rated components cannot be considered a safe strategy for 30 or 60 minutes.

BRAVENT – Large-scale fire tests (part 2): The effect of the ventilation strategy on smoke dispersion and pressure control in a mock-up school building In the overall BRAVENT project, the goal is to generate answers and documentation on current issues related to ventilation and fire by examining these with experimental fire tests. The present study aimed to evaluate how different failure mechanisms identified in sub-report 1, "Fire performance for non-fire rated ventilation components" [5], affect the pressure conditions and possible smoke spread in the building. In order to investigate the ventilation system's ability to control pressure balance and smoke spread in the event of a fire, 14 large-scale fire tests with different fuel types were carried out. The tests were conducted in a test building with several rooms representing a classroom, an office and a corridor. The fire tests were designed to investigate relevant fire scenarios for school buildings, but the findings from the tests may also be relevant for other purpose-built buildings. The building was equipped with a full-fledged damper-optimized ventilation system which was sized to serve a total of 450 m2 and thus provide a realistic basis for the fire tests. In sub-report 1, a number of component failures were uncovered when the system was exposed to hot smoke. The failure mechanisms led to the system losing its status and control over the dampers, thus also losing the overview of which air volumes passed through the dampers. The failure of the dampers then led to an imbalance in the ventilation system, also in rooms other than the fire room. In the extreme, this can lead to challenges in connection with escape by making it difficult or impossible to open doors. Smoke spread via the supply air ducts can occur due to smoke backflow if the pressure in the fire compartment exceeds the pressure in the supply air duct. In the experiments, the ventilation system increased to the maximum design airflow rate (Vmax) upon fire detection, and no smoke spread due to backflow of fire smoke in the supply air ducts was observed. It was not tested whether a backflow of smoke could have also been prevented at lower airflow rates than Vmax. Whether the ventilation system maintains sufficiently high pressure in the supply air duct to prevent backflow depends on, among other things, the capacity of the supply air fans, the density of the fire compartment and how quickly the fire develops. The smoke from the fire caused the filter in the unit to gradually clog. In three tests, the filter was clogged so much that the air handling unit could not extract enough air, which created an imbalance in the ventilation system. The new BRAVENT tests show that the size of the fire in relation to the air handling unit's capacity plays an important role in how long the exhaust airflow can be maintained. An air handling unit serving several small fire cells with limited available fuel will be able to maintain the required exhaust air volume longer than an aggregate serving a few large fire cells as the ratio between smoke volume and total air volume changes. Other factors that affect how long the unit can compensate for filter clogging are: • whether the air handling unit is designed for 100% or less simultaneity, i.e. the proportion of rooms that can have the maximum amount of airflow at the same time, • the condition of the filter at the start of the fire, completely new filters were used in each of the fire tests, • whether the ventilation system operates with Vmax or less airflow when a fire alarm is triggered. No smoke spread between the rooms via the exhaust duct was observed. However, since the exhaust airflow rate in one of the tests was significantly reduced to around 50% of the maximum design airflow rate, it is concluded that gradual clogging of the exhaust filter will increase the risk of smoke spreading via the exhaust duct. Clogging of the filters can be avoided by establishing a bypass. Nevertheless, it must then still be documented that all other functions of the ventilation system are safeguarded in the event of a fire. The tests carried out showed that the failure mechanism of some components (measurement errors in dampers, short circuits and clogging of the exhaust filters) can lead to the whole system no longer being able to maintain its function.

Dette dokumentet er en veileder. Veilederen skal benyttes av rådgivende ingeniører i brannteknisk prosjektering av ventilasjonsanlegg i skolebygg. Det fokuseres på rømningssikkerhet og skadebegrensning ved små branner i en tidlig fase av brannforløpet. Veilederen beskriver forutsetninger for bruk, krav og preaksepterte ytelser i VTEK, samt ulike typer ventilasjonsanlegg med tilhørende komponenter. Dokumentet illustrerer typiske prinsippløsninger for ulike ventilasjonsstrategier i bygg, samt forskningsresultater fra brannforsøk og undersøkelser i tidligere faser av BRAVENT-prosjektet. Veilederen tar for seg nybygg, eksisterende bygg og verneverdige bygg hvor grensesnittet mellom rådgivende ingeniør brann (RIBr), ventilasjon (RIV), elektro (RIE) og automasjon (RIAut) defineres. En sjekkliste oppsummerer viktige punkter å hensynta ved prosjektering av ventilasjonsanlegg. Rutiner for drift og vedlikehold beskrives til slutt.

The use of external fire suppression systems can reduce the risk of fire spreading between buildings. This study investigated the effectiveness and efficiency of different externally placed water-based fire suppression systems on façade fire safety. A series of large-scale experiments comprising an SP Fire 105 setup equipped with sprinklers and high-pressure water mist nozzles have been performed. A combustible facade, consisting of 2.5 cm thick oriented strand board (OSB) plates, was installed to provide challenging conditions and allow a visual assessment of the post-fire damage. The temperature profile on the façade surface was measured with 34 thermocouples, while five heat flux gauges and two fast-response plate thermocouples were used to measure the heat flux on the facade surface and emitted to the ambient. The sprinklers and the high-pressure water mist system effectively suppressed the upwards flame migration and reduced the heat flux toward adjacent buildings. It was observed that the sprinklers acted as a water curtain and kept the facade wet during the fire, promoting minor damage (the burnt area is less than 1% of the total area). The temperature and heat flux measurements demonstrated that the sprinkler system was the most effective suppression system. However, the high-pressure water mist systems achieved similar effectiveness but a much higher efficiency concerning water consumption. The sprinkler nozzles used four times as much water as the high-pressure water mist nozzles. 

Dette notatet er en delleveranse i arbeidspakke WP3 i prosjektet BRAVENT – Effektiv ventilasjon av røyk fra små branner. Notatet med tilhørende vedlegg Kontrollplan gir en beskrivelse av hvordan periodisk kontroll og daglig/ukentlig internkontroll av ventilasjonsanleggets funksjon under brann i skolebygg kan utføres. Kontrollplanen kan også være nyttig for bruk i andre offentlige formålsbygg.

Fire safety during construction and rehabilitation of buildings. This study deals with how the covering of buildings during the construction or rehabilitation of buildings affects fire safety and to what extent the regulations take this into account. The main focus has been mapping relevant requirements, recommendations, and performances related to the covering of buildings, mapping available materials, investigating the material’s fire properties, and modelling the spread of smoke within the covering. A mapping of the relevant laws and regulations applied for constructing and rehabilitating buildings has been carried out. The mapping has shown that demands are placed on owners, users, project owners, builders, businesses, employers, planners and contractors through many different laws and regulations. The people involved can have several roles, and similar roles have different names in the various regulations. For buildings in use, fire safety must be ensured for both the users and workers. It also applies that both the owner and the users are responsible for ensuring fire safety. It requires good communication and cooperation between different actors to ensure that fire safety is maintained for all involved, during the construction and rehabilitation of buildings. When covered scaffolding is used, the Regulations concerning the performance of work, use of work equipment and related technical requirements [10] require that the covering satisfy the fire requirements for materials used in escape routes (§17-20). The guideline to the Norwegian Regulations on technical requirements for construction works, TEK10, (Veiledningen til TEK10) §11-9, provides pre-accepted performance levels. For escape routes, class B-s1,d0 (In 1) is specified for walls and ceilings. There is no requirement for fire classification of the walkways in the scaffolding under the applicable laws and regulations. We believe there should be requirements for fire classification of the walkways, in the same way as for the covering, i.e., B-s1,d0 (In 1) for surfaces on walls and ceilings and Dfl-s1 (G) for surfaces on floors. The simulations of the spread of smoke from a fire inside a building during construction or rehabilitation show that the spread of smoke is affected when the scaffolding around the building is covered. Covering around the sides leads to a greater horizontal spread of smoke in the scaffolding than without covering. When the cover also has a roof, the smoke first accumulates underneath the cover's roof before it eventually also fills up with smoke down the floors of the scaffolding. The simulations showed that establishing an open field in the upper part of the cover would ventilate the smoke gases effectively, and the spread of smoke was essentially the same as for a cover without a roof. In addition, the simulation indicated that the air flow through the walkways in the scaffold could be an important factor in reducing the covering's negative effect on the spread of smoke. Of the 64 different products used for covering found in the survey, 35% had full classification according to EN 13501-1 (such as B,s1-d0). About 6% stated that the product was not flame retardant. Of the remainder, it was evenly distributed between those who stated a fire classification according to other test methods, those who did not provide any information on the fire properties and those who stated that the product was flame retardant without further specification. The mapping also indicates that the products from market leaders used by large general contractors provide products with documented fire properties. Conversations with two of Norway’s largest fire and rescue services shed light on several challenges connected to covering scaffolding and construction during firefighting activities. They pointed out that the covering could cause challenges and delays throughout their efforts. The covering gives a reduced visual overview of the spread of smoke and the location of doors and windows. This information is important for planning both extinguishing and smoke diver efforts. In addition, the covering can be an obstacle to the actual extinguishing effort, the use of an extinguishing agent and smoke divers and rescue efforts.

Fires in road vehicles are common, but a large part is associated with crashes. Of all vehicle fires registered in the USA between 2013 and 2017, 16% have occurred in a parking area and only a fraction of these involved vehicles parked in car parks. Furthermore, car park fires often involve few cars only and do not lead to fatalities. However, major car park fire incidents in the last years have shown that fires can lead to significant property and environmental damage if the fire can spread to a large enough number of adjacent vehicles. Large-scale experiments conducted in the 2000-10s have shown that it can take between 10 to 20 minutes before a car fire spreads to an adjacent car. Essential factors for the fire development are ventilation conditions within the car park, air supply to the burning car’s interior (i.e., breaking windows), and fuel involvement (i.e., breaking fuel tanks, a thermal runaway in lithium-ion batteries or gas releases from pressure release devices). Recent large-scale experiments involving a battery electric vehicle showed fire spreading 5 minutes after the first car was ignited. Thus, early detection and a quick response to the fire are essential to prevent a fire from spreading to multiple cars. Modern cars have become bigger, are thus parked closer to adjacent cars and contain more combustible material, especially plastics. A larger plastic content can increase the fire size of car fires, while an increased share of combustible material on the exterior and a decreased distance between cars may aid a faster fire spread. The increasing share of alternative fuel cars introduces new fire and explosion hazards and poses challenges for extinguishing efforts. However, early detection and quick response time still play an essential role in mitigating the associated risks