2.1 Fire Incident Data
A total of 3096 fires in car parks were reported in the United Kingdom during the years 1994–2005 [
4]. In 162 of these fires, an automatic fire sprinkler system was present. In 16 of the fires (9.9%), the sprinkler system activated and extinguished the fire and in 84 fires (51.9%), the sprinkler system activated and contained/controlled the fire. In one fire (0.6%), the sprinkler system activated but the fire was not contained/controlled. For the remaining 61 fires (37.6%), the sprinkler system did not activate, probably as the fire was too small. In the 101 fires where the sprinkler system activated, it can be concluded that it extinguished or contained the fire in 100 cases. These statistics likely only include gasoline- or diesel-fuelled vehicles as battery electric vehicles were rare during the time period.
More recently, a couple of cases that include fires in battery electric vehicles were identified. An electric vehicle fire in a car park in The Netherlands was started deliberately on September 1, 2020. The sprinkler system controlled the fire, which did not spread to the battery pack. The car park and other vehicles were undamaged [
5]. On November 21, 2021, there was a fire in the underground car park Marienplatzgarage in Ravensburg, Germany [
6]. According to initial information, an electric vehicle parked on the first parking level and connected to a charging station was probably the cause of the fire. “The sprinkler systems and other fire protection devices worked extremely well”, according to the press spokesman for the Ravensburg fire brigade. Three other vehicles were damaged by the heat. In addition to the cars, two charging stations for electric cars, several lights and cables and the concrete ceiling were damaged. As early as 2014, there was a serious fire in the particular underground car park, which resulted in closure for six years and a restoration for several million euros. The garage was reopened in 2020, this time with an automatic sprinkler system.
2.2 Fire Sprinkler Test Data
Some fire sprinkler test data that were considered useful for the work was found. In 2007 and 2008, BRE in United Kingdom conducted several multi-vehicle, full car fire tests in a parking garage mock-up having a floor area of 12 m by 6 m and a ceiling height of 2.9 m [
4]. All of the cars were used and were selected on the basis of age, size and availability and were either less than five years old, or, if older, of a current model. Gas struts, air bags and other pressurised or pyrotechnic components were left in place, but their air conditioning gas was removed. For each of the cars, the fuel tank contained 20 l of fuel. Test 2 of the test programme was conducted with automatic sprinklers. The test set-up allowed collection of combustion gases and the measurement of the heat release rate. The sprinkler system was designed and installed as representative of a system in place in a typical multi-storey or underground car park, i.e., in accordance with the recommendations for Ordinary Hazard Group 2 occupancies (OH2) of BS EN 12845:2004 [
7], that require a design density of 5 mm/min. The first sprinkler activated after 4 min, but the fire continued to grow, and eventually (after 55 min) reached a peak measured to 7 MW. All six installed sprinklers operated, the last two sprinklers after more than 45 min. The fire did not spread to the adjacent cars. The gas temperature at the ceiling peaked at almost 800°C. The report concludes that the effectiveness of sprinklers in limiting a fire to a single car was demonstrated, which supports findings reported verbally by the fire and rescue service.
In 2009, BRE conducted a single automatic sprinkler test using a two-car stacker configuration [
8]. Automatic sprinklers were positioned both at the ceiling above the stacker and above the car at the lower position. The fuel tanks of the cars were filled with 20 l of fuel. Four automatic sprinklers were installed at each ‘corner’ above the cars. The fire was ignited on the seat of the car at the lower position, with the driver’s window open. After 13:06 [min:s], a sprinkler at the ceiling activated, followed by a second sprinkler at 14:41 [min:s]. At 22:47 [min:s] a third (and final) sprinkler at the low level activated. The water flow to the system was shut-off after one hour. It was concluded that automatic sprinklers at the ceiling contained the fire to the lower car, allowed some spread to the above vehicle, but prevented it from becoming fully involved.
VdS Schadenverhütung has published several fire test methods for water mist fire protection systems. One of these is a fire test method for parking garages [
9]. The method simulates real conditions in a parking garage. The cars used in the tests are scrapped cars but must be intact with entire windscreens, etc. The car fuel tanks must be emptied of fuel. Three cars are parked side by side at a horizontal distance of 0.6 m, under a suspended ceiling. The fire is started with two trays of heptane that are placed under the middle car and ignited. The tested water mist system must be proven to have an efficiency comparable to that of traditional sprinklers. Therefore, reference fire tests are first performed with traditional sprinklers, which thus constitute the reference system. The system is designed for a water discharge density of 6.5 mm/min, which corresponds to 80 l/min per sprinkler. The test results from assignment tests are usually proprietary, but have been published with permission [
10]. For these tests, the ceiling height was 3.0 m, and the cars were manufactured in the late 1990s. The reference tests with the traditional sprinklers showed good effectiveness. Since the fire is in principle completely hidden from direct water application by the body of the car, the primary effect is in preventing the spread of fire and to reduce the gas temperature at the ceiling. The average gas temperature at the ceiling was at most around 150°C during three independent fire sprinkler tests. The surface temperatures on the part of the body facing the centremost car on both target cars were measured with four thermocouples welded to the steel body; the average surface temperature of the adjacent cars peaked at around 100°C.
Concerns about the performance of deluge water spray systems (often referred to as ‘drencher’ systems) in ro-ro cargo spaces onboard ships have been raised regarding the increased number of battery electric vehicles being transported. A straightforward comparison of the fire suppression performance of a deluge water spray system for fires involving gasoline-fuelled and battery electric vehicles in test conditions as equivalent as possible was undertaken [
11]. It was concluded that fires in the two types of vehicles are different but have similarities. A gasoline fuel spill fire develops very rapidly, peaks high but burns out fast, whilst a fire starting in the battery pack of a battery electric vehicle develops slower, is not as large but burns longer. The development of the fire in other combustibles, such as the tires, exterior and undercarriage plastic parts and inside the passenger compartment is similar. The overall conclusion from the tests was that a fire in an electric vehicle does not seem to be more challenging for the drencher system design given in MSC.1/Circ.1430 [
12] than a fire in a gasoline-fuelled vehicle of comparable size.
2.3 Design Recommendations in Sprinkler Installation Standards
The type of cargo transported on maritime vehicle carriers is broad and can be exemplified by a list of typical cargo by one of the ship operators: agriculture, automotive, boats and yachts, breakbulk, construction equipment, mining equipment, machinery, power equipment, railcars and tramways, trucks, buses, and trailers. It is obvious that the combustibles are to a large extent shielded from direct application of water from ceiling sprinklers or nozzles by the body of vehicles or cargo packaging. The amount of fuel in new vehicles transported on maritime vehicle carriers is limited, for passenger cars typically only around 5 l. The amount of fuel should be sufficient for driving the vehicles on and off the ship. It is, however, not uncommon that used vehicles are transported on maritime vehicle carriers where the amount of fuel could be larger. As the loading and unloading of the vehicles is made by dedicated staff, vehicles are positioned closer together on a maritime vehicle carrier than on ro-ro cargo and ro-ro passenger ships.
Design and installation guidelines for automatic sprinkler (wet-, dry-, or pre-action systems) and deluge water spray systems for open and closed ro-ro spaces and special category spaces is given in MSC.1/Circ. 1430 [
12] as amended in [
13,
14]. The recommendations cover aspects such as the system type, positioning of sprinklers, design densities and operating areas. As discussed, there are differences regarding the type of cargo that is transported in ro-ro and special category spaces on ro-ro cargo and ro-ro passenger ships as compared to closed ro-ro spaces on maritime vehicle carriers that should be recognized. Ro-ro cargo and ro-ro passenger ships are designed to transport freight trucks with cargo trailers or individual cargo trailers. These cargo trailers could contain high fire loads that require high water discharge densities. When the design and installation guidelines in MSC.1/Circ. 1430 were established, tests involving fires in simulated cargo trailers were conducted [
15]. These tests confirmed that controlling a fire in a cargo trailer requires a high water discharge density.
Despite the differences in cargo, the design and installation guidelines in MSC.1/Circ. 1430 were used as the starting point for the work of establishing similar guidelines for closed ro-ro spaces onboard maritime vehicle carriers. Additionally, guidance was sought in standards developed by the National Fire Protection Association (NFPA), the European Committee for Standardization (CEN) and FM Global. These standards cover protection schemes for fire hazards similar to those found on maritime vehicle carriers.
Table
1 shows the minimum required water discharge density and minimum design area given in MSC.1/Circ. 1430/Rev.2.
Table 1
The Minimum Required Water Discharge Densities and Minimum Design Areas Given in MSC.1/Circ. 1430/Rev.2
≤ 2.5 | Wet-pipe | 6.5 | 280 m2 |
Dry-pipe or pre-action | 6.5 | 280 m2 |
Deluge | 5 | 2 × 20 m x B |
˃ 2.5—≤ 6.5 | Wet-pipe | 15 | 280 m2 |
Dry-pipe or pre-action | 15 | 365 m2 |
Deluge | 10 | 2 × 20 m x B |
˃ 6.5—≤ 10 | Wet-pipe | 20 | 280 m2 |
Dry-pipe or pre-action | 20 | 365 m2 |
Deluge | 15 | 2 × 20 m x B |
Automatic sprinklers intended for spaces with a free height equal to or less than 2.5 m should have a nominal operating temperature range between 57°C and 79°C and standard-response characteristics. Standard-response refers to the reaction time of the sprinklers. Automatic sprinklers or nozzles intended for spaces with a free height in excess of 2.5 m should have a nominal operating temperature range between 121°C and 149°C and standard-response characteristics. Deluge systems should be designed for the simultaneous activation of the two adjacent deluge sections with the greatest hydraulic demand at the minimum water discharge density.
The 2022 edition of NFPA 13 [
16] provides the minimum requirements for the design and installation of automatic fire sprinkler systems and exposure protection sprinkler systems. Table
2 shows examples of occupancies that could be considered to have similarities with the fire hazard present in ro-ro spaces on maritime vehicle carriers, and the sprinkler protection criteria in terms of minimum water discharge densities and design areas. It is noted that the 2022 edition of NFPA 13 contains some applicable code changes as compared to the 2019 edition, “Automobile parking garages” is reclassified from Ordinary Hazard Group 1 to Ordinary Hazard Group 2, whilst the classification of “Automobile showrooms” remains Ordinary Hazard Group 1.
Table 2
The Classification of Occupancies Used in NFPA 13 (2022) Together with Occupancy Examples That Are Relevant for the Cargo Transported on Maritime Vehicle Carriers, Characterization of the Hazard and the Sprinkler Protection Criteria
Light Hazard occupancies | None | Spaces with low quantity and combustibility of contents | 4.1 mm/min over 140 m2 or 2.9 mm/min over 280 m2 |
Ordinary Hazard Group 1 (OH1) | Automobile showrooms | Moderate quantity and low combustibility, stockpiles < 2,4 m high | 6.1 mm/min over 140 m2 or 4.9 mm/min over 280 m2 |
Ordinary Hazard Group 2 (OH2) | Automobile parking garages, exterior loading docks, repair garages | Moderate to high quantity and combustibility of contents, stockpiles < 3,7 m | 8.1 mm/min over 140 m2 or 6.9 mm/min over 280 m2 |
Extra Hazard Group 1 (EH1) | Aircraft hangars, upholstering with plastic foams | Very high quantity and combustibility of contents, dust, lint or other similar materials present introducing the probability for rapidly developing fires | 12.2 mm/min over 230 m2 or 11.4 mm/min over 280 m2 |
Extra Hazard Group 2 (EH2) | Manufactured homes or modular building assemblies, car stackers and car lift systems with 2 cars stacked vertically | Very high quantity and combustibility of contents, substantial amounts of flammable or combustible liquids present, shielding of combustibles is extensive | 16.3 mm/min over 230 m2 or 15.5 mm/min over 280 m2 |
The design area is allowed to be reduced under certain conditions, without revising the water discharge density. The area of operation can be reduced by 25% when using high-temperature sprinklers for Extra Hazard Group 1 occupancies, which is relevant for hazards similar to ro-ro spaces. The area is, however, not allowed to be less than 185 m2. High-temperature sprinklers are defined as sprinklers having a nominal activation temperature of between 121°C and 149°C.
Fast-response sprinklers are not allowed for Extra Hazard occupancies or other occupancies where there are substantial amounts of flammable liquids or combustible dusts. The reason for this is the risk that sprinklers located outside the sprinkler operating area could activate during fast growing fires.
For dry-pipe systems, the sprinkler operating area should be increased by 30% without revising the water discharge density. In storage occupancies, NFPA 13 requires that high-temperature rated sprinklers be used for dry-pipe systems in storage occupancies.
EN 128452015+A1:2019 [
17] provides the minimum requirements for the design and installation of automatic sprinkler systems. Fire hazards similar to “Car parks” fall under Ordinary Hazard Group 2 (OH2). For such occupancies, a wet-pipe or pre-action system should be designed for a water discharge density of 5 mm/min and an area of operation of 144 m
2. A dry-pipe system should be designed for an area of operation of 180 m
2. “Car workshops” fall under OH3 where a wet-pipe or pre-action system should be designed for a water discharge density of 5 mm/min and an area of operation of 216 m
2. A dry-pipe system should be designed for an area of operation of 270 m
2. “Depots for buses, un-laden lorries and railway carriages” fall under High Hazard Production Group 2 (HHP2). For such occupancies, a wet-pipe or pre-action system should be designed for a water discharge density of 10 mm/min and an area of operation of 260 m
2. A dry-pipe system should be designed for an area of operation of 325 m
2.
The October 2021 interim revision of FM DS 3–26 [
18] by FM Global provides recommendations for fire protection using automatic sprinkler systems in non-storage occupancies, i.e., an area or building consisting of equipment, processes, and/or materials that are not maintained in a storage arrangement. Three different hazard categories are used in the document, HC-1, HC-2, and HC-3. Several of the occupancies listed in Appendix C of the document are relevant for the fire hazards found in ro-ro spaces on maritime vehicle carriers, refer to Table
3.
Table 3
Examples of Occupancies Listed in FM DS 3–26 (October 2021) That Are Relevant for Closed Ro-Ro Spaces on Maritime Vehicle Carriers
HC-1 | None |
HC-2 | None |
HC-3 | Parking garage Car parks Manufacturing/assembly of wind turbines Manufacturing /assembly of aircraft Manufacturing /assembly of boats, highway trailers, trucks, boxcars, mobile homes, or similar Manufacturing /assembly of cars |
Thermal sensitivity is the measure of how fast the thermal element (glass bulb or fusible link) operates in a standardised test. Based on this time, the Response Time Index (RTI) can be calculated. The lower the RTI of a sprinkler, the faster it activates in a fire. But a low RTI is not necessarily the best option. For dry-pipe sprinkler systems, FM DS 3–26 recommends the use of upright or dry-pendent sprinklers with standard-response characteristics and a nominal 140°C temperature rating for HC-3 occupancies. The reason that standard-response sprinklers should be used is to prevent too many sprinklers from opening before water fills the pipework.
The automatic sprinkler system designs discussed above can be compared by multiplying the recommended minimum discharge density with the design area. Table
4 compares the sprinkler protection criteria in MSC.1/Circ. 1430/Rev.2 for ro-ro spaces having a free height up to and including 2.5 m with the criteria for OH2 occupancies per NFPA 13 (2022) and EN 12845:2015+A1:2019 as well as hazard category HC-3 occupancies per FM DS 3–26 (2021). The height of the space would only allow transportation of passenger cars.
Table 4
A Comparison of the Sprinkler Protection Designs in MSC.1/Circ. 1430/Rev.2 with Those in NFPA 13, EN 12845:2015+A1:2019 and FM DS 3–26 for Spaces Where the Fire Hazard Is Similar to That Presented by Passenger Cars
MSC.1/Circ. 1430/Rev.2 (up 2,5 m) | 6.5 | 280 | 1820 | 6.5 | 280 | 1820 | 10.24 |
NFPA 13, OH2 | 8.1 | 140 | 1134 | 8.1 | 182 | 1474 | 12.1 |
EN 12845:2015, OH2 | 5 | 144 | 720 | 5 | 180 | 900 | 12.0 |
FM DS 3–26, HC-3 (up to 9 m) | 12 | 230 | 2760 | 12 | 330 | 3960 | 11.1 |
The sprinkler protection design requirements are the most stringent in FM DS 3–26 in terms of the total water flow rate, but it is noted that the design is valid for occupancies having a ceiling height up to 9 m. The design recommendations in EN 12845:2015+A1:2019 result in the lowest total water flow rate. However, it can be observed that this design appears to be sufficient according to the sprinkler statistics from fires in car parks in United Kingdom as discussed earlier in the paper.
Table
5 shows a similar comparison for EH2 occupancies per NFPA 13 (2022), HHP2 occupancies per EN 12845:2015+A1:2019 and hazard category HC-3 occupancies per FM DS 3–26 (2021). The data are relevant for spaces with a height that allows transportation of larger vehicles. The design recommendations in MSC.1/Circ. 1430 result in the highest total water flow rates and those in EN 12845:2015+A1:2019 result in the lowest total water flow rates.
Table 5
A Comparison of the Sprinkler Protection Designs in MSC.1/Circ. 1430 with Those for EH1 per NFPA 13, HHP2 Occupancies Per EN 12845:2015+A1:2019 and HC-3 per FM DS 3–26 for Spaces Where the Fire Hazard Is Relevant for Larger Vehicles
MSC.1/Circ. 1430/Rev.2 (6.5 m to 10 m) | 20 | 280 | 5600 | 20 | 365 | 7300 | 10.24 |
NFPA 13, EH2 | 16.3 | 230 | 3749 | 16.3 | 299 | 4874 | 9.3 |
EN 12845:2015, HHP2 | 10 | 260 | 2600 | 10 | 325 | 3250 | 9.0 |
FM DS 3–26, HC-3 (up to 9 m) | 12 | 230 | 2760 | 12 | 330 | 3960 | 11.1 |