article

Assessing fire development in underground metro cars

Posted: 6 May 2011 | Professor Haukur Ingason, Senior Research Scientist, SP Technical Research Institute | No comments yet

In the fire safety design process for underground metro systems, the design fire is usually an issue that requires long discussions and consensus among designers. The main reason is the complexity of fires in metro cars and lack of large scale test results which confirm the design assumptions. This article gives an overview of fire development in underground metro cars and gives an insight into a large scale test series that is planned for autumn 2011 in Sweden.
Increased demand for mass transport

Rapid advances in underground construction technology and increased demand for mass transportation of people force us to build more-and-more complex underground mass transport systems. The fire risks and conse – quences of fires usually become key issues in the design process.

In the fire safety design process for underground metro systems, the design fire is usually an issue that requires long discussions and consensus among designers. The main reason is the complexity of fires in metro cars and lack of large scale test results which confirm the design assumptions. This article gives an overview of fire development in underground metro cars and gives an insight into a large scale test series that is planned for autumn 2011 in Sweden. Increased demand for mass transport Rapid advances in underground construction technology and increased demand for mass transportation of people force us to build more-and-more complex underground mass transport systems. The fire risks and conse - quences of fires usually become key issues in the design process.

In the fire safety design process for underground metro systems, the design fire is usually an issue that requires long discussions and consensus among designers. The main reason is the complexity of fires in metro cars and lack of large scale test results which confirm the design assumptions. This article gives an overview of fire development in underground metro cars and gives an insight into a large scale test series that is planned for autumn 2011 in Sweden.

Increased demand for mass transport

Rapid advances in underground construction technology and increased demand for mass transportation of people force us to build more-and-more complex underground mass transport systems. The fire risks and conse – quences of fires usually become key issues in the design process. As underground transportation systems usually contain large volumes of passengers, it is difficult to provide them with solutions that are valid for all types of potential fire incidents. Furthermore, there is an inherent contradiction built into such fire safety design at present, as the knowledge about fire development in metro cars is limited, while at the same time this represents key information in the design process. Indeed, knowledge of fire development in metro cars is the primary information needed as input for the design of escape routes, detection, ventilation and first responder tactics.

Incidents occur

The level of fire classification of new rail vehicles transporting in underground systems is high. Naturally, this increases the total safety level and minimises the risk for large incidents to occur. The statistics of fire incidents in metro systems is historically good, although there have been several extraordinary incidents, e.g. the King’s Cross fire in 1987 in UK where 31 people were killed, the Baku fire in Azerbajdzjan in 1995 where 289 people were killed and the Daegu fire in South Korea in 2003 where 197 people were killed. There are also numerous other incidents that have occurred but where the consequences have been significantly less serious.

All of the large fires mentioned above had their own specific fire origin, and no incident is like the other. However, each of these fires revealed weak points of their metro systems. The Baku fire highlighted the problems of fires in oldtype metro cars and the effects of ventilation on the smoke and fire development. The King’s Cross fire showed the danger of having combustible materials in the escalators. The Daegu fire showed that the consequences of an arsonist are difficult to deal with and that incident management and wrong decisions can lead to disastrous consequences. These large incidents show the importance of good and safe designs of the materials that exist on the system. Minimising the combustible material existing in the transportation systems minimises the risk and consequences of serious incidents. This awareness of minimising the fire loads in underground systems appears to be well accepted by most stakeholders and actions have been taken in most cases to improve the use of suitable materials.

Requirements today

Today, all railcar and metro car manufacturers aim to build vehicles that meet the requirements of the fire classification standards. When it comes to large underground infrastructure projects, the client or fire authorities ask for a design fire for a specific type of metro car. The client himself often has access to consulting fire experts in the field. The consultant usually proposes a design fire that is assumed to be appropriate for the transportation system where the trains will run. There are no standardised design fires for railcars or metro cars available, but European standards such as CEN TS 45545 are presently under development for trains which should fill this gap to a certain degree.

These standards give the level of fire safety that is needed in order to secure the escape of passengers from a carriage in the case of a fire. They focus on fire resistance, ignitability, fire and flame propagation, smoke density, toxic productions and heat release rates. The standards cover materials such as rubber products, melamine foams, textiles, sponges, flooring, plastic products, aluminum honey – comb plates, fire coatings, fire adhesives and cables and wires. There are requirements such as that the fire should be prevented from spreading from the undercarriage to the passenger compartment or to an adjacent car. The undercarriage contains a large amount of combustible solid materials and liquids that are fire tested but in tunnels or other underground structures the undercarriage becomes important in terms of the smoke, heat produced and the fire development. Therefore it is not negligible to consider it for design and use in underground situations. Finally, the baggage passengers bring with them is something that needs to be considered in the design fire but which is presently not dealt with in the standard. Within the framework of the METRO project presented further on in this article, recently performed tests show that heat release rates from small luggage (for example a backpack or a computer bag etc.) can vary from tens of kilowatts to hundreds of kilowatts. This indicates the importance of this parameter in the design.

Fire tests

Fire tests have been carried out both on railcars and metro cars for decades now. The objectives have been to improve the fire resistance of key objects such as seats and walls, floors or ceiling linings. Most of the tests have focused on the fire behaviour of materials using small-scale tests. In many cases, small-scale tests were followed-up by large-scale tests with parts of railcar compartments. Only a few full-scale experiments with carriages have been performed, mainly because of the costs and the availability of materials. The best known full-scale test series with railcars is the EUREKA 499 tests1 carried out between 1991 and 1992. Different types of carriages, ignition sources, interior materials and ventilation conditions were applied in these tests. Experience from large- and smallscale tests shows that even if the interior materials plays an important role in the fire development, the ventilation conditions are as important, or more important, than the fire load2,3. Small ignition sources such as electrical faults, fires on a seat, or fires under a seat will not cause a large enough fire to result in a flashover of the carriage, especially if the doors are closed.

Design fires

In order to be able to create a design fire for a specific type of carriage, there are several basic pieces of information that are needed. First of all, one should try to find out what type of metro cars will be transported in the system. After that, one should consider whether the design fire is to be obtained from detailed CFD calculations or more robust methods such as full-scale testing or engineering calculations. The creation of a design fire using programmes such as FDS4 can be very tricky and requires a high level of skill for those using the programme. Usually when one creates a design fire using FDS, there is a need for some basic information concerning the interior materials used. Such information can be obtained from cone calorimeter tests and from Thermo Gravimetric Analysis (TGA) tests. These types of methods to obtain a design a fire must be supported by simple calculations such as given by Lattimer and Beyler5 to ensure that reasonable values have been obtained.

A more robust method is to consider the following parameters:

  • Total fire load in GJ inside and under the car, including fuel tanks, hydraulic oil tanks etc.
  • Type of body construction (steel, aluminium or composite material)
  • Size of windows, door openings, and dimensions of the open carriage volume
  • One side with doors open (platform), both sides open (inside tunnel)
  • Cone calorimeter data from interior material such as seats, walls, floors and ceiling linings
  • Amount of baggage carried with the passengers
  • Type of classification of windows, and how well they are fixed
  • Calculate the maximum heat release rate and time to reach maximum.

If the above list of input is accessible, this is sufficient to create a single design fire for a specific carriage. Ingason and Lönnermark6 have summarised the information available concerning the maximum heat release rates from rail and metro cars. The peak heat release rate is found to be between 7 and 43 MW while the time to reach the peak heat release rate varies from 5 to 80 minutes.

If the fire were to spread between the cars, the total heat release rate and the time to peak heat release rate would be much higher than the values given here although one cannot simply add the heat release rate for each coach to obtain an estimate of the total heat release rate. The average heat release rate per unit fuel surface of railcars and metro cars are 0.3 MW/m2 and 0.27 MW/m2, respectively6. This, together with information about combustible material inside the carriage, gives an idea of what the maximum heat release rate might be if the carriage is constructed in material that can make large openings in the body of the carriage.

If the fire is ventilation controlled, i.e. there is a flashover in the carriage (steel body); the maximum heat release rate is controlled by the size of the openings. This may be the door opening and the window opening. Usually, the maximum heat release rate is controlled by the outer boundaries, i.e. the openings that may be created, see Ingason for further information7. This however, requires sufficient combustible interior material. This is why it is so important to have fire requirements on the interior material. At the same time, however, designers must be aware of the fact that the interior material choice is not the only parameter that controls the fire development. Other parameters include the ignition source, baggage in the carriage and the ventilation conditions. In order to find out more about fire behaviour of metro cars, a nationally funded project called the METRO project has been started in Sweden. This project will consider the concept of design fires given above, among other things.

The METRO project

The METRO project is a Swedish multidisciplinary project with a unique combination of research organisations and practitioners. The project is ongoing between 2010 and 2012, and is divided into seven work packages with a total of nine participating organisations including: Mälardalen University; SP Technical Research Institute of Sweden; Lund University; the Swedish National Defense College – Crismart; the University of Gävle; the Swedish Defense Research Agency; the Greater-Stockholm Fire Brigade; the Swedish Fortifications Agency and the Stockholm Public Transport.

In September 2011, full-scale tests are planned in the Brunsberg tunnel in Sweden – a 276m-long decommissioned train tunnel. The project has access to three metro carriages, where one will be used for ignition tests, i.e. prior to a fully developed fire. The sources and place of ignition will be varied and investigated. Wood cribs and standardised ignition sources inside the carriage will be used as well as cable fires and flammable liquid. The second carriage will be tested using larger ignition sources, forcing the fires to develop to a flashover situation. One of the main results from these tests will be seminal knowledge concerning the influence of ventilation on the fire development in the carriage and which ignition sources have the potential to cause a fully developed fire. The final carriage will be tested using a combination of explosive material and flammable liquid.

In all cases, a number of different parameters will be measured and analysed including for example, heat release rate, radiation, gas temperatures, gas velocity and visibility. The final tests will also provide information about pressure duration, stagnation pressure and of course how the structure of the metro carriage is affected by an explosion. Structural damage will certainly influence both the possibilities for evacuation and the fire and rescue operations. After the tests, this can be studied in detail and provide valuable input to the various work packages in the METRO project.

In connection to the full-scale tests, there will be possibilities for qualified observers to visit the test site. More information will be presented at the project web site www.metroproject.se.

References

1. “Fires in Transport Tunnels: Report on Full-Scale Tests”, edited by Studiensgesellschaft Stahlanwendung e. V., EUREKA-Project EU499:FIRETUN, Düsseldorf, Germany, 1995.

2. White, N., “Fire Development in Passenger Trains – PhD Thesis”, Victoria University, Australia, 2009.

3. Ingason, H., “Model Scale Railcar Fire Tests”, Fire Safety Journal, 42, 4, 271-282, 2007.

4. McGrattan, K., Klein, B., Hostikka, S., and Floyd, J., “Fire Dynamics Simulator (Version 5) User Guide”, National Institute of Standards and Technology, NIST Special Publication 1019-5, USA, 2008.

5. Lattimer, B. Y., and Beyler, C. L., “Predicting Heat Release Rates of Railcars”, Fire and Materials 2005, 267-276, Fisherman’s Wharf, San Francisco, USA, 31 Jan. – 1 Feb., 2005.

6. Ingason, H., and Lönnermark, A., “Heat Release Rates in Tunnel Fires “. In The Handbook of Tunnel Fire Safety, 2nd edition (A. Beard and R. Carvel, Eds.), Thomas Telford Publishing, 2011.

7. Ingason, H., “Fire Dynamics in Tunnels”. In The Handbook of Tunnel Fire Safety (R. O. Carvel and A. N. Beard, Eds.), Thomas Telford Publishing, 231-266, London, 2005.

About the Author

Professor Haukur Ingason has been a Senior Research Scientist at SP Technical Research Institute since 1988. He has also been a PhD supervisor at Mälardalen University in the field of underground fire safety engineering since 2006.