Design Fires for Use in Fire Safety Engineering 9781848061521

Citation preview

DESIGN FIRES FOR USE IN FIRE SAFETY ENGINEERING Christopher Mayfield and Danny Hopkin

I

•

bre press

bretrust

DESIGN FIRES FOR USE IN FIRE SAFETY ENGINEERING Christopher Mayfield and Danny Hopkin

bre press

· bretrust

DESIGN FIRES FOR USE IN FIRE SAFETY ENGINEERING

This work has been funded by BRE Trust. Any views expressed are not necessarily those of BRE Trust. While every effort is made to ensure the accuracy and quality of information and guidance when it is first published, BRE Trust can take no responsibility for the subsequent use of this information, nor for any errors or omissions it may contain. The mission of BRE Trust is 'Through education and research to promote and support excellence and innovation in the built environment for the benefit of all'. Through its research programmes the Trust aims to achieve: • a higher quality built environment • built facilities that offer improved functionality and value for money • a more efficient and sustainable construction sector, with • a higher level of innovative practice. A further aim of BRE Trust is to stimulate debate on challenges and opportunities in the built environment. BRE Trust is a company limited by guarantee, registered in England and Wales (no. 3282856) and registered as a charity in England (no; 1092193) and in Scotland (no. SC039320). Registered Office: Bucknalls Lane, Garston, Watford, Herts WD25 9XX BRE Trust Garston, Watford WD25 9XX Tel: 01923 664743 Email: [email protected] www.bretrust.org.uk BRE Trust and BRE publications are available from www.brebookshop.com or IHS BRE Press Willoughby Road Bracknell RG12 8FB Tel: 01344 328038 Fax: 01 344 328005 Email: [email protected] Requests to copy any part of this publication should be made to the publisher: IHS BRE Press Garston, Watford WD25 9XX Tel: 01923 664761 Email : [email protected] Printed on paper sourced from responsibly managed forests FB 29

© Copyright BRE 2011 First published 2011 ISBN 978-1-84806-152-1

CONTENTS

CONTENTS 1 INTRODUCTION 2

IMPORTANT FACTORS INFLUENCING QUANTIFICATION OF A DESIGN FIRE

1 2

3 STEADY-STATE DESIGN FIRES

5

4 TIME-DEPENDENT DESIGN FIRES

7

4 .1 t-sq uared fi re growth curves

7

4 .2 Other calcu lation methods

8

4.3 Experimental design fires

4.3. 1 Heat release rate 4.3.2 Smoke production rate

5 FULLY DEVELOPED FIRES 5 .1 Parametric design fi res

8 9

9

10

5.2 Time-equivalence

10 11

EXPERIMENTAL FIRE DATA

13

6 OCCUPANCIES

14

6.1

Bar/Nightclub

14

6.2

Car parks

16

6.3

Carpet store

18

6.4

Clothes store

20

6.5

Indoor play area

22

6.6

Library

24

6.7

Livi ng room

26

6.8

Luggage store

28

6.9

Office

30

6.10

Prison ce li

32

6.11

Rèception

34

6.12

Retail star.e

36

6.13

Video store

37

7 COMMODITIES

39

7.1

Beds

39

7.2

Boxes

40

7.3

Buses

42

7.4

Cars

44

7.5

Chairs

46 48

7.6

Christmas trees

7.7

50

7.8

Computers Curtains

7.9

Fl ight luggage

52

7.10

Hand cart

53

51

Cont'd ...

-

DESIGN FIRES FOR USE IN FIRE SAFETY ENGINEERING

7 COMMODITIES {CONT'D) 7.11 Pallets 7.12 Pool fires 7.1 3 Soft toys 7. 14 Televisions 7.1 5 Upholstered furniture 7.1 6 Wardrobe

8 REFERENCES

55 57 60 62 63 64

66

1 INTRODUCTION

1

INTRODUCTION

The objective of this publication is to provide technical data and guidance for defining a robust, appropriate and acceptable design fire for the fire safety engineering design of a building. lt explains: • what a design fire is • how it can be determined • its limitations • the experimental data (where available) • current calculation methods used for defining a design fire. Depending on the geographical location of a building, its legislative fire safety requirements may be achieved in a number of ways. Fire safety engineering is a generally accepted approach for demonstrating that the legislative fire safety requirements of a design have been achieved. A building design which is supported by a performance-based fire safety engineered solution comprises a number of components. One of these criticai components is the selection of an appropriate and relevant design fire. In a performance-based fire safety engineered solution the design fire will determine a number of important parameters fora given space, which include: • the quantity of heat released • the quantity of smoke produced • the composition of the smoke • the fire size • the temperature of a smoke layer • the time to involvement of ali exposed combustible materials • the fire duration. Based on the values determined for the parameters, a fire engineered analysis can establish: • if predetermined tenability criteria are exceeded • if further fire protection measures are required (eg a smoke contrai system) • the specification of such fire protection measures.

in unnecessary expenditure), but nevertheless capable of meeting the life safety requirements to avoid potentially life-threatening omissions. There are a number of different approaches to defining an appropriate design fire ranging from calculation based on fuel load surveys of real buildings and quantification to experimental determination. These different approaches will be described in detail. This publication is aimed at those professionals involved in the fire safety engineering design process, either as a designer fulfilling a brief or a regulator/ approver of the design. lt is intended that this publication will provide evidence to assist the review of the foundation of the fire engineered solution as part of any approvai process. More generally, those in the position of the responsible person, as defined by the Regulatory Reform (Fire Safety Order) 2005 111 (FSO), or those delegateci as competent under the FSO by the responsible person, are likely to find this resource beneficiai when undertaking a fire safety risk assessment in both fire safety engineered and non-fire safety engineered buildings. That is, it is important that the responsible person understands the design principles of his or her buildings so that he or she can ensure that they are managed on an ongoing basis, within their design limits. Specifically, the fire load is restricted to within the limits of the assumed design fire. This is particularly important where, for example, a change of use or change of ownership might occur. Fire safety engineering design requires the identification of an appropriate fire size on which a design can be basedl 2,31. This is one of the key decisions in fire safety engineering design and requires formulation of a quantitative description of the fire. Published reliable data is scarce and fire safety engineers often resort to a simple generic description based on assumption. A summary of the most commonly used parameters in fire safety engineering are detailed as part of the summary of each of the experimental fires where available and include the following parameters:

Clearly, there is great significance associated with the selection by the fire safety engineer of an appropriate design fire to ensure it is representative of the situation considered to fulfil the life safety requirements. In addition to this, it is important to determine if the fire

• HRR

safety measures proposed by a fire safety engineered solution are proportionate (ie not overly onerous, resulting

• carbon monoxide concentration.

• • • •

heat of combustion mass of fire load optical density carbon dioxide concentration

DESIGN FIRES FOR USE IN FIRE SAFETY ENGINEERING

2

IMPORTANT FACTORS INFLUENCING QUANTIFICATION OF A DESIGN FIRE Box 1: Factors influencing the characteristics of a fire

A design fire is a simplffied approxlmation of a fire that ls considered to be representatlve of a fire lnvolvlng a specified hazard.

• Building design • Building fabric • Environmental influences • Potential ignition sources and locations • Types of combustible materials

For a design fire to be representative of a realistic fire situation, it should replicate the physical size and heat output of the fire changing with time. This allows the growing threat to occupants, property/business continuity, and firefighters to be calculated as time progresses141. Time-based calculations often form part of a holistic assessment, where the time to the onset of untenable conditions is compared with the time required for the safe evacuation of occupants of the building or the time recommended for the initiation of successful firefighting operations. To undertake a time-based assessment, fire growth curves need to be selected that are applicable to the exact circumstances of the building occupancies, fuel arrangements and suppression system performance, where appropriate'41. Where this information is available it can be integrateci into a fire engineered solution fora building following recommended fire safety engineering procedures, such as BSl's PD 79 7415 1 series of documents. Severa! important factors which influence the characteristics of a fire are summarised in Box 1 and described in the sections below. They are well documented and should be considered in the quantification of a design fire.

Building design The geometry and layout of the building should be clearly understood and defined.

Building fabric Combustible construction elements and insulation types, presence of vapour barriers in roofs, and types of roof, floor and wall constructions should be identified.

Environmental influences Internal environmental conditions should be defined in terms of ambient temperatures, oxygen concentration, air movement, presence of HVAC systems and how these could impact on pre-fire and fire conditions

• Distribution, arrangement and quantity of different fuels • Ventilation conditions • Possible interventions during the fire

Potential ignition sources and locations A hazard assessment covering both accidental and deliberate ignition events and the probabilities of each should be conducted and potential ignition sources and locations identified. Deliberate ignition can be characterised by multiple ignition sources. Depending on the hazards identified, the probability of them occurring and the resulting consequences, a design may involve fire protection measures to significantly reduce or eliminate either the hazard, probability and/or consequences of an ignition event. However, it is recognised that this may not always be achievable.

Types of combustible materials The type of occupancy and the use of the building will govern the types of combustible materials. The combustible materials may be present as construction products incorporateci into the fabric of the building or the contents of the building, or both . During the early stages of a fire, the contents will typically be of primary concern.

Distribution, arrangement and quantity of different fuels The way that the different fuels are distributed within the building will have an effect on the characteristics of the fire so ali readily available fuel should be considered including: • wall and ceiling linings • low-level localised storage of goods • high-rack, high-density storage of goods • distribution of furniture and types of furniture (eg office, residential) • materials being processed.

2 IMPORTANT FACTORS INFLUENCING QUANTIFICATION OF A DESIGN FIRE Table 1: Fire load densities for different occupancies

where:

Extract from PD 7974-1 1· 1

Occupancy

(MJ/m 2 ) Dwelling Hospital Hospital storage Hotel bedroom Offices Shops Manufacturing Manufacturi ng

t.Hc

(MJ/m 2 ) 80%

90%

95%

780 230 2000 310 420 600 300 1180

870 350 3000 400 570 900 470 1800

920 440 3700 460 670 1100 590 2240

970 520 4400 510 760 1300 720 2690

1500 285

2250 360

2550 410

and storage Libraries Schools

total heat release (kJ) tota! mass loss (kg) heat of combustion of fuel (kJ/kg)

QTOTAL mTOTAL

Fire load density Fractile Average

450

They will ali result in different distributions of fuel load per square metre of floor area which is common ly referred to as the fire load density. Traditionally, hazard assessment focuses on the building contents as the primary available fuel source(s). However, with new developments in the design of buildings leading to the use of more combustible construction, it might be necessary to consider the construction materials in the hazard assessment. This is occurring as a result of a concerted effort to use renewable materials (often cellulose-based) and highly thermally efficient materials (synthetic polymer-based). In the majority of cases, a building's fabric will be controlled by building regulation requirements (depending on the geographical location) and to the same extent any combustible elements of construction shou ld be fire-separated from the areas containing the fire hazards within a building, to prevent them from becoming involved in fire. Fire load densities for different types of occupancies are provided in Table 1 taken from PD 7974-1 151. However, caution should be exercised when using these data as they are derived from a series of fire load surveys that were carried out before 1983 and presented in a CIB W14 workshop report. For some types of occupancy in particular, advances in technology and changes in design philosophy have resu lted in significant changes, for examp le the increased use of IT and the move to openpian accommodation in offices and dwellings. Clearly, for goods stored in rack ing (eg in warehouses), the fire load per square metre of floor area could significantly exceed the va lue in Table 1. Given the data in Table 1, the maximum heat that can be rel eased from a fire, assuming that it is fu lly venti lateci will be the fire load density (MJ/m 2) x the floor area (m 2). lf the fire load density is not known or there is a mixture of uneven ly distributed fue ls within the enclosure, then it is possible to calculate the maximum HRR (assuming that all of the fuel is consumed) using either eq uations 1 or 2 below, as appropriate. (Eqn 1)

lf the fuel load is mixed, eq uation 2 can be applied as follows:

where i

= fuel

type.

Fire load density is the fire load per unit area. This is obtained by dividing QTOTAL from equation 1 or 2 by the floor area of the enclosure of fire origin. The distribution and orientation of any fire load is important in determining the rate of fire growth. Combustible materiai orientateci vertically can yield a rapid rate of fire growth as a result of the increased potential for direct flame impingement on unburnt materiai lead ing to increased levels of radi ative and convective heat transfer.

Ventilation conditions The ventilation of an enclosure within a building is of significance in determining the HRR of a fire. Assuming an enclosure of fire origin with defined dimensions and fire load . lf there is sufficient venti lation to the space within which the fire is initiated, without intervention, the fire can grow until it has consumed ali of the avai lable fue l (Figure 1 ). lf ventilation of the same enclosure with an identica! fire load is restri cted, the fire growth and heat release wi ll become limited by the available oxygen and the mass of fuel consumed by the fire wil l also be limited (Figure 2).

2

I

~

1

Q)

Fuel-controlled

I 1

"'"' Q)

~

iii Q) I

1lgnition

Time

Figure 1: Fuel-controlled fire

2

~

Q)

"'"'

/

Q)

~------ ....

\~

Fuel-controlled case

~

ro

"""'Ventilation-controlled ' , ,

Q)

I

1lgnition

.........

__ Extinction

Time

Figure 2: Ventilati on-controlled fire

-

DESIGN FIRES FOR USE IN FIRE SAFETY ENGINEER_IN_G --~----------------""

Both of the scenarios described above assume that the ventilation conditions within the enclosure of fire origin remain unchanged as a function of time. While it is important to consider the availability of oxygen in the air, it is also important to be aware of circumstances where the thermal decomposition of certain materials produces oxygen (an oxidant) in addition to that already present in the air. This could result in a combustion reaction being supported in the absence of air (or at reduced oxygen leve!). In addition, the designated function of the building may also introduce additional sources of oxygen. For example, piped oxygen and/or cylinders of compressed oxygen are found commonly across a hospital site to satisfy the ongoing needs of patients. Such circumstances are not likely to be encountered widely, but every effort should be made as part of the fire safety engineering design process to take account of the presence of oxidising materials.

Possible interventions during the fire This is basically an event or series of events that can occur during the course of a fire and alter the fire's characteristics. Such interventions could include: • breaking of windows to increase ventilation • opening of doors by occupants during evacuation or by the fire service • operation of HVAC system • operation of fire dampers • operation of smoke contrai ventilation system • operation of suppression system. AII these interventions could influence the growth of a fire and should be considered in terms of the overall characterisation and definition of a design fire. Particular care is necessary since what represents a worst case for some interventions might be best case for others.

3 STEADY-STATE DESIGN FIRES The simplest approach to specifying a design fire is based on assessing the largest size a fire is reasonably likely to reach in the situation being considered141, This is then used as the basis far the design of the fire protection systems and assumes that the HRR continues indefinitely. Figure 3 shows an example of a plot of a generic HRR versus time representation of a steady-state design fire. Steady-state design fires have historically been employed as a basis far design, as there has previously been an acknowledged lack of data in relation to time-dependent design fires121. Steady-state design fires have been used in the design of smoke control systems in a range of bui lding · occupancies, but they were initially app lied to the design of smoke control systems in enclosed and partially enclosed shopping centres. Research into smoke control in enclosed and partially enclosed shopping centres and the subsequent guidance produced1 21 used the methodology of selecting a fixed fire size. By basing the design of the smoke control system on the maximum likely fire size means that, as stateci previously, any fire size up to the maximum fire wi ll be managed by the system and a degree of flexibility can be retained . For example, even if the occupancy changes, the designer may be able to demonstrate that the maximum likely fire size is stil i within the bounds of the originai smoke control system design. lf, in the design of the fire safety precautions far a

I.s f!

GI

2500

-•-•

building, it is determined between the approver and the designer that a steady-state design fire is most appropriate far the situation being considered, then the size of the steady-state fire must be defined based on a credible fire area dependent on the fuel sources likely to be present. The basis of the design fire must be clearly explained so that its limitations are easily understood. lf there is a design fire w ithin this publication that is representative of the design scenario, the fixed fire size far design wi ll need to be defined based on consideration of the fire growth curve. Selection of a peak HRR far a steady-state design may represent an onerous design condition, but on ly in re lation to the fue l load far which the data are ava ilable . lf the design fire (transientor steady state) is going to be employed to determine the specification of a smoke control system, consideration shou ld be give n to the proportion of the total heat re leased which is convection . A number of the design fires presented in this publication have had the fraction of the total heat re leased as convection plotted or stateci. In cases where this has not been possible, it can be determined by the app lication of an appropriate correction factor based on the type of fuel load to determine the convected fraction. lt is the convective portion of the total heat re leased which re lates to the entrainment of air into the ris ing piume above the fire, thus the rate of smoke being produced, which is normally expressed in terms of mass/

-·

2000

V)

CO GI

e ivGI

1500

:e

1000 500

0 - - - - - - - - - - - - - - ; - - - - - + - - - ~- ---;,------1 600 800 1000 1200 1400 o 200 400 Time (s)

Figure 3: Example of a generi c steady-state design fire

DESIGN FIRES FOR USE IN FIRE SAFETY ENGINEERING Table 2: Examples of the fraction of the total heat released which is transferred as convection for a range of different fuel types

Materiai Ethano l Kerosene Benzene Octane Silicone Poly(methyl methacrylate) (PMMA) Douglas fir Polystyrene Polyurethane

Convective fraction 0 .74 0.65 0.40 0 .67

0.84 0.69 0 .62 0.41 0.42

volume per unit of time. Values for the proportion of the total heat released that is released as convection (convective factor) are given in the literature12 • 4 • 61. The values given in Table 2 are taken from reference [6] and are a se lection of the convective fraction values which could be applied by the designer. The remaining fraction of the total heat released which is not transferred via convection is mainly transferred via

rad iation. Radiative heat transfer has little bearing on the rate of smoke production, hence it is discounted in the design of smoke contro! systems. However, radiative heat transfer is relevant to the overall fire growth characteri stics as it is a significant factor in fl ame spread, fire spread, time-temperature analysis and also in the calculation of the temperature of the structure. Steady-state design fires are a conservative representation of a realistic fire scenario fora particular set of circumstances . However, when implemented in a transient analysis, the results may be unconservati ve in the early phases, particularly in rel ation to detection time. As such, their application in design may result in the overspecification of fire safety provisions at cost to a project. lt is probable that as the database ~f time-dependent fire cu rves increases and broadens in vari ety, together w ith adva nces in fire modell ing, the steady-state fire as a design tool may become restricted. However, calculations based on the use of steady-state design fires provide a relatively sim plistic method of giving an initial quantitative evaluation of the impact a particular fire may have on a particular set of circumstances.

------------------------4 TIME-DEPENDENT DESIGN FIRES -

4

TIME-DEPENDENT DESIGN FIRES

This type of design fire provides a closer approximation to a real fire than a steady-state design fire. lt is generally accepted that there are five stages characteristic of the development of a fire: • ignition • pre-flashover (growth) • flashover • fully developed • decay.

However, despite these uncertainties, for fully venti lateci fires, it has been found that the rate of development approximates to a parabolic growth (a t-squared growth) following an initial incubation periodl5 • 7 • 8 1. Thus: (Eqn 3) where:

Q

aI Where the fully developed and decay stages occur during the post-flashover phase of the fire, these can be · represented schematically as given in Figure 4. After ignition of the first item(s) has occurred, the speed of fire growth is dependent on the transfer of heat to adjacent combustible materials and products and the ease with which they are ignited. As a consequence, fire growth rates will vary significantly. The following sections explain how these may be approximated.

4.1

t-SQUARED FIRE GROWTH CURVES

As has already been discussed, the growth phase of a fire is sensitive to many variables, such as the: • distribution of combustible materiai within a space • properties of the materials • configuration of the materials.

lncipient phase

Growth phase

Flashover

t t0

The coefficient a I appears to lie in the range of 10-3 kW/s2 for slowly developing fires to 1 kW/s2 for rapid fire growthl71_ Four standard fire curves have been defined and validatedl 91; the values of the coefficients are set out in Table 3. The t-squared curve is quadratic without any limit; it does not have a steady state and decay period . When the coefficients a I presented in Table 3 are plotted for each of the four fire growth coefficients (slow, medium, fast and ultrafast) they appear as shown in Figure 5. These fire growth coefficients are derived from the time taken to reach the HRR of 1 MW. The fire growth coefficient is a valici way of predicting the likely rate at which a fire will develop to its maximum HRR for horizontally spread ing fires, but alone, it does not provide the user with the maximum heat release value. Users shou ld be aware that the fire growth coefficient

Fully developed phase

lgnition Time

Figure 4: Phases of fire development

HRR at any time (kW) fire growth coefficient (kW/s2) time (s) incubation time of fire (s)

Decay phase

Extinction

DESIGN FIRES FOR USE IN FIRE SAFETY ENGINEERING Table 3: Th e four standard fire growth coeffici ents. Data from Chitty & Fraser-Mitchelll'' Fire growth coefficient

Time to reach 1MW (s)

Coefficient

Slow Mediu m Fast Ultra fast

600 300 150 75

0.00293 0.011 72 0.0469 0.1876

1

Fire scenario

a.(

Densely packed paper Traditional mattress or armchai r PU mattress or PE pallets High rack storage

PU = polyu rethane, PE = polyethylene.

:::-r-----,--,----r-.--..-..-_----.-._-l --r------.1~,. .-.--._-_.,I-. . . .-..--.

-I_-_-_----,-----.

l

§" ~

1.2 .! E 1.0

5: :3 o.a °E ~

!

.j-.. .

··r··-·- -- -·--i--···-··

1.6 -·

1.4

J. I

0.6 -

··f .. I

l·~ ___L

i

·-l-···

-+- - -·

+-..

- l - - --

I

j

0.4

-

: :: ~ o

..

==;.::::;,.=;:::::;:::::::.._~ ~ ~ ~ ~·-.;.1_·· 2

4

·~·+:-·~~,--+-~~__j

~-··~·· ·. .;f. ~·~~--l-l~ --~-··_··_·

6

I-

-

Ultrafast -

8 Time (mi n)

Fast -

10 Medium -

12

14

16

Slow l

Figure 5: The t-squared fire growth curves

alone does not give an assessment of the likely duration of the fire as this is dependent on the amount of fu el available and the ventilation conditions in the vicin ity of th e fire. Wh ere this approach is used as part of th e process to develop a th eoretical fire curve, it is necessary to carry out further calculati ons (eg the size of th e fire at the onset of flashover) to set an upper limit to wh at wo uld oth erwise be an infin ite fire growth121. Th e peak HRR values and th e tim e associateci to reach the peak HRR, shown in chapters 6 and 7, have allowed a fire growth coeffi cient to be determin ed, wh ere possible, fa r the experim ental fires detailed in this pu bli cati on.

4.2

OTHER CALCULATION METHODS

There are a number of different meth ods far calculating th e HRR in an enclosure fi re, most of wh ich are based on experim ental data. As such, th ey could be described as semi-empirica! or, in some cases, fully empirica!. Due to this, th ey are all limited to th e cond iti ons relati ng to the experim ental design, in parti cular, in relation to the venti lati on conditi ons and heat transfer characteristi cs. M ore infarmation can be faund in PD 7974-1151.

4.3

EXPERIMENTAL DESIGN FIRES

In some cases, th e fire perfo rm ance of a parti cular fu el arrangement may not be kn own or q uantified. In such cases, it may be necessary and advisable to measure the parameters far the particular fu el arrangement and to use th em as the basis fa r the design fire description within a fire safety engineering design. Such an experim ent should be carefu lly designed in discussion w ith th e end user of th e data to ensure th at it is representative and appro priate. lt may also be advisable to consu lt with th e approva ls authority du ring th e planning stages to ensure that all concerns are adequately addressed. A typical checklist of factors th at wi ll be considered w ill include th ose listed in Box 2. A large-scale experiment carri ed out under an appro priately sized calori meter will produce data on parameters such as HRR, rate of production of smoke and chemical species such as CO, C0 2 (and others as speci fi ed) as a function of time. Additi onal localised measurements can also be made of parameters including te mperatures, velocities, heat transfer, chemical species concentrations and optical density. lt is important to note that any parameters determi ned experim entally are likely to be specifi c to th e test configurati on and apparatus.

4

TIME-DEPENDENT DESIGN FIRES -

Box 2: Typical checklist of factors to consider when designing an experimental fire • ldentification of types of combustible materials • Mass of each type of combustible materiai • Probability of involvement of each combustible materiai in fire • Type of fire load • Distribution of fire load • Calculation of fire load density • lgnition scenario • Param eters to be measured • Ventilation conditions, whether an enclosure is required, etc. • Large-scale calorimetry (such as th e 1O MW calori meter in BRE Global Burn Hall) will be required to measure th e parameters, heat rel ease rate, rate of smoke produced, rate of produ ction of chemical species (eg CO, CO, ... ). There are only a few large-scale calorimeters available worldwide

Figure 6: Office fire during early phases of development

• Estimation of maximum heat release and assessment of wheth er th e laboratory facility has the necessary capacity to carry out th e experiment safely (if not, re-visit th e fuel load or consider mitigation measures, or both) • Suppression system to be included or not Fi re-fighti ng provisions

4.3.1

Heat release rate

The total heat release rate

(HRR) of the gases produced

from a fire are measured by monitoring the amount of oxygen th at has been removed from a given mass of gas under a large calori meter hood. Th e method requires all

Figure 7: Fu lly developed luggage store fire

the gaseous products to be collected by a large extraction hood and duct system in which the required properties of the combustion gases are subsequently monitored . The oxygen concentration is measured as a function of time using a paramagnetic type analyser, while the concentrations of C0 2 and CO are typically measured using infrared gas analysers. The tota! heat release can also be measured using the mass loss rate of fuel. This is determined by using a load celi arrangement on which th e fu el is mounted, typically on a metal fram ework on which four load cells are mounted at eaçh corner. The load cells used are essentially strain gauges which generate a voltage according to load.

4.3.2

Smoke production rate

The tota! smoke production rate is measured within the collection hood and duct of the calori meter. lt is typically determined by m easuring the obscuration of light across th e duct. For more details, see ISO 24473 1101.

Figure 8: Deve loping fire in an indoor pl ay area

DESIGN FIRES FOR USE IN FIRE SAFETY ENGINEERING

S FULLY DEVELOPED FIRES Nominai or standard fire curves are the simplest representatio n of fu lly deve loped post-flashover fire for design applications. Such fires are normal ly adopted in the performance-based design of fire exposed structures (structural fi re engineering). They include the standard (ISO 8341 11 1) fire curve which is used in the deri vation of fire resistancel 12 , 131as we ll as more severe rep resentations of fire behaviou r such as the hydrocarbon fire. Less

Standard (ISO) fire curve:

E\ =

20

+ 34Slog10 (8t + 1)

(Eqn 4)

External fire curve : 0 = 660(1 - 0.687e-0· 32 ' - 0 .313e-3·8') 8

+ 20

Hydrocarbon fire curve: 0 = 1080(1 - 0.675e-0· 167' - 0.67Se-2·5') 8

severe representations also exist such as the external fire curve (used for elements of a structure outside the bui ld ing envelope) 11 41_ Nom inai fire curves essential ly assume that flashover occurs instantaneously at the point of ignition, the fire then continues to grow indefinitely with no cooli ng. Three fire curves taken from Eurocode 1, Part 21151are shown below in Figure 9. The use of the standard fi re (ISO 834 1131 ) curve as a design fire

+

20

(Eqn S)

(Eqn 6)

where: 0 = the co mpartment gas temperatu re (°C) 8 t = time from ign ition (min).

S.1

PARAMETRI( DESIGN FIRES

means that the performance of the structure can be directly benchmarked against regu latory fire resistance requirements. A lthough more rea listic representations of a compartment fire exist, ie parametric design fires (discussed below), the standard fire curve is stili large ly used in the performance-based design of timber

The parametric design fire curve, as specified in Eurocode 1 Part 21151 , is a step forward in complexity from nominai fire curves and re lates the time- temperature response of a compartment to the available venti lation, fire load density and thermal characteristics of the compartment boundary. Sim ilar to th e nom inai fi re

structures. The corresponding equations for each curve, taken from Eurocode 1, Part 21 1511 are shown below for

curves, the parametric approach is a post-flashover model assuming instantaneous fl ashover at ign ition and is almost exclusively adopted by structural fire engineers undertaking performance-based designs. The appli cation of parametric design fi res is largely li mited to

completeness:

1000

~

800

e

:,

~Cl) Cl.

600

E Cl) I-

400

200

o

5

10

15

20

25

30

40

35

Tlme(mln)

I - - ISO 834 Figure 9: Nominai fire curves

- - Hyd rocarbon - - Externa l

I

45

50

5 FULLY DEVELOPED FIRES

relatively small compartments and becomes inaccurate for large open-pian spaces typical in many multistorey structures. Unlike the nominai fire curves, the parametric approach includes a cooling phase and is often used to design structures to survive compartment burnout. The parametric curve is given as:

0g = 1325(1 - Q.324e·0 ·21 '

-

0.204e·l7t'

(Eqn 7)

- 0.472e·19'') where: 0 g = temperature in the fire compartment t* t.r

t

r b

o Av h A, p c A

time [O/bF/(0.04/1160) 2 -V(pcA) and shou ld li e between 1000 and 2000 opening factor (A)h/A,) area of vertical openings height of vertical openings tota I area of enclosu re density of boundary enclosure specific heat of bou ndary of enclosu re thermal conductivity of bou ndary

(OC)

(h) (h) (d imension less) U/ m2sV'K) (m V') (m2) (m) (m2) (kg/m 3 ) U/kgK) (W/mK)

, The concept of parametric t ime (t*) is used to modify the predicted time- temperature relationship. The background theory to this calcu lation approach was developed by Wickstroml161 . The valu es 0.04 and 1160 re late to the open ing factor and the thermal inertia of the standard fire compartment as used in the originai test programme. The cooling phase of the time-temperature response is assumed to be linear and is dependent on the parametric

time at which peak compartment temperatures are · reached. An example of the parametric approach is shown in Figure 1 O.

s.2

Time-equivalence

The concept of time -equivalence is used to relate the severity of real fires to the time-temperature relationship in a standard fire res istance test115• 161 • Figure 11 illustrates the concept of time-equivalence, relating the actual maximum temperature of a structural member (ie beams and columns) from an anticipateci fire severity, to the time taken for the same member to attain the same temperature when subjected to the standard fire. Generally, time-equivalence can either be determined by using a simple equation or taken from experimental data from natural and standard fire resistance tests. Although si mple to use, the time-equivalence is a crude approximate method of modell ing real fire behaviour eal fire behaviour. In and bears little relationship with r_ addition, the limitations of the method should be clearly understood. The main lim itation is that the method is on ly applicable to the types of members used in the derivation of the adopted formu lae . The method is most applicable to unprotected steel frame structures although modification factors exist for concrete and protected steel frames. The most common ly used form of the timeequ ivalence method adopted today is that of Eurcode 1 Part 21 151 w hich is shown below for comp leteness: The equivalent time of fire exposure, te,d' is calculated using: (Eq n 8) where:

q1d = design fire load density

(MJ/m 2)

1000

800

o

~ 600

I!! ::,

f

8.

!

400

200

0+-~---~---~---~~--~~- -~~--~----~---~---~0.00

0.50

Figure 10: Example of a parametric design fire

1.00

1.50 Time (hours)

2.00

2.50

3.00

DESIGN FIRES FOR USE IN FIRE SAFETY ENGINEERING

-----.,-\-~----=- -------=--'. . : -------- "'

" ,.- ---~ ....~ ~

,,: , "

,,,,,,

.............

Standard fire

..........

Member temperature

~?

--

I

/1 ,,,, ~-.,"

--

',,,,

--I:/ ---...

, ...

Realfire

-

:;" I

- - - ' - - - - Equivalent time

te,d

Time

Figure 11 : The concept of tim e-equivalence

kb

wr

conversion factor dependent on thermal properties of compartment boundaries (min .m 2/MJ), typically taken as 0.09 in the UK (as per the National Annex to BS EN 1991-1-21171)

= ventilation factor (dimensionless)

av

=

H ah

= the height of the compartm ent (m) = Ah/Af, where Ah is the area of horizontal

bv

= 12.5(1 + 10av - a } ) .;::,: 10

Av/Af, where Av and Af are the area of the vertical ventilation openings and the area of the compartment floor, respectively

ventilation openings where, w 1 is given by: w1

6 J0.3

=( H

4

·[0.62+90(0.4-a,) /(I+bv ·ah)];:,: 0.5

(Eqn 9)

(Eqn 1O)

EXPERIMENTAL FIRE DATA -

EXPERIMENTAL FIRE DATA

The fo llowing two chapters present 29 experimental fires which have been categorised into two groups.: • Occupancies (Table 4), which considers design fire scenarios that are representative of a type of building occupancy. These experimental fire scenarios can include a number of different combustible materials. The arrangement of the combustible contents was intended to represent that of a typical example of such an occupancy. • Commodities (Table 5), of which some of the examples, like the occupancies, may contain a number of different combustible materials. The fire tests are intended to document the fire performance of standalone items. For example, this resource considers a car to be a commodity and a car park to be an occupancy. A summary of the most commonly used parameters in fire safety engi neeri ng are detailed as part of the summary of each of the experimental fires where avai lable, including the following: • heat release rate • heat of combustion • mass of fire load • optical density • carbon dioxide concentration • carbon monoxide concentration. Indi cative gas concentrations and smoke production parameters are included where available. However, these parameters alone may not be appropriate fa r adoption in computational models. Full tranisent definition of many of the experimental fires contained in this publication are availab le in BRE's Design fires databasei181.

Table 4: Summary of the experimental fire occupancy scenarios included in Chapter 6 Section no.

6.1 6.2 6.3 6.4 6.5 6.6 6.7 6.8 6.9 6.10 6.11 6.12 6.13

Occupancy Bar/Nightclub Car parks Carpet stare Clothes sta re Indoo r play area Library Living room Luggage stare Office Prison cel i Reception Retai l stare Video stare

Table 5: Summary of the experimental fire commodities included in Chapter 7 Section no.

Commodity

7.1 7.2 7.3 7.4 7.5 7.6 7.7 7.8 7.9 7.10 7.11 7.12 7.13 7.14 7.1 5 7.1 6

Beds Boxes Buses Cars Chairs Christmas trees Computers Curtains Flight lu ggage H and cart (se lling flowers) Pallets Pool fires Soft toys Televisions Upholstered furn iture Wardrobe

-

DESIGN FIRES FOR USE IN FIRE SAFfTY ENGINEERING

6

OCCUPANCIES

6.1 Bar/Night club TEST TYPE Two free-burn experiments (with anci without sprinklers)

SPRINKLER SPECIFICATION Four stanciarci-response sprinklers (unless stateci) operateci manually, Total combineci flow rate of 270 1/min anci with a pressure of 0.6 bar at the sprinkler heacis. This gave at least a 12 m 2 coverage per heaci anci 5 mm/min/m 2 cielivereci water ciensity.

FIRE LOAD Description

Mass (kg)

No.of items

Upholste red bench seating constructed from chipboard, foam seat and back,

132.7

8

covered in dralon-type materiai. Solid wooden chairs and tables. Upholstered stools with materiai covers and foam fillings. Additional items in the unsprinklered test included four jackets with polyester outer shell and hollow-fibre fillings

MEASUREMENTS TAKEN Heat release rate (total anci convective), temperatures, optical ciensity, mass flow rate, raciiant heat, (0 2 anci CO concentrations

PEAK MEASURED PARAMETERS Optical density (OD/m)

Mass flow rate (kg/s)

C02 (ppm)

Unsprinklered (Test terminated at 2900 s)

2.51 (2025)

3.12 (1485)

14289 (2058)

Sprinklered (No sprinkler activated,

0.81 (108)

2.31 (648)

Test terminated at 1500 s) Numbers in pare ntheses

= time to peak parameters in seconds

FIRE DESIGN PARAMETER For O < t ~ 372 s, a= 0.0045 kW/s 2 (unsprinklered)

REFERENCE Clarke P & Smith DA. Charç1cterisation of fires for design purposes: a database for fire safety engineers. lnterflam 2001 . Proceedings of 9th Conference, Volume 1. London, lnterscie nce Communications, 2001 . p 1157

4381

CO (ppm) (498)

615 (1356) 423 (237)

6 OCCUPANCIES -

HEAT RELEASE RATE DATA 900.00 800.00

.....

l

700.00

i

À} \..,..

600.00 J

.l!l

~--------

I;! 500.00

'

-

C1)

"'"' C1)

1!

I

400.00

.ft

I

1uC1) 300.00

:r

)

200.00

I

100.00

O

200

~- . ,,,.,.

} ,.

-

1/

0.00

-- ~

i

i

i'

I 400

600

800

)

I '\. .......

_ /_

'

~.

/t\

f.J

·~

-

ìj

r-

1000 1200 1400 1600 1800 2000 2200 Time(s)

- - Total

- - Convective

Heat release rate of an unsprinklered bar

160.00

. . . . -l-

140.00

i

I

I I

120.00

§' ~

i

C1)

"'"'

100.00 80.00

C1)

1! 1uC1)

60.00

:r 40.00 20.00 0.00

o

200

400

600

- - Total

Heat release rate of a sprin kle red bar

800 Tlme(s)

1000

- - Convective

1200

1400

1600

-

DESIGN FIRES FOR USE IN FIRE SAFETY ENGINEERING

6.2 Car parks TEST TYPE Three experiments simulating an open-sided carpark (with and without sprinklers) and one experiment simulating a car park stacker

SPRINKLER SPECIFICATION Six standard-response sprinkl er heads (to BS EN 12845: 2004 specification), 5 mm/ min with coverage of 12 m 2 per head. Mean pressure of 2.5 bar with a maximum flow rate of 51 O litres/m in.

FIRE LOAD Test no.

Description Free burn: 3 cars in a 'typical' open-sided car park .

Parking space 1: large hatchback (petrol); Parking space 2: unoccupied; Parking space 3 : srna li car (petrol); Parking space 4: large estate (diesel). Fuel levels in ali ca rs approxirn ately 20 litres of either diesel or petro l. Fire service intervention after 24 rninutes. 2

Sprinkl ered burn: 3 cars in a 'typical' open-sid ed ca r park.

Parking space

1: MPV (petrol);

Parking space 2:

unoccupied; Parking space 3: srnall ca r (petrol);

Parking space 4: 4X4 (petrol). Fuel levels in all ca rs approxirnately 20 litres of either diesel or petrol. Fire service intervention after 85 rnin. Spri nkler operation: 2 heads after 4 rn inutes, 4 heads after 42 rninutes, 6 heads after 45 rninutes. 3

Free burn: 3 cars in a 'typica l' open-sided car park.

Parking space

1: MPV (petrol);

Parking space 2:

unoccupied; Parking space 3: rnid -sized estate (d iesel);

FIRE DESIGN PARAM ETERS

Parking space 1: 4 X4 (petrol).

For O < t ::::; 1269 s, a 1 = 0.0101 kW/s 2

unsprinklered

Fuel levels in ali cars approxirnately 20 litres of either

For O < t::::; 3231 s, a 2 = 0.00065 kW/s 2

sprinklered

diesel or petrol.

For O < t ::::; 600 s, a 3 = 0.0306 kW/s 2

unsprinklered

For O < t ::::; 678 s, a 11 = 0.0164 kW/s 2

unsprinklered

Fire se rvice intervention after 10.5 rninutes. 11

MEASUREMENTS TAKEN Heat release rate, te mperatu res, optical density, mass flow rate, radiant heat, C0 2 and CO concentrations

Free burn: 2 cars in a stacker configuration (one directly above the other).

Lower parking space: 4x4; Upper parking space : farnily estate.

REFERENCES

Fuel levels in ali ca rs approx irnately 20 litres of either

systems. Automatic spri nkler systems. Design, instaliation and

diesel or petro l. Fire service intervention after 24.5 rninutes.

maintenance

BSI. BS EN 12845: 2004

Shipp M, Fraser-Mitchell

+ Amendrnent 2:

J, Chitty R et al.

2009 Fixed firefighting

Fire spread in car parks; a

summary of the CLG/BRE resea rch programrne and findings. 2009. Ava ilable from http://www.info4fire.com/ in-depth-content/fu ll/firesp read-in-ca r-parks

6 OCCUPANCIES -

HEAT RELEASE RATE DATA

16000 -···------·- ·------14000 12000

·--······-··--· ··-·--·············· · - - - - - · · · · ····--·----+···-·····-----

·········--·--··-· ··--·-···

~- - - - - - + - - - - --+t

~- - - - + - - - - - + - - - - - - - + - - - - - - - - 1 - - - - - 1 - - - - ----1

I

~ 10000 G)

U)

m

e i

:I:

6000 ·······-·-

o

o

10

30

20

40

50

60

Time (minutes) -

Free-burn , 3 cars

-

Sprinklered, 3 cars

-

Free-burn, 3 cars

-

Single car, medium size, engine fire

-

Single car, MPV

-

Stacker test, 2 cars

Total heat release rates fro m car park fi res

70

80

-

DESIGN FIRES FOR USE IN FIRE SAFETY ENGINEE_RI_ NG- - - - ~ - - - - - - - - - - - - - - -

6.3 Carpet stare TEST TYPE Two free-burn experiments (with and without sprinklers)

SPRINKLER SPECIFICATION Four standard-response sprinklers (unless stated) operateci manually. Total combined flow rate of 270 1/min and with a pressure of 0.6 bar at the sprinkler heads. This gave at least a 12 m 2 coverage per head and 5 mm/ min/ m 2 delivered water density.

FIRE LOAD Mass

Description Carpet mixes included polypropylene/wool/hair, 100% polypropylene, 80% wool,

No. of

(kg)

items

117

15

wool/polypro pylene mix. Also included vin yl flooring and foa m rubber underlay. Backing materials were either hessian, felt or foam rubber.

MEASUREMENTS TAKEN Heat release rate (total and convective), temperatures, optical density, mass flow rate, radi ant heat, C0 2 and CO concentrations

PEAK MEASURED PARAMETERS Optical density (OD/m)

Mass flow rate

( 02

(kg/s)

(ppm)

CO (ppm)

Unsprinklered (Test terminateci at 1400 s)

2.23 (1116)

3.97 (11 10)

Sprinkl ered (First sprinkl er activated at 380 s,

3.72

2.79 (381)

30845 (1 134) 12858 (399)

2350 (225)

(240)

Test terminateci at 500 s)

Nurnbers in parentheses

= tirne to peak pararneters in seconds

FIRE DESIGN PARAMETER For O < t ::; 591 s, a

= 0.0238 kW/s 2 (unsprinklered)

REFERENCE Clarke P & Smith DA. Characterisation of fires for design purposes: a database for fire safety engineers. lnterfl am 2001. Proceedings of 9th Conference, Volume 1. London, lnterscience Communications, 2001. p 1157

907 (1137)

6 OCCUPANCIES -

HEAT RELEASE RATE DATA 2500.00 , - - - - - , - - - - - - , - - - - - , - - - , - - - - - , - - - - - - , - - - - - , - ---,

2000.00

I

J!! 1500.00

e! Cli

"'ns

e 1000.00 Cli

-··-

- -+---

1aCli :e

i

! ····-··---r~·· ....

500.00

I

o

200

400

600

800

1000

1200

1400

1600

Time(s)

j -Total -

Convective j

Heat release rate af an unsprinklered carpet stare

800.00

I i

Cli

I

700.00

···-····+ .. ···- ···-···-·

600.00

---·---t- --·- . ·-----+

500.00

··········-r--- -------------~--

400.00

"'ns Cli

e1a 300.00

················--··i-·····-···

I

Cli

:e 200.00

... ,

·-·---~---

100.00

!

0.00

o

100

200

400

300 Tlme(s)

j -Total -

Heat re lease rate afa sprinkle red carpet stare

Convective [

500

600

-

DESIGN FIRES FOR USE IN FIRE SAFETY ENGINEERING_ - - - ~ - - - - - - - - - - - - - - - -

6.4 Clothes store TEST TYPE Two free-burn experime nts (with anci w ithout spri nklers)

SPRINKLER SPECIFICATION Four stanciarci-response sprinklers (u nless stateci) operateci manually. Tota! combineci flow rate of 270 1/min anci with a pressure of 0.6 bar at the sprin kler heacis. This gave at least a 12 m 2 coverage per heaci anci 5 mm/min/m 2 cielivereci water ciensity.

FIRE LOAD Description

Mass

AII synthetic materi als including nylon, po lyester, acrylic and cotton mixes.

No. of

(kg)

items

144.6

363

Each item was hung on a plastic coat hanger. ltems included T-shirts, tracksuit trousers, bomber jackets and fl eece tops.

MEASUREMENTS TAKEN H eat release rate (tota ! anci convective), temperatures, optical ciensity, mass flow rate, raciiant heat, C0 2 anci CO concentrations

· PEAK MEASURED PARAMETERS Optical density (OD/m) Unsprinklered (Test terminated at 31 O s)

4 .00* (228)

Sprinklered (First sprinkl er activated at 290 s,

1.04 (336)

Test term inated at 500 s) * Limit of measuring range for instrument Numbers in parentheses

= time to peak parameters in seconds

FIRE DESIGN PARAMETER Far 150 < t :::; 310 s,

a= 0.308 kW/s 2 (u nsprinkl ereci)

REFERENCE Clarke P & Smith DA. Characterisation of fires for design purposes: a database for fire safety engineers. lnterflam 2001. Proceedi ngs of 9th Conference, Vo lum e 1. London, lnterscience Communications, 2001. p 11 57

Mass flow rate

co,

(kg/s)

(ppm)

CO (ppm)

100867 (309)

4922 (282)

26451 (3 12)

710 (315)

3.12 (297)

---------------------------6-0_CCUPANCIES -

HEAT RELEASE RATE DATA 8000.00 , -- - - - , - - - - . , . - - - - - , - - - - - - , - - - - - , , - - - - - - - ~

I

7000.00

t

6000.00

§' ::!. 5000.00

i

a,

.,UIa,

i! 1;j

4000.00

>+•• ··

3000.00

a,

:i:

2000.00 !

1000.00

...

________ i ···-· !

1 0.00 +-,-.---,-,......+~

o

,.....,......-,j,,c,;,=;=='F"l'=i=::;::::;,4~1--r-~,.....,..+~~.,....../-.---,-,-,-1

50

100

150

200

250

300

350

Time(s)

1- - Total - - Convective! Heat release rate of a n unspri nklered clothes store

1600.00 1400.00 1200.00

Is f!

1000.00

a,

800.00

i!

600.00

.,UIa, 1;j a,

:i:

400.00 200.00 0.00

o

100

200

300

400

Time(s)

1- - Total - - Convective! Heat release rate of a sprinklered clothes store

500

600

-

---~----------~-------~-

DESIGN FIRES FOR USE IN FIRE SAFETY ENGINEERING

6.S Indoor play area TEST TYPE Two free- burn experim ents (with anci w ithout sprinklers)

SPRINKLER SPECIFICATION Four stanciarci-response spri nklers (u nless stateci) operateci manually. Total combi neci fl ow rate of 270 1/min anci with a pressure of 0.6 bar at the sprinkler heacis. This gave at least a 12 m 2 coverage per heaci anci 5 mm/min/m 2 cielivereci water ciensity.

FIRE LOAD Description • Th e stee l fram ewo rk was covered in pipe lagging materiai (expanded foam), with the padded flooring constructed of plywood and foam covered w ith PVC. AII padded areas were PVC-covered foam. A GRP sl id e and polypropylene rotation moulded crawl tube were prese nt. Netting materia i was nylon.

MEASUREMENTS TAKEN Heat release rate (total anci convective), temperatures, optical ciensity, mass fl ow rate, raciiant heat, C0 2 anci CO concentrations

PEAK MEASURED PARAMETERS Optical den sity

Mass flow rate

co,

CO

(OD/ m)

(kg/s)

(ppm )

(ppm)

Unsprinklered (Test terminateci at 360 s)

4.00 * (330)

3.62 (357)

201347 (369)

8550 (369)

Sprinklered (First sprinkler activated at 440 s,

1.02 (633)

2.72 (432)

7024 (45 3)

931 (423)

Test term inateci at 800 s) *Lim it of measuring range for instrument Numbers in parentheses

= time to peak parameters in seconds

FIRE DESIGN PARAMETER For 100 < t ~ 370 s, a= 0.1 kW/s 2 (unsprinklered)

REFERENCE Clarke P & Smith DA. Characterisation of fires for design purposes: a database for fire safety engineers. ln terflam 2001. Proceed ings of 9th Conference, Volume 1. Lon don, lnterscience Communications, 2001. p 1157

_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _......._ _ _ G_o_ccuPANCIES -

HEAT RELEASE RATE DATA 8000.00 +--- - ---------:----------- ···-

7000.00 6000.00

is

··----··---·-·+·----- ···- ·-·-· ···-

5000.00

--~ ------

E 4000.00

Cli

1/1

"'Cli

e!

3000.00

_j

-; Cli

:e 2000.00

I I

1000.00 0.00

o

50

100

150

200

250

300

350

400

Time(s)

1- - Total - - Convective ! H eat release rate of an unspri nklered play area

400.00

,)\ -)r \ A

350.00 300.00

is

250.00

150.00

-; Cli

:e 100.00

·---

lt I )_ 11

1/1

"'Cli

----·

I

E 200.00 Cli

e!

-

----~

-

~~-

50.00

_.-,- ::/

~

I

0.00

o

100

200

300

400

500

- -----

-

,_ I

/"-I

I

! !

600

700

Time(s)

1- - Total - - Convective ! Heat release rate of a spririklered play area

..

800

900

-

DESIGN FIRES FOR USE IN FIRE SAFETY ENGINEERING

6.6 Library TEST TYPE Two free-burn experiments (with and without sprinklers)

SPRINKLER SPECIFICATION Four fast response heads on a 3.4 m square grid. Each sprinkler head had a coverage of 12 m2 and flow rate of 60 1/min.

FIRE LOAD Description

One wooden shelving unit fixed to the rear wall (2100 mm high, 4 bays long) filled with hardback books side by side (80%) and paperback books stored in a display fashion. One metal shelving unit fixed to the wall perpendicular to the wooden shelving (1800 mm high, 5 bays long) filled with hardback books side by side (80%) and paperback books stored in a display fashion. Two paperback racks (1800 mm long) and two video racks (60-75 videos in each) were placed in the centre of the room .

Rig hood and duci

Fibreglass curtains kept lowered lo form 'wall' behind the wooden shelving

Paperback racks 1.8m long

A small wood/plastic top table with a VDU, keyboard and chair were placed at the end of the metal shelving together with a table and three chairs.

D D

D

Video racksD (60-75 videos on each)

Book shelving metal on this run, against partition

MEASUREMENTS TAKEN Heat release rate (tota!), temperatures, optical density, mass flow rate, radiant heat, C0 2 and CO concentrations

Table and VDU with keyboard

Newspaper/magazine rack (originally proposed, bui noi included far lests)

PEAK MEASURED PARAMETERS co Unsprinklered (Test terminated at 16 mins) Sprinklered (First spri nkler activated at 3.5 mins, Test terminated at 20 mins) Numbers in parentheses

CO (ppm)

4.00 (1080) 1.40 (240)

1000 (1080) 900 (720)

= time to peak parameters in seconds

FIRE DESIGN PARAMETER For O < t :':': 580 s, a = 0.008 kW/s 2 (unsprinklered)

REFERENCE Webb J & Samme P. Th e characterisation of library fires using a sp rinklered ca lorimeter. Private communication, 1996

2

(%)

6 OCCUPANCIES -

HEAT RELEASE RATE DATA

2500

§' :. 2000 2

r:

Cl)

"''"

1500

Cl)

~

... '" Cl)

J:

1000

\ ~------

500

o

200

400

600

800

Time (s) - - Unsp rinkl ered

Total heat release rate of a li brary w ith and w ith out sprinklers

Unsprinklered tes t after 6 minutes (approx imate fi re size of 1 MW)

- - Sprinklered I

1000

1200

-

DESIGN FIRES FOR USE IN FIRE SAFETY ENGINEERING

6.7 Living room TEST TYPE One free-burn experim ent (no sprinklers operateci)

FIRE LOAD Description • Th e contents of the room primarily co nsisteci of: 3-seater settee, 1 ciining room chair, 1 beanbag, 2 small coffee tables, 1 sicieboarci, cu rtains anci books. • The centrai area of th e floor was laici with carpet tiles (see cii agram, right). • Th e seating was part-covereci by a throw rug. • Other furnishings inclucieci w icker ornaments, newspapers, magazines in a rack anci ca nciles on horizontal surfaces. • Th e curtain s were hung from a stee l pole (Note: there was no winciow). • Books anci toys were locateci o n th e sofa anci sicieboa rci. • A sma ll wastepape r bin containing crumpleci newspape r sheets was locateci betwee n the beanbag anci th e settee.

MEASUREMENTS TAKEN Heat release rate (total), temperatures, oxygen depletion, C0 2 and CO concentrations

Sideboard

Carpe! tiles

Small tables

PEAK MEASURED PARAMETERS

Unsprinklereci (Test terminateci at 24 mins) Numbe rs in parentheses

= time te peak parameters in seconds

FIRE DESIGN PARAMETER For 300 < t :s; 660 s, a

= 0.04 kW/s 2

ACKNOWLEDGEMENT Th e data for this design fire were contributed by M artin Shipp, BRE Global.

REFERENCE DeHaan JD. Kirk 's fire investigation. 6th eciition. New Jersey, Pearson Eciucation/ Prentice- Hall, 2006

0.94 (660)

CO

Minimum 0

(% )

(%)

0.026 (660)

19.9 (660)

2

6 OCCUPANCIES -

HEAT RELEASE RATE DATA

/

Flashover

5000

~4000 ~

i

i

Flames in smoke layer

3000 ~ - - · · - - - ~ -

~

e

Fire dies down

i

:e 2000

I

.............................................. ··-·······

Sideboard fully alight ··············+ ··-···················-············· ································I

1000

~-~-~--~'-,--'-..-----------~-~-...:..j

0 -1--=-.-..---.--=:;=,:,,.,..._ __ _ o 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 Time from lgnition (min)

Total heat release rate of an unsprinklered living room

-

DESIGN FIRES FOR USE IN FIRE SAFETY ENGINEERING

6.8 Luggage store TEST TYPE Four free-burn experiments (unsprinklered, fast-response sprinklers and two with standard sprinklers)

SPRINKLER SPECIFICATION Four standard-response sprinklers (unless stateci) operateci manual ly. Total combined flow rate of 270 1/min and with a pressure of 0 .6 bar at the sprinkler heads. This gave at least a 12 m2 coverage per head and 5 mm/min/m 2 delivered water density.

FIRE LOAD Description

(kg)

No. of items

92

57

Mass

Mainly manmade fibres. Nylon outer shells with PVC waterproof interliner, polyester inrier pockets and compa rtm ent dividers. Some cases had fabric she lls w ith simi lar liners to the other cases. Rucksacks were mainly nylon weave with polyester inner linings.

MEASUREMENTS TAKEN Heat release rate (total and convective), temperatures, optical density, mass flow rate, radiant heat, (0 2 and CO concentrations

PEAK MEASURED PARAMETERS Optical density (OD/m)

Mass flow rate

C02

CO

(kg/s)

(ppm)

(ppm)

Unsprinklered (Test terminateci at 175 s)

4.0* (11 O)

3.95 (175)

64994 (175)

Sprinklered (Fast-response sprinkler activated at 11 O s, Test terminateci at 500 s)

4.0* (150)

53978 (220)

5059 (175) 4322 (220)

Sprinklered (Standard-response sprin kler 1

4.0* (60)

4.05 (140)

47747 (145)

3479 (130)

4.0* (105)

3.81 (115)

40658 (155)

2730 (155)

activated at 60 s, Test terminateci at 400 s) Sprinklered (Standard-response sp rinkl er 2 activated at 100 s, Test terminateci at 460 s) * Limit of measuring range for instrument Numbers in parentheses

= tim e to peak parameters in seconds

FIRE DESIGN PARAMETER Far 75 < t :S 175 s, u = 0.6054 kW/s 2 (unsprinklered) REFERENCE Clarke P & Smith DA. Characterisation of fires for design purposes: a database for fire safety engineers. lnterflam 2001. Proceedings of 9th Conference, Volume 1. London, lnterscience Communications, 2001. p 1157

6 OCCUPANCIES

HEAT RELEASE RATE DATA 7000.00 6000.00

~~ - - - · - -····· · · · · - · è - - - - - - - - + - - - - -- -- - + - ~ -/ -·-·-·-·-·-

5000.00

I.s

4000.00 ,

f!

Cl)

!Il

ca 3000.00 Cl)

e-;

Cl) :e 2000.00

1000.00 0.00

o

50

150

100

200

Time(s)

- - Unsprinklered (Tota!) - - Unsprinklered (Convective)

Heat release rate of an unsprinklered luggage store

4000.00 -,--- - - - - , - - - - - - , - - - - - - , - - - - - , - - - - - , - - - - --, 3500.00

--

3000.00 -

~

2500.00

i

2000.00

Cli !Il

e

m 1500.00

m 1000.00 :e 500.00

o

100

200

300

400

500

Time(s)

- - Fast response (Tota!)

- - Fast response (Convective)

· - - Standard response 1 (Tota!)

- - Standard response 1 (Convective)

- - Standard response 2 (Tota!)

- - Standard response 2 (Convectivè)

Heat re lease rate of a-sprinklered luggage store

600

-

DESIGN FIRES FOR USE IN FIRE SAFETY ENGINEERING

6.9 Office TEST TYPE Two free-burn experiments (with anci without sprinklers)

SPRINKLER SPECIFICATION Four stanciarci-response sprinklers (unless stateci) operateci manually. Total combineci flow rate of 270 1/min anci with a pressure of 0.6 bar at the sprinkler heacis. This gave at least a 12 m 2 coverage per heaci anci 5 mm/min/m 2 cielivereci water ciensity.

FIRE LOAD Description

Mass

Melamine-faced chipboard desks, materiai- and foam-covered MDF dividing screens on each desk, computer monitors, keyboards and generai office items, which were mainly constructed from plastics. Upholstered office chairs with materiai covers, foam seat and back with polypropylene trim.

(kg)

No. of items

245.6

70

MEASUREMENTS TAKEN Heat release rate (total anci convective), temperatures, optical ciensity, mass flow rate, raciiant heat, C0 2 anci CO concentrations

PEAK MEASURED PARAMETERS

Unsprinklered (Test terminated at 1800 s) Sprinklered (First sprinkler activated at 950 s, Test terminated at 1200 s) Numbers in parentheses

Optical density (OD/ m)

Mass flow rate (kg/s)

co2

CO

(ppm)

(ppm)

2.21 (846) 0.84 (903)

4.13 (1575) 2.54 (924)

107157 (1779) 6063 (942)

11803 (1728) 493 (1137)

= time to peak parameters in seconds

FIRE DESIGN PARAMETER = 0.0003

For 500 < t ::s; 1200 s, a

kW/s 2 (unsprinklereci)

REFERENCE Clarke P & Smith DA. Characterisation of fires for design purposes: a database for fire safety engineers. lnterflam 2001. Proceedings of 9th Conference, Volume 1. London, lnterscience Communications, 2001. p 1157

6 OCCUPANCIES

HEAT RELEASE RATE DATA 5000.00 - , - - - - - - - , - - - - - ----,.- - - - - - ~ - - - - - -~ 4500.00

~

----- --------- --- -------- ---------- --;-------- -- ----- ----- ----

------·---+----------- ------------ -------------1- ----------------- -----·------

4000.00

i

3500.00

s - - - - - - - ------------~------------------- ·----------, -------------------- - -----------1

~

3000.00 + --- -- - -·-- - --------------t----------·----------------+------------ ·------ ------+---1- - - - - - -----

~

2500.00 + ·-- -- ·--- --- ------------------ -------------1----···-· ···---------·----- --· --------------+---------············ ··········-·----··---+

Cl)

lii

"'

Cl)

~

2000.00

~- -- - - - - - 1 - - - - - - - -- - - - -------·-- I --·-···---

1u 1500.00 Cl)

:e

1000.00 500.00 0.00

o

500

1000

1500

2000

Time(s)

1--Total - - Convectivel Heat re lease rate of an unsprinklered office

300.00

250.00

i ~

Cl)

200.00

150.00

lii

"' Cl)

~ 100.00 1u Cl)

:e

50.00

0.00

o

200

400

600

800 Tlme(s)

1--Total - - Convective j Heat re lease rate of Ùprinklered office

1000

1200

1400

DESIGN FIRES FOR USE IN FIRE SAFITT ENGINEERING

6.10 Prison cell TEST TYPE Single experiment (free burn, no sprinklers operated) . Cell contained representative prison-issue furniture and persona! items.

FIRE LOAD Description

• •

• •

Double-occupancy prison ce li measuring inte rnally 3 m x 4 m and 3 m high, givi ng a tota[ volume of 36 mJ. The room was of blockwork constructio n, inte rnally clad with plaste rboard. Fire load comprised 2 prison-issue matresses, 2 prison -issue bedside lockers, one of whi ch was placed on its side on the lower leve[ of a bunk bed. Th e locker on the bunk bed was fill ed with: 12 crisp packets, 4 boxes of cereal, a pair of jea ns, a vest, 2 toilet rolls, 2 plastic bottles, a newspaper, a magaz in e, 1 O singl e CD cases, a compu te r keyboa rd, a prison issue duvet cove r, sheet and pillow slip. Additio na lly, 2 desks, 2 chairs (one of which was plastic), 2 pairs of shoes, a towel, 3 shirts, a television and 2 pa irs of jeans. Fire was manu ally extinguished after approximate ly 18.5 minutes. Door was open th roughout the test.

++-+-+- -

Ceiling extract location

Doorway

PEAK MEASURED PARAMETERS Volume

Minimum oxygen (%)

C0 2

0.2 (mid-height)

0.165 (1074)

Numbers in parentheses

(m3/s)

CO (m 3/s) 0 .03 (396)

HCN

HCL

(ppm)

(ppm)

(m 3/s)

1030 (600)

51 (600)

20 (600)

flow rate

= time to peak parameters in seconds

MEASUREMENTS TAKEN

REFERENCE

Heat release rate (tota!), temperatures, optical density, HCN, HCL, C0 2 and CO concentrations

Annable K. Watermist systems far prison cells. 201O. Available at www. info4fi re .com/in-de pth-conte nt/fu ll/water-mist-systems-forprison-cells

FIRE DESIGN PARAMETER For O < t ::; 600 s, a = 0.0069 kW/s 2

6 OCCUPANCIES

HEAT RELEASE RATE DATA

2500

- - -•~-,

---+--·· - - -

I

I

1500

1000 --·--···-··· · ···- · --··-·· · ······--·-···- ·-·············· ·-··-··---····-

11.

f I

. r. . . . . .

i -~

I

!

500

······--·-·-··· .... ·········-··········· ········· ············· ··········-----·-·······

o

5

10

15

20 Tlme(min)

Tota! heat release rate from a typi cal prison ce li (un sprinklered)

1

! '

·······-····---····· +· ····-··-··········.

25

30

35

1111

DESIGN FIRES FOR USE IN FIRE SAFETY ENGINEERING

6.11 Reception TEST TYPE Two free-burn expe riments (with anci without sprinklers)

SPRINKLER SPECIFICATION Four stanciarci-response sprinklers (unless stateci) operateci manually. Total combineci flow rate of 270 1/min anci with a pressure of 0.6 bar at the sprinkl er heacis. This gave at least a 12 m 2 coverage per heaci anci 5 mm/min/m 2 cielivereci water ciensity.

FIRE LOAD Mass

Description Computer monitors, keyboards, TV/monitor, polypropylene wastebasket, videocassettes and recorder as well as generai office items such as ring binders and paper. Two office-style leather faced chairs with foam seat and back. Two visitors' seats material-covered w ith hollow-fibre cushions and one material-covered sofa w ith hollow fibre.

No. of

(kg)

items

86.5

19

MEASUREMENTS TAKEN Heat release rate (total & convective), temperatures, optical ciensity, mass flow rate, raciiant heat, C0 2 anci CO concentrations

PEAK MEASURED PARAMETERS

Unsprinklered (Test terminateci at 1700 s) Sprinklered (First sprinkler activated at 360 s, Test terminateci at 500 s)

Optical density (OD/m)

Mass flow rate

co,

CO

(kg/s)

(ppm)

(ppm)

4.00* (975) 2.14 (366)

4.23 (1062) 2.90 (345)

51615 (1515) 10628 (378)

12589 (1629) 642 (381)

*Lim it of measuring range for instrument Numbers in parentheses

= time to peak parameters in seconds

FIRE DESIGN PARAMETER Far 180 < t :::: 1400 s,

a = 0.003 kW/s 2 (unsprinklereci)

REFERENCE Clarke P & Smith DA. Characterisation of fires for design purposes: a database for fire safety engineers. lnterflam 2001. Proceedings of 9th Conference, Volume 1. London, lnterscience Communications, 2001.p1157

6 OCCUPANCIES -

HEAT RELEASE RATE DATA 3500.00

I

. . _.J.

3000.00

i

·········· I ·i

2500.00

I

. J_

.2! 2000.00 f! a,

I

Il)

lii a,

e'la

······--1·····--:

1500.00

......... -·- ·-·- · - . f

a, :e 1000.00

····-+

I

!

500.00 0.00

o

200

400

600

800

1000

1200

1400

1600

1800

Time(s)

1- - Total - - Convective/

Heat release rate of an unsprinkle red rece ptio n area

!~ ----,------~---------,

500.00 . - - - - - - - , - - - - - - - , 450.00

I

§"

. . -~-_r-···

350.00

i

~ 300.00

I

---L .

.! f! a, 250.00

::i

e a,

1

~-I

I :

'

200.00

m150.00 :e 100.00

I

-- -- --------- - ---·-r

50.00

o

100

200

300

400

Tlme(s)

/ - - Total - - Convective/

Heat release rate of a spri nkle red reception area

500

600

DESIGN FIRES FOR USE IN FIRE SAFETY ENGINEERING

6.12 Retail store TEST TYPE Single free-burn experim ent (without sprinklers)

FIRE LOAD Description

• Retai l store com plete with luggage, plastic manikin, bags, plastic shoes, LCD display, chairs, storage cupboard and plastic toys arranged over both the floor area and in hanging displays.

MEASUREMENTS TAKEN Heat release rate, temperatures, vo lume fl ow, rate of smoke produ ction, 0

, 2

C0

2

and CO concentrations

PEAK MEASURED PARAMETERS

co

CO

Unsp rin klered Numbers in parentheses

Smoke

Volume flow rate

(%)

(% )

(%)

production rate (m 2/s)

0.08 (1128)

1.66 (1122)

19.07 (11 13)

14.77 (11 52)

0

2

2

minimum

(m 3/s) 16(11 22)

= time to peak parameters in seconds

FIRE DESIGN PARAMETER For O < t :::; 153 s, a

= 0.1 84 kW/s 2 (unsprinklered)

HEAT RELEASE RATE DATA

4500 4000

.f...--

3500

!

·-+--~-· '

i'3000 ~ Cli

f

i---

2500

. J.

Cli

U)

Sl 2000

!

~

j

1500

J:

1000 500 0

.-.....,,::.....,...~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~--i

o

50

100

150

200 Time(s)

Tota l heat release rate of an unsprinkl ered retai l store

250

300

350

400

6 OCCUPANCIES -

6.13 Video stare TEST TYPE Single free-burn experiment (without sprinklers)

FIRE LOAD Description • Video cassettes in plastic or cardboard boxes, empty compact disc cases. Shelving formed from mou lded polypropylene video and CD trays supported by a meta l framework. The layout was intended to represent a corner aisle of a video shop. The tests represented malicious ignition of the bottom row of the centre video rack .

MEASUREMENTS TAKEN Heat release rate, temperatures, optical density, radiant heat, C0 2 and CO concentrations

·PEAK MEASURED PARAMETERS

Unsprinklered

Optical density (OD/m l

co (%)

CO (ppm)

2.5 (240)

4.1 (660)

7000 (660)

2

Numbers in parentheses = time to peak parameters in seconcls

FIRE DESIGN PARAMETER For 100 < t :e; 660 s, a = 0.02 kW/s 2 (unsprin klered)

REFERENCE Samme P & Webb J. The characterisation of video shop fires using a sprinklered ca lorimeter. Private commun ication, 1997

HEAT RELEASE RATE DATA

5000

~ ~ Cl)

"'"'

4000

Cl)

~

-; 3000 Cl)

::e

2000

o

100

200

300

400

500

Time(s)

Tota! heat release rate of an unsprinklered video store

600

700

800

900

-

6

DESIGN FIRES FOR USE IN FIRE SAFETY ENGINEERING

Notes

- - - - - - - - - - - - - - - - - - - - - - - - - - - - -7 COMMODITIES -

7

COMMODITIES

7.1 Beds FIRE LOAD Experiment

Description

Mass

No. of items

Double bed, bedding, night table Double bed, bedding, night table

(kg) 53.70 53 .70

no.

2

MEASUREMENTS TAKEN

REFERENCES

Heat release rate (tota!)

Spearpoint M. FireBaseXML database. Version 1.34. New Zealand, Unive rsity of Canterbury, 2007. Visit http ://www. civil.canterbury. ac. nz/spearpoi nt/HRR_ Database/HRR_ Database.xmI

FIRE DESIGN PARAMETERS Bed

a (kW/s 2 ) 2.22 far O< t :5: 360 s 0.82 fa r O< t :5: 230 s

2

Heat of combustion

U/kg) 18100 18100

HEAT RELEASE RATE DATA

7000 ~------~--~---------'--~-----~-----+---~~~+---~~~--+~---------- --------~-----------------------+~~~~--+-~~~~I

i'\

6000

i sooo ~

:ll 4000

e"' Q)

m3000

:i:

2000

-r---

/\

1000

1 --=-r=~=======-1-·-- -_-- --_-- -

---·1--_--

o o

200

400

600

800

1000

Time(s)

-

Bed 1

Tota! heat re lease rate of a double bed, bedding a nd night tab le

-

Bed 2

I

1200

1400

16.00

DESIGN FIRES FOR USE IN FIRE SAFETY ENGINEERING

7.2 Boxes TEST TYPE Two free -burn experiments (with anci without sprinklers)

SPRINKLER SPECIFICATION Four stanciarci-response sprinkl ers (unless stateci) operateci m anually. Tota! combineci flow rate of 270 1/min anci with a pressure of 0.6 bar at th e sprinkl er heacis. This gave at least a 12 m 2 coverage per heaci anci 5 mm/ min/m 2 cielivereci water cien sity.

FIRE LOAD Mass

Description Corrugated cardboard boxes (610 mm x 610 mm x 480 mm) fill ed w ith packing materials, mainl y polystyre ne chips and expanded foam mouldings. 16 boxes per stack each on a wooden pallet with 6 stacks in total

items

353

96

MEASUREMENTS TAKEN H eat release rate (tota! anci co nvective), temperatures, optical ciensity, mass flow rate anci raciiant heat

PEAK MEASURED PARAMETERS

Unsprinklered (Test terminated at 880 s) Sprinklered (First sprinkl er activated at 1200 s, Test terminated at 1600 s)

Optical density (OD/m) 4.00* (625) 3.07 (1470)

* Limit of measuring range for instrument Numbers in parentheses

=

time to peak pa rameters in seconds

FIRE DESIGN PARAMETER For O

< t

~ 880 s,

a = 0.0362 kW/s 2 (unsprinklereci)

REFERENCE Garrad C & Smith DA. Characterisation of fires for design. lnterflam 1999. Proceedings of 8th Confere nce, Volume 1. London, lnterscience Communications, 1999. p 555

~

No. of

(kg)

Mass flow rate (kg/s) 3.71 (880) 3.03 (1355)

7 COMMODITIES

HEAT RELEASE RATE DATA 5000.00 4500.00

····-···--·- · - · - - + - - - -- -1

i

4000.00

i e .e

·-

3500.00

- - - - ~ - - - - - - - - -- - - - · - - - - --- -

- - - - --- ••+>-•••·

·····---· ·-···---······ ·············,

.. ... ··--··1

!

:

+ - - - - -----+;_

3000.00

I

I!

· - -------:~------·-···

·····------~-~~-·-·· ----------------- ----~ ...

Cl) Il)

.,

!

Cl)

e 2000.00

. -·------ . . . . ______L__...........

'liiCl) 1500.00 :e 1000.00

_ _

------+------·--··

·+·- ----

-r-----·-··--···

-

500.00

I

O.OD

o

200

400

600

800

1000

Tlme(s)

1- -Total - - Convective l Heat re lease rate of unsprinklered boxes

800.00 I

700.00

I ·-······-·--i-1 __ _

600.00

[

.e I! ., Cl) Il)

-- -i

i i

500.00 400.00

_ _ _ : ___ _

!

____ J1-----j-- -

Cl)

e 300.00

··--· i_ _

____ ; _______ _,...._ _ _

'liiCl) :e 200.00

.

···r ·····---- ---- ----- ··i

100.00

o

200

400

....I

__

L__ ____!

____ _

].._ ____ ,

········-·-r···--

600

800

1000

1200

Tlme(s)

1- -Total - - Convectivel Heat release rate of sp'rinklered boxes

1400

1600

1800

1!11

DESIGN FIRES FOR USE IN FIRE SAFETY ENGINEERING- - ~ - - - - - - - - - - ~ - - - - - - -

7.3 Buses TEST TYPE A single free-burn experiment to simulate a fire in the rear luggage compartment

FIRE LOAD D escription

• A 13 m si ngle-deck Volvo coach with 49 passenger seats. Th e bus was used for development purposes and was never used in publi c traffic. Gear box fitted with a retarder. Fire ignited in the rear luggage compartmen t using a propane burner. Calorimeter placed to measure heat release rate from rear two-thirds of the bus length.

MEASUREMENTS TAKEN H eat re lease rate (total), t emperatures, visibility, CO, C0 2 , H C I, HCN, HF, HBr, 50 2, NOa and N0 2 concentrations

Photos courtesy of Bjorn Sundstrom, SP Techn ica l Resea rch lnstitute of Sweden

PEAK MEASURED PARAMETERS (inside passenger compartment after 8 minutes) HCN HF HBr NOa so, (ppm )

( pm)

( pm )

(p m)

(

65

"'

800.00

1;j

600.00

Q)

i!

i

Q)

J:

-·--

400.00

-

f -· ----·-··----···- ·----------

200.00 1 0. 00 -h"""F~ - +~ ~-r-l-r--r-~-i-~-,-.--+-,-~-,-+-,~ ~-+~ ~--,---,1-r--r-r-r-1

o

200

600

400

1000

800

1400

1200

1600

Time(s)

- - Tota!

- - Convective

Heat release rate of unsprinklered soft toys

1400.00 . - - - - , - --

! a>

~

1200.00

- ----·-·-(-

1000.00

] .

...,--

--.---- - . - - - --,-----,----

-+-···- 1· ...

--,-----·-··--·

:-- --- I_-···

- , - - - -- --,

j fI

-- -- - -

! !

800.00

Q)

U>

:;'l

·-· ..........

600.00

°ffi

J:

~----

!

i!

: i

[---

400.00 200.00

o

50

100

150

200

250

300

Time(s) I

Heat release rate of sprinklered soft toys

- - Tota! - - Convective!

350

400

450

-

DESIGN FIRES FOR USE _IN_F_IR_ ES _A_FETY __ EN_G_ IN_ EE_RI_NG_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ __

7.14 Televisions FIRE LOAD Experiment

Description

no. 28 inch TV in a plastic case. lgnition with a 1 kW propane gas flame

2 3

Mass

No. of

(kg)

items

31.83

25 inch TV in a plastic case. lgn it ion with a 1 kW propane gas flame

24.42

28 inch TV in a plastic case. lgn ition with a 1 kW propane gas flame

30.53

MEASUREMENTS TAKEN

REFERENCES Hietaniemi J, Mangs J & Hakkarainen T. Burning of electrical

Heat release rate (tota!)

househo/d appliances. VTT Research Note 2084. Espoo, VTT Technica l Research Centre of Finland, 2001

FIRE DESIGN PARAMETERS Television

Spearpoint M . Fi reBaseXML database. Version 1.34. New Zealand,

a

Heat of combustion

University of Canterbury, 2007. Visit http://www.civil.canterbury.

(kW/s 2 )

U/kg)

ac. nz/spearpoi nt/H RR _ Database/H RR_ Database.xm I

0.057 for 77