NT FIRE 049 COMBUSTIBLE PRODUCTS: COMMODITY CLASSIFICATION FIRE TEST PROCEDURE TABLE OF CONTENTS NORDTEST METHOD NT FIRE (16)

Transcription

1 NORDTEST METHOD NT FIRE COMBUSTIBLE PRODUCTS: COMMODITY CLASSIFICATION FIRE TEST PROCEDURE Key words: Combustible products, commodity classification, fire test procedure 1(16) NT FIRE 049 Edition 2 Approved TABLE OF CONTENTS 1 INTRODUCTION 2 2 SCOPE AND FIELD OF APPLICATION 2 3 REFERENCES 2 4 DEFINITIONS 3 5 TEST PRINCIPLES 3 6 CALORIMETER SYSTEM Hood and exhaust system Instrumentation in the exhaust duct Volume flow rate Gas analysis 4 7 STORAGE ARRAY Rack storage arrangement Ignition source 5 8 WATER APPLICATION SYSTEM Water applicator Flow rate and pressure measurements 5 9 TEST COMMODITIES 5 10 TESTING Initial conditions and preparation Test procedure Repeat of tests 6 11 CLASSIFICATION General Principle of classification Evaluation of the test results classification 7 12 TEST REPORT 8 ANNEX A PRINCIPLE DESIGN OF THE CALORIMETER SYSTEM 9 ANNEX B PRINCIPLE DESIGN OF THE WATER APPLICATOR 10 ANNEX C CLASSIFICATION TABLES 11 ANNEX D CLASSIFICATION GRAPHS (Informative) 14 Nordic Innovation Centre Stensberggata 25, 0170 OSLO Telephone Fax info@nordicinnovation.net ISSN: Project:

2 NORDTEST METHOD NT FIRE INTRODUCTION The fire characteristics of commodities stored in a warehouse are important parameters when the design and capacity of a sprinkler system are determined. Unfortunately, these parameters are often difficult to assess. Actual fire testing and classification of the type of commodity to be stored provides considerable information on its fire and suppressability characteristics. This Nordtest method provides a methodology for commodity classification by fire testing at an intermediate scale. Fire testing involves arranging a four-pallet load of the particular commodity in a rack storage arrangement. A water applicator consisting of a matrix of spray nozzles is arranged over the commodity and the entire set-up is positioned underneath a calorimeter to facilitate the measurement of the heat. With the data obtained from the tests, classification criteria for the four main commodity categories according to the European sprinkler standard, EN [1] (categories I, II, III and IV) can be established. It is also possible to determine whether or not a commodity shall be treated as a special hazard. The commodity classification methodology, and its resulting classification limits, mean that any particular type of commodity can be classified with considerably better accuracy and safety than by using simple tabular assessment in accordance with EN A full classification of a particular commodity requires three fire tests to be conducted. This methodology was originally developed by Factory Mutual Research Corporation for classification according to the US commodity classification scheme, but has been modified to correspond to the European commodity classification scheme. Note: Users of this test method should observe the following warning: SUITABLE PRECAUTIONS MUST BE TAKEN TO SAFEGUARD HEALTH, REQUIRING THAT THE ATTENTION OF ALL ENGAGED IN THE FIRE TESTS IS DRAWN TO THE POSSIBILITY THAT TOXIC OR HARMFUL GASES MAY BE EVOLVED DURING COMBUSTION OF TEST COMMODITIES. 2 SCOPE AND FIELD OF APPLICATION This Nordtest method is intended to provide a basis for classification of commodities. On the basis of this classification, relevant sprinkler protection can be achieved according to the European sprinkler standard, EN The classification test procedure is highly standardised. The aim is to determine the hazard level of a commodity by comparing the test results with data from identical tests on well-defined standard commodities, with fire characteristics similar to commodities of Category I, II, III and IV. These tests, forming the basis of this philosophy, are described in SP Report 2003:03 [2]. Additional information is provided in SP Report 1993:70 [3]. If a tested product shows fire characteristics similar to one of the standard commodities, it is assumed that the same protection requirements are adequate for the tested commodities. A rack storage configuration is used in the classification tests. However, the results may also be applied to freestanding, palletized rack and post-pallet storage configurations, using the design criteria given for these configurations in EN It is important to note that the commodity classification tests do not directly provide the protection requirements e.g. design water density for the commodity tested although reference is made to a range of water discharge densities in the tests. The purpose is to determine the hazard level of the commodity. It should also be noted that this classification procedure has been developed for ordinary combustible hazards. Commodities such as exposed plastics, aerosols and containers with flammable liquids may represent a hazard that exceeds that of Category IV commodities, and should therefore be considered as special hazards. Requirements for the protection of special hazards are given in Annex G, Protection of special hazards of EN REFERENCES [1] EN 12845, Fixed fire fighting systems, Automatic sprinkler systems, Design, installation and maintenance, (Approved by CEN on November 29, 2002), August 2003 [2] Arvidson, Magnus and Lönnermark, Anders, Commodity Classification Tests of Selected Ordinary Combustible Products, SP Report 2002:03, Swedish National Testing and Research Institute, Borås, 2002 [3] Persson, Henry, Commodity Classification A more objective and applicable methodology, SP Report 1993:70, Swedish National Testing and Research Institute, Borås, 1993 [4] ISO 9705:1993(E), Fire tests Full-scale room test for surface products, International Organisation for Standardization, First edition [5] Dahlberg, Martin, The SP Industry Calorimeter for Rate of Heat Release Measurements up to 10 MW, SP Report 1992:43, Swedish National Testing and Research Institute, Borås, 1992 [6] Dahlberg, Martin, Error analysis for heat release measurements with the SP Industry Calorimeter, SP Report 1994:29, Swedish National Testing and Research Institute, Borås, 1994 [7] Heskestad, Gunnar, A Fire Products Collector for Calorimetry in the MW Range, Factory Mutual Research Corporation, FMRC J.I. OC2E1.RA, June 1981 [8] Croce, P. A., A Method for Improved Measurement of Gas Concentration Histories in Rapidly Developing Fires, Combustion Science and Technology, Vol. 14, pp , 1976 [9] Chicarello, P. J. and Troup, J. M. A., Fire Products Collector Test Procedure for Determining the Commodity Classification of Ordinary Combustible Products, Technical Report, FMRC J.I. OROE5.RR, August 1990

3 NORDTEST METHOD NT FIRE DEFINITIONS For the purpose of this test method, the following definitions apply: Commodity: A commodity is the combination of a product, packaging material, container, and material handling aids (e.g. pallet), upon which the commodity classification is based. HRRconv: The convective part of the heat measured during a test on the basis of the gas temperature and mass flow rate in the calorimeter system. HRRtot: The total heat measured during a test on the basis of oxygen depletion in the calorimeter system, which is comprised of both the convective and radiative heat release. Mean total rank: The average of the mean unit rank values obtained in tests specified in this test method. Mean unit rank: Average of the rank values obtained for each of the parameters V1-V2-V3-V4 on the basis of the tests results obtained for one of the specified water discharge densities or the free burn test. Pallet load: The quantity (size and volume) of the commodity stored on a standard 1200 mm by 1000 mm pallet (including the pallet). Pallet load height: The height of the pallet load, measured from the bottom of the pallet to the top of the commodity. Rank value: A numerical value (0,25 4,50) related to each of the parameters V1-V2-V3-V4. The rank values are obtained for each test by comparing the test results with the rank values tabulated in Tables C1 C4 of Annex C. Storage array: The test arrangement of four pallet loads in a 2 by 1 by 2 rack storage arrangement. V1: Maximum one-minute average of the total heat release rate (see also Clause 11). V2: Maximum one-minute average of the convective part of the total heat (see also Clause 11). V3: Effective convective heat, defined as the convective heat averaged over the five-minute interval of most intense fire (see also Clause 11). V4: Convective energy, defined as the amount generated during the most intense ten-minute interval of the fire (MJ) (see also Clause 11). Water discharge density: The water density applied to the top of the storage array, using a specified water applicator. 5 TEST PRINCIPLES Fire testing involves arranging a four-pallet load of the particular commodity in a rack in a 2 by 1 by 2 storage arrangement, see Figure 1. A water applicator consisting of a matrix of spray nozzles is arranged over the commodity. The water applicator provides a uniform water density on top of the storage array, simulating the discharge density from a sprinkler. The entire set-up is positioned underneath a calorimeter to facilitate the measurement of the heat release rate. The commodity is ignited at the centreline of the flue space, and water is applied when the fire reaches a convective heat of 2 MW. At this point, the fire normally involves the whole upper tier of the commodity. The measurement of the heat continues for 25 minutes after ignition. Three fire tests are conducted for each type of commodity, at three different rates of water application. However, in some of the cases, the first two fire tests are followed by a third freeburning test. The water application rates shall be either 2,5, 5,0 or 7,5 mm/min (Note. mm/min equals (L/min)/ m 2 ). The first two tests are always conducted at 5,0 mm/min and 2,5 mm/min, respectively, and based on the results, the decision is to be made whether the water application rate should be increased to 7,5 mm/min or a free burn test conducted. Based on the heat measurements, four parameters, V1 V4 are calculated. The values of these parameters are translated into a rank value, which, when averaged, is used as the basis for the classification. Figure 1. A schematic drawing of the test set-up and the water applicator. The nominal dimensions given in the figure may need to be adjusted due to the size of the commodity. 6 CALORIMETER SYSTEM 6.1 Hood and exhaust system The calorimeter system consists of a large hood connected to an exhaust system with a capacity of at least 25 m 3 /s at 20ºC. Exhaust system design based on natural convection is not permitted. The hood shall be at least 6 m in diameter with its lower rim about 8 m to 9,5 m above floor level. Additional information can be found in Annex A.

4 NORDTEST METHOD NT FIRE Instrumentation in the exhaust duct The following specifications are minimum requirements. Additional information can be found in Annex A Volume flow rate The volume flow rate in the exhaust duct shall be measured with an accuracy of at least ±5% Gas analysis Sampling line The gas samples shall be taken into the exhaust duct at a position where the combustion products are uniformly mixed along the entire diameter of the duct. The sampling line tubes shall be made of a material which does not influence the concentration of the gaseous species to be analysed. Oxygen analyser The oxygen depletion shall be measured with an accuracy of at least ±0,01% by volume of oxygen. A suitable output range is 0 21% by volume. The time delay from sampling to analysis, including the time constant of the instrument, shall be reduced as much as possible and accurately controlled. Carbon monoxide and carbon dioxide analyser The gas species shall be measured with an accuracy of at least ±0,1% by volume for carbon dioxide and ±0,02% by volume for carbon monoxide. A suitable output range is 0 1% for carbon monoxide and 0 6% for carbon dioxide. The time delay from sampling to analysis, including the time constant of the instrument, shall be reduced as much as possible and accurately controlled. Smoke obscuration measurement system (not compulsory) Smoke obscuration may be determined either by measuring the light attenuation with a system consisting of a lamp, lenses, an aperture and a photocell or a laser equipment. The system shall be constructed such that soot deposits during a test do not reduce the light transmission by more than 5%. Such systems are described in ISO 9705 [4]. burner or a methanol fire of relevant size. The heat release rate calculated from the metered gas input or weighed mass loss, respectively, and the measured heat based on oxygen consumption shall then be compared. Equations for the calculations are given in ISO 9705 [4] and in SP Reports 1992:43 and 1994:29 [5, 6]. The calibration of the convective part of the total heat shall be based on methanol pool fires assuming the average convective heat flux fraction (of the theoretical heat release rate) to be 0,80 as described in the report FMRC J.I OC2E1.RA [7]. The difference between the time average value, measured over a period of one minute, calculated from the measured oxygen consumption and the heat calculated from the metered gas input or weighed mass loss, respectively, must not exceed 10%. These measurements shall only be made when steady state conditions have been reached. The calibration procedure shall also involve stepwise changes of the heat to verify that the time lag of the gas concentration measurements is properly accounted for to obtain results closely representing the temporal concentration variations in the exhaust duct. The overall time lag associated with each analyser consists of two components; the gas transport time from the gas sampling port to the analyser s detection cell and the response lag due to the analyser itself. The former lag is linear and can be corrected by performing simple time shift. The latter lag depends on the analyser s characteristics and is generally non-linear. The characteristic response time of most gas analysers is quite significant compared with the variation rate of fire development. The non-linear lag can be corrected electronically in real time or corrected in post-test data processing using appropriate schemes, e.g. like those proposed by Croce [8]. The concentration deduced from simple time shift may deviate significantly from actual concentrations if the analyser is not responsive enough. Simple time shift is only acceptable for steady or moderately varying fire development if the mean value is of primary interest. A basic calibration also involves measurements of the flow profile in the exhaust duct, determination of temperature losses in the exhaust duct, etc. as described in references [5, 6] and [7]. 6.3 Calibration The accuracy of the heat release measurements is of great importance and a quality control system for the use of the calorimeter system including calibration routines should therefore be applied. A basic calibration shall be performed on a newly installed calorimeter system or on any other occasion when this is considered necessary. A simplified calibration shall be made prior to each test or continuous test series. The simplified calibration of the instrumentation in the exhaust duct shall be performed by using e.g. a propane gas 7 STORAGE ARRAY 7.1 Rack storage arrangement A two tier, single rack storage segment is used to hold the test commodity during the tests (see also Clause 9). The rack storage segment is placed centrally below the calorimeter hood. The lower beam shall be located 300±50 mm above the floor. The upper beam shall be adjusted such that the vertical distance between the top of the commodity on the first tier and the top of the beam is 200±50 mm, as shown in Figure 1.

5 NORDTEST METHOD NT FIRE Ignition source Two igniters shall be used to ignite the commodity. Each igniter shall be made from a porous cellulose material, e.g. pieces of insulating fibre board. The igniters may be either square or cylindrical in shape, 60 mm across for a square or 75 mm in diameter for a cylinder. The length shall be 75 mm. The igniters shall be mounted at the end of a steel rod, fixed to a stand. The overall height of the steel rod and the stand shall such that the igniters are in the correct position, as described below. Immediately prior to the test, the igniters shall be soaked in 120 ml of n-heptane and wrapped in a polyethylene foil bag. The igniters on the steel rod shall be placed on a support near the centre flue space of the storage array at the lower tier. The igniters shall be positioned close (10±5 mm) to the commodity, level with the top surface of the wood pallet, as shown in Figure 1. 8 WATER APPLICATION SYSTEM 8.1 Water applicator A water applicator is used for delivering a known flow of water directly onto the burning storage array. The system consists of four parallel, double-jacketed, stainless steel pipes fitted with six water spray nozzles along each pipe to form a 6 by 4 matrix of nozzles, 450 mm apart. The corresponding total coverage area is 4,86 m 2, which results in some degree of overlap, and some extension outside the pallet loads. The nozzles are installed close to the top of the storage array. The following nozzles and total water flow rates shall be used for the nominal water discharge densities specified in the test method (see Clause 10). Table 1. Water discharge densities, total water flow rates, pressures and the associated spray nozzles. Nominal water discharge density (mm/min) 1) Total water flow rate Approximate water pressure Nozzles 2) (L/min) (bar) 2,5 12±0,5 1, ,0 24±1 2, ,5 36±1,5 3, ) The rate mm/min equals L/min/m 2. 2) The nozzles are manufactured by Lechler GmbH, Germany. On activation, the system shall be capable of reaching 90% of the steady-state flow rate within 10 seconds. This implies that the water delivery system to the applicator is equipped with one adjustable valve for the flow adjustment prior to the test and an on-off valve to be used for the activation. The applicator must also be equipped with an air relief system. Note: The detailed design of the water applicator is shown in Annex B. If other pallet sizes than European standard (EUR) pallets with the dimension 1000 mm by 1200 mm are used for the tested commodities, the design of the applicator might have to be modified. The reproducibility and classification criteria when different pallet sizes and/or design of the water applicator are used shall be verified by comparative tests. 8.2 Flow rate and pressure measurements The total water flow rate and static water pressure shall be measured and the following equipment and performance are required. Flow meter with a suitable range and accuracy within the tolerances specified in Clause 8.1. Pressure gauge with an accuracy within ±0,05 bar within the range 0 5 bar. 9 TEST COMMODITIES The overall size of a single pallet load shall conform as closely as possible to the dimensions given below. If this condition cannot be met, the test results achieved may not be accurate. The commodity shall be stored on European standard (EUR) wood pallets with the dimension 1000 mm by 1200 mm. The total height of one pallet load shall be 1,15 m including the height of the wood pallet. However, for practical reasons these dimensions might be difficult to achieve, and the following tolerances are normally considered as acceptable: Horizontal dimensions: The horizontal dimensions are important because of the water applicator design. If the horizontal dimensions vary by more than ±25 mm of the nominal pallet size for which the applicator is designed, the amount of water spraying outside the storage array (overshoot) may vary considerably and thereby affect the results and the classification. Testing of cylindrical or irregular shaped commodities might pose problems because of this and the results must be considered carefully. Pallet load height: Reasonable tolerances for the pallet load height are considered to be 1,15±0,10 m. The moisture content of the commodity shall be within the range expected during normal storage. Moisture of cardboard cartons shall be controlled within 5 10% by weight. The commodity shall be carefully documented, the weight of the various parts determined (product(s), packaging material, carton, etc.) If the tested commodity is expected to become unstable during the tests, it must be supported as a collapse might influence the test results and thereby the subsequent classification. A suitable support method is to use wires wrapped around the storage array or reinforcement netting placed around the storage array.

6 NORDTEST METHOD NT FIRE TESTING 10.1 Initial conditions and preparation The ambient temperature inside the test hall shall be 20±5ºC. The horizontal wind draught measured at a distance of 1 m from the storage array may not exceed 0,5 m/s. All measurement channels for the calorimeter and water application system shall be scanned and checked for ambient conditions. The flow rate, pressure and the filling sequence of the water applicator shall be adjusted to correspond to the selected water density for the test. The discharge from the nozzles shall be visually examined to ensure that all nozzles are free of blockage. The water applicator system shall be allowed to completely drain out. Ensure that all air relief devices are in an open position. The test commodity shall be placed in the rack storage segment. The flue space shall be carefully adjusted to 150±10 mm. The height of the water applicator shall be adjusted so that the clearance between the nozzles and the top of the storage array is 250±50 mm. If the top surface of the storage array is not completely even, the distance shall be measured along the rim of the storage array. Prepare the igniters and place them as described in Clause 7.2. Means for manual extinguishment of a fully developed fire shall be ready available Test procedure The commodity shall be measured, weighed and photographed before testing. All recording and measuring devices shall be started and data taken at least two minutes prior to ignition. Ensure steady state temperature conditions in the test hall and calorimeter system and that the flow through the hood system is at least 25 m 3 /s. Ignite the storage array. The fire is allowed to develop naturally until the convective heat reaches 2 MW. The water applicator is activated to deliver the calibrated flow rate of water onto the burning commodity. If necessary, adjustments are made during the test to ensure that the specified tolerances in Clause 8.1 are complied with. A photographic and/or video recording shall be made during a test. A clock shall appear in all photographic records, giving time to nearest one second. During a test, records of the following events including times when they occur shall be taken: Ignition of the storage array. Activation time of the water application system. Any occurrence of melting and dripping. Any formation of a pool fire under the storage array. A general description of the burning behaviour. Any collapses from the storage array. Any other event of interest. Usually, the test is continued for 25 minutes after ignition, but may be terminated earlier if the commodity is completely consumed. The extent of damage to the commodity shall be documented after the test. A judgement is made of each of the four pallet loads and the proportion consumed by the fire (by volume) is estimated. The total damage to the commodity is then calculated Repeat of tests The first test shall be made according to Clause 11.2, using a nominal water discharge density of 5,0 mm/min. The second test shall be made at a nominal water discharge density of 2,5 mm/min. Based on the calculated mean unit rank of the first two tests, the decision is to be made whether the water application rate shall be increased to 7,5 mm/min or whether a free burn test shall be conducted. 11 CLASSIFICATION 11.1 General The classification test procedure is highly standardised. The aim is to determine the hazard level of a commodity by comparing the test results with data from identical tests on well-defined standard commodities, with fire characteristics similar to commodities of Category I, II, III and IV. The tests, forming the basis of this philosophy, are described in SP Report 2003:03 [2]. Additional information is provided in SP Report 1993:70 [3]. If a tested product shows fire characteristics similar to one of the standard commodities, it is assumed that the same protection requirements are adequate for the tested commodity. It is important to note that the commodity classification tests do not directly provide the protection requirements e.g. design water density for the commodity tested although reference is made to a range of water discharge densities in the tests. The purpose is to determine the hazard level of the commodity. A rack storage configuration is used in the classification tests. However, the results may also be applied to freestanding, palletized rack and post-pallet storage configurations, using the design criteria given for these configurations in EN

7 NORDTEST METHOD NT FIRE Principle of classification Four parameters are used to characterize the level of hazard of the tested commodity. These parameters refer to both the total heat and the convective part of the total heat as both these play a major role regarding sprinkler operation and the protection of the building construction. The four parameters are: V1 Maximum one-minute average of the total heat release rate. V2 Maximum one-minute average of the convective part of the heat. V3 Effective convective heat, defined as the convective heat averaged over the most intense five-minute interval of the fire. V4 Convective energy, defined as the amount generated during the ten-minute period of most intense fire. These are considered to be important variables which can be measured with accuracy and give large enough differences in the measured values from one hazard level to the next, to provide a credible assessment of the hazard. The significance of each variable is discussed in detail in the report FMRC J.I. OROE5.RR [9] and can be summarized as follows: V1 Maximum one-minute average of the total heat The maximum total heat is an important measure of the potential for fire spread and is also an overall fundamental measure of fire severity. The total heat release rate can be divided into two parts, the convective part and the radiative part. Normally one-third of the energy is released by radiation. Radiation is the primary mechanism for fire spread across aisles and other open spaces to adjoining combustibles and is also, in part, responsible for lateral fire spread throughout a large storage array. The total heat is based on gas analysis of the combustion gases and the one-minute average value is used to avoid the influence of spikes during the measurements due to environmental changes, nonuniformity of storage packages, electrical noise, etc. V2 Maximum one-minute average of the convective heat The maximum convective heat is one of the most important measures for characterizing fire severity. Approximately two-thirds of the energy generated from a fire is released as convective energy. Both the gas velocity and temperature within a fire plume are related to the convective heat and these two parameters are very important for the activation of sprinklers and the penetration of the water droplets. The higher the velocity and temperature, the lower the portion of water that penetrates the fire plume. The one-minute average is used for the same reason as mentioned above. V3 Effective convective heat (The convective heat averaged over the most intense five-minute interval of the fire) The convective energy released from the fire is to a large part responsible for heating of the ceiling construction and the activation of sprinklers. For this purpose, the maximum rate of heat release is not relevant, as it is a matter of heat transfer for which duration time is very important. A very intense but short-lived fire might be less severe than a fire of lower intensity but with a longer duration time. When assessing the fire severity it is therefore necessary to determine the heat for a longer period of time. The effective convective heat is, therefore, defined as the convective heat averaged over the most intense five-minute interval of the fire. V4 The convective energy generated during the most intense ten-minute interval of the fire The total convective energy released during a fire is an important measure of the potential for causing thermal damage to a construction. The higher the convective energy, the greater the damage potential. Once again, a product with a lower intensity but with a long fire duration time, which thereby releases more energy in total, might cause more severe damage to a construction. The convective energy reported from these classification tests is defined as the amount generated during the most intense ten-minute period of the fire or during the entire test if the fire duration time is less. The 10 minute value is based on the experience that most of the energy from the commodities used in the tests will be released during this period of time Evaluation of the test results classification In tables C1 through C4 of Annex C, rank values are listed in one-quarter increments for the V1 V4 parameters at each of the water discharge densities and the fire burn test. For information purposes, a graphical presentation of the tables C1 through C4 is given in Annex D. For each test, a mean unit rank shall be calculated as the arithmetic average of the rank values obtained for each of the parameters V1 V4. The mean total rank shall be calculated as the arithmetic average of the mean unit rank values obtained in each of the three tests. Based on the mean total rank, the commodity is classified according to Table 2. Table 2. Classification of commodity. Mean total rank Less than 1,0 Equal to or greater than 1,0 but less than 2,0 Equal to or greater than 2,0 but less than 3,0 Equal to or greater than 3,0 but less than 4,0 Equal to or greater than 4,0 Classification of commodity Category I Category II Category III Category IV Special hazard

8 NORDTEST METHOD NT FIRE In addition to Table 2, the following special requirements apply: If a tested commodity has a mean unit rank which exceeds the mean total rank by more than 1,00, the commodity should be classified according to the highest mean unit rank value. If a single, or several, individual rank values exceed 4,50 the commodity shall be considered a special hazard. In Table C5, a commodity classification fire test analysis protocol is provided. 12 TEST REPORT The test report shall contain the following information: a) Name and address of the testing laboratory b) Date and identification number of the report c) Name and address of the client d) Purpose of the test e) Method of sampling f) Name of manufacturer or supplier of the commodity g) Name or other identification marks and description of the commodity h) Dimensions, numbers, weight, moisture content, etc of the main component in the commodity i) Conditioning of the commodity j) Date of test k) Test method l) Dimensions and description of the storage array arrangement, support if any, etc m) Test results (see also Annex C) m:1 Total heat histories (graphs) m:2 Convective heat histories (graphs) m:3 Mass flow as a function of time in the exhaust duct m:4 Maximum one-minute average of the total heat (V1) m:5 Maximum one-minute average of the convective part of the heat (V2) m:6 Effective convective heat (V3) m:7 Convective energy (V4) m:8 Rank values, mean unit rank for each test and mean total rank according to Clause 11 and the tables in Annex C m:9 Calibration results according to Clause 6.3 When appropriate also: m:10 The production rate of carbon monoxide as a function of time m:11 The production rate of carbon dioxide as a function of time m:12 The production rate of light obscuration smoke as a function of time n) Deviations from the test method, if any o) When not identified in the test method, equipment and instruments used.

9 NORDTEST METHOD NT FIRE ANNEX A ANNEX A DESIGN PRINCIPLE OF THE CALORIMETER SYSTEM The combustion gases from the burning storage array are collected by a hood connected to an exhaust system. The flow capacity shall be at least according to the specification in Clause 6.1. Note: A large amount of smoke may be generated during the fire tests described in this test method and it is strongly recommended that a suitable smoke cleaning system is connected to the exhaust system. As the conditions might be different in various fire laboratories it is not relevant to specify the detailed design of the calorimeter system. However, there are some calorimeter systems in use which are designed to comply with the requirements specified in this test method. Two such systems are described in SP Reports 1992:43 and 1994:29 [5, 6] and in the report FMRC J.I OC2E1.RA [7]. Some general guidance regarding the design of a calorimeter system can also be obtained in ISO 9705 [4]. A general description of the instrumentation and a detailed description of the calculation equations for volume flow, total heat based on oxygen consumption, production rates of combustion gases and light obscuration are given in ISO 9705 [4]. Further information is also given in references [5, 6] and in reference [7].

10 NORDTEST METHOD NT FIRE ANNEX B ANNEX B DESIGN PRINCIPLE OF THE WATER APPLICATOR The water applicator is used for delivering a specific rate of water directly onto the burning storage array. The water applicator consists of four parallel, double-jacketed, stainless steel pipes fitted with six spray nozzles along each pipe to form a 6 by 4 matrix of nozzles. The full-cone type nozzles are spaced 450 mm apart to provide uniform coverage over a 4,86 m 2 area. The detailed design of the applicator is shown in Figure B1. Figure B1. Design of the water applicator and the water delivery system to be used in the commodity classification tests. The feed line is equipped with a flow meter, pressure transducer and a control valve in order to adjust the flow rate corresponding to the desired water discharge density during the water calibration, prior to the test. The water flow is then turned on and off by using the solenoid valves. The suppression water is fed from both ends into the pipes. In order to reduce the fill-up time, air relief devices 1 ) are installed at the midpoint of the pipes. This allows the air in the pipes to bleed. The relief devices are automatically shut off as soon as the pipes are completely filled with water. In order to further reduce the fill-up time, a special charge line should also be connected. This shall be controlled with a time relay and shut off at the same moment as the pipes are filled with water. This charge time has to be adjusted for each flow rate during the water calibration, prior to the test. In order to avoid blockage of the nozzles during the test, the feed line shall be equipped with necessary filter arrangements. Preferably, a filter shall also be installed before each water spray nozzle. The applicator is water cooled in the annular area of the double jacked pipes to protect it from the flames. The cooling water is fed from one end and discharged through the other. The air relief devices must also be protected from the flames by insulation. If the horizontal dimensions of the tested commodity differ considerably from the specified dimensions the results and the classification may be affected. Testing of cylindrical or other irregular commodities might pose a problem because of this and the results have to be considered carefully. 1) A check valve for water, turned in the opposite flow direction and with a light weight valve cone and the spring-load removed might be used.

11 NORDTEST METHOD NT FIRE ANNEX C ANNEX C CLASSIFICATION TABLES Table C1. Rank values: Free burn test. Rank V1 V2 V3 V4 Maximum one-minute total heat Maximum one-minute convective heat Effective convective heat Convective energy 0, , , , , , , , , , , , , , (MJ) Table C2. Rank values: Nominal discharge density of 2,5 mm/min. Rank V1 V2 V3 V4 Maximum one-minute total heat SH = Special hazard, the commodity needs special considerations. Maximum one-minute convective heat Effective convective heat Convective energy 0, , , , , , , , , , , , , , , , , , SH > > 6376 > 5276 > 3026 (MJ)

12 NORDTEST METHOD NT FIRE ANNEX C Table C3. Rank values: Nominal discharge density of 5,0 mm/min. Rank V1 V2 V3 V4 Maximum one-minute total heat SH = Special hazard, the commodity needs special considerations. Maximum one-minute convective heat Effective convective heat Convective energy 0, , , , , , , , , , , , , , , , , , SH > 8451 > 4751 > 3901 > 2201 (MJ) Table C4. Rank values: Nominal discharge density of 7,5 mm/min. Rank V1 V2 V3 V4 Maximum one-minute total heat SH = Special hazard, the commodity needs special considerations. DNA = Does not apply. Maximum one-minute convective heat Effective convective heat Convective energy 0,25 DNA DNA DNA DNA 0,50 DNA DNA DNA DNA 0,75 DNA DNA DNA DNA 1,00 DNA DNA DNA DNA 1,25 DNA DNA DNA DNA 1,50 DNA DNA DNA DNA 1,75 DNA DNA DNA DNA 2, , , , , , , , , , , SH > 6076 > 3126 > 2526 > 1376 (MJ)

13 NORDTEST METHOD NT FIRE ANNEX C Table C5. Commodity classification fire test analysis protocol. COMMODITY CLASSIFICATION FIRE TEST ANALYSIS Client: Date: Description of tested commodity: Nominal discharge density (mm/min)* Free burn test 2,5 mm/min 5,0 mm/min 7,5 mm/min Notes: V1 V2 V3 V4 Mean unit rank** V1 V2 V3 V4 Mean unit rank** V1 V2 V3 V4 Mean unit rank** V1 V2 V3 V4 Mean unit rank** Mean total rank*** Test result Rank value *) Three tests are conducted either without the application of water (free burn), and at 2,5 mm/min and 5,0 mm/min or, alternatively, at 2,5 mm/min, 5,0 mm/ min and 7,5 mm/min. **) Average of the rank values for V1 V4 obtained with one specific water discharge density. ***) The average of the mean unit rank values obtained in the three tests.

14 NORDTEST METHOD NT FIRE ANNEX D ANNEX D CLASSIFICATION GRAPHS (Informative) V Special hazard HRRtot III IV 4000 II I Nominal discharge density (mm/min) 6000 V2 Special hazard 5000 HRRconv III II IV 2000 I Nominal discharge density (mm/min)

15 NORDTEST METHOD NT FIRE ANNEX D 5000 V3 Effective HRRconv III II IV Special hazard I Nominal discharge density (mm/min) 3000 V4 Convective energy (MW) Special hazard IV III II I Nominal discharge density (mm/min)

16 NORDTEST METHOD NT FIRE